Technical Field

The invention relates to protein chemistry and molecular biology.

Background

Members of the cystine-knot superfamily are dimeric proteins that share a structurally related "cystine-knot" conformation. The cystine knot structural motif contains three disulfides (six cysteine residues in close proximity in a protein backbone), with one of the disulfides passing through a ring formed by the other two disulfide bonds (Lenguyen et al, 1990,

Biochimie 72(6/7):431; Sun et al, 1995, Annu. Rev. Biophys. Biomol. Struct. 24:269).

Neublastin, also known as artemin and enovin, is a 24 kDa homodimeric, secreted protein that promotes the outgrowth and survival of neurons of the peripheral and central nervous system (Baudet et al, 2000, Development, 127:4335; Masure et al, 1999, Eur. J. Biochem., 266:892; Rosenblad et al, 2000, Mol. Cell Neurosci., 15(2):199). Neublastin is a member of the transforming growth factor (TGF) beta superfamily, which constitutes a superfamily within the cystine-knot superfamily, whose members are characterized by the presence of seven conserved cysteine residues with similar spacing that form the structure of a cystine knot (Saarma, 1999, Microsc. Res. Tech., 45:292). As for the cystine-knot superfamily, each TGF beta superfamily monomer contains two disulfide bonds that form a closed loop structure encircling the third disulfide to form a tight knot structure. The seventh cysteine contained within each monomer forms an intermolecular disulfide bond, covalently linking the monomers to form the final dimer product (Rattenholl et al, 2000, J. Mol. Biol, 305:523).

Cystine knots play critical roles in the structural stability and efficacy of many protein biopharmaceuticals. The accurate characterization of cystine knots is important to ensure consistent stability and efficacy profiles of a biopharmaceutical throughout clinical development.

Summary

Enzymatic digestion of Neublastin, a cystine knot superfamily member, was found to introduce disulfide scrambling among some of the disulfides in the cystine knot structure, which results in significant method-induced artifacts and mistaken assignment of some disulfide linkages

The present invention is based, at least in part, on the discovery that a partial reduction of the protein, followed by blocking the free cysteines prior to enzyme digestion, protects the disulfide bonds involved in cystine knots that are particularly susceptible to scrambling during enzyme digestion. The methods described herein minimize potential scrambling and other method-induced artifacts in the characterization of cystine knots in proteins.

The invention features a method for analyzing a protein, comprising: providing a sample comprising a preparation of a dimeric protein, wherein the dimeric protein is a cystine knot superfamily member comprising a first monomer and a second monomer linked by an inter-chain disulfide bond; contacting the sample with a reductant, wherein at least some of the dimeric protein content of the sample is partially reduced by the reductant to yield partially reduced monomers that have the inter-chain disulfide bond cleaved but maintain the cystine knot- structure; contacting the partially reduced monomers with a cysteine blocking agent that blocks free cysteines in the partially reduced monomers to prevent disulfide bond formation, yielding partially reduced cysteine-blocked monomers; contacting the partially reduced cysteine-blocked monomers with an enzyme that digests the partially reduced cysteine-blocked monomers, to result in enzymatic peptides; and analyzing the enzymatic peptides for the presence of disulfide- linked digested peptides.

A "cystine knot superfamily member" is a dimeric protein containing two monomers linked by a disulfide bond, wherein each of the monomers contains a cystine knot having a structural motif with three disulfides (six cysteine residues in close proximity in a protein backbone), with one of the disulfides passing through a ring, formed by the other two disulfide bonds.

A "reductant" is an agent that cleaves a disulfide bond. In the methods described herein, the reductant is contacted with the sample at a concentration, temperature, and for a duration that causes the desired partial reduction of the cystine knot superfamily member protein.

In some embodiments, the reductant is Tris(2-carboxyethyl)phosphine hydrochloride

(TCEP). In some embodiments, the TCEP reaction concentration is 0.1 mM to 10 mM (e.g., about 1 mM). In some embodiments, the TCEP incubation time is 1 to 20 minutes (e.g., about 20 minutes).

In some embodiments, the reductant is 2-mercaptoethanol or dithiothritol. Additional reductants that can be used in the methods include, but are not limited to, dithioerythriol (DTE), 2-mercaptoethylamine, tributylphosphine (TBP), hexafluoroisopropanol (HFIP), dimethyl sulfoxide (DMSO), sodium dodecyl sulfate (SDS), glutathione, and hydroquinone. A "cysteine blocking agent" is an agent that modifies free cysteine amino acids to block disulfide bond formation between them. In the methods described herein, the cysteine blocking agent is contacted with the partially reduced monomers at a concentration, temperature, and for a duration that causes the free cysteines in the partially reduced monomers to be blocked such that they are not capable of forming disulfide bonds.

In some embodiments, the cysteine blocking agent is an alkylation reagent. An

"alkylation agent" is an agent that alkylates free cysteine amino acids to block disulfide bond formation between them.

In some embodiments, the alkylation reagent is N-ethylmaleimide (NEM). In some embodiments, the NEM reaction concentration is 5 mM to 500 mM (e.g., about 50 mM). In some embodiments, the NEM incubation time is 40 to 80 minutes (e.g., about 60 minutes).

In some embodiments, the alkylation reagent is 4-vinylpyridine, iodoacetamide, or acrylamide. Additional alkylation reagents that can be used in the methods include, but are not limited to, monobromobimane, iodoacetic acid, methyl methanethiosulfonate, and mercury orange.

