WO2013101014A2

WO2013101014A2 - High pressure treatment of aggregated interferons

Info

Publication number: WO2013101014A2
Application number: PCT/US2011/067698
Authority: WO
Inventors: David Zeng; Matthew B. Seefeldt; Mary S. Rosendahl; Shawn Kane; Jeffrey L. Cleland
Original assignee: Nuron Biotech, Inc.
Priority date: 2011-12-29
Filing date: 2011-12-29
Publication date: 2013-07-04
Also published as: WO2013101014A3

Abstract

Aspects of the instant invention provide methods that may be used to make nonglycosylated IFN-β more amenable to high pressure treatment. Such methods include, for example, pre-treatment of aggregated IFN-β that involves solubilizing the protein in a low pH solution. Additional aspects provide pharmaceutical compositions containing nonglycosylated interferon.

Description

HIGH PRESSURE TREATMENT OF AGGREGATED INTERFERONS

CROSS REFERENCE TO RELATED APPLICATIONS

This application is a continuation-in-part of U.S. Application Ser. No. 12/287,262, filed October 6, 2008, entitled "HIGH PRESSURE TREATMENT OF AGGREGATED

INTERFERONS," which claims the benefit of U.S. Provisional Application Serial No.

60/997,782, filed October 5, 2007, entitled "HIGH PRESSURE TREATMENT OF

AGGREGATED INTERFERONS", and U.S. Provisional Application Serial No. 61/130,208, filed May 29, 2008, entitled "HIGH PRESSURE TREATMENT OF AGGREGATED

INTERFERONS", each of which are incorporated herein by reference in their entirety.

FIELD OF THE INVENTION

Aspects of the present invention relate to using high pressure to treat aggregated interferons, particularly recombinant human interferon-β (rhIFN-β).

BACKGROUND OF THE INVENTION

Therapeutic proteins provide enormous potential for the treatment of human disease.

Dozens of protein therapeutics are currently available commercially, with hundreds more in clinical development. However, protein misfolding and aggregation are common problems that arise in all phases of recombinant protein production, specifically during fermentation, purification, and long-term storage.

During production of recombinant protein via fermentation, protein instability commonly leads to extensive aggregation. Within prokaryote host cells such as E. coli, the reducing environment within the cytoplasm prevents the proper formation of disulfide bonds and commonly results in the formation of insoluble inclusion bodies of protein in non-native structures. Inclusion body formation is additionally fostered by the overexpression of the recombinant protein of interest in host cells.

Chemical protein denaturants (chaotropes such as urea or guanidine HC1 or ionic surfactant such as sodium dodecyl sulfate, SDS) have been traditionally used to solubilize and refold proteins from inclusion bodies. The refolding process can be difficult to optimize and control and in many cases refolding is not viable due to the formation of aggregate-prone intermediates and subsequent re-aggregation.

Proper disulfide bond formation is another component of a refolding process that needs to occur to generate a biologically active pharmaceutical composition. The formation of native disulfide bond(s) can often be confounded by competing non-native disulfide bond(s) that can lead to aggregates. Disulfide shuffling agents (reduced/oxidized glutathione, cysteine/cystine, and cysteamine/cystamine) have been used extensively for the refolding of proteins that contain multiple disulfide bonds.

High hydrostatic pressure (c.a. 2000 bar) has also been shown to be an effective refolding tool, enabling refolding at relatively high protein concentrations and with high yield.

See U.S. Pat. Nos. 7,064,192 and 6,489,450, the contents of which are incorporated herein by reference. In contrast to traditional chaotrope-based refolding, high pressure techniques can dissociate aggregates under conditions that favor the protein's native conformation.

Additionally, high pressure-induced refolding can be conducted in the absence of chaotropes or strong protein-binding surfactants, facilitating subsequent downstream purification of the protein.

The interferons are a family of glycoproteins whose secretion from cells is induced by a number of signals, including viruses, double-stranded RNAs, other polynucleotides, antigens, and mitogens. Interferons exhibit multiple biological activities, including antiviral, antiproliferative, and immunomodulatory activities. At least three distinct types of human interferons, α, β, and γ, have been identified.

Human interferon-beta (hIFN-β) and variants thereof are therapeutic proteins used for the treatment of multiple sclerosis. Human IFN-β tends to be glycosylated when purified from natural sources, but can be de-glycosylated. Human IFN-β made via recombinant techniques with expression in E. coli or via chemical synthesis tends to be non-glycosylated.

