O-linked glycoforms of faktor vii and method to manufacture them

FIELD OF THE INVENTION

The present invention relates to compositions comprising Factor VII polypeptides having altered patterns of O-linked glycosylation.

BACKGROUND OF THE INVENTION

The biological activity of many glycoproteins is highly dependent upon the presence or absence of particular oligosaccharide structures attached to the glycoprotein. The glycosylation pattern of a therapeutic glycoprotein can affect numerous aspects of the therapeutic efficacy, such as, e.g, solubility, resistance to proteolytic attack, thermal inactivation, immunogenicity, half-life, bioactivity, bioavailability, and stability.

Glycosylation is a complex post-transitional modification that is cell dependent. Following translation, proteins are transported into the endoplasmic reticulum (ER), glycosylated and sent to the Golgi for further processing and subsequent targeting and/or secretion. During glycosylation, either N-linked or O-linked glycoproteins are formed.

Serum proteins involved in coagulation or fibrinolysis, including, e.g., Factor VII and Factor IX are proving to be useful therapeutic agents to treat a variety of pathological conditions. Accordingly, there is an increasing need for formulations comprising these proteins that are pharmaceutically acceptable and exhibit a uniform and predetermined clinical efficacy.

Because of the many disadvantages of using human plasma as a source of pharmaceutical products, it is preferred to produce these proteins in recombinant systems. The clotting proteins, however, are subject to a variety of co- and post-translational modifications, including, e.g., asparagine-linked (N-linked) glycosylation; serine- or threonine-linked (O-linked) glycosylation; and γ-carboxylation of glu residues. These modifications may be qualitatively or quantitatively different when heterologous cells are used as hosts for large-scale production of the proteins. In particular, production in heterologous cells often results in a different array of glycoforms, which identical polypeptides are having different covalently linked oligosaccharide structures.

In different systems, variations in the oligosaccharide structure of therapeutic proteins have been linked to, inter alia, changes in immunogenicity and in vivo clearance. Thus, there is a need in the art for compositions and methods that provide preparations comprising recombinant human Factor VII or modified Factor VII or Factor VII-related polypeptides that contain predetermined glycoform patterns.

SUMMARY OF THE INVENTION

The present invention relates to a pharmaceutical composition comprising (a) a preparation comprising human Factor VII polypeptides that exhibit predetermined serine linked glycoform patterns; and b) a pharmaceutically acceptable carrier.

As used herein, a glycoform pattern refers to the distribution within the preparation of oligosaccharide chains having varying structures that are covalently linked to a serine residue located in an EGF-like domain in the amino acid backbone of the polypeptide.

The human Factor VII polypeptide contains a Cys-X1-SerX2-Pro-Cys motif and wherein said serine is located at position Ser52 and forms part of a Glc-O-Ser covalent bond wherein at least 85% of said serine 52 residues in the preparation are glycosylated with Xyl-Xyl-Glc.

In one embodiment of the invention, at least 90%, at least 95%, or at least 98% of said serine residues are glycosylated with Xyl-Xyl-Glc.

In another aspect the invention provides methods for making preparations of Factor VII as described herein. The methods are useful for remodelling or altering the glycosylation pattern present on a Factor VII polypeptide upon its initial expression.

Described herein is a general enzymatic methodology for the modification of glycans (in particular O-linked glycans) of Factor VII polypeptide, in order to improve or enhance their pharmaceutically properties. One method involves treatment of the Factor VII polypeptide with xylosidases in order to remove any terminal xylose residues; other methods includes attachment of xylose residues to the exposed glucose or xylose residues on the Factor VII polypeptide by treatment with xylosyltransferases; a third method includes attachment of glucose residues to serine and/or threonine amino acid residues in the Factor, VII polypeptide backbone thereby creating a glycosylated Factor VII polypeptide. The invention in particular provides a method as set out in claim 3.

BRIEF DESCRIPTION OF THE DRAWINGS

<ul id="ul0001" list-style="none" compact="compact"><li><figref idref="f0001">Fig. 1</figref> shows the serine 52 glycosylation of wt-Factor VII.</li><li><figref idref="f0001">Fig. 2</figref> shows an O-glycosylatlon mapping of Factor VII.</li><li><figref idref="f0002">Fig. 3</figref> shows a reaction scheme for the making of a preparation of glycoproteins exhibiting a predetermined serine/threonine-linked glycosylation.</li><li><figref idref="f0003">Fig. 4</figref> shows a chromatogram from first HIC cycle showing fractions "A" and "B".</li><li><figref idref="f0004">Fig. 5</figref> shows a chromatogram obtained by reloading fraction "A" onto the HIC column; Glc-O-Ser52-FVII was identified in the peak fraction, fraction 10.</li><li><figref idref="f0005">Fig. 6</figref> shows a chromatogram obtained by reloading fraction "B" onto the HIC column; Xyl-Xyl-Glc-O-Ser52-FVII was identified in the peak fraction, fraction 15.</li><li><figref idref="f0006">Fig. 7A</figref> shows a tryptic peptide map of the peak fraction, fraction 10; the arrow indicates the Gic-O-Ser52 O-glycopeptide.</li><li><figref idref="f0006">Fig. 7B</figref> shows a tryptic peptide map of the peak fraction, fraction 15; the arrow indicates the Xyl-Xyl-Glc-O-Ser52 O-glycopeptide.</li><li><figref idref="f0007">Fig. 8A</figref> shows a total mass analysis of the peak fraction, fraction 10; the arrow indicates the Glc-O-Ser52-rFVIIa O-glycoform.</li><li><figref idref="f0007">Fig. 8B</figref> shows a total mass analysis of the peak fraction, fraction 15; the arrow indicates the Xyl-Xyl-Glc-O-rFVIIa O-glycoform.</li></ul>

DETAILED DESCRIPTION

The following abbreviations are used herein: <dl id="dl0001" compact="compact"><dt>Glc</dt><dd>= glucosyl</dd><dt>Xyl</dt><dd>= xylosyl</dd><dt>Ser</dt><dd>= serine (one letter code: S)</dd><dt>Thr</dt><dd>= threonine (one letter code: T)</dd><dt>Pro</dt><dd>= proline (one letter code: P)</dd><dt>Cys</dt><dd>= cysteine (one letter code: C)</dd><dt>FVII</dt><dd>= Factor VII</dd><dt>FVIIa</dt><dd>= activated (two-chain) Factor VII</dd><dt>FIX</dt><dd>= Factor IX</dd><dt>FIXa</dt><dd>= activated (two-chain) Factor IX</dd></dl>

As used herein, a "glycoform pattern" (or "glycosylation pattern") refers to the distribution within the preparation of oligosaccharide chains having varying structures that are covalently linked to a serine residue located in the amino add backbone of the polypeptide.

"Homogeneity" refers to the structural consistency across a population of polypeptides with conjugated glycans. Thus, a glycoprotein preparation is said to be about 100% homologous if all contained glycoprotein molecules contain identical glycans attached to the relevant glycosylation site. For example, a preparation of Factor VII polypeptides is said to be at least 90% homologous if at least 90% of the Factor VII polypeptide molecules contain the glycan of interest attached to serine 52 (e.g., Xyl-Xyl-Glc-O-Ser52).

"Uniform glycoform" or "Uniform glycosylation" or "Uniform glycosylation pattern", when referring to a glycopeptide species, refers to the percentage of acceptor moieties, i.e., serine residues, that are glycosylated by the glycan of interest. For example, in the case of Factor VII, a substantially uniform glycosylation patterns exists if substantially all (as defined below) of the serine residues in position 52 are glycosylated with the glycan of interest. It is understood by one skilled in the art that the starting material may contain glycosylated serine residues that are glycosylated with a species having the same structure as the glycan of interest. Thus, the calculated percent glycosylation includes serine residues that are glycosylated with the glycan of interest according to the invention, as well as those serine residues already glycosylated with the glycan of interest in the starting material.

At least about 85%, such as at least about 90%, at least about 95%, or at least about 98% of the serine residues in the Factor VII polypeptide is glycosylated with a predetermined, specific glycan or glycan of interest. The glycosylation pattern is typically determined by one or more methods known to those skilled in the art, such as, e.g., tryptic digestion followed by high performance liquid chromatography (HPLC), liquid-chromatography-mass spectrometry (LC-MS), matrix assisted laser desorption mass time of flight spectrometry (MALDITOF), capillary electrophoresis, and the like.

The term "acceptor moiety" is intended to encompass the group or moiety to which a desired oligo- or mono-saccharide group is transferred such as, without limitation, the serine residue located within a Cys-X1-Ser-X2-Pro-Cys motif, a Glc-residue covalently linked to such a serine residue, or a Xyl-residue covalently linked to a Glc-residue or a Xyl-residue in a Glc-O-Ser or Xyl-Glc-O-Ser moiety, respectively.

The term "saccharide donor moiety" is intended to encompass an activated saccharide donor molecule (e.g., a desired oligo- or mono-saccharide structure such as, for example, a xylosyl-xylosyl-donor, xylosyl-donor, or glycosyl-donor) having a leaving group (e.g., xylose-UDP or glucose-UDP) suitable for the donor moiety acting as a substrate for the relevant catalysing enzyme (e.g. glycosyltransferase, xylosidase or xylosyltransferase).

Oligosaccharides are considered to have a reducing and a non-reducing end, whether or not the saccharide at the reducing end is in fact a reducing sugar. In accordance with accepted nomenclature, oligosaccharides are depicted herein with the non-reducing end on the left and the reducing end on the right (e.g., Xyl-Xyl-Glc-O-Ser)

As used herein, the term "glycan" or, interchangeable, "sugar chain", "oligosaccharide chain" or "oligosaccharide moiety" refers to the entire oligosaccharide structure that is covalently linked to a single serine residue. The glycan may comprise one or more saccharide units; examples of glycans include, e.g., Glc-, Xyl-Glc-, and Xyl-Xyl-Glc-.

The term "O-glycosylation site" is intended to indicate the glycosylation site at serine (i.e., Ser) located within the motif Cys1-X1-Ser-X2-Pro-Cys2 where Cys1 and Cys2 are the first and second conserved cysteines of the EGF repeat and X1 and X2 independently is any amino acid. These include the glycosylation site at position Ser-52 (S52) of human wt-FVII and the corresponding residues in homologous polypeptides such as, without limitation, FVII sequence variants and FIX polypeptides. The term "corresponding residues" is intended to indicate the Ser amino acid residue corresponding to the Ser52 residue of wild-type Factor VII (see <figref idref="f0001">Fig.1</figref>) when the sequences are aligned. Amino acid sequence homology/identity is conveniently determined from aligned sequences, using a suitable computer program for sequence alignment, such as, e.g., <nplcit id="ncit0001" npl-type="s"><text>the ClustalW program, version 1.8, 1999 </text></nplcit>(<nplcit id="ncit0002" npl-type="s"><text>Thompson et al., 1994, Nucleic Acid Research, 22:4673-4680</text></nplcit>). For example, the wt-factor VII Ser52-residue corresponds to the Ser53-residue of wt-Factor IX. It is further to be understood that polypeptide variants may be created containing non-naturally occurring Cys-X1-Ser-X2-Pro-Cys motifs and thereby containing non-naturally occurring O-glycosylation sites that can be glycosylated in accordance with the present invention. In one embodiment of the invention, the O-glycosylation site is a serine-glycosylation site and the motif is Cys1-X1-Ser-X2-Pro-Cys2.

The term "terminal glucose" is intended to encompass glucose residues linked as the terminal sugar residue in a glycan, or oligosaccharide chain, i.e., the terminal sugar of each antenna is glucose. The term "terminal xylose" is intended to encompass xylose residues linked as the terminal sugar residue in a glycan, or oligosaccharide chain.

Enzymes

Protein O-glycosyltransferase may be prepared as described, e.g., in <nplcit id="ncit0003" npl-type="s"><text>Shao et al. (Glycobiology 12(11): 763-770 (2002</text></nplcit>)).

The alpha-xylosidase enzymes may be prepared, e.g., as described by <nplcit id="ncit0004" npl-type="s"><text>Monroe et al. (Plant Physiology and Biochemistry 41:877-885 (2003</text></nplcit>)).