Additional cysteine blocking agents that can be used in the methods include, but are not limited to, thiol-reactive compounds such as cystine, cystamine, dithioglycolic acid, oxidized glutathione, iodine, hydrogen peroxide, dihydro ascorbic acid, tetrathionate, O-iodosobenzoate, and oxygen in the presence of a metal ion

The enzyme or enzyme combination that is used to digest the partially reduced cysteine- blocked monomers is used at a concentration, temperature, and for a duration to generate enzymatic peptides. In some embodiments, the enzyme is Asp-N, trypsin, or the combination of Asp-N and trypsin. The Asp-N to substrate ratio can optionally be in the range of 75 : 1 to 125 : 1 (e.g., about 100: 1). The trypsin to substrate ratio can optionally be in the range of 20: 1 to 60: 1 (e.g., about 40:1). Asp-N and/or trypsin incubation time can optionally be in the range of 2 to 6 hours (e.g., about 4 hours).

Additional enzymes that can be used to digest the partially reduced cysteine-blocked monomers include, but are not limited to, endoproteinase Glu-C, pepsin, chymotrypsin, endoproteinase Arg-C, and endoproteinase Lys-C. These enzymes can be used individually or in combination. In some embodiments, mass spectrometry is used to analyze the enzymatic peptides for the presence of disulfide- linked digested peptides. The mass spectrometry can be, for example, liquid chromatography/ultraviolet/mass spectrometry (LC/UV/MS).

In the methods described herein. The cystine knot superfamily member dimeric protein can be either a homodimeric protein or a heterodimeric protein.

In some embodiments, the cystine knot superfamily member is a transforming growth factor (TGF) beta superfamily member, such as a glial cell line-derived neurotrophic factor (GDNF) ligand family protein (e.g., a GDNF protein, a Neurturin protein, a Persephin protein, or a Neublastin protein).

In embodiments where the cystine knot superfamily member is a Neublastin protein, the amino acid sequence of the first monomer and/or the second monomer can optionally be at least 90% identical (e.g., at least 95% or 98% identical) to amino acids 15-113 of SEQ ID NO:l. In some embodiments, the amino acid sequence of the first monomer and/or the second monomer contains or consists of amino acids 10-113 of SEQ ID NO: l, amino acids 15-113 of SEQ ID NO: l, amino acids 15-113 of SEQ ID NO:2, amino acids 15-113 of SEQ ID NO:3, amino acids 15-113 of SEQ ID NO:4, amino acids 15-113 of SEQ ID NO:5, amino acids 15-113 of SEQ ID NO:8, or amino acids 15-113 of SEQ ID NO:9. For example, the amino acid sequence of the first monomer and/or the second monomer can contain or consist of the amino acid sequence of SEQ ID NO: l, the amino acid sequence of SEQ ID NO:2, the amino acid sequence of SEQ ID NO:3, the amino acid sequence of SEQ ID NO:4, the amino acid sequence of SEQ ID NO:5, the amino acid sequence of SEQ ID NO:8, or the amino acid sequence of SEQ ID NO:9.

In some embodiments where the cystine knot superfamily member is a Neublastin protein, the first monomer of the Neublastin protein comprises amino acid residues 15-113 of SEQ ID NO: 1. In some embodiments where the cystine knot superfamily member is a

Neublastin protein, the first monomer of the Neublastin protein comprises the amino acid sequence set forth in SEQ ID NO: 1. In some embodiments where the cystine knot superfamily member is a Neublastin protein, the first monomer of the Neublastin protein comprises amino acid residues 10-113 of SEQ ID NO:l. In some embodiments where the cystine knot superfamily member is a Neublastin protein, the amino acid sequence of the first monomer of the Neublastin protein consists of amino acid residues 10-113 of SEQ ID NO: 1. In some

embodiments where the cystine knot superfamily member is a Neublastin protein, the first monomer and the second monomer of the Neublastin protein comprises amino acid residues 10- 113 of SEQ ID NO: 1. In some embodiments where the cystine knot superfamily member is a Neublastin protein, the amino acid sequence of the first monomer and the second monomer of the Neublastin protein consists of amino acid residues 10-113 of SEQ ID NO: 1.

In some embodiments, the TGF beta superfamily member is an Activin/Inhibin family protein (e.g., Activin A, Activin B, Activin AB, Activin C, Activin AC, Inhibin alpha, Inhibin beta E, or Inhibin beta C).

In some embodiments, the TGF beta superfamily member is a TGF-beta family protein (e.g., TGF-beta 1, TGF-beta 2, TGF-beta 3, TGF-beta 4, or TGF-beta 5).

In some embodiments, the TGF beta superfamily member is a Bone Morphogenetic Protein (BMP) family protein (e.g., BMP10, BMP15, BMP2, BMP3, BMP4, BMP5, BMP6, BMP7, BMP 8 A, or BMP8B).

In some embodiments, the TGF beta superfamily member is a Growth Differentiation Factor (GDF) family protein (e.g., GDF1, GDF10, GDF11, GDF15, GDF2, GDF3, GDF3A, GDF5, GDF6, GDF7, GDF 8, or GDF9).

In some embodiments, the cystine knot superfamily member is a glycoprotein hormone

(GH) family member (e.g., follitropin (FSH), Iutropin (LH), thyrotropin, (TSH), or chorionic gonadotropin (CG)).

In some embodiments, the cystine knot superfamily member is a Neurotrophin family member (e.g., Nerve growth factor (NGF), Brain-derived neurotrophic factor (BDNF),

Neurotrophin-3, or Neurotrophin-4).

In some embodiments, the cystine knot superfamily member is a Platelet-derived growth factor (PDGF) family member (e.g., PDGF-AA, PDGF-AB, PDGF-BB, Vascular endothelial growth factor (VEGF)-A, VEGF-B, VEGF-C, or VEGF-D).

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the exemplary methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present application, including definitions, will control. The materials, methods, and examples are illustrative only and not intended to be limiting. Other features and advantages of the invention will be apparent from the following detailed description, and from the claims.