A commercially important variant of human IFN-β modifies the native amino acid sequence in two ways. First, the cysteine residue at the 17 position is replaced with serine. Second, the methionine residue at the N-terminus is deleted. In native human IFN-β, the cysteine- 17 residue does not form a disulfide bond, hence its replacement with a serine residue is beneficial in that it can prevent the formation of competing non-native disulfide bond during the refolding process. The removal of the methionine residue at the N-terminus is a consequence of producing the protein in E. coli cells where the endogenous enzyme methionine aminopeptidase hydrolyzes the N-terminal methionine residue.

The use of non-glycosylated versions of human IFN-β or variants thereof as a therapeutic agent is desirable. Expression in E. coli, which tends to produce non-glycosylated IFN-β, is significantly easier and less expensive than expression in mammalian cell systems, which tend to produce glycosylated forms. One major obstacle that must be overcome in the use of non-glycosylated human IFN-β or variants thereof from E. coli as a therapeutic agent, however, concerns solubilization and refolding of aggregated IFN- β in inclusion bodies.

Proteins expressed in prokaryotic host cells such as E. coli tend to be entrapped in inclusion bodies in aggregated, misfolded, and insoluble state. It is necessary to solubilize and refold the protein to convert it into native and therapeutically useful protein.

One conventional process for the refolding and production of non-glycosylated IFN-β is described in U.S. Pat. No. 4,462,940. A significant disadvantage of such a conventional refolding method, however, is that it relies substantially upon the cationic surfactant sodium dodecyl sulfate (SDS) throughout the refolding and purification process. SDS has long been known to be a denaturing surfactant, enabling non-native and aggregated proteins to remain in solution. Since SDS can solubilize most proteins, the refolding method is prone to retaining large amounts of E. coli host cell proteins as impurities. The denaturing effects of SDS also result in re-aggregation once the denaturant is removed. This results in the formation of soluble aggregates that are difficult to purify and can contain objectionable amounts of residual SDS.

There are publications of clinical complications associated with IFN-β product with significant IFN-β aggregates. Protein aggregates are often not recognized as "natural" by the recipient's immune system (possibly by exposure of new epitopes on the protein in the aggregate which are not found in non-aggregated protein, or by a hapten effect that stimulates B-cells, with the result that the immune system is sensitized to the administered recombinant protein aggregate). In many instances, the immune system produces antibodies that bind to the aggregates, which do not neutralize the therapeutic effect of the protein. However, in some cases, IFN^-binding antibodies are produced that bind to the recombinant protein and interfere with the therapeutic activity thereby resulting in declining efficacy of the therapy.

Furthermore, in some instances, repeated administration of a recombinant protein with aggregates can cause acute and chronic immunologic reactions. Neutralizing antibodies have been shown to develop in patients treated with IFN-β with aggregates, likely due to the presence of aggregates in the pharmaceutical product. Soluble aggregates in IFN-β product could be the source of the decreased efficacy and immunogenicity issues.

Another complication associated with IFN-β product containing HSA is that HSA can contain aggregates and poses a risk of viral contamination. The HSA used in pharmaceutical compositions is often obtained from human donors and purified using Cohn fractionation and thus poses a constant risk of viral contamination to this product. Furthermore, the viral inactivation treatment (heating at 60°C for 10 hours) used for the HSA protein can cause its aggregation.

An improved, HSA-free formulation of non-glycosylated IFN-β has been described in U.S. Patent No. 7,371,373. However the aggregate content of this material is greater than 6% and can be even higher depending upon factors including pH, ionic strength, and other agents present in the formulation. U.S. Patent No. 7,544,354 describes a chaotrope-based refolding method that leads to completely monomeric protein. However, no mention is made of oxidation methods, and low pHs are used during refolding which quench disulfide bond formation. Neither was any mention of the refolded interferon resembling the native protein and being biologically active. Consequently, the method described in these patents could provide mostly monomeric protein, but not active material with the appropriate disulfide bonding. U.S. Pat. No. 4,530,787 discusses the need for oxidation and describes the use of the oxidative agent iodosobenzoic acid for the formation of disulfide bonds.

SUMMARY OF THE INVENTION

In certain aspects, the present invention provides a method of preparing a composition comprising non-glycosylated interferon to minimize the presence of inclusion bodies and aggregated non-glycosylated interferon. In certain embodiments, the method comprises solubilizing the aggregated interferon in a solubilizing media having a pH less than 7, incorporating the solubilized interferon into a refolding admixture, applying an amount of pressure to the refolding admixture; and precipitating impurities from the refolding admixture using a salt or acid precipitation. Additional aspects of the invention relate to pharmaceutical compositions comprising a nonglycosylated interferon. In certain aspects, the interferon comprises less than about 5 weight percent of protein aggregation. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows the amino acid sequence for SEQ ID No. 2.

