WO2013037520A1 - Modified hgf-1k1 polypeptide - Google Patents

Abstract

The present invention relates to modified HGF-1K1 polypeptides (such as that shown in Seq ID No:4), and certain mutations for the disruption of an identified glycosylation site, DNA sequences encoding those proteins (such as those sequences in Seq ID Nos: 5, 6 and 7) and various methods for their production. The invention also relates to pharmaceutical compositions containing the modified HGF-1K1 polypeptides and to their use in the treatment of various diseases such as COPD, broncheolitis obliterans, acute lung injury (ALI), idiopathic pulmonary fibrosis (IPF), myocardial infarction, liver fibrosis and kidney fibrosis.

Description

MODIFIED HGF-lKl POLYPEPTIDE

Field of the Invention

The present invention relates to modified HGF-lKl polypeptides, DNA sequences encoding them and various methods for their production. The invention also relates to pharmaceutical compositions containing the modified HGF-lKl polypeptides and to their use in the treatment of various diseases such as COPD, broncheolitis obliterans, acute lung injury (ALI), idiopathic pulmonary fibrosis (IPF), heart failure, e.g. myocardial infarction, liver fibrosis and kidney fibrosis.

Background to the Invention

The polypeptide growth factor hepatocyte growth factor/scatter factor (HGF/SF) (Gherardi et al., 1989; Miyazawa et al., 1989; Nakamura et a/., 1989; Stoker et a/., 1987) and its receptor MET, the product of the c-MET protoncogene (Bottaro et a/., 1991), play essential roles in the development of epithelial organs such as the placenta and liver (Schmidt eta/., 1995; Uehara et a/., 1995) and in the migration of myogenic precursor cells (Bladt et a/., 1995) and motor neurons (Caton eta/., 2000; Ebens eta/., 1996).

HGF/SF and MET are also involved in the spreading of a variety of epithelial tumours as a result of MET chromosomal rearrangements (Yu eta/., 2000), somatic and/or germline mutations in the MET kinase (Schmidt et a/., 1997) or, more often, over expression in tumour cells of an unrearranged and unmutated MET gene (reviewed in Jeffers eta/., 1996).

HGF/SF has a unique domain structure that resembles that of the blood proteinase precursor plasminogen and consists of six domains: an N-terminal (N) domain, homologous to plasminogen activation peptide, four copies of the kringle (K) domain and a catalytically inactive serine proteinase domain (Donate et al., 1994). Two products of alternative splicing of the primary HGF/SF transcript encode NK1, a fragment containing the N and the first K domain, Kl, (Cioce et al., 1996), and NK2, a fragment containing the N, Kl and second kringle, K2, domains (Chan et al., 1991; Hartmann et al., 1992; Miyazawa et al., 1991). Both NK1 (Lokker and Godowski, 1993) and NK2 (Chan et al., 1991) were initially characterized as MET antagonists, although experiments in transgenic mice have subsequently indicated that NK1 behaves in vivo as a bona fide receptor agonist (Jakubczak et al., 1998).

There is an important difference in the mechanism of receptor binding and activation by

HGF/SF and NK1. HGF/SF is fully active in cells lacking heparan sulphate, while NK1 is only active in cells that display heparan sulphate or in the presence of soluble heparin (Schwall et al., 1996). Thus NK1, but not HGF/SF, resembles FGF (Rapraeger et al., 1991; Yayon et al. 1991) in terms of a requirement for heparan sulphate for receptor binding and/or activation.

Early domain deletion experiments indicated that the N domain is important for heparin binding (Mizuno et al., 1994) and site directed mutagenesis identified residues in this domain essential for binding (Hartmann et al., 1998). This reverse charge mutation of R73 and R76 decreased the affinity of HGF/SF for heparin by more than 50 fold (Hartmann et al., 1998). A role for several other positively-charged residues, such as K58, K60 and K62, was suggested from the solution structure of the N domain, as these residues are clustered in close proximity of R73 and R76 (Zhou et al., 1998), and recent NMR experiments have provided experimental support for an involvement of K60, K62, R73, R76, R78 and several other residues in heparin binding to the N domain (Zhou et al., 1999).

Despite this progress, the mechanism through which heparin and heparan sulphate confer agonistic activity to NK1 remains incompletely understood. NK1 crystallizes as a dimer in the absence of heparin (Chirgadze et al., 1999; Ultsch et al., 1998), and the features of this dimer suggested that it could represent the biologically active form of NK1 (Chirgadze et al., 1999).

WO2002/088354 discloses two mutated variants of the NK1 splice variant, designated HGF- 1K1 and HGF-1K2, which show increased activity relative to the naturally occurring NK1 splice variant in certain assays disclosed therein. Variants designated HP11 and HP12 showed decreased activity relative to the naturally occurring NK1 splice variant in the same assays. The HGF-1K1 variant, having mutations K132E R134E (with respect to wild type, full length HGF) has the amino acid sequence shown in Fig 1.

Brief Description of the Invention

In a first aspect of the present invention, there is provided an HGF-1K1 variant polypeptide, wherein the polypeptide chain consists of the amino acids of SEQ ID No:4. In another aspect of the present invention, there is provided an HGF-1K1 variant polypeptide, wherein the polypeptide chain consists of amino acids 1 to 179 of SEQ ID No:4.

In a second aspect of the present invention, there is provided an HGF-1K1 variant polypeptide, wherein the polypeptide chain consists of the amino acids of Seq ID No:4, apart from one or two (e.g. one) amino acid substitutions at Thrl64 and/or Asnl66. In one embodiment, the substitutions are independently selected from Thrl64Ala, Thrl64Asn, Thrl64Val, Asnl66Asp, Asnl66Ser and Asnl66His.

In certain embodiments of the present invention, the HGF-1K1 variant polypeptide is glycosylated. In certain embodiments, the polypeptide is glycosylated with one or more hexose units. In certain embodiments, the polypeptide is glycosylated at Serl27 and/or Serl80. In another embodiment the HGF-1K1 variant polypeptide contains a Serl80Ala mutation.

In certain embodiments, the HGF-1K1 variant polypeptide is not glycosylated. In certain embodiments, the polypeptide is not glycosylated at Asnl66. In other embodiments, the polypeptide is glycosylated at Serl27 and/or Ser 180, but is not glycosylated at Asnl66.

In a third aspect of the present invention, there is provided a polynucleotide encoding an

HGF-1K1 variant polypeptide of the invention. In certain embodiments, the DNA sequence comprises Seq ID No: 5, Seq ID No:6 or Seq ID No:7. In a fourth aspect of the present invention, there is provided a plasmid comprising a polynucleotide of the invention.

In a fifth aspect of the present invention, there is provided a host cell comprising a plasmid of the present invention. In certain embodiments, the host cell is a Pichia pastoris cell, for example a strain selected from X-33, SMD1168H, SMD1136, YCC120, GS115 and KM71H.

In a sixth aspect of the present invention, there is provided a process for the preparation of an HGF-1K1 variant polypeptide of the invention comprising the steps of:

a. Inserting a plasmid of the invention into a host cell;

b. Expressing the protein;

c. Recovering the protein; and then optionally

d. Purifying the protein.

In a seventh aspect of the present invention, there is provided a pharmaceutical composition comprising an HGF-1K1 variant of the invention and a pharmaceutically acceptable carrier or diluent. In certain embodiments, the composition is adapted for inhaled or intravenous administration. In certain embodiments, the composition is adapted for inhaled administration and is either a spray dried powder formulation or a liquid formulation for nebulising.

In a eighth aspect of the present invention, there is provided an HGF-1K1 variant polypeptide of the invention for use in therapy.

In a ninth aspect of the present invention, there is provided an HGF-1K1 variant polypeptide of the invention for use in the treatment of a disease or disorder selected from COPD, broncheolitis obliterans, acute lung injury (ALI), Adult Respiratory Distress Syndrome (ARDS), idiopathic pulmonary fibrosis (IPF), myocardial infarction, liver fibrosis and kidney fibrosis. In certain embodiments, the disease is COPD and optionally wherein the polypeptide is inhaled.

In a tenth aspect of the present invention, there is provided an HGF-1K1 variant polypeptide of the invention for the manufacture of a medicament for the treatment of a disease or disorder selected from COPD, broncheolitis obliterans, acute lung injury (ALI), Adult Respiratory Distress Syndrome (ARDS), idiopathic pulmonary fibrosis (IPF), myocardial infarction, liver fibrosis and kidney fibrosis.

In an eleventh aspect of the present invention, there is provided a method of treating of a disease or disorder selected from COPD, broncheolitis obliterans, acute lung injury (ALI), Adult Respiratory Distress Syndrome (ARDS), idiopathic pulmonary fibrosis (IPF), myocardial infarction, liver fibrosis and kidney fibrosis comprising administering to a human patient in need thereof an effective amount of the HGF-1K1 polypeptide variant of the invention. In certain embodiments, the disease is COPD and optionally wherein the polypeptide is inhaled. Brief Description of the Figures

Figure 1 shows the amino acid sequence of HGF-1K1 as disclosed in WO2002/088354 (Seq ID No: l).

Figure 2 shows the mammalian codon-optimised DNA sequence (Seq ID No: 2) corresponding to the HGF-1K1 as disclosed in WO2002/088354 (Seq ID No:l).

Figure 3 shows the pichia codon-optimised DNA sequence (Seq ID No:3) corresponding to the HGF-1K1 as disclosed in WO2002/088354 (Seq ID No:l).

Figure 4 shows the amino acid sequence of an HGF-1K1 variant of the present invention (Seq ID No:4). The putative reverse glycosylation site is identified in bold and underlined at position 163 to 165.

Figure 5 shows the mammalian codon-optimised DNA sequence (Seq ID No:5) corresponding to the HGF-1K1 construct of Seq ID No:4.

Figure 6 shows an alternative mammalian codon-optimised DNA sequence (Seq ID No:6) corresponding to the HGF-1K1 construct of Seq ID No:4, which is modified due to a cloning artefact.

Figure 7 shows the pichia codon-optimised DNA sequence (Seq ID No:7) corresponding to the HGF-1K1 construct of Seq ID No:4.

Figure 8 shows the amino acid sequence of another HGF-1K1 variant (Seq ID No:8, not encompassed by the present invention).

Figure 9 shows the mammalian codon-optimised DNA sequence (Seq ID No: 9) corresponding to the HGF-1K1 construct of Seq ID No: 8.

Figure 10 shows the pichia codon-optimised DNA sequence (Seq ID No: 10) corresponding to the HGF-1K1 construct of Seq ID No:8.

Figure 11 shows the HGF-1K1 variant as disclosed in WO2002/088354 (Seq ID No: l) purified by affinity chromatography on a heparin column and analysed by mass spectrometry. The spectrogram shows a peak of intact protein mass and three other significant peaks of different N- terminal cleaved proteins.

Figure 12 shows a representative example of a successful expression assay. In this case, 10 colonies of Pichia KM71H were tested for expression of the YVEG Pichia codon optimized construct (Seq ID No: 10). The band of interest, around 21 KDa, is framed in the red rectangle.

Figure 13 shows HGF-1K1 expression assays. The percentage of colonies which expressed the desired protein for each construct/strain combination was calculated using the number of colonies that showed a visible band of the protein of interest in the relevant SDS-PAGE gels. The intensity of the bands was very variable, but, as an average, constructs EAEAQRK (Seq ID No: 6, Pichia host strain KM71H), YAEG (Seq ID No:3, Pichia host strain KM71H; Seq ID No:2, Pichia host strains KM71H and X33) and YVEG (Seq ID No:9, Pichia host strains KM71H and X33) showed the highest level of expression. Figure 14: Heparin purification titres. The heparin titres were calculated from the A₂₈₀ of the purified protein pool. The titres are ranked from highest to lowest value. Those titres which are asterixed had a culture volume of 800 ml as opposed to 400 ml for the others.