The enzyme, UDP-D-xylose: β-D-glucoside α-1,3-D-xylosyltransferase can be prepared from HepG2 cells as described by <nplcit id="ncit0005" npl-type="s"><text>Omichi et al. (Eur. J. Biochem. 245:143-146 (1997</text></nplcit>)).

The enzyme, UDP·D-xylose: α-D-xyloside α1,3-xylosyltransferase can be prepared from HepG2 cells as described by <nplcit id="ncit0006" npl-type="s"><text>Minamida et al. ((J.Biochem. (Tokyo) 120: 1002-1006 (1996</text></nplcit>)).

UDP-beta-D-glucose is commercially available from, e.g., Sigma (Sigma U4625)

UDP-D-xylose is commercially available from, e.g., Sigma (Sigma U5875)

Glycoproteins

The motif: Cys-X1-Ser/Thr-X2-Pro-Cys appears to be primarily found in epidermal growth factor (EGF) domains of multi-modular proteins such as coagulation and fibrinolytic factors. The motif is a consensus sequence for O-glucose modification whereby a serine-glucose (Gic-O-Ser) or threonine-glucose (Glc-O-Thr) bond is formed. Coagulation factors VII, IX, X and XII as well as plasma Protein Z, Fibrillin and thrombospondin have all been shown to contain the Cys-X1-Ser/Thr-X2-Pro-Cys consensus sequence. Of these, Factors VII and IX and Protein Z have been described to contain the consensus sequence Cys-X1-Ser-X2-Pro-Cys.

Factor VII (proconvertin) is a vitamin K-dependent serine protease which participates in the blood clotting process by participating in activation of prothrombin into thrombin. FVII is activated into FVIIa by contact with exposed tissue factor (TF) at sites of injury of the vessel wall.

Factor VII polypeptides and Factor VII-related polypeptides

As used herein, the terms "Factor VII polypeptide" or "FVII polypeptide" means any protein comprising the amino acid sequence 1-406 of wild-type human Factor VIIa (i.e., a polypeptide having the amino acid sequence disclosed in <patcit id="pcit0001" dnum="US4784950A"><text>U.S. Patent No. 4,784,950</text></patcit>), variants thereof as well as Factor VII-related polypeptides, Factor VII derivatives and Factor VII conjugates. This includes FVII variants, Factor VII-related polypeptides, Factor VII derivatives and Factor VII conjugates exhibiting the same or improved biological activity relative to wild-type human Factor VIIa.

The term "Factor VII" is intended to encompass Factor VII polypeptides in their uncleaved (zymogen) form, as well as those that have been proteolytically processed to yield their respective bioactive forms, which may be designated Factor VIIa. Typically, Factor VII is cleaved between residues 152 and 153 to yield Factor VIIa. Such variants of Factor VII may exhibit different properties relative to human Factor VII, including stability, phospholipid binding, altered specific activity, and the like.

As used herein, "wild type human FVIIa" is a polypeptide having the amino add sequence disclosed in <patcit id="pcit0002" dnum="US4784950A"><text>U.S. Patent No. 4,784,950</text></patcit>.

The term "Factor VII derivative" as used herein, is intended to designate a FVII polypeptide exhibiting the same or improved biological activity relative to wild-type Factor VII, in which one or more of the amino adds of the parent peptide have been genetically and/or chemically and/or enzymatically modified, e.g. by alkylation, glycosylation, PEGylation, acylation, ester formation or amide formation or the like. This includes but is not limited to PEGylated human Factor VIIa, cysteine-PEGylated human Factor VIIa and variants thereof. Non-limiting examples of Factor VII derivatives includes GlycoPegylated FVII derivatives as disclosed in <patcit id="pcit0003" dnum="WO0331464A"><text>WO 03/31464</text></patcit> and US Patent applications <patcit id="pcit0004" dnum="US20040043446A"><text>US 20040043446</text></patcit>, <patcit id="pcit0005" dnum="US20040063911A"><text>US 20040063911</text></patcit>, <patcit id="pcit0006" dnum="US20040142856A"><text>US 20040142856</text></patcit>, <patcit id="pcit0007" dnum="US20040137557A"><text>US 20040137557</text></patcit>, and <patcit id="pcit0008" dnum="US20040132640A"><text>US 20040132640</text></patcit> (Neose Technologies, Inc.); FVII conjugates as disclosed in <patcit id="pcit0009" dnum="WO0104287A"><text>WO 01/04287</text></patcit>, <patcit id="pcit0010" dnum="US20030165996A"><text>US patent application 20030165996</text></patcit>, <patcit id="pcit0011" dnum="WO0158935A"><text>WO 01/58935</text></patcit>, <patcit id="pcit0012" dnum="WO0393465A"><text>WO 03/93465</text></patcit> (Maxygen ApS) and <patcit id="pcit0013" dnum="WO0202764A"><text>WO 02/02764</text></patcit>, <patcit id="pcit0014" dnum="US20030211094A"><text>US patent application 20030211094</text></patcit> (University of Minnesota).

The term "improved biological activity" refers to FVII polypeptides with i) the same or increased proteolytic activity compared to recombinant wild type human Factor VIIa or ii) to FVII polypeptides with substantially the same or increased TF binding activity compared to recombinant wild type human Factor VIIa or iii) to FVII polypeptides with substantially the same or increased half life in blood plasma compared to recombinant wild type human Factor VIIa. The term "PEGylated human Factor VIIa" means human Factor VIIa, having a PEG molecule conjugated to a human Factor VIIa polypeptide. It is to be understood, that the PEG molecule may be attached to any part of the Factor VIIa polypeptide including any amino acid residue or carbohydrate moiety of the Factor VIIa polypeptide. The term "cysteine-PEGylated human Factor VIIa" means Factor VIIa having a PEG molecule conjugated to a sulfhydryl group of a cysteine introduced in human Factor VIIa.

Non-limiting examples of Factor VII variants having the same or increased proteolytic activity compared to recombinant wild type human Factor VIIa include S52A-FVIIa, S60A-FVIIa (<nplcit id="ncit0007" npl-type="s"><text>Lino et al., Arch. Biochem. Biophys. 352: 182-192, 1998</text></nplcit>); FVIIa variants exhibiting increased proteolytic stability as disclosed in <patcit id="pcit0015" dnum="US5580560A"><text>U.S. Patent No. 5,580,560</text></patcit>; Factor VIIa that has been proteolytically cleaved between residues 290 and 291 or between residues 315 and 316 (<nplcit id="ncit0008" npl-type="s"><text>Mollerup et al., Biotechnol. Bioeng. 48:501-505, 1995</text></nplcit>); oxidized forms of Factor VIIa (<nplcit id="ncit0009" npl-type="s"><text>Kornfelt et al., Arch. Biochem. Biophys. 363:43-54, 1999</text></nplcit>); FVII variants as disclosed in <patcit id="pcit0016" dnum="DK0200189W" dnum-type="L"><text>PCT/DK02/00189</text></patcit> (corresponding to <patcit id="pcit0017" dnum="WO02077218A"><text>WO 02/077218</text></patcit>); and FVII variants exhibiting increased proteolytic stability as disclosed in <patcit id="pcit0018" dnum="WO0238162A"><text>WO 02/38162</text></patcit> (Scripps Research Institute); FVII variants having a modified Gla-domain and exhibiting an enhanced membrane binding as disclosed in <patcit id="pcit0019" dnum="WO9920767A"><text>WO 99/20767</text></patcit>, <patcit id="pcit0020" dnum="US6017882A"><text>US patents US 6017882</text></patcit> and <patcit id="pcit0021" dnum="US6747003B"><text>US 6747003</text></patcit>, <patcit id="pcit0022" dnum="US20030100506A"><text>US patent application 20030100506</text></patcit> (University of Minnesota) and <patcit id="pcit0023" dnum="WO0066753A"><text>WO 00/66753</text></patcit>, US patent applications <patcit id="pcit0024" dnum="US20010018414A"><text>US 20010018414</text></patcit>, <patcit id="pcit0025" dnum="US2004220106A"><text>US 2004220106</text></patcit>, and <patcit id="pcit0026" dnum="US200131005B"><text>US 200131005</text></patcit>, <patcit id="pcit0027" dnum="US6762286B"><text>US patents US 6762286</text></patcit> and <patcit id="pcit0028" dnum="US6693075B"><text>US 6693075</text></patcit> (University of Minnesota); and FVII variants as disclosed in <patcit id="pcit0029" dnum="WO0158935A"><text>WO 01/58935</text></patcit>, <patcit id="pcit0030" dnum="US6806063B"><text>US patent US 6806063</text></patcit>, <patcit id="pcit0031" dnum="US20030096338A"><text>US patent application 20030096338</text></patcit> (Maxygen ApS), <patcit id="pcit0032" dnum="WO0393465A"><text>WO 03/93465</text></patcit> (Maxygen ApS), <patcit id="pcit0033" dnum="WO04029091A"><text>WO 04/029091</text></patcit> (Maxygen ApS), <patcit id="pcit0034" dnum="WO041083361A"><text>WO 041083361</text></patcit> (Maxygen ApS), and <patcit id="pcit0035" dnum="WO04111242A"><text>WO 04/111242</text></patcit> (Maxygen ApS), as well as in <patcit id="pcit0036" dnum="WO04108763A"><text>WO 04/108763</text></patcit> (Canadian Blood Services).