Brief Description of the Drawings

Fig. 1 is a depiction of the neublastin amino acid sequence (104 amino acid human neublastin) as well as the location of the six cysteines that form the cystine knot disulfide bonds and the one cysteine that forms the inter-chain disulfide bond.

Fig. 2 is a depiction of the neublastin amino acid sequence (104 amino acid human neublastin) and the disulfide- linked peptides resulting from digestion by trypsin or

trypsin+AspN.

Fig. 3 is a depiction of the neublastin amino acid sequence (104 amino acid human neublastin) and the disulfides resulting partial reduction and alkylation with N-ethylmaleimide.

Fig. 4 is a depiction of the neublastin amino acid sequence (104 amino acid human neublastin) and the disulfide- linked peptides resulting from partial reduction and alkylation with N-ethylmaleimide followed by digestion by trypsin or trypsin+AspN.

Fig. 5 is a liquid chromatography/mass spectrometry (LC/MS) chromatogram after partial reduction and alkylation with N-ethylmaleimide followed by digestion by trypsin+AspN.

Figs. 6A-6C are characterizations of the anticipated and observed disulfide linkages for the Cys7-Cys72 disulfide with Cys71 alkylated with N-ethylmaleimide (6A), in the ETD-MS2 and CID-MS2 spectra (6B), and in the CID and ETD-MS2 spectra using Orbitrap FTMS (6C).

Fig. 7 is a characterization of the anticipated and observed Cys34-Cysl00 and Cys38- Cysl02 disulfide linkages.

Fig. 8 is a characterization of the anticipated and observed Cys71-Cys71 and Cys7-Cys72 disulfide linkages.

Fig. 9 is an alignment of wild type human (SEQ ID NO: 10), mouse (SEQ ID NO: 11), and rat (SEQ ID NO: 12) pre pro neublastin polypeptides. The left and right vertical lines indicate, respectively, the start of the mature 113 amino and 104 amino acid forms. The RRXR heparin binding motif is boxed. Detailed Description

The present invention provides methods for analyzing cystine knot superfamily member proteins. As disclosed in the accompanying examples, the disulfide linkages of Neublastin become unstable and re-shuffled to form mis-matched disulfides upon disruption of the cystine knot structure in each monomer. Partial reduction of Neublastin, followed by blocking free cysteines prior to enzymatic digestion and mass spectrometry analysis, was found to reduce the production of mis-matched disulfide bonds involved in cystine knots and allow for effective monitoring of the Neublastin disulfides. Because Neublastin has a cysteine knot structure common to members of the cystine knot superfamily, the partial reduction and cysteine blocking methods described herein are expected to be effective for analyzing disulfides in other proteins belonging to the cystine knot superfamily.

Cystine Knot Superfamily

Members of the cystine-knot superfamily are dimeric proteins that have a structurally related "cystine-knot" conformation. The active forms of these proteins are dimers (either homodimers or hetero dimers). The amino acid sequences of members of this superfamily contain seven cysteine residues with a conserved spacing pattern. Six of the cysteines form intrachain disulphide bridges that form the cystine knot. The remaining cysteine residue forms an interchain disulfide bond that links the two monomers (each having a cystine knot) together to form a dimer. The cystine knot structure constitutes a polypeptide fold that is conserved across all members of this superfamily.

There are four primary subfamilies of related proteins within the cystine-knot superfamily: the transforming growth factor (TGF) beta superfamily; the glycoprotein hormone (GH) family; the neurotrophin family; and the platelet-derived growth factor (PDGF) family. Each of these subfamilies comprises members sharing greater amino acid sequence conservation between them than that which is present between members of the cystine-knot superfamily as a whole.

Polypeptides having the amino acid sequence of a wild-type cystine-knot superfamily member (e.g., a wild-type human cystine-knot superfamily member) or a biologically active variant of thereof can be used in the methods described herein. A variant cystine-knot superfamily member polypeptide can contain one or more additions, substitutions, and/or deletions. In some embodiments, a variant has at least 70%, 80%, 85%, 90%, 95%, or 98% sequence identity to the full length wild type protein (e.g., wild-type human protein) and retains the biological activity of the wild type protein. In some embodiments, a variant has at least 70%, 80%), 85%o, 90%), 95%o, or 98%> sequence identity to the mature wild type protein ((e.g., mature wild-type human protein) and retains the biological activity of the wild type protein.

The Transforming growth Factor (TGF) Beta Superfamily

The TGF beta superfamily includes several subfamilies, as follows: the glial cell line- derived neurotrophic factor (GDNF) ligand family (which includes Neublastin, GDNF,

Neurturin, and Persephin); the Activin/Inhibin family (which includes Activin A, Activin B, Activin AB, Activin C, Activin AC, Inhibin alpha, Inhibin beta E, and Inhibin beta C); the TGF- beta family (which includes TGF-beta 1, TGF-beta 2, TGF-beta 3, TGF-beta 4, and TGF-beta 5); the Bone Morpho genetic Protein (BMP) family (which includes BMP 10, BMP 15, BMP2, BMP3, BMP4, BMP5, BMP6, BMP7, BMP 8 A, and BMP8B); and the Growth Differentiation Factor (GDF) family (which includes GDF1, GDF10, GDF11, GDF15, GDF2, GDF3, GDF3A, GDF5, GDF6, GDF7, GDF8, and GDF9).

The Glycoprotein Hormone (GH) Family

The glycoprotein hormone (GH) family includes the following members: follitropin (FSH), Iutropin (LH), thyrotropin, (TSH), and chorionic gonadotropin (CG). FSH, LH, and TSH are anterior pituitary-released hormones, whereas CG is a hormone produced by the placenta.