Figure 15: Percentage of intact construct species after Heparin and Ion Exchange purification. The percentages were calculated considering the ratio of different species in a given peak, as determined by Mass Spectrometry. All peaks with an area equal or inferior to 15% of the total area were discarded. Note: the Mass Spectrometry is not a quantitative technique. These results are not to be taken as absolute values. Detailed Description of the invention

The present invention provides new variants of HGF-1K1, which may provide a number of benefits over the previously disclosed HGF-1K1 sequence in WO2002/088354, in particular improved expression in Pichia pastoris as well as simpler and cheaper purification, both of which allow for large-scale production of the polypeptide. The enzymatic activity of the polypeptide remains substantially unaltered. Furthermore, the polypeptide of the present invention has a higher N terminal homogeneity than previously disclosed species, which provides for more consistent batch production and therefore less variability in the product. A further disadvantage of variable and large N-terminal heterogeneity is a resulting reduction in the final purified product yield from expression titres, which would not be compatible with material production at manufacturing scale.

As used herein, "patient" refers to a human or other animal. In one embodiment, the patient is a human.

As used herein, "treatment" means: (1) the amelioration or prevention of the condition being treated or one or more of the biological manifestations of the condition being treated, (2) the interference with (a) one or more points in the biological cascade that leads to or is responsible for the condition being treated or (b) one or more of the biological manifestations of the condition being treated, or (3) the alleviation of one or more of the symptoms or effects associated with the condition being treated. One skilled in the art will appreciate that "prevention" is not an absolute term. In medicine, "prevention" is understood to refer to the prophylactic administration of a drug to substantially diminish the likelihood or severity of a condition or biological manifestation thereof, or to delay the onset of such condition or biological manifestation thereof.

As used herein, "effective amount" means an amount of the active ingredient sufficient to significantly induce a positive modification in the condition to be treated but low enough to avoid serious side effects (at a reasonable benefit/risk ratio) within the scope of sound medical judgment. An effective amount of an active ingredient will vary with the particular variant chosen (e.g. consider the potency, efficacy, and half-life of the variant); the route of administration chosen; the condition being treated; the severity of the condition being treated; the age, size, weight, and physical condition of the patient being treated; the medical history of the patient to be treated; the duration of the treatment; the nature of concurrent therapy; the desired therapeutic effect; and like factors, but can nevertheless be routinely determined by one skilled in the art.

The term "operably linked" refers to a juxtaposition wherein the components described are in a relationship permitting them to function in their intended manner. A control sequence "operably linked" to a coding sequence is ligated in such a way that expression of the coding sequence is achieved under conditions compatible with the control sequences.

According to a first aspect, an HGF-1K1 polypeptide variant of the present invention has a polypeptide chain consisting of the amino acid sequence of Seq ID No: 4. It is understood that in various expression systems, the mature amino acid sequence may be glycosylated. Glycosylated forms of the amino acid sequence of Seq ID No: 4 are within the scope of the present invention.

In the HGF-1K1 polypeptide variants of the present invention, the sequence may be glycosylated at either an N-linked or O-linked glycosylation site, or at both, or the variant may not be glycosylated.

N-linked glycosylation sites are typically at asparagine residues. It is known that the attachment of a glycan to a particular asparaginyl residue is governed by protein sequence and it has been determined, from studies involving amino acid substitutions of the +2 residue following the asparagine residue, that a proximal hydrogen-bond acceptor is necessary for rendering the asparagine residue sufficiently nucleophilic to displace the GlcNAc2-Man9Glc3 from the dolichol donor. Therefore, the canonical N-linked consensus sequence motif, NX(S/T), where X is not Proline, has been identified as a putative N-linked glycosylation site. Another previously identified N-linked glycosylation motif is NXC (see John F. Valliere-Douglass, eta/., J. Biol. Chem., vol. 285, no. 21, pp. 16012-16022, May 21, 2010). Interestingly, the reversal of the motif NX(S/T) results in an atypical non-consensus reverse glycosylation sequence with the motif (T/S)XN. This highly atypical modification has been shown to be present at very low levels of ~0.5-2.0% in a number of recombinant antibodies (see John F. Valliere-Douglass et al., J. Biol. Chem., vol. 284, no. 47, pp. 32493-32506, November 20, 2009).

Hence, within the amino acid sequence of the HGF-1K1 polypeptide variants of the present invention, a non-consensus reverse N-linked glycosylation site is thought to exist, corresponding to the reverse glycosylation motif TXN at positions 164 to 166 (TSN) of Seq ID No:4. In one embodiment, the HGF-1K1 polypeptide variants of the present invention may be glycosylated at an N-linked glycosylation site, which is Asnl66. In one embodiment, the glycan structure is hexose (alternatively known as mannose). In another embodiment, the HGF-1K1 polypeptide variants are glycosylated at position Asnl66 with three, four, five or six hexose units, such as three or six, in particular three hexose units. In one embodiment, the hexose units are linked directly to Asnl66. In another embodiment, the hexose units may be linked via one or two Glc-NAc groups. Other possible glycosylation structures include branched mannose structures. In another embodiment, the HGF-lKl polypeptide variants of the present invention may be mutated in order to disrupt the identified non-consensus reverse N-linked glycosylation site. One or more, e.g. one of the amino acids at positions 164 to 166 (TSN) of Seq ID No:4 may be substituted. In one embodiment, the substitution is made at Thrl64. In another embodiment, the substitution is made at Asnl66. In another embodiment, the substitution is made at both Thrl64 and Asnl66. Particular substitutions are independently selected from Thrl64Ala, Thrl64Asn, Thrl64Val, Asnl66Asp, Asnl66Ser and Asnl66His. It is envisaged that the removal of at least one of the glycosylation sites may result in reduced immunogenicity of the proteins when administered as an active ingredient, in particular, the removal of the non-consensus reverse N-linked glycosylation site may provide this advantage.

O-linked glycosylation sites are typically at serine or threonine residues. In the HGF-lKl polypeptide variants of the present invention, one O-linked glycosylation site has been identified. Another O-linked glycosylation site exists near the C-terminus of the polypeptide.

Hence, within the amino acid sequence of the HGF-lKl polypeptide variants of the present invention, an O-linked glycosylation site exists at the serine residue corresponding to serine 127 (Serl27) of Seq ID No:4. A further O-linked glycosylation site exists at the serine residue corresponding to serine 180 (Serl80) of Seq ID No:4. In one embodiment, the HGF-lKl polypeptide variants of the present invention may be glycosylated at an O-linked glycosylation site, which is Serl27 and/or Serl80. In another embodiment, the HGF-lKl polypeptide variant is glycosylated at an O-linked glycosylation site near the C terminus. In one embodiment, the glycan structure is hexose, e.g. glucose. In another embodiment, the HGF-lKl polypeptide variants are glycosylated at position Serl27 with one, two or three hexose (e.g. glucose) units, in particular one hexose (e.g. glucose) unit. In another embodiment, the HGF-lKl polypeptide variants are glycosylated at position Serl80 with one, two, three, four, five, six, seven eight or nine hexose (e.g. mannose) units, in particular three or six hexose, such as three (e.g. mannose) units. In one embodiment, the hexose (e.g. glucose or mannose) unit(s) is linked directly to Serl27 and/or Serl80. In another embodiment, the hexose units may be linked via one Glc-NAc group. Other possible glycosylation structures include glucose, glucosamine, xylose, galactose, fucose, or mannose structures as the initial sugar bound to the serine residue. In a further embodiment, protein modifications may also be made to disrupt O-linked glycosylation sites, in particular at serine 127 and/or Serl80, such as mutation to alanine at Serl80.

Hence in one embodiment, the HGF-lKl polypeptide variants of the present invention are not glycosylated at the non-consensus reverse N-linked glycosylation site at Asnl66 of Seq ID No:4, but are glycosylated at one or both of the O-linked glycosylation sites at Serl27 and/or Serl80 of Seq ID No:4. In another embodiment, the HGF-lKl polypeptide variants of the present invention are glycosylated at both the non-consensus reverse N-linked glycosylation site at Asnl66 of Seq ID No:4, and both of the O-linked glycosylation sites at Serl27 and Serl80 of Seq ID No:4. In a further embodiment, the HGF-1K1 polypeptide variants of the present invention are glycosylated at the non- consensus reverse N-linked glycosylation site at Asnl66 of Seq ID No:4, but are not glycosylated at one or both of the O-linked glycosylation sites at Serl27 and/or Serl80 of Seq ID No:4. In another embodiment, the HGF-1K1 polypeptide variants of the present invention are glycosylated at Serl27, and the variant comprises a Serl80Ala mutation. In another embodiment, the HGF-1K1 polypeptide variants of the present invention comprise a polypeptide chain consisting of amino acids 1 to 179 of Seq ID No:4.

HGF-1K1 polypeptide variants of the present invention may be prepared by expression in a Pichia pastoris system (or, alternatively in Saccharomyces cerevisiae). Alternatively, the variants may be chemically synthesised, e.g. in a stepwise manner.

DNA sequences encoding the HGF-1K1 polypeptide variants of the invention may be readily prepared by the skilled person using the information and references contained herein and techniques known in the art (for example, see Sambrook, Fritsch and Maniatis, "Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory Press, 1989, and Ausubel et al., Short Protocols in Molecular Biology, John Wiley and Sons, 1992). Particular DNA sequences are described in Seq ID No: 5, Seq ID No:6 and Seq ID No: 7.

Polynucleotides of the invention can be incorporated into a recombinant replicable vector. The vector may be used to replicate the nucleic acid in a compatible host cell. Thus, in a further embodiment, the invention provides a method of making polynucleotides of the invention by introducing a polynucleotide of the invention into a replicable vector, introducing the vector into a compatible host cell, and growing the host cell under conditions which bring about replication of the vector. The vector may be recovered from the host cell. Suitable host cells are described below.

In one embodiment, a polynucleotide of the invention in a vector is operably linked to a control sequence which is capable of providing for the expression of the coding sequence by the host cell, i.e. the vector is an expression vector.

Suitable vectors can be chosen or constructed, containing appropriate regulatory sequences, including promoter sequences, terminator fragments, polyadenylation sequences, enhancer sequences, marker genes and other sequences as appropriate. Vectors may be plasmids, viral (e.g. phage, phagemid) or baculoviral, cosmids, YACs, BACs or PACs as appropriate. In one embodiment, the vector is compatible with both an £ coii and P. pastoris expression host cell. A particular vector for use in the present invention is a pPicZ alpha vector (Invitrogen).

The vectors may be provided with an origin of replication, optionally a promoter for the expression of the said polynucleotide and optionally a regulator of the promoter. The vectors may contain one or more selectable marker genes, for example an ampicillin resistance gene in the case of a bacterial plasmid or a Zeocin resistance gene for a pichia vector. Vectors may be used in vitro for example for the production of RNA or used to transfect or transform a host cell. Systems for cloning and expression of a polypeptide in a variety of different host cells are well known. Suitable host cells include bacteria (e.g. £ coli) or eukaryotic cells, such as yeast systems (e.g. Saccharomyces cerevisiae, such as a PEP4 knockout) or mammalian systems, (e.g. CHO). A particular yeast system useful in the expression of the present invention is Pichia pastoris (such as a PEP4 knockout). Particular Pichia pastoris strains of interest include, but are not limited to: X-33, SMD1168H, SMD1136, YCC120, GS115 and KM71H.

Promoters and other expression regulation signals may be selected to be compatible with the host cell for which the expression vector is designed. For example, yeast promoters include P. Pastoris, AOX1 and AOX2 promoters, S. cerevisiae GAL4 and ADH promoters, S. pombe nmtl and adh promoter. All these promoters are readily available in the art.

The vectors may include other sequences such as promoters or enhancers to drive the expression of the inserted nucleic acid, nucleic acid sequences so that the polypeptide is produced as a fusion and/or nucleic acid encoding secretion signals so that the polypeptide produced in the host cell is secreted from the cell.

Plasmids may be introduced into a suitable host cell as described above to provide for expression of a polypeptide of the invention. Thus, in a further aspect the invention provides a process for preparing HGF-1K1 variant polypeptides according to the invention which comprises the steps of:

a. Inserting a plasmid into a host cell;

b. Expressing the protein;

c. Recovering the protein; and then optionally

d. Purifying the protein.

A further embodiment of the invention provides host cells carrying the plasmids for the replication and expression of polynucleotides of the invention. The cells will be chosen to be compatible with said plasmid and are, in one embodiment yeast. In another embodiment, the yeast is Pichia pastoris. In another embodiment, the yeast is Saccharomyces cerevisiae. In a further embodiment, the plasmid is compatible with both £ coii and P. pastoris.