Non-limiting examples of FVII variants having increased biological activity compared to wild-type FVIIa include FVII variants as disclosed in <patcit id="pcit0037" dnum="WO0183725A"><text>WO 01/83725</text></patcit>, <patcit id="pcit0038" dnum="WO0222776A"><text>WO 02/22776</text></patcit>, <patcit id="pcit0039" dnum="WO02077218A"><text>WO 02/077218</text></patcit>, <patcit id="pcit0040" dnum="DK0200635W" dnum-type="L"><text>PCT/DK02/00635</text></patcit> (corresponding to <patcit id="pcit0041" dnum="WO03027147A"><text>WO 03/027147</text></patcit>), Danish patent application PA <patcit id="pcit0042" dnum="DK200201423"><text>2002 01423</text></patcit> (corresponding to <patcit id="pcit0043" dnum="WO04029090A"><text>WO 04/029090</text></patcit>), Danish patent application PA <patcit id="pcit0044" dnum="DK200101627"><text>2001 01627</text></patcit> (corresponding to <patcit id="pcit0045" dnum="WO03027147A"><text>WO 03/027147</text></patcit>); <patcit id="pcit0046" dnum="WO0238162A"><text>WO 02/38162</text></patcit> (Scripps Research Institute); and FVIIa variants with enhanced activity as disclosed in <patcit id="pcit0047" dnum="JP2001061479B"><text>JP 2001061479</text></patcit> (Chemo-Sero-Therapeutic Res Inst.). Examples of variants of factor VII include, without limitation, L305V-FVII, L305V/M306D/D309S-FVII, L305I-FVII, L305T-FVII, F374P-FVII, V158T/M298Q-FVII, V158D/E296V/M298Q-FVII, K337A-FVII, M298Q-FVII, V158D/M298Q-FVII, L305V/K337A-FVII, V158D/E296V/M298Q/L305V-FVII, V158D/E296V/M298Q/K337A-FVII, V158D/E296V/M298Q/L305V/K337A-FVII, K157A-FVII, E296V-FVII, E296V/M298Q-FVII, V158D/E296V-FVII, V158D/M298K-FVII, and S336G-FVII, L305V/K337A-FVII, L305V/V158D-FVII, L305V/E296V-FVII, L305V/M298Q-FVII, L305V/V158T-FVII, L305V/K337A/V158T-FVII, L305V/K337A/M298Q-FVII, L305V/K337A/E296V-FVII, L305V/K337A/V158D-FVII, L305V/V158D/M298Q-FVII, L305V/V158D/E296V-FVII, L305V/V158T/M298Q-FVII, L305V/V158T/E296V-FVII, L305V/E296V/M298Q-FVII, L305V/V158D/E296V/M298Q-FVII, L305V/V158T/E296V/M298Q-FVII, L305V/V158T/K337A/M298Q-FVII, L305V/V158T/E296V/K337A-FVII, L305V/V158D/K337A/M298Q-FVII, L305V/V158D/E296V/K337A-FVII, L305V/V158D/E296V/M298Q/K337A-FVII, L305V/V158T/E296V/M298Q/K337A-FVII, S314E/K316H-FVII, S314E/K316Q-FVII, S314E/L305V-FVII, S314E/K337A-FVII, S314E/V158D-FVII, S314E/E296V-FVII, S314E/M298Q-FVII, S314E/V158T-FVII, K316H/L305V-FVII, K316H/K337A-FVII, K316H/V158D-FVII, K316H/E296V-FVII, K316H/M298Q-FVII, K316H/V158T-FVII, K316Q/L305V-FVII, K316Q/K337A-FVII, K316Q/V158D-FVII, K316Q/E296V-FVII, K316Q/M298Q-FVII, K316Q/V158T-FVII, S314E/L305V/K337A-FVII, S314E/L305V/V158D-FVII, S314E/L305V/E296V-FVII, S314E/L305V/M298Q-FVII, S314E/L305V/V158T-FVII, S314E/L305V/K337A/V158T-FVII, S314E/L305V/K337A/M298Q-FVII, S314E/L305V/K337A/E296V-FVII, S314E/L305V/K337A/V158D-FVII, S314E/L305V/V158D/M298Q-FVII, S314E/L305V/V158D/E296V-FVII, S314E/L305V/V158T/M298Q-FVII, S314E/L305V/V158T/E296V-FVII, S314E/L305V/E296V/M298Q-FVII, S314E/L305V/V158D/E296V/M298Q-FVII, S314E/L305V/V158T/E296V/M298Q-FVII, S314E/L305V/V158T/K337A/M298Q-FVII, S314E/L305V/V158T/E296V/K337A-FVII, S314E/L305V/V158D/K337A/M298Q-FVII, S314E/L305V/V158D/E296V/K337A-FVII, S314E/L305V/V158D/E296V/M298Q/K337A-FVII, S314E/L305V/V158T/E296V/M298Q/K337A-FVII, K316H/L305V/K337A-FVII, K316H/L305V/V158D-FVII, K316H/L305V/E296V-FVII, K316H/L305V/M298Q-FVII, K316H/L305V/V158T-FVII, K316H/L305V/K337A/V158T-FVII, K316H/L305V/K337A/M298Q-FVII, K316H/L305V/K337A/E296V-FVII, K316H/L305V/K337A/V158D-FVII, K316H/L305V/V158D/M298Q-FVII, K316H/L305V/V158D/E296V-FVII, K316H/L305V/V158T/M298Q-FVII, K316H/L305V/V158T/E296V-FVII, K316H/L305V/E296V/M298Q-FVII, K316H/L305V/V158D/E296V/M298Q-FVII, K316H/L305V/V158T/E296V/M298Q-FVII, K316H/L305V/V158T/K337A/M298Q-FVII, K316H/L305V/V158T/E296V/K337A-FVII, K316H/L305V/V158D/K337A/M298Q-FVII, K316H/L305V/V158D/E296V/K337A-FVII, K316H/L305V/V158D/E296V/M298Q/K337A-FVII, K316H/L305V/V158T/E296V/M298Q/K337A-FVII, K316Q/L305V/K337A-FVII, K316Q/L305V/V158D-FVII, K316Q/L305V/E296V-FVII, K316Q/L305V/M298Q-FVII, K316Q/L305V/V158T-FVII, K316Q/L305V/K337A/V158T-FVII, K316Q/L305V/K337A/M298Q-FVII, K316Q/L305V/K337A/E296V-FVII, K316Q/L305V/K337A/V158D-FVII, K316Q/L305V/V158D/M298Q-FVII, K316Q/L305V/V158D/E296V-FVII, K316Q/L305V/V158T/M298Q-FVII, K316Q/L305V/V158T/E296V-FVII, K316Q/L305V/E296V/M298Q-FVII, K316Q/L305V/V158D/E296V/M298Q-FVII, K316Q/L305V/V158T/E296V/M298Q-FVII, K316Q/L305V/V158T/K337A/M298Q-FVII, K316Q/L305V/V158T/E296V/K337A-FVII, K316Q/L305V/V158D/K337A/M298Q-FVII, K316Q/L305V/V158D/E296V/K337A-FVII, K316Q/L305V/V158D/E296V/M298Q/K337A-FVII, K316Q/L305V/V158T/E296V/M298Q/K337A-FVII, F374Y/K337A-FVII, F374Y/V158D-FVII, F374Y/E296V-FVII, F374Y/M298Q-FVII, F374Y/V158T-FVII, F374Y/S314E-FVII, F374Y/L305V-FVII, F374Y/L305V/K337A-FVII, F374Y/L305V/V158D-FVII, F374Y/L305V/E296V-FVII, F374Y/L305V/M298Q-FVII, F374Y/L305V/V158T-FVII, F374Y/L305V/S314E-FVII, F374Y/K337A/S314E-FVII, F374Y/K337A/V158T-FVII, F374Y/K337A/M298Q-FVII, F374Y/K337A/E296V-FVII, F374Y/K337A/V158D-FVII, F374Y/V158D/S314E-FVII, F374Y/V158D/M298Q-FVII, F374Y/V158D/E296V-FVII, F374Y/V158T/S314E-FVII, F374Y/V158T/M298Q-FVII, F374Y/V158T/E296V-FVII, F374Y/E296V/S314E-FVII, F374Y/S314E/M298Q-FVII, F374Y/E296V/M298Q-FVII, F374Y/L305V/K337A/V158D-FVII, F374Y/L305V/K337A/E296V-FVII, F374Y/L305V/K337A/M298Q-FVII, F374Y/L305V/K337A/V158T-FVII, F374Y/L305V/K337A/S314E-FVII, F374Y/L305V/V158D/E296V-FVII, F374Y/L305V/V158D/M298Q-FVII, F374Y/L305V/V158D/S314E-FVII, F374Y/L305V/E296V/M298Q-FVII, F374Y/L305V/E296V/V158T-FVII, F374Y/L305V/E296V/S314E-FVII, F374Y/L305V/M298Q/V158T-FVII, F374Y/L305V/M298Q/S314E-FVII, F374Y/L305V/V158T/S314E-FVII, F374Y/K337A/S314E/V158T-FVII, F374Y/K337A/S314E/M298Q-FVII, F374Y/K337A/S314E/E296V-FVII, F374Y/K337A/S314E/V158D-FVII, F374Y/K337A/V158T/M298Q-FVII, F374Y/K337A/V158T/E296V-FVII, F374Y/K337A/M298Q/E296V-FVII, F374Y/K337A/M298Q/V158D-FVII, F374Y/K337A/E296V/V158D-FVII, F374Y/V158D/S314E/M298Q-FVII, F374Y/V158D/S314E/E296V-FVII, F374Y/V158D/M298Q/E296V-FVII, F374Y/V158T/S314E/E296V-FVII, F374Y/V158T/S314E/M298Q-FVII, F374Y/V158T/M298Q/E296V-FVII, F374Y/E296V/S314E/M298Q-FVII, F374Y/L305V/M298Q/K337A/5314E-FVII, F374Y/L305V/E296V/K337A/S314E-FVII, F374Y/E296V/M298Q/K337A/5314E-FVII, F374Y/L305V/E296V/M298Q/K337A-FVII, F374Y/L305V/E296V/M298Q/S314E-FVII, F374Y/V158D/E296V/M298Q/K337A-FVII, F374Y/V158D/E296V/M298Q/S314E-FVII, F374Y/L305V/V158D/K337A/S314E-FVII, F374Y/V158D/M298Q/K337A/S314E-FVII, F374Y/V158D/E296V/K337A/S314E-FVII, F374Y/L305V/V158D/E296V/M298Q-FVII, F374Y/L305V/V158D/M298Q/K337A-FVII, F374Y/L305V/V158D/E296V/K337A-FVII, F374Y/L305V/V158D/M298Q/S314E-FVII, F374Y/L305V/V158D/E296V/S314E-FVII, F374Y/V158T/E296V/M298Q/K337A-FVII, F374Y/V158T/E296V/M298Q/S314E-FVII, F374Y/L305V/V158T/K337A/S314E-FVII, F374Y/V158T/M298Q/K337A/S314E-FVII, F374Y/V158T/E296V/K337A/S314E-FVII, F374Y/L305V/V158T/E296V/M298Q-FVII, F374Y/L305V/V158T/M298Q/K337A-FVII, F374Y/L305V/V158T/E296V/K337A-FVII, F374Y/L305V/V158T/M298Q/S314E-FVII, F374Y/L305V/V158T/E296V/S314E-FVII, F374Y/E296V/M298Q/K337A/V158T/S314E-FVII, F374Y/V158D/E296V/M298Q/K337A/S314E-FVII, F374Y/L305V/V158D/E296V/M298Q/S314E-FVII, F374Y/L305V/E296V/M298Q/V158T/S314E-FVII, F374Y/L305V/E296V/M298Q/K337A/V158T-FVII, F374Y/L305V/E296V/K337A/V158T/S314E-FVII, F374Y/L305V/M298Q/K337A/V158T/S314E-FVII, F374Y/L305V/V158D/E296V/M298Q/K337A-FVII, F374Y/L305V/V158D/E296V/K337A/S314E-FVII, F374Y/L305V/V158D/M298Q/K337A/S314E-FVII, F374Y/L305V/E296V/M298Q/K337A/V158T/S314E-FVII, F374Y/L305V/V158D/E296V/M298Q/K337A/S314E-FVII, S52A-Factor VII, S60A-Factor VII; R152E-Factor VII, S344A-Factor VII, T106N-FVII, K143N/N145T-FVII, V253N-FVII, R290N/A292T-FVII, G291N-FVII, R315N/V317T-FVII, K143N/N145T/R315N/V317T-FVII; and FVII having substitutions, additions or deletions in the amino add sequence from 233Thr to 240Asn; FVII having substitutions, additions or deletions in the amino acid sequence from 304Arg to 329Cys; and FVII having substitutions, additions or deletions in the amino acid sequence from 153Ile to 223Arg.

Factor VII variants having substantially the same or improved biological activity relative to wild-type Factor VIIa encompass those that exhibit at least about 25%, such as, e.g., at least about 50%, at least about 75%, at least about 90%, at least about 120%, at least about 130% or at least about 150% of the specific activity of wild-type Factor VIIa that has been produced in the same cell type, when tested in one or more of a clotting assay, proteolysis assay, or TF binding assay as described above.

The biological activity of Factor VIIa in blood clotting derives from its ability to (i) bind to tissue factor (TF) and (ii) catalyze the proteolytic cleavage of Factor IX or Factor X to produce activated Factor IX or X (Factor IXa or Xa, respectively). For purposes of the invention, Factor VIIa biological activity may be quantified by measuring the ability of a preparation to promote blood clotting using Factor VII-deficient plasma and thromboplastin, as described, e.g., in <patcit id="pcit0048" dnum="US5997864A"><text>U.S. Patent No. 5,997,864</text></patcit>. In this assay, biological activity is expressed as the reduction in clotting time relative to a control sample and is converted to "Factor VII units" by comparison with a pooled human serum standard containing 1 unit/ml Factor VII activity. Alternatively, Factor VIIa biological activity may be quantified by (i) measuring the ability of Factor VIIa to produce of Factor Xa in a system comprising TF embedded in a lipid membrane and Factor X. (<nplcit id="ncit0010" npl-type="s"><text>Persson et al., J. Biol. Chem. 272:19919-19924, 1997</text></nplcit>); (ii) measuring Factor X hydrolysis in an aqueous system (see, <i>"General Methods"</i> below); (iii) measuring its physical binding to TF using an instrument based on surface plasmon resonance (<nplcit id="ncit0011" npl-type="s"><text>Persson, FEBS Letts. 413:359-363, 1997</text></nplcit>) (iv) measuring hydrolysis of a synthetic substrate (see, <i>"General Methods"</i> below); and (v) measuring generation of thrombin in a TF-independent in vitro system (see, <i>"General Methods"</i> below).