The Neurotrophin Family

The Neurotrophin family includes the following members: Nerve growth factor (NGF), Brain-derived neurotrophic factor (BDNF), Neurotrophin-3, and Neurotrophin-4.

The Platelet-Derived Growth Factor (PDGF) Family

The platelet-derived growth factor (PDGF) family includes the following members:

PDGF-AA, PDGF-AB, PDGF-BB, Vascular endothelial growth factor (VEGF)-A, VEGF-B, VEGF-C, and VEGF-D. Neublastin Polypeptides

As described above, Neublastin is a member of the TGF beta superfamily. The working examples section describes analytical methods performed using a Neublastin protein. The following are examples numerous wild-type and variant Neublastin protein sequences that can be used in the methods described herein.

A mature form of wild-type human neublastin is 113 amino acids in length and has the following amino acid sequence: AGGPGSRARAAGARGCRLRSQLVPVRALGLGHRSDELV RFRFCSGSCRRARSPHDLSLASLLGAGALRPPPGSRPVSQPCCRPTRYEAVSFMDVNSTW RTVDRLSATACGCLG (SEQ ID NO: l). Polypeptides having the amino acid sequence of SEQ ID NO: 1 or biologically active variants of thereof can be used in the methods described herein. A variant neublastin polypeptide can contain one or more additions, substitutions, and/or deletions, as detailed in the following sections. Wild-type neublastin polypeptides and biologically active variants thereof are collectively referred to herein as "neublastin

polypeptides."

A variant neublastin polypeptide can vary in length from the corresponding wild-type polypeptide. Although the mature human neublastin polypeptide of SEQ ID NO: 1 consists of the carboxy terminal 113 amino acids of pre pro neublastin (SEQ ID NO: 10), not all of the 113 amino acids are required to achieve useful neublastin biological activity. Amino terminal truncation is permissible. Thus, a variant neublastin polypeptide can contain, for example, the carboxy terminal 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, or 113 amino acids of SEQ ID NO: l (i.e., its length can be 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, or 113 amino acids).

A variant neublastin polypeptide can also vary in sequence from the corresponding wild- type polypeptide. In particular, certain amino acid substitutions can be introduced into the neublastin sequence without appreciable loss of a neublastin biological activity. In exemplary embodiments, a variant neublastin polypeptide (i) contains one or more amino acid substitutions, and (ii) is at least 70%, 80%, 85%, 90%, 95%, 98% or 99% identical to SEQ ID NO: l (or 70%, 80%, 85%, 90%, 95%, 98% or 99% identical to amino acids 15-113 of SEQ ID NO: l). A variant neublastin polypeptide differing in sequence from SEQ ID NO: l (or differing in sequence from amino acids 15-113 of SEQ ID NO: l) may include one or more amino acid substitutions (conservative or non-conservative), one or more deletions, and/or one or more insertions. Fig. 9 is an alignment of the wild type human, mouse, and rat pre pro neublastin polypeptides. The vertical lines in Fig. 9 indicate the start of the mature 113 amino acid form (left vertical line) and 104 amino acid form (right vertical line) of neublastin. The RRXR heparin binding motif is boxed. This alignment of bioactive forms of neublastin indicates specific exemplary residues (i.e., those that are not conserved among the human, mouse, and rat forms) that can be substituted without eliminating bioactivity.

Percent identity between amino acid sequences can be determined using the BLAST 2.0 program. Sequence comparison can be performed using an ungapped alignment and using the default parameters (Blossom 62 matrix, gap existence cost of 11, per residue gap cost of 1 , and a lambda ratio of 0.85). The mathematical algorithm used in BLAST programs is described in Altschul et al, 1997, Nucleic Acids Research 25:3389-3402.

A conservative substitution is the substitution of one amino acid for another with similar characteristics. Conservative substitutions include substitutions within the following groups: valine, alanine and glycine; leucine, valine, and isoleucine; aspartic acid and glutamic acid;

asparagine and glutamine; serine, cysteine, and threonine; lysine and arginine; and phenylalanine and tyrosine. The non-polar hydrophobic amino acids include alanine, leucine, isoleucine, valine, proline, phenylalanine, tryptophan and methionine. The polar neutral amino acids include glycine, serine, threonine, cysteine, tyrosine, asparagine and glutamine. The positively charged (basic) amino acids include arginine, lysine and histidine. The negatively charged (acidic) amino acids include aspartic acid and glutamic acid. Any substitution of one member of the

above-mentioned polar, basic or acidic groups by another member of the same group can be deemed a conservative substitution.

Non-conservative substitutions include those in which (i) a residue having an

electropositive side chain (e.g., Arg, His or Lys) is substituted for, or by, an electronegative residue (e.g., Glu or Asp), (ii) a hydrophilic residue (e.g., Ser or Thr) is substituted for, or by, a hydrophobic residue (e.g., Ala, Leu, He, Phe or Val), (iii) a cysteine or proline is substituted for, or by, any other residue, or (iv) a residue having a bulky hydrophobic or aromatic side chain (e.g., Val, He, Phe or Trp) is substituted for, or by, one having a smaller side chain (e.g., Ala, Ser) or no side chain (e.g., Gly).

A biologically active variant neublastin polypeptide, when dimerized, binds to a ternary complex containing GFRa3 and RET. Any method for detecting binding to this complex can be used to evaluate the biological activity a variant neublastin polypeptide. Exemplary assays for detecting the ternary complex-binding ability of a variant neublastin polypeptide are described in WO00/01815 (the content of which is incorporated herein by reference).

A variant neublastin polypeptide can also be assessed to evaluate its ability to trigger the neublastin signaling cascade. For example, the Kinase Receptor Activation (KIRA) assay can be used to assess the ability of a variant neublastin polypeptide to induce RET autophosphorylation {See also, Sadick et al, 1996, Anal. Biochem., 235(2):207).