The introduction of plasmids into a host cell may be followed by causing or allowing expression from the nucleic acid, e.g. by culturing host cells (which may include cells actually transformed although more likely the cells will be descendants of the transformed cells) under conditions for expression of the gene, so that the encoded polypeptide is produced. If the polypeptide is expressed coupled to an appropriate signal leader peptide (e.g. a-mating factor leader sequence) it may be secreted from the cell into the culture medium. Following production by expression, a polypeptide may be isolated and/or purified from the host cell and/or culture medium, as the case may be, and subsequently used as desired, e.g. in the formulation of a composition which may include one or more additional components, such as a pharmaceutical composition which includes one or more pharmaceutically acceptable excipients, vehicles or carriers (e.g. see below). A further aspect of the present invention provides a host cell containing a plasmid and/or nucleic acid as disclosed herein. The polynucleotides and plasmids of the invention may be integrated into the genome (e.g. chromosome) of the host cell. Integration may be promoted by inclusion of sequences which promote recombination with the genome, in accordance with standard techniques.

In one embodiment, the DNA sequence of interest is first cloned into £ Coll cells, in order to confirm the sequence and that the construct has inserted into the vector correctly. In one embodiment, the expression of the polypeptide occurs in a Pichia pastoris host cell. Alternatively, the expression occurs in a Saccharomyces cerevisiae host cell.

In one embodiment, the HGF-1K1 polypeptide variants of the present invention are purified.

Hence, the polypeptide of the invention may be purified to at least about 70%, 75%, 80%, 85%, 87%, 90%, 92%, 95%, 96%, 97%, 98%, or 99% purity. Purity may be determined by assays known in the art, such as SDS-PAGE, SEC, RP-HPLC and mass spectrometry, particularly SEC. Purification techniques are well-known to those skilled in the art, and include protein purification columns, such as with a heparin column, and ion exchange column such as Capto SPImpRes^(TM), size-exclusion chromatography (SEC), such as using Superdex 75 resin, or other purification techniques, such as multi-modal resin technology (e.g. Capto MMC^(TM)), known to those skilled in the art. In another aspect of the present invention, the HGF-1K1 polypeptide variants of the present invention may be provided in form of a pharmaceutical composition. The HGF-1K1 polypeptide variants of the present invention may be administered alone or as part of a pharmaceutical composition.

Pharmaceutical compositions of the present invention comprise an effective amount of an HGF-1K1 polypeptide variant of the invention as an active ingredient, which amount will depend upon a number of factors. For example, the species, age, and weight of the recipient, the precise condition requiring treatment and its severity, the nature of the composition, and the route of administration are all factors to be considered. The therapeutically effective amount ultimately should be at the discretion of the attendant physician. Regardless, an effective amount of an HGF- 1K1 polypeptide variant of the invention for the treatment of humans generally should be in the range of about 0.0001 to about 100 mg/kg body weight of recipient per day.

There is thus provided a pharmaceutical composition comprising an HGF-1K1 polypeptide variant of the invention and a pharmaceutically acceptable carrier or diluent. In particular embodiments, an HGF-1K1 polypeptide variant of the invention is administered topically to the eye, ear (as ear drops), skin, or may be administered via pulmonary delivery, such as by inhalation (e.g., intrabronchial, intranasal or oral inhalation, intranasal such as by drops) or by systemic delivery (e.g., parenteral, intravenous, intramuscular, intraperitoneal, intraarterial, intrathecal, intraarticular, subcutaneous, vaginal or rectal administration). In another embodiment, an HGF-1K1 polypeptide variant of the invention is administered to the eye e.g. by topical administration, as eye drops, particulate polymer system, gel or implant, or by intraocular injection e.g. into the vitreous humour. A particular mode of administration is inhaled (e.g. as a nebulised liquid or as a spray dried powder and intravenous administration).

In one aspect of the invention, an aqueous solution containing an HGF-1K1 polypeptide variant of the invention, which may be buffered at physiological pH (e.g. pH 4 to 8), in a form ready for injection, is prepared. The compositions will commonly comprise a solution of an HGF-1K1 polypeptide variant of the invention dissolved in a pharmaceutically acceptable carrier, which is, in one embodiment, an aqueous carrier or diluent. A variety of aqueous carriers may be employed, e.g. 0.9% saline, and the like. The aqueous component of the compositions of the invention may be a high grade quality of water such as water for injection. These solutions are sterile and generally free of particulate matter. The solutions may be sterilized by conventional, well known sterilization techniques (e.g. filtration). Compositions may contain pharmaceutically acceptable auxiliary substances, as required, to approximate physiological conditions such as pH adjusting and buffering agents, etc.

The skilled person would readily appreciate that some auxiliary substances may perform more than one function, depending on the nature and number of auxiliary substances used in that composition and the particular properties of the active ingredient contained therein.

One or more tonicity adjusting agent(s) may be included to achieve isotonicity with body fluids e.g. with the blood or skin, which may result in reduced levels of irritancy. Examples of pharmaceutically acceptable tonicity adjusting agents include, but are not limited to, sodium chloride, dextrose, xylitol and calcium chloride. In one embodiment, the composition includes a tonicity adjusting agent which is sodium chloride.

The compositions of the invention may be buffered by the addition of suitable buffering agents such as sodium acetate (which may be hydrated, e.g. as the trihydrate), sodium citrate, citric acid, trometarol, phosphates such as disodium phosphate (for example the dodecahydrate, heptahydrate, dihydrate and anhydrous forms) or sodium phosphate and mixtures thereof. In one embodiment, the composition includes a buffering agent which is sodium phosphate.

Compositions may include one or more stabilising agents for both preventing surface adsorption and as stabilizers against protein aggregation due to denaturation at interfaces like liquid/air and liquid/container interfaces. Examples of pharmaceutically acceptable stabilising agents include, but are not limited to, sugars (such as sucrose and/or trehalose), fatty alcohols, esters and ethers, such as polyoxyethylene (80) sorbitan monooleate (Polysorbate 80), macrogol ethers and poloxamers. In one embodiment, the composition includes a stabilising agent which is sucrose and/or trehalose. Chelating agents such as disodium ethylenediaminetetraacetate (EDTA) effectively scavenge free metal ions in a solution before they oxidize the proteins and may be used to inhibit metal catalyzed oxidation.

Solubility enhancers such as arginine are solvent additives for the enhancement of protein solubility and suppression of protein aggregation. In one embodiment, the composition includes a solubilizer which is an arginine base.

The concentration of the HGF-1K1 polypeptide variant of the invention in such pharmaceutical compositions can vary widely, i.e. from about 15 to about 150 mg/mL, e.g. about 50 to about 150 mg/mL, such as about 60 to about 140 mg/mL, e.g. about 70 to about 130 mg/mL, for example about 80 to about 120 mg/mL. In one embodiment, the concentration of an HGF-1K1 polypeptide variant of the invention in the composition is from about 90 to about 115 mg/mL, or about 90 to about 110 mg/mL. Active ingredient concentrations will generally be selected primarily based on fluid volumes, viscosities, etc., as required.

Dosage forms for nasal or inhaled administration may conveniently be formulated as aerosols, solutions, suspensions drops, gels or dry powders. Compositions for inhaled administration include aqueous, organic or aqueous/organic mixtures, dry powder or crystalline compositions administered to the respiratory tract by pressurised pump or inhaler, for example, reservoir dry powder inhalers, unit-dose dry powder inhalers, pre-metered multi-dose dry powder inhalers, nasal inhalers or pressurised aerosol inhalers, nebulisers or insufflators

In one embodiment, for compositions suitable and/or adapted for inhaled administration, the active ingredient has a small particle size to facilitate delivery to the lungs. In one embodiment, the particle size of the active ingredient is defined by a D₅₀ value of about 0.5 to about 10 microns (for example as measured using laser diffraction). Compositions adapted for administration by inhalation include particle dusts or mists. Suitable compositions wherein the carrier is a liquid for administration as a nasal spray or drops include aqueous or oil solutions/suspensions of the active ingredient which may be generated by means of various types of metered dose pressurised aerosols, nebulizers or insufflators.

Aerosol formulations, e.g. for inhaled administration, can comprise a solution or fine suspension of the active ingredient in a pharmaceutically acceptable aqueous or non-aqueous solvent. Aerosol formulations may be presented in single or multidose quantities in sterile form in a sealed container, which may take the form of a cartridge or refill for use with an atomising device or inhaler. Alternatively, the sealed container may be a unitary dispensing device such as a single dose nasal inhaler or an aerosol dispenser fitted with a metering valve (metered dose inhaler) which is intended for disposal once the contents of the container have been exhausted.

Where the dosage form comprises an aerosol dispenser, it may take the form of a pump- atomiser. The pressurised aerosol may contain a solution or a suspension of the active ingredient. This may require the incorporation of additional excipients e.g. co-solvents and/or surfactants to improve the dispersion characteristics and homogeneity of suspension formulations. Solution formulations may also require the addition of co-solvents such as ethanol. Other excipient modifiers may also be incorporated to improve, for example, the stability and/or taste and/or fine particle mass characteristics (amount and/or profile) of the formulation.

For pharmaceutical compositions suitable and/or adapted for inhaled administration, the pharmaceutical composition may be a dry powder inhalable composition. Such a composition may comprise a powder base such as lactose, glucose, trehalose, mannitol or starch, the active ingredient and optionally a performance modifier such as L-leucine or another amino acid, cellobiose octaacetate and/or metals salts of stearic acid such as magnesium or calcium stearate.

Aerosol formulations may be arranged so that each metered dose or "puff" of aerosol contains a particular amount of the active ingredient. Administration may be once daily or several times daily, for example 2, 3 4 or 8 times, giving for example 1, 2 or 3 doses each time. The overall daily dose and the metered dose delivered by capsules and cartridges in an inhaler or insufflator will generally be double those with aerosol formulations.

Tablets and capsules for oral administration may contain conventional excipients such as binding agents, for example syrup, acacia, gelatin, sorbitol, tragacanth, mucilage of starch, cellulose or polyvinyl pyrrolidone; fillers, for example, lactose, microcrystalline cellulose, sugar, maize starch, calcium phosphate or sorbitol; lubricants, for example, magnesium stearate, stearic acid, talc, polyethylene glycol or silica; disintegrants, for example, potato starch, croscarmellose sodium or sodium starch glycollate; or wetting agents such as sodium lauryl sulphate. The tablets may be coated according to methods well known in the art.

The HGF-lKl polypeptide variants of the invention may be used in the treatment of various diseases.

Therefore, according to another aspect of the present invention, there is provided an HGF- lKl polypeptide variant of the invention for use in therapy. Also provided are methods of treatment of various diseases and disorders comprising administering to a human patient in need thereof an effective amount of an HGF-lKl polypeptide variant of the invention. The invention further provides the use of an HGF-lKl polypeptide variant of the invention in the manufacture of a medicament for the treatment of various diseases and disorders. The invention further provides an HGF-lKl polypeptide variant of the invention for use in the treatment of various diseases and disorders. The various diseases and disorders will be described in more detail as follows.

The HGF-lKl polypeptide variants of the present invention may be used to treat respiratory diseases such as Acute Lung Injury (ALI) and/or Adult Respiratory Distress Syndrome (ARDS), pulmonary emphysema, resection, allergic airway inflammation, bronchiolitis obliterans, vocal fold scarring, Chronic Obstructive Pulmonary Disease (COPD) and asthma. Furthermore, the HGF-1K1 polypeptide variants of the present invention may be used in the treatment of heart failure (both acute and chronic), particularly myocardial infarction (including cardiac dysfunctions after myocardial infarction), coronary artery disease, cardiac allograft vasculopathy, dilated cardiomyopathy. The HGF-1K1 polypeptide variants of the present invention may also be used in the treatment of cardiac hypertrophy and diastolic dysfunctions.

Furthermore, the HGF-1K1 polypeptide variants of the present invention may be used in the treatment of certain kidney diseases, such as acute renal failure, acute kidney injury, acute renal inflammation, septic acute renal failure, chronic renal failure, chronic renal disease, renal injury, renal fibrosis, immune mediated nephritis, primary focal segmental sclerosis, diabetic nephropathy, glomerulonephritis and chronic allograft nephropathy.