O-linked Glycosylation

In practicing the present invention, the pattern of oligosaccharides may be determined using any method known in the art, including, without limitation: high-performance liquid chromatography (HPLC); capillary electrophoresis (CE); nuclear magnetic resonance (NMR); mass spectrometry (MS) using ionization techniques such as fast-atom bombardment, electrospray, or matrix-assisted laser desorption (MALDI); gas chromatography (GC); and treatment with exoglycosidases in conjunction with anion-exchange (AIE)-HPLC, size-exclusion chromatography (SEC), or MS. See, e.g., <nplcit id="ncit0012" npl-type="s"><text>Weber et al., Anal. Biochem. 225:135 (1995</text></nplcit>); <nplcit id="ncit0013" npl-type="s"><text>Klausen et al., J. Chromatog. 718:195 (1995</text></nplcit>); <nplcit id="ncit0014" npl-type="b"><text>Morris et al., in Mass Spectrometry of Biological Materials, McEwen et al., eds., Marcel Dekker, (1990), pp 137-167</text></nplcit>; <nplcit id="ncit0015" npl-type="s"><text>Conboy et al., Biol. Mass Spectrom. 21:397, 1992</text></nplcit>; <nplcit id="ncit0016" npl-type="s"><text>Hellerqvist, Meth. Enzymol. 193:554 (1990</text></nplcit>);<nplcit id="ncit0017" npl-type="s"><text> Sutton et al., Anal. Biochem. 318:34 (1994</text></nplcit>); <nplcit id="ncit0018" npl-type="s"><text>Harvey et al., Organic Mass Spectrometry 29:752 (1994</text></nplcit>). A method for carrying out the O-glycosylation of small peptides is described in <patcit id="pcit0049" dnum="WO9518232A"><text>WO A 9518232</text></patcit>

The relative content of O-glycoforms can be determined, for example, by tryptic peptide mapping. In short, the Factor VII polypeptide is digested with trypsin and the polypeptides containing the O-glycosylation site are separated according to the glycan structure by RP-HPLC chromatography, mass spectrometry or another suitable analytical separation technique. If necessary in order to obtaining a suitable separation, the Factor VII polypeptide can prior to the digestion with trypsin be reduced and alkylated and the polypeptide chain containing the O-glycosylation site is purified by, e.g. RP-HPLC chromatography. Then the purified Factor VII polypeptide is subjected to tryptic digestion followed by analysis as described above.

Methods for producing Factor VII polypeptide preparations having a predetermined pattern of O-linked oligosaccharides

The origin of the acceptor Factor VII polypeptide is not a critical aspect of the invention. Typically, the Factor VII polypeptide will be expressed in a cultured prokaryote cell or eukaryote cell such as a mammalian, yeast insect, fungal or plant cell. The protein, however, may also be isolated from a natural source such as plasma, serum or blood. The Factor VII polypeptide can either be a full length protein or a fragment.

The invention provides compositions that include human <i>Factor VII polypeptide</i> species that have a substantially uniform glycosylation pattern. The methods are useful for remodelling or altering the glycosylation pattern present on a <i>Factor VII polypeptide</i> upon its initial expression. Thus, the methods of the invention provide a practical means for large-scale preparation of glycoforms having pre-selected or pre-determined uniform derivatization patterns. The methods are particularly well suited for modification of therapeutic peptides, including but not limited to, <i>Factor VII polypeptides</i> that are incompletely glycosylated during production in cell culture cells or transgenic animals. However, the compositions of the invention may also be prepared by purification of natural sources, such as plasma, serum or blood, or cell culture fluids and isolating the desired glycoforms therefrom.

The Factor VII polypeptides to be re-modelled in accordance with the invention are typically prepared by cell culture processes. Suitable host cells include, without limitation, human cells expressing an endogenous gene such as, e.g., a Factor VII gene. In these cells, the endogenous gene may be intact or may have been modified <i>in situ</i>, or a sequence outside the endogenous gene may have been modified <i>in situ</i> to alter the expression of the endogenous Factor VII polypeptide gene. Any human cell capable of expressing an endogenous Factor VII polypeptide gene may be used. Other, included host cells are heterologous host cells programmed to express a Factor VII polypeptide, from a recombinant gene. The host cells may be vertebrate, insect, or fungal cells. Preferably, the cells are mammalian cells capable of the entire spectrum of mammalian N-linked glycosylation; O-linked glycosylation; and γ-carboxylation. See, e.g., <patcit id="pcit0050" dnum="US4784950A"><text>U.S. Patent Nos. 4,784,950</text></patcit>. Preferred mammalian cell lines include the CHO (ATCC CCL 61), COS-1 (ATCC CRL 1650), baby hamster kidney (BHK) and HEK293 (ATCC CRL 1573; <nplcit id="ncit0019" npl-type="s"><text>Graham et al., J. Gen. Virol. 36:59-72, 1977</text></nplcit>) cell lines. A preferred BHK cell line is the tk^- ts13 BHK cell line (<nplcit id="ncit0020" npl-type="s"><text>Waechter and Baserga, Proc.Natl.Acad.Sci.USA 79:1106-1110, 1982</text></nplcit>), hereinafter referred to as BHK 570 cells. The BHK 570 cell line is available from the American Type Culture Collection, 12301 Parklawn Dr., Rockville, MD 20852, under ATCC accession number CRL 10314. A tk^- ts13 BHK cell line is also available from the ATCC under accession number CRL 1632. In addition, a number of other cell lines may be used, including Rat Hep I (Rat hepatoma; ATCC CRL 1600), Rat Hep II (Rat hepatoma; ATCC CRL 1548), TCMK (ATCC CCL 139), Human lung (ATCC HB 8065), NCTC 1469 (ATCC CCL 9.1) and DUKX cells (CHO cell line) (<nplcit id="ncit0021" npl-type="s"><text>Urlaub and Chasin. Proc. Natl. Acad. Sd. USA 77:4216-4220, 1980</text></nplcit>). (DUKX cells also referred to as CXB11 cells), and DG44 (CHO cell line) (<nplcit id="ncit0022" npl-type="s"><text>Cell, 33:405, 1983</text></nplcit>, and <nplcit id="ncit0023" npl-type="s"><text>Somatic Cell and Molecular Genetics 12:555, 1986</text></nplcit>). Also useful are 3T3 cells, Namalwa cells, myelomas and fusions of myelomas with other cells. Suitable host cells include BHK 21 cells that have been adapted to grow in the absence of serum and have been programmed to express Factor VII. The cells may be mutant or recombinant cells that express a qualitatively or quantitatively different spectrum of glycosylation enzymes (such as, e.g., glycosyl transferases and/or glycosidases) than the cell type from which they were derived. The cells may also be programmed to express other heterologous peptides or proteins, including, e.g., truncated forms of Factor VII. The host cells may also be CHO cells that have been programmed to co-express both the Factor VII polypeptide of interest (i.e., Factor VII or a Factor-VII-related polypeptide) and another heterologous peptide or polypeptide such as, e.g., a modifying enzyme or a Factor VII fragment.

<u>Methods</u>: The present invention encompasses methods for producing a preparation comprising a predetermined serine-linked glycoform pattern as described above. Also described herein are methods for optimizing the glycoform distribution of a <i>Factor VII polypeptide</i> (see <figref idref="f0002">Figure 3</figref>). The individual process steps described can be applied in different combinations in order to obtain the desired glycoform pattern. Non-limiting reference examples are given below.

In one aspect, these methods are carried out by the steps of: <ol id="ol0001" compact="compact" ol-style=""><li>(a) obtaining a preparation of a Factor VII polypeptide containing a Cys-X1-Ser-X2-Pro-Cys motif and wherein said serine forms part of a Glc-O-Ser covalent bond from a cell in which it is prepared; e.g., from an engineered cell (cell culture) or by isolating the Factor VII polypeptide from a natural source;</li><li>(b) contacting the Factor VII polypeptide preparation with an activated donor of the desired mono- or oligosaccharide moiety and an enzyme suitable for transferring the desired mono- or oligo-saccharide group under conditions appropriate for transferring the mono- or oligo-saccharide group from the donor moiety to the acceptor moiety, thereby producing the Factor VII polypeptide having an altered glycosylation pattern.</li></ol>

In another aspect, these methods are carried out by the steps of: <ul id="ul0002" list-style="none" compact="compact"><li>(aa) obtaining a preparation of a Factor VII polypeptide containing a Cys-X1-Ser-X2-Pro-Cys motif and wherein said serine forms part of a Glc-O-Ser covalent bond from a cell in which it is prepared; e.g., from an engineered cell (cell culture) or by isolating the Factor VII polypeptide from a natural source;</li><li>(bb) contacting the Factor VII polypeptide preparation with an enzyme suitable for removing the terminal mono- or oligo-saccharide group under conditions appropriate for removing said mono- or oligo-saccharide group, thereby producing the Factor VII polypeptide having an altered glycosylation pattern.</li></ul>

The methods described herein may comprise a combination of steps (b) and (bb). The methods further comprise a step of isolating the Factor VII polypeptide having an altered glycosylation pattern.

The methods described herein may comprise a further step of:

Analyzing the structure of the oligosaccharides linked to the Factor VII polypeptides to determine a glycoform pattern, and optionally, repeating steps (b) and/or (bb) until a desired glycoform pattern is achieved.

These methods may further comprise the step of subjecting preparations having predetermined glycoform patterns to at least one test of bioactivity (including, e.g., clotting, Factor X proteolysis, or TF binding) or other functionality (such as, e.g., pharmacokinetic profile or stability), and correlating particular glycoform patterns with particular bioactivity or functionality profiles in order to identify a desired glycoform pattern.

The desired glycoform pattern is a substantially uniform <i>xylose-xylose-glucose-O-serine glycosylation</i>: in this embodiment, the method <i>(METHOD A2</i>) comprises the steps of: <ol id="ol0002" compact="compact" ol-style=""><li>(a) obtaining a preparation of a Factor VII polypeptide containing a Cys-X1-Ser-X2-Pro-Cys motif and wherein said serine is located at position 52 and forms part of a Glc-O-Ser covalent bond; e.g., from an engineered cell (cell culture) or by isolating the Factor VII polypeptide from a natural source;</li><li>(b) contacting the preparation obtained in step (a) with UDP-D-xylose: β-D-glucoside α-1,3-D-xylosyltransferase and an activated xylosyl donor under conditions appropriate for transferring a xylose residue from the xylose donor moiety to the acceptor moiety, thereby producing the Factor VII polypeptide having an altered glycosylation pattern.</li><li>(c) contacting the preparation obtained in step (b) with UDP-D-xylose: α-D-xyloside α-1,3-xylosyltransferase and an activated xylosyl donor under conditions appropriate for transferring a xylose residue from the xylose donor moiety to the acceptor moiety, thereby producing the Factor VII polypeptide having an altered glycosylation pattern.</li></ol>

In one embodiment, the method further includes the step of isolating the preparation obtained in step (b) prior to subjecting the preparation to step (c).

In one embodiment, the method further includes the step of isolating the Factor VII polypeptide prepared in step (c) having a Xyl-Xyl-Glc-O-Ser glycosylation.

In one embodiment, the method further includes the step of analysing the structure of the oligosaccharides linked to the Factor VII polypeptides to determine a glycoform pattern, and, optionally, repeating step (b) and/or step (c) until the desired glycoform pattern is achieved.

The Factor VII polypeptide exhibits substantially uniform Xyl-Xyl-Glc-O-Ser glycosylation; Ser being the serine of the contained Cys-X1-Ser-X2-Pro-Cys motif (X1 and X2 independently being any amino acid residue).

In preferred embodiments, the Factor VII polypeptide preparation is a human FVII polypeptide preparation selected from the list of: <ul id="ul0003" list-style="none" compact="compact"><li>Factor VII polypeptides exhibiting substantially uniform Xyl-Xyl-Glc-O-Ser52 glycosylation,</li><li>Factor VII-related polypeptides exhibiting substantially uniform Xyl-Xyl-Glc-O-Ser52 glycosylation</li><li>Factor VII variants exhibiting substantially uniform Xyl-Xyl-Glc-O-Ser52 glycosylation</li></ul>

It is to be understood that oligosaccharides such as Xyl-Xyl- may also be transferred to the acceptor Glc-O-Ser/Thr moiety by using a suitable transferring enzyme and an activated Xyl-Xyl- donor.