Substitutions at one or more of the following amino acid residues are expected to result in a variant neublastin polypeptide having reduced or absent heparin binding ability as compared to wild type neublastin: Arg 48, Arg 49, Arg 51, Ser 46, Ser 73, Gly 72, Arg 39, Gin 21, Ser 20, Arg 68, Arg 33, His 32, Val 94, Arg 7, Arg 9, or Arg 14. Reference to a neublastin amino acid reside by position number refers to the numbering of residues relative to SEQ ID NO: 1.

A neublastin amino acid residue designated for substitution (e.g., an arginine residue at position 48, 49, and/or 51) can be substituted with a non-conservative amino acid residue (e.g., glutamic acid) or a conservative or amino acid residue. Exemplary amino acids that can be substituted at a residue identified herein (e.g., position 48, 49, and/or 51) include glutamic acid, aspartic acid, and alanine.

Examples of variant neublastin polypeptides that exhibit reduced or absent heparin binding are disclosed in Table 1 and in WO 2006/023781 (the content of which is incorporated herein by reference). Amino acid residues of the variant neublastin polypeptides that are mutated as compared to the corresponding wild type position are bolded and underlined in Table 1. In addition, the neublastin polypeptide (e.g., 113, 99, or 104 amino acids in length) used as the background for the substitution is depicted in Table 1.

Variant Neublastin Polypeptides

A polypeptide can optionally contain heterologous amino acid sequences in addition to a neublastin polypeptide. "Heterologous," as used when referring to an amino acid sequence, refers to a sequence that originates from a source foreign to the particular host cell, or, if from the same host cell, is modified from its original form. Exemplary heterologous sequences include a heterologous signal sequence (e.g., native rat albumin signal sequence, a modified rat signal sequence, or a human growth hormone signal sequence) or a sequence used for purification of a neublastin polypeptide (e.g., a histidine tag).

Neublastin polypeptides can be isolated using methods known in the art. Naturally occurring or recombinantly produced neublastin polypeptides can be isolated from cells or tissue sources using standard protein purification techniques. Alternatively, mutated neublastin polypeptides can be synthesized chemically using standard peptide synthesis techniques. The synthesis of short amino acid sequences is well established in the peptide art. See, e.g., Stewart, et al, Solid Phase Peptide Synthesis (2d ed., 1984).

In some embodiments, neublastin polypeptides are produced by recombinant DNA techniques. For example, a nucleic acid molecule encoding a neublastin polypeptide can be inserted into a vector, e.g., an expression vector, and the nucleic acid can be introduced into a cell. Suitable cells include, e.g., mammalian cells (such as human cells or CHO cells), fungal cells, yeast cells, insect cells, and bacterial cells (e.g., E. coli). When expressed in a recombinant cell, the cell is preferably cultured under conditions allowing for expression of a neublastin polypeptide. The neublastin polypeptide can be recovered from a cell suspension if desired. As used herein, "recovered" means that the mutated polypeptide is removed from those components of a cell or culture medium in which it is present prior to the recovery process. The recovery process may include one or more refolding or purification steps. Buffers and methods for inducing folding of a denatured neublastin polypeptide are described in, e.g., WO 2006/023782.

Variant neublastin polypeptides can be constructed using any of several methods known in the art. One such method is site-directed mutagenesis, in which a specific nucleotide (or, if desired a small number of specific nucleotides) is changed in order to change a single amino acid (or, if desired, a small number of predetermined amino acid residues) in the encoded variant neublastin polypeptide. Many site-directed mutagenesis kits are commercially available. One such kit is the "Transformer Site Directed Mutagenesis Kit" sold by Clontech Laboratories (Palo Alto, CA). Partial Reduction and Cysteine Blocking

The methods entail partial reduction of a cystine knot superfamily member protein (e.g. Neublastin). The partial reduction is achieved by contacting a sample comprising a preparation of the cystine knot superfamily member protein with a reductant. The partial reduction causes at least some of the cystine knot superfamily member protein content of the sample to be partially reduced, to yield partially reduced monomers that have their inter-chain disulfide bond broken but maintain the cystine knot-structure. The reduction step is not intended to be performed under conditions that cause a complete reduction of the cystine knot superfamily member protein content of the sample, such that all disulfide bonds are broken.

Any agent that cleaves a disulfide bond can be used as a reductant in these methods.

Examples of reductants include tris(2-carboxyethyl)phosphine hydrochloride (TCEP),

2-mercaptoethanol, dithiothreitol (DTT), dithioerythriol (DTE), 2-mercaptoethylamine, tributylphosphine (TBP), hexafluoroisopropanol (HFIP), dimethyl sulfoxide (DMSO), sodium dodecyl sulfate (SDS), glutathione, and hydroquinone. The reductant is contacted with the sample at a concentration, temperature, and for a duration that causes the desired partial reduction of the cystine knot superfamily member protein.

The methods also entail, subsequent to the partial reduction, blocking free cysteines (produced by the partial reduction) in the partially reduced monomers to prevent the free cysteines from forming disulfide bonds with other cysteines.

Any agent that blocks free cysteines to prevent disulfide bond formation can be used as a cysteine blocking agent in these methods. For example, any agent that alkylates free cysteine amino acids to block disulfide bond formation can be used. Examples of alkylation agents include N-ethylmaleimide, 4-vinylpyridine, iodoacetamide, acrylamide, monobromobimane, iodoacetic acid, methyl methanethiosulfonate, and mercury orange. Other cysteine blocking agents include thiol-reactive compounds such as cystine, cystamine, dithioglycohc acid, oxidized glutathione, iodine, hydrogen peroxide, dihydro ascorbic acid, tetrathionate, O-iodosobenzoate, and oxygen in the presence of a metal ion. The cysteine blocking agent is contacted with the partially reduced monomers at a concentration, temperature, and for a duration that causes the free cysteines in the partially reduced monomers to be blocked such that they are not capable of forming disulfide bonds. Enzymatic Digestion and Disulfide Characterization

After partial reduction and cysteine blocking of the cystine knot superfamily member protein (e.g. Neublastin) has been achieved, the protein is subjected to enzymatic digestion and disulfide analysis.