Fibrotic diseases which may be treated using the HGF-1K1 polypeptide variants of the present invention include liver fibrosis, kidney fibrosis, lung fibrosis, acute hepatitis, fulminant hepatitis, cholestasis, liver cirrhosis, alcoholic steatosis, alcoholic steatohepatisis, late onset hepatic failure, total parenteral nutrition-associated liver disease and renal tubule regeneration.

Other diseases in which the use of the HGF-1K1 polypeptide variants of the present invention may be useful include wound healing, Crohn's disease, diabetic ulcers, burns, scarring e.g. keloids, preconditioning of organs for transplantation (in particular kidney, liver, heart and/or lungs), gastrointestinal, ulcerative colitis, gastric ulcer, gastric injury, and in peripheral arterial diseases including critical limb ischemia, neointimal hyperplasia, cerebral ischemia, stroke, peripheral nerve injury, spinal cord injury, amyotrophic lateral sclerosis, multiple sclerosis, hydrocephalus, retinal injury, proliferative vitreoretinopathy, photoreceptor degeneration, retinitis pigmentosa, difficulty in hearing, tinnitus, articular cartilage injury, rheumatoid arthritis, ligament injury. Another disease of potential interest is skeletal muscle disease.

See T. Nakamura et a/., "Hepatocyte growth factor twenty years on: Much more than a growth factor", Journal of Gastroenterology and Hepatology, Vol 26, Suppl 1, p.188-202, 2011.

In one embodiment, the diseases to be treated are selected from COPD, broncheolitis obliterans, acute lung injury (ALI), Adult Respiratory Distress Syndrome (ARDS), idiopathic pulmonary fibrosis (IPF), myocardial infarction, liver fibrosis and kidney fibrosis. In another embodiment, the disease to be treated is COPD. In another embodiment, the disease is myocardial infarction. In a further embodiment, the disease is liver and/or kidney fibrosis. In a further embodiment, the disease is skeletal muscle disease.

In another embodiment the disease or disorder is selected from liver fibrosis, kidney fibrosis, renal tubule regeneration, spinal cord injury, amytrophic lateral sclerosis, multiple sclerosis, critical limb ischemia, stroke (e.g. ischemic stroke), total parenteral nutrition-associated liver disease, proliferative vitreoretinopathy and scarring (e.g. keloids). The present invention is further illustrated by the following figures and examples without being limited to these specific embodiments of the invention.

Examples

As used herein, the media BMGY is 2% Peptone, 1% Yeast Extract, lOOmM Potassium

Phosphate pH 6.0, 1.34% Yeast Nitrogen Base (w/o AA), 0^g/ml_ Biotin, 1% Glycerol.

BMMY is 2% Peptone, 1% Yeast Extract, lOOmM Potassium Phosphate pH 6.0, 1.34% Yeast Nitrogen Base (w/o AA), 0^g/ml_ Biotin, 1% methanol.

YP is yeast peptone dextrose.

LC/MS is liquid chromatography-mass spectroscopy.

OD is optical density.

MeOH is methanol.

Example 1: Small scale Pichia expression with DNA codon optimised for Pichia expression

The protein construct of an HGF-lKl variant according to the present invention, having an amino acid sequence as defined in Seq ID No:4 will be compared with the HGF-lKl construct of SEQ ID No:8. The HGF-lKl construct of Seq ID No:4 has been compared with the HGF-lKl construct of SEQ ID No: l.

The DNA sequences corresponding to Seq ID No:3 and Seq ID No:7 were cloned into the pPICzalpha vector with the alpha mating factor secretory leader and were cloned from synthesized DNA template which had been codon optimised for pichia expression. The plasmids were transformed into three Pichia strains: KM71h (Mut⁵), X33 (Mut⁺) and YCC120 (Mut⁺) which is a protease knock-out strain based on Invitrogen's SMD1163 (pep4prblhis4) with an additional protease knocked out. This additional protease is Yeast Methylotropic Protease (YMP)2. High throughput small scale expression studies were performed using 2 ml culture in 24 well plates in duplicate for each construct.

Seed cultures were prepared by inoculating a single colony into 2 ml BMGY medium and incubating overnight at 30 °C, 260rpm. At T=0h, OD₆oo of seed cultures were measured and expression cultures were inoculated using the seed culture to an OD₆₀o of 1.0 in 2 ml BMMY medium. These cultures were incubated at 30 °C, 260 rpm.

The cultures were fed with 0.5% methanol twice a day and samples were taken every 24h for analysis of OD₆₀o to measure growth rates. Cultures were harvested at 72h and culture supernatants were analysed qualitatively for expression levels of secreted protein by SDS-PAGE. Protein identity was confirmed by peptide mass fingerprinting using trypsin and chymotrypsin enzymatic digestion (data not shown).

For all constructs several individual colonies were tested in KM71h, X33 and YCC120 strains. The HGF-1K1 construct of Seq ID No:4 was consistently expressed at higher levels than the HGF-1K1 construct of SEQ ID No: l and in all three cell strains as used in this experiment.

Example 2: Small scale Pichia expression with DNA codon optimised for mammalian expression Example 2A: Small scale

Constructs using DNA template which had been codon optimised for mammalian expression had also been tested, and similar or improved levels of secreted expression for the HGF-1K1 construct of Seq ID No:4 were observed by SDS-PAGE.

For the HGF-1K1 construct of SEQ ID No: l, 10 colonies in each of the two strains KM71h and X33 were selected from plates with differing amounts of Zeocin ranging from 100 μg/ml Zeocin to 600 μg/ml Zeocin and re-streaked onto fresh plates to ensure single colonies for expression testing.

An initial culture of 50 ml BMGY was inoculated with each single colony and grown at 30 °C and 250 rpm for 24 h. Every colony was also plated on a master plate for future reference. After 24 hours, the cultures were centrifuged at 4500 rpm for 5 min. The pellets were resuspended in 10 ml (KM71H strain) and 100 ml (X33 strain) BMMY. The resulting cultures were incubated for 72 h at 30 °C and 350 rpm. The cultures were fed with methanol every 24 h, to a final concentration of 1% v/v at 24h and 1.5% v/v at 48h. Cultures were harvested at 72h. For each time point, a 500 μΙ aliquot was collected. The harvest sample was centrifuged at 13000 rpm for 5 min. The resulting pellets and supernatants were stored at -80 °C for visual comparison by SDS-PAGE analysis, loading 20 μΙ per well. The harvest OD₅₉₅s were all around 35-40. With this construct, only one clone showed any expression which was colony 19, produced in the KM71h strain.

For the HGF-1K1 construct of Seq ID No:4, 100 and 15 ml of BMGY were inoculated with a single colony in KM71H and X33 strains, respectively. The cultures were grown at 30 °C and 250 rpm for ~ 20 h, after which they were harvested at 4500 rpm for 5 min. They were then resuspended in 25 ml of BMMY and allowed to grow at 350 rpm and 30°C. After 24, 48 and 72 h the cultures were induced with 1, 2 and 3% of methanol (v/v, final concentration).

Samples were collected at 0, 24, 48, and 72 h. The samples were centrifuged and the supernatants and pellets were stored at -20 °C for SDS-PAGE analysis. The OD₅₉₅ of the cultures was also measured at the same time points.

The supernatants from 72h timepoint were analysed by SDS-PAGE with 20 μΙ loaded per well. The levels of expression of the HGF-1K1 construct of Seq ID No:4 were better than the HGF- 1K1 construct of SEQ ID No: l and appeared to be higher with mammalian codon optimised DNA than with Pichia, based on qualitative data from SDS-PAGE gel analysis. The protein yields obtained with the two best expressers in Pichia host strain KM71h were 12 and 20 mg/L, based on the titres obtained after first step purification on a 1 ml heparin column. The highest yields obtained in the X33 host strain were 10 and 15 mg/L. Example 2B: Larger scale expression of the HGF-1K1 construct of Seq ID No:4

The HGF-1K1 construct of Seq ID No:4, made using DNA which had been codon optimised for mammalian expression, was analysed for expression at slightly larger scale using 300ml and 500 ml cultures in the KM71h strain.

125 ml of BMGY were inoculated with a single colony of the construct. The culture was grown at 30°C and 250 rpm for ~ 20 h, after which it was centrifuged at 4500 rpm for 5 min. It was then resuspended in 800 ml of BMMY, divided into 500 and 300 ml aliquots in 2L baffled glass shake flasks and incubated at 350 rpm and 30 °C. The initial OD was 1.5. The culture was fed twice per day with 0.5% methanol (v/v, final concentration) for 72h.

Samples were taken every 24h for analysis. After 72h, the OD₅₉₅ of the culture was measured and was between 24 -26 AU. The samples were centrifuged and the supernatants were analysed by SDS-PAGE analysis, 20μΙ loaded per well. This construct appeared to express well.

Example 2C: Effect of methanol feeding on expression of a mammalian codon-optimised HGF-1K1 construct of Seq ID No: 4

Expression of the HGF-1K1 construct of Seq ID No:4, mammalian codon optimised construct, was compared using different levels of MeOH feeding with X33 and KM71h strains and expression levels were quantified using the Caliper system with protein standard curve.

Seed cultures were set up from a single colony grown on YPDS agar in 250 ml flasks containing 20 ml BMGY medium. Cultures were incubated overnight at 30 °C, 250 rpm. OD₆oo of seed cultures was measured and used to inoculate microtitre plates containing BMMY to a final OD₆₀o of 1.0 in a final volume of 2 ml per well. Cultures were incubated at 30 °C, 260 rpm for 72h. Feeding with 0, 0.25, 0.5 and 1% v/v MeOH was performed twice a day with samples taken for analysis every 24h. At 72h, final OD₆oo was read and cultures were harvested by centrifugation at 4500 rpm for 5 min. Supernatant was removed for analysis.

From the growth curves (data not shown) it could be concluded that MeOH concentration had little effect on the growth of the two strains. The biomass at 0% MeOH was less than that observed for the other conditions, as expected since MeOH was the only carbon source for growth. In the case of the KM71h strain, the wild-type and the chromosomally integrated transformant ("C6") demonstrated similar growth at all MeOH concentrations. For the X33 strain, at higher MeOH concentrations, the wild type strain had a significantly higher biomass (OD₆₀o = 70-80) compared to the C6 transformant strain (OD₆oo = 30-40), suggesting growth is slowed in favour of protein expression.

Selected samples from the experiment were analysed by SDS-PAGE at 72h timepoint (gel not shown). Protein bands of the expected size are only observed with the strains carrying the chromosomally integrated construct, "C6". No significant protein degradation is occurring as the protein is still present and observed at 72h. It should be noted that the protein construct should have a size of 21 kDa however it runs at a slightly higher size on the gel than expected (~26 kDa). For quantification of the expression, samples were run on a Caliper GXII lab chip system along with a serially diluted standard protein ranging from 500 μg/ml to 7.8 μg/ml. Protein expression was determined from this standard curve and normalised for the culture cell density i.e. mg/L/OD₆₀₀.

Expression levels of this construct in a high throughput microtitre plate experiment were demonstrated to be measurable using the Caliper system. Expression was determined from a standard curve and normalised for the culture cell density i.e. mg/L/OD₆₀o in the graph. Expression levels of X33 C6 were highest at 0.25% MeOH concentrations (26.6 mg/L), whereas expression levels of KM71h C6 were highest at 1% MeOH concentrations (14.2 mg/L) in the microtitre plate system.

Example 2D: Expression levels of the HGF-1K1 construct of Seq ID No:4 at 50ml scale

Expression data from shake flask experiments had shown that at 50 ml scale, the KM71h strain gave higher levels of secreted protein than X33 strain with MeOH feed concentrations of at least 1%. At this feed level, KM71h expression was also higher than X33 in microtitre plate scale expression studies.

The effect of MeOH feeding was investigated at shake flask scale in KM71h strain.