<i>Chromatographic method</i>: Described herein are hydrophobic interaction chromatographic methods for producing a preparation comprising a predetermined serine-linked glycoform pattern as described above, and for purifying a O-glycosylated Factor VII polypeptide having a desired glycoform pattern from a composition comprising said Factor VII polypeptide and Factor VII polypeptides having unwanted glycoform patterns.

In one aspect, the method comprises the following steps: <ol id="ol0003" compact="compact" ol-style=""><li>(a) obtaining a preparation of a Factor VII polypeptide containing a Cys-X1-Ser-X2-Pro-Cys motif and wherein said serine forms part of a Gic-O-Ser covalent bond from a cell in which it is prepared; e.g., from an engineered cell (cell culture) or by isolating the Factor VII polypeptide from a natural source;</li><li>(b) binding the Factor VII polypeptide to an hydrophic interaction material using using a solution comprising water, optionally a salt component, and optionally a buffer,</li><li>(c) optionally washing the hydrophobic interaction material using a solution comprising water, optionally a salt component, and optionally a buffer so as to elute contaminants from the hydrophobic interaction material;</li><li>(d) washing the hydrophobic interaction material using a solution comprising an organic modifier, water, optionally a salt component, and optionally a buffer, at a linear or step gradient or isocratically in salt component so as to separate Factor VII polypeptide having a desired glycoform pattern from Factor VII polypeptide not having the desired glycoform from the hydrophobic interaction material;</li><li>(e) collecting the fraction containing the Factor VII polypeptide having the desired glycoform pattern.</li></ol>

The above-described methods may further include the step of repeating steps (a) to e) by subjecting the preparation obtained in step (e) to steps (a) to (e). This further step may be repeated more than once if deemed necessary.

It is to be understood that the preparations described herein may also be prepared by a process comprising a combination of purification steps whereby Factor VII polypeptide species having the desired glycosylation are captured from the cell culture liquid or natural source of origin and the above-described enzymatic methods.

The above-described methods may further comprise the step of subjecting preparations having predetermined glycoform patterns to at least one test of bioactivity (including, e.g., clotting, Factor X proteolysis, or TF binding) or other functionality (such as, e.g., pharmacokinetic profile or stability), and correlating particular glycoform patterns with particular bioactivity or functionality profiles in order to identify a desired glycoform pattern.

Further enzymatic treatments may be used in connection with the above methods to modify the oligosaccharide pattern of N- or O-linked glycans of a preparation; such treatments include, without limitation, treatment with one or more of sialidase (neuraminidase), galactosidase, fucosidase; galactosyl transferase, fucosyl transferases, and/or sialyltransferase, in a sequence and under conditions that achieve a desired modification in the distribution of oligosaccharide chains having particular terminal structures. Glycosyl transferases are commercially available from Calbiochem (La Jolla, CA) and glycosidases are commercially available from Glyko, Inc., (Novato, CA).

Factor VII polypeptide preparations

As used herein, a "Factor VII polypeptide preparation" refers to a plurality of Factor VII glycoforms that have been separated from the cell in which they were synthesized. The Factor VII polypeptide preparation is a human Factor VII polypeptide preparation and includes activated forms, functionally related polypeptides such as, e.g., variants and chemically modified forms, that have been separated from the cell in which they were synthesized.

For example, as used herein, a "Factor VII preparation" refers to a plurality of Factor VII polypeptides, Factor VIIa polypeptides, or Factor VII-related polypeptides, including variants and chemically modified forms, that have been separated from the cell in which they were synthesized or isolated from a natural source.

Separation of Factor VII polypeptides from their cell of origin may be achieved by any method known in the art, including, without limitation, removal of cell culture medium containing the desired product from an adherent cell culture; centrifugation or filtration to remove non-adherent cells; and the like.

Optionally, the polypeptides may be further purified. Purification may be achieved using any method known in the art, including, without limitation, affinity chromatography, such as, e.g., on an anti-Factor VII (see, e.g., <nplcit id="ncit0024" npl-type="s"><text>Wakabayashi et al., J. Biol. Chem. 261:11097, 1986</text></nplcit>; and <nplcit id="ncit0025" npl-type="s"><text>Thim et al., Biochem. 27:7785, 1988</text></nplcit>); hydrophobic interaction chromatography; ion-exchange chromatography; size exclusion chromatography; electrophoretic procedures (<i>e.g.</i>, preparative isoelectric focusing (IEF), differential solubility (e.g., ammonium sulfate precipitation), or extraction and the like. See, generally, <nplcit id="ncit0026" npl-type="b"><text>Scopes, Protein Purification, Springer-Verlag, New York, 1982</text></nplcit>; and <nplcit id="ncit0027" npl-type="b"><text>Protein Purification, J.-C. Janson and Lars Ryden, editors, VCH Publishers, New York, 1989</text></nplcit>. Following purification, the preparation preferably contains less than about 10% by weight, more preferably less than about 5% and most preferably less than about 1%, of non-related proteins derived from the host cell.

Factor VII and Factor VII-related polypeptides respectively, may be activated by proteolytic cleavage, using Factor XIIa or other proteases having trypsin-like specificity, such as, e.g., Factor IXa, kallikrein, Factor Xa, and thrombin. See, e.g., <nplcit id="ncit0028" npl-type="s"><text>Osterud et al., Biochem. 11:2853 (1972</text></nplcit>); <patcit id="pcit0051" dnum="US4456591A"><text>Thomas, U.S. Patent No. 4,456,591</text></patcit>; and <nplcit id="ncit0029" npl-type="s"><text>Hedner et al., J. Clin. Invest. 71:1836 (1983</text></nplcit>). Alternatively, Factor VII, may be activated by passing it through an ion-exchange chromatography column, such as Mono Q® (Pharmada) or the like. The resulting activated Factor VII polypetide may then be formulated and administered as described below.

Functional properties of Factor VII polypeptide preparations

The preparations of Factor VII polypeptides having predetermined oligosaccharide patterns described herein (including Factor VII polypeptides, Factor VII-related polypeptides exhibit improved functional properties relative to reference preparations. The improved functional properties may include, without limitation, a) physical properties such as, e.g., storage stability; b) pharmacokinetic properties such as, e.g., bioavailability and half-life; c) immunogenicity in humans, and d) biological activity, such as, e.g., clotting activity.

A reference preparation refers to a preparation comprising a polypeptide that is identical to that contained in the preparation of the invention to which it is being compared (such as, e.g., wild-type Factor VII or a particular variant or chemically modified form) except for exhibiting a different pattern of serine linked glycosylation.

Storage stability of a Factor VII polypeptide preparation may be assessed by measuring (a) the time required for 20% of the bioactivity of a preparation to decay when stored as a dry powder at 25°C and/or (b) the time required for a doubling in the proportion of (e.g., Factor VIIa) aggregates of said Factor VII polypeptide in the preparation.

In some embodiments, the preparations exhibit an increase of at least about 30%, preferably at least about 60% and more preferably at least about 100%, in the time required for 20% of the bioactivity to decay relative to the time required for the same phenomenon in a reference preparation, when both preparations are stored as dry powders at 25°C. Bioactivity measurements may be performed using any of a clotting assay, proteolysis assay, TF-binding assay, or TF-independent thrombin generation assay.

In some embodiments, the preparations exhibit an increase of at least about 30%, preferably at least about 60%, and more preferably at least about 100%, in the time required for doubling of aggregates relative to a reference preparation, when both preparations are stored as dry powders at 25°C. The contents of aggregates may be determined according to methods known to the skilled person, such as, e.g., gel permeation HPLC methods. For example, the content of Factor VII aggregates is determined by gel permeation HPLC on a Protein Pak 300 SW column (7.5 x 300 mm) (Waters, 80013) as follows. The column is equilibrated with Eluent A (0.2 M ammonium sulfate, 5 % isopropanol, pH adjusted to 2.5 with phosphoric add, and thereafter pH is adjusted to 7.0 with triethylamine), after which 25 µg of sample is applied to the column. Elution is with Eluent A at a flow rate of 0.5 ml/min for 30 min, and detection is achieved by measuring absorbance at 215 nm. The content of aggregates is calculated as the peak area of the Factor VII aggregates/total area of Factor VII peaks (monomer and aggregates).

"Bioavailability" refers to the proportion of an administered dose of a (e.g., Factor VII or Factor VII-related) Factor VII polypeptide preparation that can be detected in plasma at predetermined times after administration. Typically, bioavailability is measured in test animals by administering a dose of between about 25-250 µg/kg of the preparation; obtaining plasma samples at predetermined times after administration; and determining the content of (e.g., Factor VII or Factor VII-related) glycosylated Factor VII polypeptides in the samples using one or more of a clotting assay (or any bioassay), an immunoassay, or an equivalent. The data are typically displayed graphically as Factor VII polypeptide v. time and the bioavailability is expressed as the area under the curve (AUC). Relative bioavailability of a test preparation refers to the ratio between the AUC of the test preparation and that of the reference preparation.

In some embodiments, the preparations exhibit a relative bioavailability of at least about 110%, preferably at least about 120%, more preferably at least about 130% and most preferably at least about 140% of the bioavailability of a reference preparation. The bioavailability may be measured in any mammalian species, preferably dogs, and the predetermined times used for calculating AUC may encompass different increments from 10 min- 8 h.

"Half-life" refers to the time required for the plasma concentration of (e.g., Factor VII polypeptides of Factor VII-related polypeptides) the Factor VII polypeptide to decrease from a particular value to half of that value. Half-life may be determined using the same procedure as for bioavailability. In some embodiments, the preparations exhibit an increase in half-life of at least about 0.25 h, preferably at least about 0.5 h, more preferably at least about 1 h, and most preferably at least about 2 h, relative to the half-life of a reference preparation.

"Immunogenicity" of a preparation refers to the ability of the preparation, when administered to a human, to elicit a deleterious immune response, whether humoral, cellular, or both. Factor VIIa polypeptides and Factor VIIa-related polypeptides are not known to elicit detectable immune responses in humans. Nonetheless, in any human sub-population, there may exist individuals who exhibit sensitivity to particular administered proteins. Immunogenicity may be measured by quantifying the presence of anti-Factor VII antibodies and/or Factor VII-responsive T-cells in a sensitive individual, using conventional methods known in the art. In some embodiments, the preparations exhibit a decrease in immunogenicity in a sensitive individual of at least about 10%, preferably at least about 25%, more preferably at least about 40% and most preferably at least about 50%, relative to the immunogenicity for that individual of a reference preparation.

Pharmaceutical compositions and Methods of Use

The preparations may be used to treat any syndrome responsive to the relevant glycoprotein. Factor VII-, FIX and FX-responsive syndromes, respectively, include syndromes such as, e.g., bleeding disorders, including, without limitation, those caused by clotting factor deficiencies (e.g., haemophilia A and B or deficiency of coagulation factors XI or VII); by thrombocytopenia or von Willebrand's disease, or by clotting factor inhibitors, or excessive bleeding from any cause. The preparations may also be administered to patients in association with surgery or other trauma or to patients receiving anticoagulant therapy.

Pharmaceutical compositions comprising the Factor VII and Factor VII-related preparations according to the present are primarily intended for parenteral administration for prophylactic and/or therapeutic treatment. Preferably, the pharmaceutical compositions are administered parenterally, i.e., intravenously, subcutaneously, or intramuscularly. They may be administered by continuous or pulsatile infusion.

Pharmaceutical compositions or formulations according to the invention comprise a preparation in combination with, preferably dissolved in, a pharmaceutically acceptable carrier, preferably an aqueous carrier or diluent. A variety of aqueous carriers may be used, such as water, buffered water, 0.4% saline, 0.3% glycine and the like. The preparations of the invention can also be formulated into liposome preparations for delivery or targeting to the sites of injury. Liposome preparations are generally described in, e.g., <patcit id="pcit0052" dnum="US4837028A"><text>U.S. Patents Nos. 4,837,028</text></patcit>, <patcit id="pcit0053" dnum="US4501728A"><text>4,501,728</text></patcit>, and <patcit id="pcit0054" dnum="US4975282A"><text>4,975,282</text></patcit>. The compositions may be sterilised by conventional, well-known sterilisation techniques. The resulting aqueous solutions may be packaged for use or filtered under aseptic conditions and lyophilised, the lyophilised preparation being combined with a sterile aqueous solution prior to administration.