FIG. 2 shows the gene sequence for SEQ ID No. 3, IFN-β C17S with the following underlined recognition sequences: Ndel , Hindlll, and EcoRI.

FIG. 3 provides a Box-Beckham SED optimization of zwittergent concentration, protein concentration, and refolding time on process yield. The plot demonstrates the relationship of zwittergent and protein concentration at 2 hours.

FIG. 4 provides SDS-PAGE results of samples before (lane 2) and after (lane 3) clarification to remove acid- induced precipitate. MW markers are located in lane 1. The pH 6.0 treatment induced precipitation of high molecular weight impurity species without a sub stantial lo s s of IFN- β .

FIG. 5 provides Blue Sepharose column elution chromatogram where - protein impurities are eluted on the leading edge of the IFN-β elution peak.

FIG. 6 provides a representative chromatogram of Cu IMAC elution.

FIG. 7 provides an SP Sepharose Column elution chromatogram.

FIG. 8 provides a flow chart diagram of the refold/clarification conditions.

FIG. 9 provides a flow chart diagram of the Blue Column Purification Process.

FIG. 10 provides a flow chart diagram of the IMAC Column Purification Process.

FIG. 11 provides a flow chart diagram of the SP Sepharose Column Process.

FIG. 12 SDS-PAGE image of interferon beta- lb samples at various stages of purification. The lane numbers correspond to (1) MW markers with molecular weights in kilo daltons indicated, (2) to (4) interferon beta- lb standard, (5) Clarified Refold, (6) Blue Column Pool, (7) IMAC Column Pool, (8) SP Column Pool, (9) Ultrafiltration- 1 Retentate, (10) Diafiltration Retentate, (11) Ultrafiltration-2 Retentate, and (12) Final Purified interferon beta- lb.

FIG. 13 provides a reverse phase HPLC chromatogram of a final purified interferon beta-lb. The interferon beta-lb had a purity of 97.1% by the RP-HPLC analysis. FIG. 14 provides a size exclusion HPLC chromatogram of a final purified interferon beta- lb that was >99 monomeric (eluted at 12.3 min).

DETAILED DESCRIPTION OF THE INVENTION

The present invention, in certain aspects, relates to using high pressure to treat aggregated interferons, particularly recombinant human interferon-β (IFN-β). More particularly, aspects of the present invention relate to improved conditions for solubilization of IFN-β in inclusion bodies and improved media for high pressure refolding to reduce the aggregate content of interferon material, particularly recombinant human interferon- β. Highly pure, soluble, monomeric, and biologically active recombinant interferon-β is prepared and purified in representative embodiments. The processes of the present invention are useful in both the small scale and large scale manufacturing of IFN-β, including but not limited to bulk process manufacturing.

In one embodiment, the present invention relates to methods of preparing a composition of non-glycosylated interferon that maximizes the solubilization of inclusion bodies and aggregated non-glycosylated interferon. The composition is then adjusted into a refolding admixture for the high pressure refolding of interferon. Solubilization of inclusion bodies and interferon preferably occurs in a solubilizing media having a pH below about 7, at a pH between about 1 and about 4, or about 2.8. The solubilization buffer may also include at least one surfactant, which may be a zwitterionic surfactant such as Zwittergent SB 3-14. In further embodiments, the solubilizing media may optionally include at least one chemical protein denaturant such as, for example, urea. Once solubilized, the interferon may be incorporated into a refolding mixture. The refolding mixture is preferably provided at a pH between about 7 and about 11, and in certain embodiments is between about 7.5 and about 9.5, which is controlled using one or more known buffering compositions. In certain embodiments, the refolding mixture includes at least one oxidizing/reducing agent that is adapted to promote the formation of disulfide bond. Such agents may include one or more disulfide shuffling agents such as cysteine (a reducing agent that works with dissolved oxygen as an oxidizing agent), a cysteine/cystine pair, or one or more agents otherwise defined herein.

The amount of pressure applied to the refolding mixture is at least about 3000 bars and the resulting yield of non-glycosylated, disaggregated interferon is about 40%. In certain embodiments, the yield of non-glycosylated interferon is about 75% or about 90%. Based on these methods the resulting composition has less than about 5 percent of aggregated IFN-β protein after high pressure refolding, as determined by size exclusion high performance liquid chromatography (SE-HPLC).

After high pressure refolding, the protein can be further purified before being placed into a composition suitable for pharmaceutical use. Purification of IFN-β may be

accomplished using any of the methods known to one skilled in the art. In one embodiment, the protein impurities are precipitated by adjusting the pH with an acid, such as, but not limited to, phosphoric acid, citric acid, succinic acid, acetic acid, or one or a combination of acids provided herein. The precipitated impurities may be removed by centrifugation or filtration. In alternative or additional embodiments, impurities or the protein of interest may be "salted out" of the solution, in accordance with the teachings herein.