Two colonies (colony 45 and 30) were selected to test the effect of different methanol feeding regimes on protein expression

5 ml BMGY were inoculated with a single colony of the HGF-1K1 construct of Seq ID No:4 and allowed to grow overnight at 30 °C and 200 rpm. The OD₅₉₅ of the culture was around 12 after 16 h. The culture was then centrifuged and resuspended in BMMY to a final OD₅₉₅ of 1. Three 50 ml cultures were grown for each colony. The cultures were grown for 24 h at 30 °C and 250 rpm before being fed with methanol. The three methanol regimes are detailed below:

For condition 1, the cultures were fed twice per day with 0.5% methanol final concentration (v/v).

For condition 2, cultures were fed once each day with 0.5% methanol final concentration

(v/v).

For condition 3, cultures were fed once each day with increasing amounts of methanol from 1% to 2% concentration (v/v) at 24h and 48h respectively.

After 72 h, the cultures were centrifuged and all supernatants were filtered using a 0.22 μιτι filter. These samples were then analyzed by HPLC using an ion exchange S0₃ CIM disc column to quantify the amount of protein in the supernatant. The volume injected was 100 μΙ. The samples were analyzed at 280 nm. Quantification was performed using a standard curve ranging from 0 to 100 mg/L in BMMY media. Samples were also analysed using SDS-PAGE, loading 15 μΙ per well. The final OD₅₉₅s are detailed in Table 1. Construct/colony Feeding (% Methanol) Harvest OD

0.5% twice daily 39

Construct having amino acid of

0.5% once daily 42

Seq ID No:4, colony 30

1% and 2% 59

0.5% twice daily 25

Construct having amino acid of

0.5% once daily 20

Seq ID No: 4, colony 45

1% and 2% 43

Table 1 shows harvest cell densities of the HGF-1K1 construct of Seq ID No: 4, the colonies and feeding regimes, measured by absorbance at 595nm (OD₅₉₅).

Even though the cell densities at harvest time vary within the different regimes, the levels of expression seem quite uniform within constructs in the SDS-PAGE gels. The best expresser is colony 30, followed by colony 45. The levels of secreted protein evaluated by HPLC are shown in the table in Table 2.

Table 2 shows expression levels of the constructs evaluated by HPLC Example 3: N-terminal homogeneity

The different constructs were also compared for N-terminal homogeneity during expression using mass spectrometry.

The results show that the HGF-1K1 construct as disclosed in WO2002/088354 (Seq ID No: l), after initial isolation from supernatant using heparin affinity chromatography, was heterogeneous with at least four different N-terminal sequences. The HGF-1K1 construct of Seq ID No:8 was shown by mass spectrometry to have a range of protein masses eluting from the heparin column but these were separated by ion exchange with intact protein recovery of 50%. The HGF- 1K1 construct of Seq ID No:4 was the least heterogeneous construct with the most fully intact protein.

Three colonies were selected for the HGF-1K1 construct of Seq ID No: l. These colonies were colonies 19 (X33 strain), 33 (KM71h strain) and 1 (X33 strain). The colonies were selected from preliminary expression tests using small scale production. 10ml of BMGY were inoculated with a single colony. The cultures were grown at 30 °C and 250 rpm. After approximately 20h, the cultures were centrifuged and resuspended in 100 ml BMMY (timepoint Oh) such that all cultures had an OD₅₉₅ of 1. These cultures were allowed to grow at 175 rpm and 28 °C. The cultures were fed twice per day with 0.5% Methanol (v/v, final concentration).

Samples were collected at 0, 24, 48, and 66 hours. The samples were centrifuged and the supernatants and pellets were stored at -20°C. For analysis, the supernatants were thawed and 50ml were loaded onto a heparin column, column volume of lml. The column was washed with buffer A (PBS) until the A280 and A215 were stable. The protein was then eluted with 100%B (PBS with 1.5 M salt) and analysed by mass spectrometry. All colonies and strains which expressed enough protein for analysis, gave similar spectrographs. A typical spectrograph of the HGF-1K1 construct of Seq ID No: l recovered from the culture of colony 1 in X33 strain is shown in Figure 11 and demonstrates N-terminal heterogeneity with one of the main peaks having a mass of 21068 Da which, allowing for equipment calibration error, corresponds to the expected intact protein mass of 21071 Da allowing for the loss of 10 Da owing to the formation of 5 di-sulphide bonds. The mass difference of the other peaks allows the assignment of their N-terminal sequences.

For the HGF-1K1 construct of Seq ID No:8 in GS115 strain, 50 ml BMGY were inoculated from a glycerol stock of transformed GS115 Pichia Pastoris yeast strain in vector pPIC9K in 15% glycerol in YP with secretion leader alpha-factor signalling peptide from Saccharomyces cerevisiae (MRC, UK) and incubated overnight at 30 °C with shaking at 250 rpm. This starter culture was then used to inoculate 1 L BMGY (divided into two aliquots of 500 ml in 2 L shake flasks) and incubated at 30 °C with shaking at 250 rpm. At an OD₆oo of 2.0 the culture was centrifuged at 3000 rpm for 20 min. The pellet was resuspended with 40 ml of BMMY and used to inoculate a total volume of 4 L BMMY (divided into eight 2L flasks). One drop of antifoam was added to each flask. The flasks were incubated at 30 °C, 250 rpm overnight with 0.5% MeOH induction twice during the next day and harvest after 24h. Cultures were harvested with centrifugation at 7000 rpm for 45 min at 4 °C. The supernatant was filtered through a 0.8 μιτι filter before purification.

The supernatant was loaded onto a 50 ml pre-equilibrated heparin column at a flow rate between 3 ml/min and 5 ml/min. The column was then washed with 10 CV of PBS until the UV absorbance returned to baseline. The protein was eluted with a gradient of 0 - 100% buffer B (PBS with additional NaCI to a final salt concentration of 2M) over 50 min at 5 ml/min. Fractions of 4 ml were collected throughout the gradient. Buffer B was held at 100% for at least 3 CV before the column was re-equilibrated in buffer A. Peak fractions were pooled and dialysed overnight in 5 L of 50 mM MES, 150 mM NaCI, pH 6.0 at 4 °C in preparation for cation exchange chromatography. The pool was then loaded at 4 ml/min onto a 10 ml Source 15S IEX column pre-equilibrated with 50 mM MES, 150 mM NaCI, pH 6.0. After loading the column was washed with at least 5 CV of 50 mM MES, 150 mM NaCI, pH 6.0 at 4 ml/min and the protein was eluted with a gradient to 2M NaCI in 50 mM MES pH 6.0 over 30 min at 4 ml/min. 2 ml fractions were collected. Separated protein fractions were analysed by SDS-PAGE which showed the heterogeneity of the 1K1 protein. This was confirmed by mass spectrometry. Of the eight fractions collected, only two, B9 and B10, showed a single main peak by mass spectrometry of 21100/21099 Da. The expected mass for intact, correctly processed HGF-1K1 construct of Seq ID No:8 would be 21100/21099 Da taking into account loss of 10 Da through formation of 5 di-sulphide bonds.

For an HGF-1K1 construct of Seq ID No:4, four colonies were selected from preliminary expression tests using small scale production. These colonies were as follows: colony 30 (KM71h strain), colony 46 (KM71h strain), colony 42 (X33 strain) and colony 24 (X33 strain). 10 ml of BMGY were inoculated with a single colony of each construct and incubated at 30 °C, 250 rpm. The cultures were grown for ~20h before being centrifuged and pellets were resuspended in 0.5 L BMMY to an OD₅₉₅ of 1 (timepoint Oh). The cultures were incubated at 175 rpm, 28 °C and fed twice per day with 0.5% methanol (v/v, final concentration).

Samples were collected at 0, 24, 48 and 66 h. The samples were centrifuged and supernatants were stored at -20 °C until analysis. Analysis was performed by thawing the samples before loading 50 ml onto a 1 ml heparin column. The column was then washed with PBS until the A₂8o and A₂i5 were stable. The protein was then eluted with 100%B (PBS with 1.5 M salt) and analysed by mass spectrometry. The molecular weights of the proteins from each construct is depicted in Table 3, The mass spectrograms showed a major peak of 21052 for colonies 24 (in X33) and 30 (in KM71h). The expected mass is 21051 Da for the HGF-1K1 construct of Seq ID No:4, taking into account the loss of 10 Da owing to the formation of 5 di-sulfide bonds.

Table 3: shows a table of protein masses observed with the HGF-1K1 construct of Seq ID No:4, for each colony and strain measured by mass spectrometry.

Example 4: Comparison of heterogeneity after cation exchange chromatography for the HGF-1K1 construct of Seq ID No:8 and the HGF-1K1 construct of Seq ID No:4

To obtain intact protein of the HGF-1K1 construct of Seq ID No:8 for comparison, cultures were harvested after 24h and purified by heparin affinity chromatography as previously described (see Example 3). This protein pool and the post heparin affinity purified HGF-1K1 construct of Seq ID No:4 described in Example 5 below were further purified by cation exchange chromatography and fractions were analysed by SDS-PAGE. From the gels, the HGF-1K1 construct of Seq ID No:4 appeared to be less heterogeneous than the HGF-1K1 construct of Seq ID No:8, although conditions were not directly comparable as the HGF-1K1 construct of Seq ID No:8 was harvested after 24h, whereas the HGF-1K1 construct of Seq ID No:4 was harvested after 66h. However, for both proteins, pure intact protein was obtained, which supports the theory that certain N-terminal extensions to the 1K1 protein can improve expression, stability during upstream processing and N- terminal homogeneity.

100 ml of BMGY in a 250ml shake flask was inoculated with the HGF-lKl construct of Seq ID No:8 in GS115 strain straight from glycerol stock and then incubated at 25 °C, 150 rpm over a weekend. The OD₆oo was measured as 14 at this time. 50 ml of this culture was then used to inoculate each of two 1L flasks each containing 500 ml BMGY (1 L total). The flasks were incubated at 30 °C, 250 rpm for 14h at which point the OD₆₀o was measured as 8.5.

The 1 L culture was centrifuged at 3000 rpm for 30 min and the pellet resuspended in 120 ml BMMY. 12 2 L shake flasks, each containing 500 ml BMMY (6 L total) were inoculated with 10 ml suspended pellet. One drop of antifoam was added to each flask. The flasks were incubated at 30 °C, 250 rpm overnight.

The cultures were then further induced by the addition of 0.5% MeOH at 17.5h and 22h before being harvested at 24h post resuspension in induction media (BMMY). Cultures were harvested by centrifugation at 7000rpm for 45 min at 4°C and the supernatants were filtered through a 0.8 μιτι filter for purification.

A protein capture step was performed using heparin affinity chromatography on a 50 ml prepacked heparin column which was loaded with protein supernatant using a flow rate of 5 ml/min. The column was washed with 10 CV of PBS at 5 ml/min until baseline stabilised. Elution was performed using a gradient elution of 0 -100% 2M NaCI in PBS over 50 min at 5 ml/min. 4 ml fractions were collected throughout the gradient. The column was washed with 100% 2M NaCI in PBS for 3 CV before re-equilibration with PBS.

Peak fractions were pooled and yield was calculated as 7 mg/L. The pool was dialysed into 50 mM MES, 150 mM NaCI pH 6.0 before further purification using cation exchange chromatography.

The dialysed pool of fractions was then loaded onto the pre-equilibrated Sourcel5S IEX column at a flow rate of 4 ml/min. The column was then washed until UV₂8o stabilised. The protein was eluted by raising the concentration of Buffer B (50 mM MES, 2 M NaCI pH6.0) from 0-100% over 30 min flowing at 4 ml/min. 2 ml fractions were collected throughout the gradient.

For the HGF-lKl construct of Seq ID No:4, peak fractions from the elution from the heparin affinity column were pooled and dialyzed overnight against buffer A (50 mM MES, 250 mM NaCI, pH6.0) for the ion exchange purification.

The dialyzed protein was loaded onto a Source 15S column (GE Healthcare), CV of 10 ml, and the column was washed until A280 and A215 were stable. The protein was then eluted using a gradient 0-100% B (50mM MES, 2M NaCI, pH6.0) for 12 CVs. The flow rate was 4 ml/min and 2 ml fractions were collected.

The fractions were analysed by SDS-PAGE which showed that some fractions contained single bands of purified protein. The HGF-lKl construct of Seq ID No:4 appeared to be more homogeneous than the HGF-lKl construct of Seq ID No:8 although culture conditions were different for both. Fractions were further analysed by mass spectrometry (data not shown) showing a single mass of the correct size for these fractions and the samples were fully active in a cMET receptor phosphorylation assay.