The compositions may contain pharmaceutically acceptable auxiliary substances or adjuvants, including, without limitation, pH adjusting and buffering agents and/or tonicity adjusting agents, such as, for example, sodium acetate, sodium lactate, sodium chloride, potassium chloride, calcium chloride, etc.

The concentration of Factor VII or Factor VII-related polypeptides in these formulations can vary widely, i.e., from less than about 0.5% by weight, usually at or at least about 1% by weight to as much as 15 or 20% by weight and will be selected primarily by fluid volumes, viscosities, etc., in accordance with the particular mode of administration selected.

Thus, a typical pharmaceutical composition for intravenous infusion could be made up to contain 250 ml of sterile Ringer's solution and 10 mg of the preparation. Actual methods for preparing parenterally administrable compositions will be known or apparent to those skilled in the art and are described in more detail in, for example, <nplcit id="ncit0030" npl-type="b"><text>Remington's Pharmaceutical Sciences, 18th ed., Mack Publishing Company, Easton, PA (1990</text></nplcit>).

The compositions of the present invention can be administered for prophylactic and/or therapeutic treatments. In therapeutic applications, compositions are administered to a subject already suffering from a disease, as described above, in an amount sufficient to cure, alleviate or partially arrest the disease and its complications. An amount adequate to accomplish this is defined as "therapeutically effective amount". Effective amounts for each purpose will depend on the severity of the disease or injury as well as the weight and general state of the subject. In general, however, the effective amount will range from about 0.05 mg up to about 500 mg of the preparation per day for a 70 kg subject, with dosages of from about 1.0 mg to about 200 mg of the preparation per day being more commonly used. It will be understood that determining an appropriate dosage may be achieved using routine experimentation, by constructing a matrix of values and testing different points in the matrix.

Local delivery of the preparations, such as, for example, topical application, may be carried out, e.g., by means of a spray, perfusion, double balloon catheters, stents, incorporated into vascular grafts or stents, hydrogels used to coat balloon catheters, or other well established methods. In any event, the pharmaceutical compositions should provide a quantity of the preparation sufficient to effectively treat the subject.

The pharmaceutical compositions of the invention may further comprise other bioactive agents, such as, e.g., non-Factor VII-related coagulants or anticoagulants.

EXPERIMENTALS

General methods

α-xylosidase assay

The α-xylosidase assays are conducted in an appropriate buffer, e.g. 50 mM sodium acetate, pH 4.5, containing a suitable substrate, e.g. the O-glycopeptides that can be obtained from the O-glycopeptide map of the relevant glycoprotein (e.g., rFVIIa). The reaction is stopped after an appropriate time that can be determined experimentally, by e.g. addition of trifluoroacetic acid, and the assay mixtures are analysed by HPLC.

α-xylosyltransferase assay

The α-xytosyltransferase assays are conducted in an appropriate buffer, e.g. 10 mM Hepes, pH 7.2, 0.1% Triton X-100, 0.5 mM UDP-Xylose (Sigma U5875), containing a suitable substrate, e.g. the O-glycopeptides that can be obtained from the O-glycopeptide map of the relevant glycoprotein (e.g., rFVIIa) or the pyridyl-aminated oligosaccharides prepared as described in Minamida et al. (<nplcit id="ncit0031" npl-type="s"><text>Minamida et.al., Detection of UDP-D-xylose: α-D-xyloside α1-3xylosyltransferase activity in human hepatoma cell line Hep2. J. Biochem. 120 1002-1006, 1996</text></nplcit>). The reaction is stopped after an appropriate time, that can be determined experimentally, by e.g. addition of trifluoroacetic acid, and the assay mixtures are analysed by HPLC.

The α-xylosidase and α-xylosyltransferase assays are optimized for time and, optionally for temperature and pH.

O-glucosyltransferase assay.

The O-glucosyltransferase assays are conducted, e.g., as described by <nplcit id="ncit0032" npl-type="s"><text>Shao et al. (Glycobiology 12(11) 763-770 2002</text></nplcit>).

Factor VII assays

A suitable assay for testing for factor VIIa activity and thereby selecting suitable factor VIIa variants can be performed as a simple preliminary <i>in vitro</i> test. The assay is also suitable for selecting suitable factor VIIa variants.

In Vitro

Hydrolysis Assay

Native (wild-type) factor VIIa and factor VIIa variant (both hereafter referred to as "factors VIIa") may be assayed for specific activities. They may also be assayed in parallel to directly compare their specific activities. The assay is carried out in a microtiter plate (MaxiSorp, Nunc, Denmark). The chromogenic substrate D-Ile-Pro-Arg-p-nitroanilide (S-2288, Chromogenix, Sweden), final concentration 1 mM, is added to factor VIIa (final concentration 100 nM) in 50 mM Hepes, pH 7.4, containing 0.1 M NaCl, 5 mM CaCl₂ and 1 mg/ml bovine serum albumin. The absorbance at 405 nm is measured continuously in a SpectraMax™ 340 plate reader (Molecular Devices, USA). The absorbance developed during a 20-minute incubation, after subtraction of the absorbance in a blank well containing no enzyme, is used to calculate the ratio between the activities of variant and wild-type factor VIIa: <maths id="math0001" num=""><math display="block"><mi>Ratio</mi><mo>=</mo><mfenced separators=""><msub><mi mathvariant="normal">A</mi><mrow><mn>405</mn><mspace width="1ex" /><mi>nm</mi></mrow></msub><mspace width="1ex" /><mi>factor</mi><mspace width="1ex" /><mi>VIIa</mi><mspace width="1ex" /><mi>variant</mi></mfenced><mo>/</mo><mfenced separators=""><msub><mi mathvariant="normal">A</mi><mrow><mn>405</mn><mspace width="1ex" /><mi>nm</mi></mrow></msub><mspace width="1ex" /><mi>factor</mi><mspace width="1ex" /><mi>VIIa</mi><mspace width="1ex" /><mi>wild-type</mi></mfenced><mo>.</mo></math><img file="EP1745141B2_D0001.tif" /></maths>

Based thereon, factor VIIa variants with an activity comparable to or higher than native factor VIIa may be identified, such as, for example, variants where the ratio between the activity of the variant and the activity of native factor VII (wild-type FVII) is around, versus above 1.0.

The activity of factor VIIa or factor VIIa variants may also be measured using a physiological substrate such as factor X, suitably at a concentration of 100-1000 nM, where the factor Xa generated is measured after the addition of a suitable chromogenic substrate (eg. S-2765). In addition, the activity assay may be run at physiological temperature.

In Vitro Proteolysis Assay

Native (wild-type) Factor VIIa and Factor VIIa variant. (both hereafter referred to as "Factor VII") are assayed in parallel to directly compare their specific activities. The assay is carried out in a microtiter plate (MaxiSorp, Nunc, Denmark). Factor VIIa (10 nM) and Factor X (0.8 microM) in 100 microL 50 mM Hepes, pH 7.4, containing 0.1 M NaCl, 5 mM CaCl2 and 1 mg/ml bovine serum albumin, are incubated for 15 min. Factor X cleavage is then stopped by the addition of 50 microL 50 mM Hepes, pH 7.4, containing 0.1 M NaCl, 20 mM EDTA and 1 mg/ml bovine serum albumin. The amount of Factor Xa generated is measured by addition of the chromogenic substrate Z-D-Arg-Gly-Arg-p-nitroanilide (S-2765, Chromogenix, Sweden), final concentration 0.5 mM. The absorbance at 405 nm is measured continuously in a SpectraMax™ 340 plate reader (Molecular Devices, USA). The absorbance developed during 10 minutes, after subtraction of the absorbance in a blank well containing no FVIIa, is used to calculate the ratio between the proteolytic activities of variant and wild-type Factor VIIa: <maths id="math0002" num=""><math display="block"><mi>Ratio</mi><mo>=</mo><mfenced separators=""><mi mathvariant="normal">A</mi><mn>405</mn><mspace width="1ex" /><mi>nm</mi><mspace width="1ex" /><mi>Factor</mi><mspace width="1ex" /><mi>VIIa</mi><mspace width="1ex" /><mi>variant</mi></mfenced><mo>/</mo><mfenced separators=""><mi mathvariant="normal">A</mi><mn>405</mn><mspace width="1ex" /><mi>nm</mi><mspace width="1ex" /><mi>Factor</mi><mspace width="1ex" /><mi>VIIa</mi><mspace width="1ex" /><mi>wild-type</mi></mfenced><mo>.</mo></math><img file="EP1745141B2_D0002.tif" /></maths>

Based thereon, factor VIIa variants with an activity comparable to or higher than native factor VIIa may be identified, such as, for example, variants where the ratio between the activity of the variant and the activity of native factor VII (wild-type FVII) is around 1, versus above 1.0.

Thrombin generation assay:

The ability of factor VII or factor VII-related polypeptides (e.g., variants) to generate thrombin can be measured in an assay comprising all relevant coagulation factors and inhibitors at physiological concentrations and activated platelets (as described on p. 543 in <nplcit id="ncit0033" npl-type="s"><text>Monroe et al. (1997) Brit. J. Haematol. 99, 542-547</text></nplcit> which is hereby incorporated as reference).

Clot assays.

1st generation assay

The activity of the Factor VII polypeptides may also be measured using a one-stage clot assay essentially as described in <patcit id="pcit0055" dnum="WO9215686A"><text>WO 92/15686</text></patcit> or <patcit id="pcit0056" dnum="US5997864A"><text>US 5,997,864</text></patcit>. Briefly, the sample to be tested is diluted in 50 mM Tris (pH 7.5), 0.1% BSA and 100 µL is incubated with 100 µL of Factor VII deficient plasma and 200 µL of thromboplastin C containing 10 mM Ca2+. Clotting times are measured and compared to a standard curve using a reference standard or a pool of citrated normal human plasma in serial dilution.

2nd generation assay:

Essentially the same, except that recombinant human tissue factor is used instead for thromboplastin C.

Factor IX assay

Test for factor IX activity:

Suitable assays for testing for factor IX activity, and thereby providing means for selecting suitable factor IX variants for use in the present invention, can be performed as simple <i>in vitro</i> tests as described, for example, in <nplcit id="ncit0034" npl-type="s"><text>Wagenvoord et al., Haemostasis 1990;20(5):276-88 </text></nplcit>

Factor IX biological activity may also be quantified by measuring the ability of a preparation to correct the clotting time of factor IX-deficient plasma, e.g., as described in Nilsson et al., 1959.(<nplcit id="ncit0035" npl-type="s"><text>Nilsson IM, Blombaeck M, Thilen A, von Francken I., Carriers of haemophilia A - A laboratory study, Acta Med Scan 1959; 165:357</text></nplcit>). In this assay, biological activity is expressed as units/ml plasma (1 unit corresponds to the amount of FIX present in normal pooled plasma.

Examples

The following examples are intended as non-limiting illustrations of the present invention.

Example 1

Preparation of α-xylosidase by extraction and purification

The enzyme, α-xylosidase, can be prepared from various sources, e.g. from plant material as described by <nplcit id="ncit0036" npl-type="s"><text>Monroe et al. (Plant Physiology and Biochemistry 41:877-885 (2003</text></nplcit>)). For example, plant tissues from e.g. Arabidopsis thaliana are ground in a mortar and pestle with quartz sand in two volumes of Buffer A (40 mM Hepes, pH 7.0, 1 M NaCl), and the filtered extract is centrifuged at 15000 x g for 15 min. Ammonium sulfate is added to for example 80% saturation. Precipated proteins are collected by centrifugation at 15000 x g for 15 min and redissolved in Buffer A. The α -xylosidase is purified by chromatography, for example on a Concanavalin A-Sepharose column, on an anion-exchange column and/or on other chromatographic columns known for the skilled person. Fractions are collected during elution and the fractions containing the α-xylosidase enzyme are identified by use of the α-xylosidase assay.