The partially reduced cysteine-b locked monomers are contacted with an enzyme that digests them to result in enzymatic peptides. Examples of enzymes that can be used include trypsin, endoproteinase Asp-N, endoproteinase Glu-C, pepsin, chymotrypsin, endoproteinase Arg-C, and endoproteinase Lys-C. Combinations of enzymes can be used in the digestion, including but not limited to the trypsin- Asp N combination used in the working examples. The enzyme or enzyme combination is contacted with the partially reduced cysteine-blocked monomers at a concentration, temperature, and for a duration to generate enzymatic peptides.

Subsequent to the enzymatic digestion, the enzymatic peptides are analyzed for the presence of disulfide- linked digested peptides. Such analysis can be performed, for example, by mass spectrometry methods such as liquid chromatography/mass spectrometry (LC/MS) or liquid chromatography/ultraviolet/mass spectrometry (LC/UV/MS). The disulfide-linked peptides observed in the analysis can then be compared to the disulfide linkages anticipated for the cystine knot superfamily member protein. The resulting disulfide information can be used in assessing the stability and efficacy of the cystine knot superfamily member protein in the analyzed sample.

The following are examples of the practice of the invention. They are not to be construed as limiting the scope of the invention in any way.

Examples

Example 1 : Neublastin Disulfide Determination

Six cysteines of Neublastin (Cys7-Cys72, Cys34-Cysl00, and Cys38-Cysl02) form a cystine knot in each monomer, while the un-paired Cys71 in the molecule connects through an inter- linked disulfide with another Neublastin molecule at Cys71 to form a homodimer (Fig. 1). Thus, Neublastin contains two cystine knots, which are linked by Cys71-Cys71.

Partial Reduction with Subsequent Enzymatic Digestion

When Neublastin was subjected to enzymatic digestion directly (by trypsin or by trypsin+Asp-N digestion, without partial reduction), two anticipated disulfides (Cys34-Cysl00 and Cys38-Cysl02) were clearly identified, while the third disulfide bond (Cys7-Cys72) was nearly undetectable and mis-matched disulfides such as Cys34-Cys38 and Cysl00-Cysl02 were observed (Fig. 2). In these cases, the formation of a highly abundant intra-linked disulfide (Cys71 connected to Cys72) was found to support the formation of the kinetically favorable disulfide between the two adjacent cysteines. However, if the formation of intra-linked disulfide (Cys71 connected to Cys72) were to occur in the original Neublastin sample (i.e., before digestion), the molecular weight of Neublastin would be only one half of the proposed structure (since the inter-chain Cys71-Cys71 would be broken). From the mass measurement of the intact molecule, as well as the migration in SDS-PAGE, Neublastin' s molecular weight was found to correspond to the molecule as proposed in Fig. 1 , thereby supporting the hypothesis that the kinetically favorable disulfide between the two adjacent cysteines (Cys71-Cys72) forms only after breaking the cystine knot. Therefore, it was expected that the mis-matched disulfides (Cys71-Cys72) with their related mis-matched forms do not occur in the original sample (without enzymatic digestion or breaking the cystine knots) and are likely an artifact induced after breaking the knots.

Because it would be difficult to characterize the exact disulfide linkages of Neublastin without enzymatic digestion, a partial reduction step was developed to be applied prior to the enzymatic digestion, to stabilize the molecule and minimize re-shuffling in the characterization. Since Cys71- Cys71 is the only inter-chain linked disulfide (and does not participate in the cystine knot disulfides), a preferential reaction would be able to reduce mainly Cys71-Cys71, which could then be blocked (via alkylation) prior to enzymatic digestion such that Cys71 could not react with the intra-linked disulfides (i.e., the cystine knot disulfides). This partial reduction step was based upon the assumption that the cystine knot bonds are stronger (more stable) than the non-cystine knot bond.

To ensure that the partial reduction preferentially reduced and alkylated the inter-chain Cys71-Cys71 disulfide at desired levels, a range of reduction conditions were developed, evaluated, and optimized by varying the time and amount of reductant (TCEP), the amount of alkylation reagent (NEM), and the enzymatic digestion conditions under the protocol. In this process the optimal conditions were determined by monitoring the mass of Neublastin monomer with only one NEM addition (Fig. 3), and also by monitoring its corresponding enzymatic peptides (after partial reduction and enzymatic digestion) containing a disulfide- linked Cys7- Cys72 peptide with Cys71 alkylated with one NEM (Fig. 4).

The reason for adding only one NEM to the monomer was that, if run correctly, the partial reduction step should result in breaking only the inter-chain linked disulfide (Cys71- Cys71), and therefore opening the molecule to only one potential addition at each monomer. For example, if more than one NEM were added to the Neublastin monomer, the reduction condition would have been too harsh in breaking other intra-chain linked (cystine knot) disulfides.

Following the same approach, the enzymatically digested molecule should also be observed with only a disulfide- linked Cys7- Cys72 peptide, with Cys71 alkylated with one NEM (further proving the partial reduction occurring at the Cys71 position).