Example 4B: c-MET receptor phosphorylation assay

The A549 cell line (derived from human alveolar epithelium) is plated into 96-well plates at 2e⁴ cells/well and allowed to adhere overnight in growth medium +10% foetal calf serum. Following overnight incubation, the media is replaced with serum-free media for at least 5 h to starve. Cells are then treated with the HGF-lKl constructs (in serum-free media + 0.1%BSA + 2 IU/ml sodium heparin) for 15 min at 37 °C followed by cell lysis. Percentage c-MET phosphorylation is then determined by assaying lysates for both total c-MET and phosphorylated c-MET using the MesoScale Discovery phospho/total c-MET duplex assay kit according to manufacturer's instructions (MSD Catalogue #K15126D).

The HGF-lKl construct as disclosed in WO2002/088354 (corresponding to SEQ ID No:l) may be prepared and analysed in the same manner for comparison.

Example 5: Comparison of the HGF-lKl construct of Seq ID No:8 and an HGF-lKl construct of Seq ID No:4 for stability during upstream process

The HGF-lKl construct of Seq ID No:8 was expressed at 500 ml scale in shake flasks. The construct was received as a glycerol stock of transformed GS115 Pichia Pastoris yeast strain in vector pPIC9K in 15% glycerol in YP with secretion leader alpha-factor signalling peptide from Saccharomyces cerevisiae (MRC, UK).

The HGF-lKl construct of Seq ID No:4 was expressed at the same scale in a separate experiment and it was demonstrated by SDS-PAGE analysis that this construct was not as prone to proteolytic degradation during the upstream process as the HGF-lKl construct of Seq ID No:8.

Induction conditions of 0.5% MeOH feeding twice per 24h were used for expressing both constructs.

50 ml BMGY (in 250 ml shake flask) was inoculated straight from the glycerol stocks of the HGF-lKl construct of Seq ID No:8. These were left to grow at 30 °C with shaking at 250 rpm. 1 L BMGY (500 ml per 2 L flask) was inoculated with these cultures at OD₆oo of 2.6 to a final OD₆oo of 0.016. The cultures were left overnight at 30 °C with shaking at 250 rpm.

After 16h an OD₆oo of 13 was measured for the HGF-lKl construct of Seq ID No:8. The cultures were centrifuged and cell pellets resuspended in 500 ml of BMMY (500 ml of BMMY/IL pellet). 40 ml of BMMY resuspended cells were added to each one of ten 2 L flasks containing 460 ml BMMY to reach a starting OD₆₀o of 1.0. One drop of antifoam was added per flask. The flasks were incubated at 30 °C with shaking at 250 rpm. Cultures were fed twice per 24h with 0.5% MeOH until harvested. Samples were taken at 24h and 48h and supernatants were analysed by SDS-PAGE. 15 μΙ supernatant was loaded per well of 4 -12% gradient Novex gel which was run at 200 V for 40 min with MES running buffer.

For the HGF-1K1 construct of Seq ID No:4, colony 30 (strain KM71H) had been selected from preliminary expression tests for a larger-scale production. 125 ml BMGY were inoculated with a single colony of the construct. The culture was grown at 30 °C and 250 rpm. After ~ 20h, 10 ml of this culture were used to inoculate two 2 L flasks containing 0.5 L BMGY. These cultures were incubated at 30 °C and 250 rpm for an additional 30h, after which they were centrifuged and the cell pellets were resuspended in 60 ml BMMY. Ten 2 L flasks, each containing 0.5 L BMMY, were inoculated with 5 ml each of the resuspension culture (timepoint Oh). The OD₅₉₅ for the ten cultures was 1.5. These cultures were incubated at 175 rpm and 28 °C and were fed twice per day with 0.5% Methanol (v/v, final concentration).

Samples were collected at 0, 25, 40, 48, and 66 h. The samples were centrifuged and the supernatants and pellets were stored at -20 °C for SDS-PAGE analysis. The OD₅₉₅ of the cultures was also measured at the same time points. 20 μΙ of the thawed supernantants were loaded onto the gel.

For the HGF-1K1 construct of Seq ID No:8, after 24h visible bands were observed, but after 48h the protein band was no longer visible. This suggests that the HGF-1K1 construct of Seq ID No:8 is susceptible to protease activity in culture conditions.

The HGF-1K1 construct of Seq ID No:4 was visible by SDS-PAGE until 48h post induction, although protein was lost after 66h. This suggests that the HGF-1K1 construct of Seq ID No:4 is slightly more stable to culture conditions during upstream processing.

The protein titres after this degradation were measured and compared after purification by heparin affinity chromatography and the HGF-1K1 construct of Seq ID No:8 pool was analysed by mass spectrometry which confirmed that the protein had been proteolytically degraded.

To perform this analysis, the culture was continued with induction up to 72h. Harvest was performed by centrifuging the cultures at 7000 rpm for 45 min at 4 °C. The supernatant (approximately 4.5 L) was then filtered through a 0.8 μιτι filter unit. The filtered supernatant was used to purify the protein.

After harvesting at 72h, the filtered supernatant was loaded on a 50 ml XK50 Heparin column. Column loading was performed at a rate between 3 ml/min and 5 ml/min. The column was washed with at least 10 CV of buffer A (PBS) at 5 ml/min until the A₂₈₀ reading stabilised. The protein was eluted by increasing the concentration of buffer B (PBS with additional NaCI to a final salt concentration of 2M) from 0-100% over 50 min at 5 ml/min. Fractions of 4 ml were collected throughout the gradient. Buffer B was held at 100% for at least 3 CV before the column was re- equilibrated in buffer A. The peak fractions were pooled and dialysed overnight in 5 L of IEX buffer A (50 mM MES, 150 mM NaCI, pH 6.8) at 4 °C. The harvested and filtered culture supernatant containing the HGF-1K1 construct of Seq ID No:4 was captured by heparin affinity chromatography in the same way.

Protein yield for the HGF-1K1 construct of Seq ID No:8 was measured by A₂₈₀ and an expression titre of 2 mg/L was determined, whereas protein expression titre for the HGF-1K1 construct of Seq ID No:4 was determined to be 5 mg/L. For both these constructs, however, maximum yield was not obtained as both were harvested after protein degradation and loss had occurred.

The HGF-1K1 construct of Seq ID No:4 was further purified by cation exchange chromatography and several peaks were separated. All protein fractions were analysed by mass spectrometry and some fractions contained pure intact protein. Cation exchange purified fractions of the HGF-1K1 construct of Seq ID No:8 protein were analysed by mass spectrometry and all were found to have masses coinciding with protein that had been cleaved at the N-terminus to varying degrees with no intact material remaining. Data not shown

The HGF-1K1 construct as disclosed in WO2002/088354 (corresponding to SEQ ID No: l) may be prepared and analysed in the same manner for comparison.

Example 6: Characterisation of qlycosylation for the HGF-1K1 construct of Seq ID No:4.

Example 6A: Enrichment for glycosylated HGF-1K1 construct of Seq ID No:4 on Concanavalin A Sepharose.

Purified HGF-1K1 construct of Seq ID No:4, 46.5 mg in 9 ml dialysed into 20 mM Tris, 0.5 M

NaCI, 1 mM CaCI₂, 1 mM MnCI₂ pH 7.4, was batch bound overnight to 1.5 ml of Concanavalin A Sepharose (GE healthcare). The resin was recovered in a gravity column, washed with 6 CV equilibration buffer and eluted with 0.2 M Methyl alpha-D-mannopyranoside in equilibration buffer. Aproximately 0.7 mg of protein was recovered in the eluate. LC/MS analysis identified species with mass increases corresponding to the addition of between 2 and 7 hexose residues, with 3 (+486) and 6 (+972) residues being the most prevalent.

Example 6B: Fractionation of HGF-1K1 construct of Seq ID No:4 by reverse phase

For preparation of HGF-1K1 construct of Seq ID No:4 modified with a single hexose residue, protein was fractionated by reverse phase (Vydac C4 column, 1 ^χ 25cm) on an Agilent 1200 HPLC. 800 μg of purified 1K1 in 75 μΙ was injected onto the column equilibrated in 0.06%(w/v)TFA, 10% acetonitrile and separated using a gradient to 33% acetonitrile. Fractions containing protein showing a mass increase of +162 were identified by LC/MS (Q-Tof API) and dried down under vacuum (Savant speed vac). The dried protein was resuspended in 6 M Urea, 50 mM Tris pH 8, 10 mM glycine, 0.2 M NaCL. Example 6C: Preparation of protein for peptide mapping

Protein samples were prepared for peptide mapping by SDS-PAGE under reducing conditions using Novex® NuPAGE reagents and gels. Protein was mixed with sample buffer, DTT added to a final concentration of 50mM and the sample heated to 70 °C for 5 min. 10 μg samples of protein were applied to each lane of a 4-12% Bis-Tris gel. After electrophoresis gels were stained with Instant Blue coomassie stain (Novexin). Gels were destained with water.

In another experiment, of the EAEAQRK constructs of Seq ID No:4, approximately 75% of the purified material was not glycosylated. Of the 25% material which was glycosylated, approximately half had a single glucose and the other half had one glucose and a variety of mannose glycosylation (data not shown).

Example 7: Cloning different versions of the HGF gene

Six different versions of the HGF gene (Seq ID 2, 3, 5, 6, 8 and 9) were cloned from plasmid 0958433_lKl_Pichia_CodUpt_pM or 0958432_lKl_Human_CodUpt_pM (GeneArt) into the pPICZ-a- vector (Invitrogen), which includes the a-mating factor leader_, sequence and the corresponding cleavage signal sequences for expression in Pichia Pastoris. The different sequences comprise the YVEG, YAEG and EAEA molecules, both mammalian and pichia codon optimized. The final amino acid sequences correspond to Seq ID Nos: l, 4 and 8.

The synthetic genes were assembled from synthetic oligonucleotides and/or PCR products.

The product was cloned into pMA (ampR) using Sad and Kpnl restriction sites. The plasmid DNA was purified from transformed bacteria and concentration determined by UV spectroscopy (GeneArt). The original plasmid (0958433 or 0958433) was PCR amplified with the respective primers, provided by Invitrogen. The forward primers included a Xhol restriction site and the reverse primers included a Notl restriction site and a double stop codon after the last amino acid residue.

PCR was performed using the Pwo Polymerase system (Roche) according to the manufacturer's instructions. One μΙ of a 1: 10 dilution in water of the GeneArt plasmid DNA was used as a template. The PCR cycles were set following the manufacturer's instructions, with a total volume of 4x50 μΙ for each PCR reaction. The PCR reactions were purified using the Qiaquick PCR purification kit. Double Xhol and Notl digestions were then performed using 15 μΙ of the pPICZ vector (Invitrogen) or 15 μΙ of each of the 1K1 PCR fragments from the previous step. The reactions, in Buffer III (NEB) and BSA, were incubated at 37 °C for 5 h. Each reaction was again purified with the PCR purification Kit, Qiagen.

The digested inserts were then ligated into the digested pPICZ-a vector. The procedure was carried out according to manufacturer's instructions, using the Roche Rapid DNA ligation kit and approximately 100 and 250 ng, respectively, of plasmid and insert DNA. 5-10 μΙ of the ligation mixture were transformed into E. Coli one shot TOP10 (Invitrogen) following the manufacturer's instructions. The cells were allowed to recover at 37 °C for 2 h and then 75-150 μΙ of each reaction were plated on LB low salt + zeocin plates and incubated overnight.

5 ml low salt LB zeocin media were inoculated with at least ten colonies and grown overnight for plasmid DNA extraction. The samples were sent to sequence using the 5'- and 3'-AOX primers (Invitrogen). Once their identity was verified, one selected colony from each molecule was grown in 100 ml LB low salt + zeocin for a larger DNA extraction. The purified plasmid DNA was digested with Pmel, therefore making it linear and ready for transformation into competent Pichia X33 and KM71H cells. These strains were selected as they are readily commercially available. X33 is a mut⁺, whereas KM17h is mut^s and therefore use different AOX promoters and metabolise methanol differently. The cells were prepared and electroporated with 0.5-1 μg of DNA, according to standard procedures (Invitrogen). The cells were plated on LB low salt plates containing 100, 300, 600 and 900 μg/ml zeocin.