Example 2

Preparation of α-xylosidase by cloning and expression in E. coli and purification

Genes encoding α-xylosidases, which can hydrolyse alpha xylosidic bonds, have been cloned and characterized previously and genes showing significant homology to characterized α-xylosidases have been annotated in the genomes from several prokaryotic and eukaryotic organisms. The gene sequences are available in databases such as SWISS-PROT or NCBI and can be amplified by PCR from genomic DNA from the respective organisms. Several candidates were chosen for cloning and expression in <i>E. coli</i> after searching protein databases for the presence of α-xylosidase proteins. The following candidate genes were selected on the basis of already existing annotation in databases (A), previous published characterization(P) or based on homology analysis to known α-xylosidases(H): gene tm0308 (Thermotoga maritima: A); gene bt3085 (2139 bp) and gene bt3659(2475 bp) (Bacteroides thetaiotaomicron: A); gene bf0551 (2238 bp) and gene bf1247 (2538 bp) (Bacteroides fragilis: A); gene b102681 (2310 bp)(Bacillus licheniformis: H); gene bh1905 (2328 bp)(Badllus halodurans: H), gene xylS (2196 bp)(Sulfolobus solfataricus: P); gene yicl (2319 bp) (Escherichia coli: P)

Strategy for cloning and expression of α-xylosidases in E. coli

The SignalP software (<nplcit id="ncit0037" npl-type="s"><text>Bendtsen,J.D.et al..J. Mol. Biol., 340:783-795, 2004</text></nplcit>) is used to evaluate whether a signal peptide is potentially present in the N-terminal of the candidate enzymes. BF0551, BF1247, BT3085, BT3659 are presumably secreted as indicated by a strong prediction of a signalpeptidase I cleavage site. A methionine codon encoding a start-methionine is included in front of the first amino acid following the predicted cleavage site.

Purified genomic DNA from <i>Bacteroides thetaiotaomicron</i> (ATCC 29148D), <i>Bacteroides fragilis</i> (ATCC 25285D), Bacillus haludurans(ATCC 21591 D&amp;BAA-125D), Sulfolobus solfataricus (ATCC 35092D), Thermotoga maritima (ATCC 43589D) is obtained from American Type Culture Collection. In case of E. coli (strain K-12 derivative) and Bacillus licheniformis (ATCC 28450), genomic DNA is prepared from bacterial cells cultivated overnight in LB medium using the DNeasy tissue kit (Qiagen) according to the manufactures instructions.

Forward and reverse primers for PCR amplification are designed with an extension in the 5'-ends comprising the restriction enzyme cleavage sites NdeI(or XbaI) and XmaI, respectively. PCR is performed using the following condition:1) 95°C for 3 min: denaturation, 2) 94°C for 30 sec: denaturation, 3) 55°C or 60°C for 30 sec: annealing, 4) 72°C for 2 min: elongation. Step 2-4 is repeated for 15 cycles. PCR products are separated on 1% ethidium bromide agarose gels and bands showing the correct predicted sizes are excised from the gels and purified using the GFX DNA purification kit (Amersham Pharmacia). Purified PCR products are cloned into the pCR2.1TOPO vector according to the instructions of the manufacturer (Invitrogen). Clones showing the correct restriction enzyme cleavage profile are sequenced to evaluate the DNA sequence. The insert representing the α-xylosidase genes are released from the pCR2.1TOPO vector using the relevant restriction enzymes. A pET11a E. coli expression vector (Novagen) containing a NdeI (and XbaI) and a XmaI site is cleaved with relevant restriction enzymes and the vector part is purified as described for the PCR products. Vector and inserts are ligated together using the Rapid Ligation Kit (Roche) according to the manufacturer's instructions.

Ligation products are transformed into E.coli TOP10 (Invitrogen) cells by means of chemical transformation or heat shock methods known to the skilled persons. Cells are plated on LB/ampecillin(Amp)-medium culture plates overnight. Single colonies are selected from plates and grown overnight in LB/Amp medium. Purified pET plasmids from each colony are screened for the presence of correct inserts using restriction cleavage enzymes and evaluation of sizes of released inserts.

E. Coli Rosetta DE3 (Novagen) is transformed with pET plasmids containing the α-xylosidase genes and plated on chloramphinicol(Cam)/Amp LB plates. Cells from overnight plates are resuspended in liquid Cam/Amp LB medium and diluted to OD₆₀₀ = 0.1. Cells in liquid medium are propagated until OD₆₀₀ = 0.4-0.8. Cells are then equilibrated to a temperature of 18°C for 30 min. and protein induction is induced with 0.5 mM IPTG o/n at 18°C. Cells are harvested and pellets are re-suspended in a buffer (e.g., 25mM Tris HCl pH 7 or 10 mM potassium phosphate buffer pH 7) to a cell density corresponding to OD₆₀₀=∼10. Cells are sonicated on ice for 3-7 times 15-30 sec with interruptions of 30 sec on ice. Cell debris is removed by centrifugation and supernatants are assayed for activity.

Assay for α-xylosidase activity

Supernatants resulting from sonication are evaluated on <i>p</i>-nitrophenyl α-D xylopyranoside (Sigma) for presence of α-xylosidase activity. Crude enzyme is incubated with 5 mM <i>p</i>-nitrophenyl α-D xylopyranoside at 37°C for 1-2 hours in a buffer (e.g., 10mM potassium buffer pH 7 or a 25 mM Tris HCl pH 7 buffer). Crude enzymes are also assayed on a fragment of human FVII comprising the Xyl-Xyl-Glc-O-Ser52 glycosylation (peptide fragment consisting of amino acid residues 39-62 of FVII) to evaluate whether the enzyme can cleave the alpha-1,3 xylosidic bonds. The incubation with peptide is performed for 3 hours or overnight at 37°C. Peptide samples incubated with or without α-xylosidase are then evaluated by MALDI MS directly after incubation to evaluate whether the enzyme can remove zero, one or two xylose sugars from the glycopeptide.

Purification olt-xylosidase

A partial purification of the expressed α-xylosidase is performed prior to incubation with rFVII. Supernatants (from approximately 20-50 ml cell culture) obtained after cell disruption in a suitable buffer (eg. a 10mM phosphate buffer pH 7). In case of enzymes coming from thermophiles (eg. tm0308, BH1905, XylS), supernatants are also heated at 50-70°C for 30 min, cooled on ice for 10 min and precipitate is removed by centrifugation for 15 min at 15.000 G in order to remove thermo-labile <i>E. coli</i> contaminants..

The supernatants are sterile filtrated and applied to a 1 ml DEAE FF column (Amersham Pharmacia). The purification is performed with the AKTA explorer (Amersham Pharmacia) FPLC with the following buffers: Buffer A: 25 mM sodium phosphate pH 7, Buffer B: 25 mM sodium phosphate pH 7 and 1 M NaCl. After the application is loaded, unbound sample is washed out with buffer A for 5 CV. A gradient from 0-100% buffer B is used for 20 CV during which the target protein is eluted in fractions. After purification, fractions comprising the main peak in the resulting chromatogram are assayed by incubation on <i>p</i>-nitrophenyl α-D xylopyranoside or by SDS PAGE. The fractions containing the α-xylosidase activity are diluted in a 20 mM Tris HCl pH 7, 2 mM CaCl₂ buffer and concentrated on Vivaspin 20 50.000 MWCO columns (Vivascience) by centrifugation at 2900 rpm.

O-glycoforms of rFVIIa with exclusively glucose at serine 52

The O-glycoforms of rFVIIa with exclusively glucose at serine 52 are obtained by incubation of rFVIIa in an appropriate buffer, e.g. glycylglycine or 20 mM Tris HCl pH 7.0, 2 mM CaCl₂, with purified α-xylosidase for an appropriate time, that can be determined experimentally. Mass spectra visualizing the deglycosylation are obtained by analysing rFVII α-xylosidase incubations ESI-MS (Q-STAR).

The resulting glycan-remodeled rFVIIa is purified from the α-xylosidase enzyme by for example anion-exchange chromatography or gel filtration or suitable combinations. The purity of the prepared O-glycoform of rFVIIa is verified by the O-glycopeptide map of rFVIIa.

Example 3

Preparation of α-xylosidase by cloning and expression of T. maritima putative α-xylosidase gene (tm0308) in E. coli and purification

The above strategy (see Example 2) was followed all the way to a conclusion for tm0308. The <i>T. maritima</i> putative α-xylosidase gene (<i>tm0308</i>) was PCR amplified and cloned into a <i>E</i>. <i>coli</i> pET11a vector. Soluble <i>tm0308</i> could be obtained after expression in an <i>E. coli</i> Rosetta (DE3) expression strain and evaluation of a crude TM0308 preparation on a <i>p</i>-nitrophenyl α-D xylopyranoside, clearly indicated α-xylosidase activity. The α-xylosidase was partly purified using DEAE FF chromatography followed by up-concentration by ultra filtration. The partly purified enzyme was incubated with FVII in a 25 mM Tris pH 7, 2 mM CaCl₂ buffer at different enzyme/FVII ratios for 3 hours at 50°C or overnight at 37°C. Controls with identical compositions of α-xylosidase and rFVII, to which synthetic substrate was added, showed that the enzyme was active under these conditions and it was possible to visualize FVII with and without xylosidase treatment on SDS-gels and by ESI-MS. However, no significant removal of the xylose sugars linked to Glc-O-Ser52 could be detected in this first experiment. In contrast, removal of xylose from a purified reduced and alkylated FVIIa peptide comprising Xyl-Xyl-Gic-O-Ser52 was observed.

Example 4

Preparation of α-xylosidase by cloning and expression in E. coli

The following constructs have been cloned into the pET expression vectors in accordance with the strategy described in Example 2: Gene bl02681(2310 bp)( Bacillus licheniformis: H); gene bl1905 (2328 bp)(Bacillus halodurans: H), gene xylS (2196 bp)(Sulfolobus solfataricus: P); gene yicl (2319 bp) (Escherichia coli: P). The constructs will be expressed in Rosetta, isolated, purified, and evaluated for α-xylosidase activity in accordance with the above-described strategy. Each α-xylosidase will be incubated with rFVIIa in an appropriate buffer, e.g. glycylglycine or 20 mM Tris HCl pH 7.0, 2 mM CaCl₂, for an appropriate time that can be determined experimentally and MS Spectra visualizing the deglycosylation will be obtained by analysing rFVII α-xylosidase incubations ESI-LC-MS (Q-STAR).

The resulting glycan-remodeled rFVIIa will be purified from the α-xylosidase enzyme by for example anion-exchange chromatography or gel filtration or suitable combinations thereof. The purity of the prepared O-glycoform of rFVIIa is verified by the O-glycopeptide map of rFVIIa.

Example 5

Preparation of truncated α-xylosidase by cloning and expression in E. coli

The crystal structure of Yicl was recently solved. Thus, cloning of a truncated α-xylosidase enzyme representing an active, catalytical domain of the Yicl protein (or other similar α-xylosidases) may be possible and is being planned, since a smaller enzyme, if active, may better access the Xyl-Xyl-Glc-O-Ser52 present in native rFVIIa. A domain comprising the active site in the enzyme is predicted from the structure. Gene sequence encoding this part of the Yicl sequence is prepared from the already existing Yicl pET11a plasmid for an example by PCR amplification of relevant areas of the Yicl gene, The primers used for PCR will have extensions with restriction enzyme sites that can be used for ligation of the truncated Yicl gene into the pET11a vector. The truncated enzyme will after expression and purification be evaluated for its potential for deglycosylation of rFVIIa as described above.

Example 6

Preparation of rFVIIa with exclusively xylose-glucose at serine 52 or exclusively xylose-xylose-glucose at serine 52 by α-xylosyltransferase treatment

Preparation of α-xylosyltransferase

The enzyme, UDP-D-xylose: β-D-glucoside α-1,3-D-xylosyltransferase, can be prepared from HepG2 cells as described by Omichi et al. (1997). In short, HepG2 cells are grown in a medium supplemented with 10% fetal calf serum. The microsomal fraction is prepared by homogenisation of the cells followed by centrifugation. The α-xylosyltransferase enzyme is purified by chromatography, for example on an anion-exchange column and/or on other chromatographic columns known for the skilled person. Fractions are collected during elution and the fractions containing the α-xylosyltransferase enzyme are identified by use of the α-xylosyltransferase assay.