Based on these criteria, a protocol was developed and defined from testing various combinations of partial reductions, along with the optimization of digestion efficiency. The conditions and steps of the protocol are detailed at the conclusion of this section. The resulting disulfide-containing peptides reflected the anticipated disulfides (i.e., Cys7- Cys72 disulfide with Cys71 alkylated with one NEM) as indicated in the chromatogram (Figs. 5 A and 5B). Further, the optimal conditions were repeated with two different preparations, with reproducible results (with the majority of disulfides measured within 10% variation) (Figs. 5 A and 5B and Table 2). The enzymatic digested peptides and disulfide-linked peptides were further assigned in the chromatogram (Fig. 5A), with the observed m/z, charges, and retention times indicated in Table 3. A detailed characterization of the anticipated disulfide linkages was performed for the Cys7- Cys72 disulfide with Cys71 alkylated with NEM (Fig. 6A), for the Cys34- CyslOO and Cys38-

Cysl02 disulfides (Fig. 7), and for the Cys71-Cys71 and Cys7-Cys72 disulfides (Fig. 8). The observed monoisotopic mass matched the expected peptide mass with a disulfide and also with a cysteine modified with one NEM (Fig. 6A) (i.e., a loss of 2H from the formation of a disulfide and the addition of 125.0471 Da from one NEM modification).

Table 2: Reproducibility in Generation of Disulfide-linked peptides of Neublastin (After Partial

Reduction and Enzymatic Digestion; monitored by QTOF MS)

Table 3 : Neublastin peptides (after partial reduction by NEM and digestion by Trypsin+AspN enzymes)

In the related ETD-MS2 spectrum (Fig. 6B), the disulfide bond was preferentially dissociated as expected, resulting in two dissociated peptides designated as PI and P2, thereby confirming that the two peptides (Cys7 and Cys72 linked peptides) were linked together. In the corresponding CID-MS2 spectrum (Fig. 6B), the disulfide linkage sites (Cys72- Cys7 and Cys71 modified with NEM) could be conclusively assigned by the observations of y2 and y3 ions. In addition, a repeat experiment using the high resolution-accurate mass measurement, measured by the CID and ETD-MS2 spectra using Orbitrap FTMS (Fig. 6C), provided further assurance of the linkage assignments as described in Fig. 6B. Similarly, the ETD and CID-MS2 spectra (Fig. 7) provided the assignments of Cys34- CyslOO and Cys38- Cysl02 disulfide linkages, and confirmed the assignments of Cys7- Cys72 and Cys71- Cys71 disulfide linkages (Fig. 8).

Further, the mis-cleavage forms of the disulfide-linked peptide as shown in Fig. 8 are seen, representing the same disulfide linkage but with different lengths of backbone elongation.

The protocol conditions were as follows: Neublastin Samples

Recombinant neublastin was produced by Chinese hamster ovary (CHO) cells stably transfected with a plasmid encoding the carboxyl-terminal, 104 amino acids of human neublastin (hNBN) linked to a signal peptide. hNBN is expressed as a 103/104 amino acid homodimer. Each neublastin monomer contains one asparagine-linked glycosylation site, therefore, as expressed in CHO cells, the neublastin dimer has a potential for two, one, or no glycosylations. Reagents

Guanidine hydrochloride (Sigma-Aldrich, cat# G3272-500G), Tris(2- carboxyethyl)phosphine hydrochloride (TCEP) (Thermo Scientific Pierce, cat# 20490), N- ethylmaleimide (NEM) (Sigma-Aldrich, cat# 04260-5G-F), Dimethyl sulfoxide anhydrous (Sigma-Aldrich, cat# 276855-lOOML), Formic acid optima LC/MS grade (Fisher Scientific, cat# Al 17-50), LC/MS grade Acetonitrile (Fisher Scientific, cat# A996-4), LC/MS grade water (Fisher Scientific, cat# W7-4), Trifluoroacetic acid (Fisher Scientific, cat# AC293811000), Sodium acetate anhydrous (Sigma-Aldrich, cat# S2889-250G), Tris buffer (Fisher Scientific, cat# 17-1321-01), 1M hydrochloric acid solution (Sigma-Aldrich, cat# 318949-500ML), Dithiothreitol (DTT) (Sigma-Aldrich, cat# 43815-1G), Iodoacetamide (IAM) (Sigma-Aldrich, cat# II 149-5G), and Ammonium bicarbonate (Sigma-Aldrich, cat# 09830-500G).

Endoproteinase Asp-N (Sigma-Aldrich, cat# P3303-1VL), Sequence grade modified trypsin (Promega, cat# V5111), and PNGaseF (New England Bio labs, cat# P0705L). Amicon centrifugal filters (10 kDa molecular weight cutoff (EMD Millipore, cat# UFC501096). Agilent C-18 column from Agilent (SN USRA001598) (300Extend-C18, 3.5 μιη, 2.1 x 150 mm).

Partial Reduction Procedures

A 6 M GnHCl in 0.75 M sodium acetate buffer was prepared, at pH 4.6 (adding GnHCl salt into 0.75M sodium acetate buffer, pH 4.6. fresh prepare before use). 25 uL of 6 M GnHCl in 0.75 M sodium acetate buffer was added (pH 4.6) to the 1.25 uL Neublastin sample (79.7 ug/uL).

A preparation of 0.1M TCEP in 6M GnHCl/ 0.75 M sodium acetate buffer was prepared. A 0.1M TCEP stock solution was diluted to 20 mM TCEP with 6M GnHCl/0.75 M sodium acetate buffer. 1.38 uL of the 20 mM TCEP solution was added to the Neublastin sample solution (final concentration of TCEP at 1 mM) and the sample was incubated at 37 °C for 20 min.

A 2M NEM in DMSO solution was prepared. 0.71 uL of 2M NEM solution was added to the sample (to achieve a final concentration of NEM at 50 mM), and the sample was incubated at 37 °C for 60 min in the dark.