Example 8: Final expression assays

Seed cultures were set up from single colonies grown on YPD media with different concentrations of zeocin. The cultures were set up in 50 ml flasks containing 10 ml BMGY medium and they were incubated overnight at 30 °C and 220 rpm. The OD (A₆₀o) was then measured and the cultures were harvested at 700 RCF (xg) for 10 min. The pellets were resuspended to a final OD of 1 in 400 ml of BMMY containing Antifoam 204 (Sigma Aldrich). The cultures were incubated for an additional 30-36 h at 30 °C and 220rpm. 2 ml methanol (Sigma) were added to each flask (0.5% methanol v/v) every twelve hours to induce protein expression. After the induction time, the samples were harvested at 700 RCF (xg) for 10 min and the supernatants were stored at -80 °C. 15 - 18 μΙ of each supernatant were loaded onto SDS PAGE gels to check for expression of the different HGF-1K1 variants. A representative example of a successful screening is depicted in Figure 12.

A total number of 20 colonies for each construct/strain combination were screened, with a total of 200 colonies. The screening assays were performed as in Example 2A and 2D hereinabove. Ten colonies were screened in these experiments. The screening results are shown in Figure 13. Most constructs were expressed properly and higher percentages of colonies which expressed the desired protein were obtained. However, the intensity of the bands observed on the gels was very variable. Constructs EAEAQRK (Seq ID No:6, Pichia host strain KM71H), YAEG (Seq ID No:3, Pichia host strain KM71H; Seq ID No: 2, Pichia host strains KM71H and X33) and YVEG (Seq ID No:9, Pichia host strains KM71H and X33) showed the highest levels of expression. Example 9 - Preliminary Heparin Purifications

The best expressers of the different HGF 1K1 constructs were selected for a preliminary characterization that included heparin purification and Mass Spectrometry analysis. The supernatants were thawed and filtered (0.22 μΜ). Approximately 30 ml of the harvested material were loaded onto a 1 ml heparin column (GE Amersham) equilibrated with PBS. The column was washed with PBS until the A₂is and A₂₈o were stable. The protein was then eluted with a 20 CV gradient ranging from 0 to 100% buffer B (PBS with NaCI 1.5M). The relevant protein fractions were pooled and analyzed by Mass Spectrometry (as described in Example 11B below) before the second purification step, Ion Exchange.

A minimum of five colonies and a maximum of eight were cultured, purified and characterized for each construct in each strain. The yields varied significantly among the different clones for every strain and construct combination, ranging between, approximately, 0.2 and 2.2 mg/L (supernatant). The Mass Spectrometry profiles for each Seq ID Nos: l, 4 and 8 were consistent throughout the whole experiment and did not show any significant variation with the strain or codon optimization variations, see Table 4. The purified proteins from two to four clones of each construct in each strain were selected for cMet phosphorylation assays. The differences observed in the EC₅₀ and maximum phosphorylation levels were minimal in most cases. The best clones for each construct and strain were tested again in parallel in a final assay (Tables 5, 6 and 7). Purified material from constructs YVEG and EAEAQRK (Seq ID Nos: 10 and 6) was used as control. None of the Pichia codon optimized constructs (Table 5) showed equal or higher activities than the controls. Clones Seq ID NO:3 KM71H clone 2, Seq ID No:3 X33 clone 2 and Seq ID No: 10 KM71H were selected as the best leads for each construct/strain. The EC₅o and maximum phosphorylation values obtained with the mammalian codon optimized constructs were again very variable (Tables 6 and 7). However, most constructs were not as active as the controls. Clones Seq ID No: 9 X33 clone 1, Seq ID No: 2 KM71H clone 1, Seq ID No: 10 KM71H clone 2, Seq ID NO: 2 X33 clone 2, Seq ID No: 10 X33 clone 1, Seq ID No:7 KM71H clone 2 and Seq ID No:7 X33 clone 1 were selected as the best leads for each construct/strain.

Table 4: The main species for each of the HGF-1K1 constructs as detected by Mass

Spectrometry. The numbers in parentheses indicate the ratios of the different species. This table is a summary of the most commonly observed profiles and it does not represent each of the results individually i.e. some of the EAEAQRK profiles may lack one or two species. Clones Seq ID No:3 KM71H clone 2, Seq ID No:3 X33 clone 2 and Seq ID No: 10 KM71H were selected as the best leads for each construct/strain.

Seq ID No: 3 X33 554.55 2.57

Seq ID No: 3 X33 clone 2 2.70 44.76

Seq ID No:3 KM71H 9.13 36.10

Seq ID No:3 KM71H clone 2 3.05 40.82

Seq ID No: 10 KM71H 4.95 42.47

Seq ID No: 10 KM71H clone 2 7.79 41.73

Table 5: cMet phosphorylation assays results for Pichia codon optimized constructs. Purified material from constructs YVEG and EAEAQRK (Seq ID Nos: 10 and 6) was used as control. Clones Seq ID No:3 KM71H clone 2, Seq ID No:3 X33 clone 2 and Seq ID No: 10 KM71H were selected as the best leads for each construct/strain.

Table 6: cMet phosphorylation assays results for Pichia and Mammalian codon optimized constructs. Purified material from constructs YVEG and EAEAQRK (Seq ID Nos: 10 and 6) was used as control. Clones Seq ID No:9 X33, Seq ID No:2 KM71H and Seq ID No: 10 KM71H clone 2 were selected as the best leads for each construct/strain.

Table 7: cMet phosphorylation assays results for Mammalian codon optimized constructs. Purified material from constructs YVEG and EAEAQRK (Seq ID Nos: 10 and 6) was used as control. Clones Seq ID No:2 X33 clone 2, Seq ID No:9 X33, Seq ID No:6 KM71H clone 2 and Seq ID No:6 X33 were selected as the best leads for each construct/strain.

Example 10: Final expression assays

Seed cultures were set up from YPD glycerol stocks produced from a single colony grown on

YPD media. The cultures were set up in 250 ml flasks containing 50 ml BMGY medium and they were incubated overnight at 30 °C and 220 rpm. The OD (A₆oo) was then measured and the cultures were harvested at 700 RCF (xg) for 10 min. The pellets were resuspended to a final OD of 1 in 40 ml BMMY containing Antifoam 204 (Sigma Aldrich). The cultures were incubated for an additional 30-36h at 3 0°C and 220 rpm. 2 ml methanol (Sigma) were added to each flask (0.5% methanol v/v) every twelve hours to induce protein expression. After the induction time, the samples were harvested at 3000 rpm for 10 min and the supernatants were stored at -80 °C.

100 μΙ samples were collected at 18 and 36h. They were centrifuged and the supernatants and pellets were stored at -20 °C for SDS-PAGE analysis. The OD of the cultures was also measured at the same time points. The final OD₆oo values for all cultures ranged between 14 and 20. The protein of interest was expressed in most supernatants. The best expressers were YVEG X33 Mammalian codon optimised (Seq ID No:9) and EAEAQRK KM71H Mammalian codon optimised (Seq ID No:6).

Owing to low yields, the expression assays were modified for the following constructs: EAEAQRK KM71H and X33, Pichia codon optimised (Seq ID No:7), YAEG KM71H and X33, Mammalian codon optimised (Seq ID No:2), and YVEG X33 Pichia codon optimised (Seq ID No: 10). These constructs were less stable throughout the expression assays and were extensively degraded by proteolytic activity after 36h. Therefore, to allow for enough protein to be obtained for further characterisation, the induction times were reduced to 24h.

Example HA: Heparin purification - capture step

The supernatants were thawed and filtered (0.22 μΜ). Approximately 380 ml of the harvested material were loaded onto a 1 ml heparin column at a flow rate of 1 ml/min (GE Amersham). The column had been equilibrated with PBS. The supernatant volume was doubled for those constructs with low expression levels ie EAEAQRK KM71H and X33, Pichia CO (Seq ID No:7), YAEG KM71H, Mammalian CO (Seq ID No: 2), and YVEG X33 Pichia CO (Seq ID No: 10). For these constructs, two different cultures were grown and pooled to double the volume of supernatant. The column was washed with PBS at a flow rate of 1 ml/min until the A₂is and A₂₈o were stable. The protein was then eluted with a 20 CV gradient ranging from 0 to 100% buffer B (PBS with NaCI 1.5M). The protein peak fractions were pooled and characterised by mass spectrometry before the second purification step, Ion Exchange. The highest titre (see Figure 14) was that of EAEAQRK KM71H Mammalian codon optimised (Seq ID No: 6);, followed by that of YVEG X33 Mammalian codon optimised (Seq ID No: 9). Most other titres were roughly 40% lower than that of Seq ID NO:6. Constructs EAEAQRK X33 Pichia codon optimised (Seq ID No: 7), YAEG X33 Mammalian codon optimised (Seq ID No: 2) and YAEG KM71H Pichia codon optimised (Seq ID No:3) were the worst expressers, with titres 80% lower than that of Seq ID No: 6.

Example 11B: Mass spectrometry analysis

Analysis by mass spectrometry was performed on protein samples of approximately 10 picomoles using the following protocol. Samples were injected onto a Zorbax Poroshell 300SB-C8 guard column (2.1 mm ^χ 12.5 mm), desalted by washing with 0.1% formate in 5% acetonitrile, then eluted with 0.1% formate in 90% acetonitrile at a flow rate of 0.5 mlmin^"1. The eluant was split such that a flow of 0.2 mlmin^"1 was directed to the ESI interface (a standard Z-spray source fitted with an electrospray probe) of a Micromass Q-Tof API-US mass spectrometer controlled from a PC running MassLynx (version 4.1) software. The source temperature and desolvation temperature were set to 100 °C and 150 °C respectively. The capillary voltage was 3.0 kV and the sample cone voltage was 35 V. A calibrant mixture was injected directly into the source at the start of each sample analysis to enable mass correction of the charge envelope against external standards. The mass spectrometer was routinely calibrated against myoglobin or sodium iodide to ensure that peaks in the mass/charge spectrum accurately represent the samples. Raw data were externally mass corrected and deconvoluted to the parent mass spectrum using the MaxEnt 1 algorithm of MassLynx.

The main species for both the YVEG and EAEAQRK constructs (Seq ID Nos:8 and 4, respectively) were the intact molecules, with observed molecular weights of, respectively, 21104 and 21029/21030 (Table 8). These values are in accordance with the expected molecular weights of 21109 and 21060 KDa, taking into consideration the five disulfide bonds, a pyroglutamate modification for the Seq ID No:4 and the possible experimental error. The main species for the YAEG constructs (Seq ID No: l) were the intact protein and modified protein with a molecular weight corresponding to proteolytically cleaved protein with an amino terminal sequence starting EG. The observed molecular weights of these proteins were 21076-80 Da and 20840-42 Da respectively. These were again in accordance with the expected molecular weights of 21081 and 20846 Da, respectively.

Although it may appear from the MS data alone that the constructs of EAEAQRK are more heterogeneous than the YVEGQRK constructs, as can be seen in Example 12 below, the EAEAQRK intact species are more easily separated from degradant by-products, resulting in a higher overall yield after secondary purification step.

YAEG samples (Seq ID No: l) YAEG ~ EG > NTI

EAEAQRK samples (Seq ID No:4) EAEAQRK » EAQRK > QRK > NTI

Table 8 shows approximate rankings of species observed by MS. Note, the MS data doesn't allow for quantitative but rather for qualitative or semi-quantitative analysis. Therefore, all rankings have been done on the basis of visual inspection of signal intensity and the instrument counts. Example 12: Ion Exchange Chromatography

The pooled purified material from the capture step was thawed and loaded onto a 1 ml Resource S column (GE Amersham) equilibrated with MES 50 mM. The column was washed with MES 50 mM, pH 6, until the A₂is and A₂₈o were stable. The column was then washed for 5 CV with 20% Buffer B (MES 50 mM, NaCI 1M, pH 6). The protein was eluted with a 20 CV gradient ranging from 20 to 80% buffer B. Finally, the column was washed with 100% Buffer B for 5 CV. The relevant protein fractions were analysed by mass spectrometry.