The enzyme, UDP-D-xylose: α-D-xyloside α1,3-xylosyltransferase, can be prepared from HepG2 cells as described by Minamida et al. (1996). In short, HepG2 cells are grown in a medium supplemented 10% fetal calf serum. The microsomal fraction is prepared by homogenisation of the cells followed by centrifugation. The α-xylosyltransferase enzyme is purified by chromatography, for example on an anion-exchange column and/or on other chromatographic columns known for the skilled person. Fractions are collected during elution and the fractions containing the α-xylosyltransferase enzyme are identified by use of the α-xylosyltransferase assay.

α-xylosyltransferase assay

The α-xylosyltransferase assays are conducted in an appropriate buffer, e.g. 10mM Hepes, pH 7.2, 0.1% Triton X-100, 0.5 mM UDP-Xylose (Sigma U5875), containing a suitable substrate, e.g. the O-glycopeptides that can be obtained from the O-glycopeptide map of rFVIIa or the pyridylaminated oligosaccharides prepared as described in Minamida et al. (<nplcit id="ncit0038" npl-type="s"><text>Minamida et.al., Detection of UDP-D-xylose: α-D-xyloside α1-3xylosyltransferase activity in human hepatoma cell line HepG2. J. Biochem. 120 1002-1006, 1996</text></nplcit>). The reaction is stopped after an appropriate time, that can be determined experimentally, by e.g. addition of trifluoroacetic acid, and the assay mixtures are analysed by HPLC.

O-glycoforms of rFVIIa with exclusively xylose-glucose- at serine 52

The O-glycoforms of rFVIIa with exclusively xylose-glucose at serine 52 are obtained by (1) treatment of rFVIIa with xylosidase as described above, (2) purification of the xylosidase treated rFVIIa from the xylosidase by for example anion-exchange chromatography, and (3) by incubation of xylosidase-treated rFVIIa in an appropriate buffer, e.g. glycylglycine, pH 7.0, 10 mM calcium chloride, with purified UDP-D-xylose: P-D-glucoside α-1,3-D-xylosyltransferase and UDP-D-xylose for an appropriate time, that can be determined experimentally. The resulting glyco-remodelled rFVIIa is purified from the UDP-D-xylose: β-D-glucoside α-1,3-D-xylosyltransferase enzyme by for example anion-exchange chromatography. The purity of the prepared O-glycoform of rFVIIa is verified by the O-glycopeptide map of rFVIIa.

O-glycoforms of rFVIIa with exclusively xylose-xylose-glucose- at serine 52

The O-glycoforms of with xylose-xylose-glucose at serine 52 are obtained by (1) treatment of rFVIIa with xylosidase as described above, (2) purification of the xylosidase treated rFVIIa from the xylosidase by for example anion-exchange chromatography, (3) further treatment with UDP-D-xylose: β-D-glucoside α-1,3-D-xylosyltransferase and UDP-D-xylose as described above, and (4) by incubation of the product in an appropriate buffer, e.g. glycylglycine, pH 7.0, 10 mM calcium chloride, with purified UDP-D-xylose: α-D-xyloside α1,3-xylosyltransferase and UDP-D-xylose for an appropriate time, that can be determined experimentally. The resulting glyco-remodelled rFVIIa is purified from the UDP-D-xylose: α-D-xyloside α1,3-xylosyltransferase enzyme by for example anion-exchange chromatography. The purity of the prepared O-glycoform of rFVIIa is verified by the O-glycopeptide map of rFVIIa.

Example 7

Analysis of O-glycoform pattern of rFVIIa

Tryptic peptide mapping of the rFVIIa light chain

The relative content of the O-glycoforms of rFVIIa is determined by tryptic peptide mapping of the rFVIIa light chain. The rFVIIa is reduced and alkylated and the rFVIIa light chain is purified on a RP-HPLC column eluted with an acetonitrile gradient in water:trifluoroacetic acid. The purified rFVIIa light chain is buffer-exchanged to Tris buffer, pH 7.5 and digested with trypsin. The tryptic digest of the rFVIIa light chain is analysed on a RP-HPLC column (for example Nucleosil C18, 5 µ, 300 Å, 4.0 x 250 mm, Macherey-Nagel 720065) eluted with an acetonitrile gradient (0%-45% acetonitrile in 100 min) in water: trifluoroacetic acid (see <figref idref="f0001">Figure 2</figref>). Flow is 1.0 ml/min and detection is UV at 215 nm.

The peaks containing the O-glycopeptides of rFVIIa are eluted after approx. 60-65 min where the 1st and the 3rd peak contain O-glycopeptides with a xylose-xylose-glucose-linked to serine 52, and the 2nd and 4th peak contain O-glycopeptides with a glucose linked to serine 52.

Similarly, the 1st and the 2nd peak contain O-glycopeptides with a tetrasaccharide linked to serine 60, and the 3rd and the 4th peak contain O-glycopeptides with a fucose linked to serine 60.

Tryptic peptide mapping of rFVIIa

The O-glycoform pattern can be analysed by tryptic peptide mapping of rFVIIa. The rFVIIa is buffer-exchanged to Tris buffer, pH 7.5, and digested with trypsin. The tryptic digest of the rFVIIa is analysed on a RP-HPLC column (for example Nucleosil C18, 5 µ, 300 Å, 4.0 x 250 mm, Macherey-Nagel 720065) eluted with an acetonitrile gradient (0%-45% acetonitrile in 100 min) in water: trifluoroacetic acid. Flow is 1.0 ml/min and detection is UV at 215 nm. The peaks containing the O-glycopeptides of rFVIIa are eluted after approx. 67-70 min where the 2nd peak contains O-glycopeptides with a xylose-xylose-glucose linked to serine 52, and the 1st peak contains O-glycopeptides with a glucose linked to serine 52.

Total mass analysis of rFVIIa

The O-glycoform pattern can be analysed by total mass analysis of rFVIIa. The rFVIIa is desalted on a Millipore ZipTip C4 column equilibrated with 1 % formic acid and eluted with 3 % formic acid in 90 % methanol. The eluted sample is analysed by the nanospray technique on a Qstar XL mass spectrometer. The major peak at approximately 50500 Da represents rFVIIa O-glycoforms with a glucose linked to serine 52 and the major peak at approximately 50800 Da represents rFVIIa O-glycoforms with a xylose-xylose-glucose linked to serine 52.

Example 8

Purification of Glc-O-Ser52-FVII and Xyl-Xyl-Glc-O-Ser52-FVII

Glc-O-ser52-FVII and Xyl-Xyl-Glc-O-Ser52-FVII was purified using two cycles of hydrophobic interaction chromatography (HIC). The column (1.0 cm in inner diameter x 7.0 cm length = 5.5 ml column volume (CV)) packed with Toso Haas TSK-Gel phenyl 5 PW, was equilibrated with 5 CV 10 mM histidine, 10 mM CaCl2, 2.0 M NH4-acetate, pH 6.0. The column was loaded with approximately 2.5 mg of FVII pr. ml resin. To the load solution 2.0 M NH4-acetate and 10 mM CaCl2 was added prior to load. The column was washed with 5 CV 10 mM histidine, 10 mM CaCl2, 2.0 M NH4-acetate, pH 6.0. Elution was performed using a 20 CV linear gradient from 10 mM histidine, 10 mM CaCl2, 2.0 M NH4-acetate, pH 6.0 to 10 mM histidine, 10 mM CaCl2, pH 6.0. The purification was performed at a flow rate of 6 CV/h and at a temperature of 5°C. Fractions were collected during elution.

The FVII eluted in two overlapping major peaks (see <figref idref="f0003">Figure 4</figref>: Chromatogram from first HIC cycle). Fractions containing the first peak were pooled (fraction "A", <figref idref="f0003">Figure 4</figref>) and further purified by a second cycle of HIC, using the same chromatographic procedure as for the first HIC cycle (see <figref idref="f0004">Figure 5</figref>: Chromatogram obtained by reloading fraction "A" onto the HIC column). Fractions containing the second major peak (fraction "B", <figref idref="f0003">Figure 4</figref>) were pooled as well and further purified by a second cycle of HIC, using the same chromatographic procedure as for the first HIC cycle (see <figref idref="f0005">Figure 6</figref>: Chromatogram obtained by reloading fraction "B" onto the HIC column).

Purified Glc-O-Ser52-FVII was identified in the peak fraction, fraction 10 (<figref idref="f0004">Figure 5</figref>), obtained by reloading fraction "A" onto the second HIC step. Purified Xyl-Xyl-Glc-O-Ser52-FVII was identified in the peak fraction, fraction 15 (<figref idref="f0005">Figure 6</figref>), obtained by reloading fraction "B" onto the second HIC step. The identification was obtained by tryptic peptide mapping of rFVIIa as described in Example 7 (<figref idref="f0006">Figure 7A and 7B</figref>) and by total mass analysis of rFVIIa as described in Example 7 (<figref idref="f0007">Figure 8A and 8B</figref>). Both analyses showed a high content of Glc-O-Ser52-rFVIIa and a low content of Xyl-Xyl-Glc-O-Ser52-rFVIIa in the peak fraction, Fraction 10, and a low content of Glc-O-Ser52-rFVIIa and a high content of Xyl-Xyl-Glc-O-Ser52-rFVIIa in the peak fraction, Fraction 15. A quantitation of the content of the O-glycoforms in the two peak fractions could not be obtained due to relatively low rFVIIa content in the fractions (<figref idref="f0006">Figure 7A and 7B</figref>: Tryptic peptide mapping: Other peptide fragments of rFVIIa co-eluted with or eluted close to the O-glycopeptides, and the content of O-glycopeptides in low amounts could therefore not be determined.) (<figref idref="f0007">Figure 8A and 8B</figref>: Total mass analysis: Other O- and/or N-glycoforms of rFVIIa, for example N-glycoforms of rFVIIa lacking one N-acetylneuraminic acid, appeared in the mass spectra, and the content of O-glycoforms of rFVIIa in low amounts could therefore not be determined).

The specific activities of the peak fractions obtained from the HIC (Table 1) were determined by the 1st generation clotting assay. It was found that the Glc-O-Ser52-rFVIIa O-glycoform had a low specific activity while the Xyl-Xyl-Glc-OrSer52-rFVIIa O-glycoform had a high specific activity. <tables id="tabl0001" num="0001"><table frame="all"><title>Table 1. Specific activities determined using the 1st generation clotting assay for the peak fractions obtained from HIC. The content of rFVIIa was determined by HPLC.</title><tgroup cols="2"><colspec colnum="1" colname="col1" colwidth="89mm" /><colspec colnum="2" colname="col2" colwidth="77mm" /><thead><row><entry valign="top">Sample</entry><entry valign="top">Specific activity</entry></row></thead><tbody><row><entry>PS5002-014 Frak. 10</entry><entry>44 IU/µg</entry></row><row><entry>PS5002-015 Frak. 15</entry><entry>61 IU/µg</entry></row><row><entry>PS5002-014/015 starting material</entry><entry>53 IU/µg</entry></row></tbody></tgroup></table></tables>

Example 9

Purification by hydrophobic interaction chromatography

Highly purified Glc-O-Ser52-rFVIIa preparations and highly purified Xyl-Xyl-Glc-O-Ser52-rFVIIa preparations can be obtained by repeated purification on the hydrophobic interaction chromatography as described above. Highly purified Glc-O-Ser52-rFVIIa and Xyl-Xyl-Glc-O-Ser52-rFVIIa preparations with higher rFVIIa content can be obtained by increasing the amount of starting material for the hydrophobic interaction chromatography performed as described above. The content of each O-glycoform of rFVIIa in the highly purified preparations with higher rFVIIa content can be quantitated by tryptic peptide mapping of the rFVIIa light chain as described in Example 7. The specific activities of the highly purified preparations can be determined by the 1st generation clotting assay as above.