Desalting Procedures

An HPLC fractionation approach was used for desalting on an Agilent 300Extend-C18 column with 1200 series HPLC pump coupled online with an UV detector (G1315D photodiode array and monitor at 280 nm) at room temperature. A Gilson FC 203B fraction collector (with 200 uL sample loop) was used to collect the eluent. Buffer A was 0.1% formic acid in water and buffer B was 0.1%> formic acid in acetonitrile. The desalting procedures were as follows.

The partial reduction mixture was injected into the column to first desalt the mixture in the mobile phase A (98% buffer A) for at least 1 h (to eliminate GnHCl, NEM, and other salt).

The gradient was started (from 10%B to 80%B in 7 min and held at 80%B for another 3 min) and the anticipated fraction was collected (with retention between 7 to 9 min). After the collection, the sample was dried with a speed vacuum (consider splitting the collected fraction into multiple fractions, such as 2 x 0.6 mL, to save the drying time).

50 uL of 100 mM Tris-HCl buffer (pH 7.2) was added to each dry tube and combined the sample solution. Enzyme Digestion Procedures (Trypsin+Asp-N) after Partial Reduction and Desalting Steps

An Asp-N solution was prepared by adding 8 uL 100 mM Tris-HCl (pH 7.2) buffer to a 2 ug Asp-N solution. 4 uL of the prepared Asp-N solution was added to the sample (final Asp-N concentration at 0.25 ug/uL), the protein to Asp-N ratio (w/w) was around 100: 1. 2.5 uL of trypsin (1 ug/uL) was further added to the sample to obtain a protein to trypsin ratio of around 40: 1 (w/w).

After 4 h at 37°C, a second dose of Asp-N plus Trypsin was added (in the same amount), and the sample was incubated at room temperature overnight (between 12 to 16 hrs).

The reaction was stopped by adding 5% formic acid in water. The sample was then diluted to 1 μg/μL, and 2 μg of the sample was injected for nano-LC-MS analysis.

LC-MS System for Disulfide Characterization (System 1)

Analysis of enzymatic peptides was performed on an Thermo Ultimate 3000 nano-LC pump (Mountain View, CA) and a self-packed CI 8 column (Magic CI 8, 200 A pore and 5 μιη particle size, 75 μιη i.d. x 15 cm) (Michrom Bioresources, Auburn, CA) or equivalent C18 column with similar dimension from New Objectives (Woburn, MA). The column was coupled online to a Thermo LTQ-Orbitrap-Elite-ETD mass spectrometer (San Jose, CA) through a New Objective nanospray ion source (Woburn, MA). Mobile phase A (0.1% formic acid in water) and mobile phase B (0.1% formic acid in acetonitrile) were used for the gradient consisting of (i) 20 min at 2% B for sample loading and 5 min for desalting at 0.3 I min; (ii) linear from 2 to 5% B for 2 min; (iii) linear from 5 to 40% B for 60 min; (iv) linear from 40 to 90% B for 10 min; and finally (v) isocratic at 90% B for 5 min. The flow rate of the column was maintained at 0.2 μΏ min. The LTQ-Orbitrap-Elite-ETD mass spectrometer was operated initially in the data- dependent mode as follows: survey full-scan MS spectra (m/z 300-2000) were acquired in the Orbitrap with a mass resolution of 60,000 at m/z 400, followed by operation in the data- dependent mode to switch automatically between CID-MS2 (scan 2 at the LTQ), and ETD-MS2 (scan 3 at the LTQ). After surveying the MS spectrum from m/z 300 to m/z 2000, subsequent CID-MS2 and ETD-MS2 steps were performed on the same precursor ion with a ±2.5 m/z isolation width. Any inadequate information (assignment) in the CID-MS2 and ETD-MS2 spectra was repeated by targeting the desired ions, i.e., the same precursor but with different charge state, in order to gain additional linkage information. These targeted approaches were repeated (i.e., targeting multiple charges of a precursor ion or the same disulfide- linked peptide but with different enzymatic cleavages or miscleavages) until the linkage information was adequate. In addition, a targeted CID-MS3 after ETD for the ions of interest was further performed as necessary. A 2 μg of the sample (the enzyme-digest) was used per LC-MS analysis.

Disulfide Assignment

The anticipated disulfide-linked tryptic or multi-enzyme digested peptide masses with different charges were first calculated, and then matched to the observed masses in the LC-MS chromatogram. The matched masses (with < 5 ppm mass accuracy) were further confirmed by the corresponding CID-MS2 and ETD-MS2 fragmentation, as well as by CID-MS3

fragmentation as needed.

LC-UV-MS System for Routine Monitoring (System 2)

Enzymatic peptides were monitored by an Agilent 1200 HPLC pump and an Agilent 300SB-C18 HPLC column (3.5 μιη particle size, 2.1 mm i.d. x 15 cm) coupled online with an Agilent UV detector (G1315D photodiode array) and an Agilent 6520 quadrupole time-of-flight (QTOF) mass spectrometer (Santa Clara, CA). The mobile phase A was 0.1% formic acid in water, and mobile phase B was 0.1% formic acid in acetonitrile. The separation was carried out by a linear gradient: starting from 2% B to 40% B in 40 min, increased to 80% B in 5 min, and then to 80% B in 2 min. The flow rate was maintained at 200 μΕ/ηιίη. The QTOF mass spectrometer was set to run in positive-ion mode with capillary voltage of 4 KV, gas temp at 325 °C, drying gas at 11 L/min, nebulizer at 45 psi, and m/z scan range of 350-1800 with acquisition rate of 7 spectra/sec. A 20 μg of the enzyme-digested samples was used per LC-MS analysis.

Other Embodiments

While the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and

modifications are within the scope of the following claims.