A summary from the results obtained after Ion Exchange chromatography is shown in Figure 15 and Table 9. In general terms, the full resolution of different species wasn't achieved. Most peaks contained at least two species as determined by mass spectrometry. The heterogeneity of the purified constructs seemed to follow the trend EAEAQRK > YVEG > YAEG. The major species for EAEAQRK (Seq ID No:4) and YVEG (Seq ID No: 8; only in strain X33) was the intact protein. The mammalian codon optimised version in strain KM71H was the most homogenous variant of the EAEAQRK construct. In most cases, the EAEAQRK intact species was the only protein species present in the two main purification peaks. That seemed to be the case for YVEG as well but the integrity of this construct was very dependent on the strain. Indeed, the main species for Seq ID Nos:9 and 10 in strain KM71H was NTI (19045 KDa). YAEG (Seq ID No: l) was mostly clipped or present as a mixture of YAEG and EG in the main peaks.

The yield of intact species was very similar for both YVEG X33 Seq ID No:9 and EAEAQRK KM71H Seq ID No:6. The percentage of expressers and the levels of expression in the expression screening assays were also very similar for these two constructs. However, the EAEAQRK Seq ID No:6 was chosen as the lead molecule because of the large scale experiments (Examples 3, 4A and 5). Seq ID No:6 presented lower levels of heterogeneity and it was also more stable during the upstream process.

YVEG X33 Pichia CO 64

YVEG KM71H Pichia CO <5

EAEAQRK X33 Mammalian CO 48

EAEAQRK KM71H Mammalian CO 46

EAEAQRK X33 Pichia CO 45

EAEAQRK KM71H Pichia CO 17

Table 9. Percentage of intact construct species after Heparin and Ion Exchange purification. The percentages were calculated considering the ratio of different species in a given peak, as determined by Mass Spectrometry. All peaks with an area equal or inferior to 15% of the total area were discarded. The yields of intact species were calculated considering the Heparin purification titres and are therefore not the final yields, obtained after second step purification.

Example 13: cMET receptor phosphorylation assay

The A549 cell line (derived from human alveolar epithelium) was plated into 96-well plates at 2e⁴ cells/well and allowed to adhere overnight in growth medium +10% foetal calf serum. Following overnight incubation, the media was replaced with serum-free media for at least 5 hours to starve. Cells were then treated with the HGF-1K1 constructs (in serum-free media + 0.1%BSA + 2 IU/ml sodium heparin) for 15 min at 37 °C followed by cell lysis. Percentage c-MET phosphorylation was then determined by assaying lysates for both total c-MET and phosphorylated c-MET using the MesoScale Discovery phospho/total c-MET duplex assay kit according to manufacturer's instructions (MSD Catalogue #K15126D).

There were no significant differences within the EC₅o and maximum phosphorylation values among the different constructs and the controls, YVEG (Seq ID No:8) and EAEAQRK (Seq ID No:4), see Table 10. Most constructs showed a similar degree of activity. The two exceptions were constructs EAEAQRK (Seq ID No:7, Pichia Codon Optimized; Pichia host strain KM71H) and YAEG (Seq ID No: 3, Pichia Codon Optimized; Pichia host strain X33), which showed a lower activity. The concentration of the purified Seq ID No: 3 at the time of the assay was much lower than that calculated after purification, which explains its lower activity. There was no apparent reason for the low activity of Seq ID No:7, as the construct was active in strain KM71H. The construct showed similar low levels of activity in this strain throughout the tests so it cannot be attributed to an experimental error. Further experiments would need to be performed to determine the cause of this low activity.

Seq ID No:7 X33 0.74 40.16

Seq ID No: 2 KM71H 0.99 37.85

Seq ID No:9 KM71H 0.49 41.93

Seq ID No:6 KM71H 0.92 38.72

Seq ID No: 2 X33 0.87 45.39

Seq ID No:9 X33 0.47 42.98

Seq ID No:6 X33 0.44 45.86

Seq ID No: 10 GS115 0.83 45.73

Seq ID No:6 KM71H 0.98 40.58

Table 10: cMet Receptor Binding assay results. The EC₅o and maximum phosphorylation values are shown. Most constructs showed a similar degree of activity. Constructs EAEAQRK (Seq ID 7, Pichia Codon Optimized; Pichia host strain KM71H) and YAEG (Seq ID 3, Pichia Codon Optimized; Pichia host strain X33) showed lower activity than average.

Example 14: Reverse Phase Characterization (Glycosylation profiles^')

The glycosylation profile of the different 1K1 constructs was characterized by running the IEX purified material on Reverse Phase (HPLC). Low protein recovery for some constructs meant that they could not be tested. The tested constructs were Seq ID Nos: 2, 3, 6, 7, 9, and 10, as detailed below:

YAEG* KM71H Mammalian CO (Seq ID No:2)

YAEG* X33 Pichia CO (Seq ID No: 3)

YVEG X33 and KM71H* Pichia CO (Seq ID No: 10)

YVEG X33 and KM71H* Mammalian CO (Seq ID No:9)

EAEAQRK X33 and KM71H Pichia CO (Seq ID No:7)

EAEAQRK X33 Mammalian CO (Seq ID No: 6)

Where possible, the intact species was selected. In some cases (*), a clipped or mixed version of the construct had to be used (insert reference to table from IEX).

The HPLC system was run at a flow rate of 1 ml/min, using a Vydac C4 column (214TP54), 250 x 4.6mm. The temperature was set at 40 °C. The gradient between Buffer A (0.06% TFA in water) and Buffer B (0.06% TFA in HPLC acetonitrile) is detailed below. The separation phase occurs between 25 and 35% B.

Time (minutes) %B

0 10

4 10

5 25

45 35

46 80

51 80 52 10

56 10

The different glycosylated species were fully separated during the run (data not shown). The levels of glycosylation did not vary significantly with the different strains or codon optimizations, with the possible exception of EAEAQRK, which was less glycosylated in strain X33.

The lack of some combinations of constructs/strains makes it difficult to reach a definitive conclusion on the comparison between constructs. In general, the amount of non-glycosylated EAEAQRK species appeared greater than that of YVEG, except for the EAEAQRK KM71H Pichia codon optimized construct. The fact that the percentage of non glycosylated species is so similar in both KM71H and X33 wild type Pichia strains for all the constructs could potentially indicate that the glycosylation occurs at the same amino acid positions in all constructs.

Claims

Claims

1. An HGF-lKl variant polypeptide, wherein the polypeptide chain consists of the amino acids of SEQ ID No:4.

2. An HGF-lKl variant polypeptide, wherein the polypeptide chain consists of amino acids 1 to 179 of SEQ ID No:4.

3. An HGF-lKl variant polypeptide, wherein the polypeptide chain consists of the amino acids of Seq ID No: 4, apart from one or two (e.g. one) amino acid substitutions at Thrl64 and/or Asnl66.

4. An HGF-lKl variant polypeptide according to claim 3, wherein the substitutions are independently selected from Thrl64Ala, Thrl64Asn, Thrl64Val, Asnl66Asp, Asnl66Ser and

Asnl66His

5. An HGF-lKl variant polypeptide according to any one of claims 1 to 4 which is glycosylated.

6. An HGF-lKl variant polypeptide according to claim 5, which is glycosylated with one or more (e.g. one to six or one to nine) hexose units.

7. An HGF-lKl variant polypeptide according to claim 5 or claim 6 which is glycosylated at one or two O-linked glycosylation sites, preferably at Serl27 and/or Serl80.

8. An HGF-lKl variant polypeptide according to claim 7 which is glycosylated at Serl27, and preferably which is not glycosylated at Serl80.

9. An HGF-lKl variant polypeptide according to claim 7 which is glycosylated at Serl80 with mannose.

10. An HGF-lKl variant polypeptide according to claim 9 which is glycosylated with three or six, preferably three mannose units.

11. An HGF-lKl variant polypeptide according to any one of claims 1 to 4 which is not glycosylated.

12. An HGF-lKl variant according to any one of claims 1 to 4 or any one of claims 7 to 10, which is not glycosylated at Asnl66.

13. A polynucleotide encoding an HGF-lKl variant as defined in any one of claims 1 to 4.

14. A polynucleotide according to claim 13, wherein the DNA sequence comprises Seq ID No: 5, Seq ID No:6 or Seq ID No:7.

15. A plasmid comprising a polynucleotide as defined in claims 13 or claim 14.

16. A host cell comprising a plasmid as defined in claim 15.

17. A host cell according to claim 16, wherein the host cell is a Pichia pastoris cell or a Saccharomyces cerevisiae cell, preferably a Pichia pastoris strain selected from X-33, SMD1168H, SMD1136, YCC120, GS115 and KM71H.

18. A process for the preparation of an HGF-lKl polypeptide as defined in any one of claims 1 to 12 comprising the steps of:

a. Inserting a plasmid (preferably as defined in claim 15) into a host cell;

b. Expressing the protein;

c. Recovering the protein; and then optionally d. Purifying the protein.

19. A pharmaceutical composition comprising an HGF-lKl variant as defined in any one of claims 1 to 12 and a pharmaceutically acceptable carrier or diluent.

20. A pharmaceutical composition according to claim 19 wherein the composition is adapted for inhaled or intravenous administration.

21. A pharmaceutical composition according to claim 20, which is adapted for inhaled administration and is either a spray dried powder formulation or a liquid formulation for nebulising.

22. An HGF-lKl variant polypeptide as defined in any one of claims 1 to 12 for use in therapy.

23. An HGF-lKl variant polypeptide as defined in any one of claims 1 to 12 for use in the treatment of a disease or disorder selected from COPD, broncheolitis obliterans, acute lung injury (ALI), Adult Respiratory Distress Syndrome (ARDS), idiopathic pulmonary fibrosis (IPF), myocardial infarction, liver fibrosis and kidney fibrosis.

24. An HGF-lKl variant polypeptide as defined in any one of claims 1 to 12 for use in the treatment of a disease or disorder selected from liver fibrosis, kidney fibrosis, renal tubule regeneration, spinal cord injury, amytrophic lateral sclerosis, multiple sclerosis, critical limb ischemia, stroke (e.g. ischemic stroke), total parenteral nutrition-associated liver disease, proliferative vitreoretinopathy and scarring (e.g. keloids).

25. An HGF-lKl variant polypeptide for the use according to claim 23, wherein the disease is COPD and optionally wherein the polypeptide is inhaled.

26. An HGF-lKl variant polypeptide as defined in any one of claims 1 to 12 for the manufacture of a medicament for the treatment of a disease or disorder selected from COPD, broncheolitis obliterans, acute lung injury (ALI), Adult Respiratory Distress Syndrome (ARDS), idiopathic pulmonary fibrosis (IPF), myocardial infarction, liver fibrosis and kidney fibrosis.

27. HGF-lKl variant polypeptide as defined in any one of claims 1 to 12 for the manufacture of a medicament for the treatment of a disease or disorder selected from liver fibrosis, kidney fibrosis, renal tubule regeneration, spinal cord injury, amytrophic lateral sclerosis, multiple sclerosis, critical limb ischemia, stroke (e.g. ischemic stroke), total parenteral nutrition- associated liver disease, proliferative vitreoretinopathy and scarring (e.g. keloids).

28. A method of treating of a disease or disorder selected from COPD, broncheolitis obliterans, acute lung injury (ALI), Adult Respiratory Distress Syndrome (ARDS), idiopathic pulmonary fibrosis (IPF), myocardial infarction, liver fibrosis and kidney fibrosis comprising administering to a human patient in need thereof an effective amount of the HGF-lKl polypeptide variant as defined in any one of claims 1 to 12.

29. A method of treating of a disease or disorder selected from liver fibrosis, kidney fibrosis, renal tubule regeneration, spinal cord injury, amytrophic lateral sclerosis, multiple sclerosis, critical limb ischemia, stroke (e.g. ischemic stroke), total parenteral nutrition-associated liver disease, proliferative vitreoretinopathy and scarring (e.g. keloids) comprising administering to a human patient in need thereof an effective amount of the HGF-lKl polypeptide variant as defined in any one of claims 1 to 12.

30. A method of treatment according to claim 28, wherein the disease is COPD and optionally wherein the polypeptide is inhaled.