Further purification may be achieved using a column chromatography or a series of column chromatographic methods. A series of three columns, Blue Sepharose Affinity Column, Cu Immobilized Metal Ion Affinity Chromatography (IMAC) Column, and SP or CM Sepharose Cation Exchange Column, is one illustrative purification technique. In such an example, the Blue affinity column captures the interferon out of the refolding solution, and the other two columns help to further improve the purity of the interferon. These purification columns may be used in accordance with the methods discussed herein or in different orders. In particularly preferred methods, a Blue Sepharose affinity column is used for elution and is pre-washed with a low ionic strength buffer prior to elution. Applicants have surprisingly found that the low ionic strength pre-wash removes impurities, such as host cell proteins (HCP), resulting in a highly purified product.

As a consequence of using the pre- solubilization discussed herein, disaggregated interferon have been generated from inclusion bodies with yields of > 90%. As a consequence of using the high pressure treatment discussed herein, refolded monomeric and biologically active interferon have been produced. The purification techniques, as discussed herein, substantially improve impurity removal, such as host cell proteins, beyond techniques that are presently used. Additional advantages will be apparent to one of ordinary skill in the art based upon the disclosure and examples provided herein.

The following definitions are used in this specification: "Aggregated" with respect to a protein refers to protein material composed of a multiplicity of protein molecules wherein non-covalent interactions and/or intermolecular covalent bonds such as disulfide bonds hold the protein molecules together. Typically, but not always, an aggregate contains sufficient number of molecules so that it is insoluble in aqueous medium at physiological pH. Inclusion bodies are a type of aggregate of particular interest, to which aspects of the present invention may be applicable.

"Atmospheric pressure" (ambient) is approximately 15 pounds per square inch (psi) or

1 bar.

"Biologically active" means a protein or variant thereof has at least 10% of maximal known specific activity as measured in an assay that is generally accepted in the art to be correlated with the known or intended utility of the protein. For proteins intended for therapeutic use, the assay of choice is one accepted by a regulatory agency to which data on safety and efficacy of the protein must be submitted. For interferon-β, the assay of choice is Cytopathic Effect (CPE) assay which is listed in European Pharmacopoeia monograph method 5.6, Assay of Interferon. Interferon-β protects certain cells from viral infection. In this assay, A549 cells are treated with the test sample and then challenged with encephalomyocarditis virus. The cell viability is determined after viral challenge and the biological activity is calculated in millions of international units (MIU) per mg of interferon protein. A standard curve is generated using a reference standard such as the reference standard 00/574 from National Institute for Biological Standards and Control (NIBSC). A protein having greater than 10% of maximal known specific activity is "biologically active".

"Chaotropic agent" is a compound, including, without limitation, guanidine

hydrochloride (guanidinium chloride, GdmCl), sodium thiocyanate, urea and/or anionic detergents such as SDS which disrupts the non-covalent intermolecular bonding within the protein, permitting the peptide chain to assume a substantially random conformation.

"Denatured" as applied to a protein in the present context, means that native secondary and tertiary structures are disrupted to an extent that the protein does not have biological activity.

"Glycosylated" describes the process or result of covalently linking saccharides to proteins. Two types of glycosylation can exist: N-linked glycosylation to the amide nitrogen of asparagine side chains and O-linked glycosylation to the hydroxy oxygen of serine and threonine side chains.

"Heterologous proteins" are proteins which are normally not produced by a particular host cell. Recombinant DNA technology has permitted the expression of relatively large amounts of heterologous proteins (for example, growth hormone) from transformed host cells such as E. coli cells. These proteins are often sequestered in insoluble inclusion bodies in the cytoplasm and/or periplasm of the host cell. The inclusion bodies or cytoplasmic aggregates contain, at least in part, the heterologous protein to be recovered. These aggregates often appear as bright spots under a phase contrast microscope.

"Host cell" refers to a microbial cell such as bacteria and yeast or other suitable cell including animal or a plant cell which has been transformed to express the heterologous protein of interest. Host cells which are contemplated by aspects of the present invention are those in which the heterologous protein expressed by the cell is sequestered in refractile bodies. An exemplary host cell is E. coli K12, strain W3110G [pBGHI], which has been transformed to effect expression of the desired heterologous protein.

"IFN or IFNs" refers to the family of secreted proteins known as interferons, which are cytokines with antiviral, anti-protozoal, immunomodulatory, and cell growth regulatory activities. IFNs were originally classified by their sources: leukocytes (IFN-a-1 and IFN-a-2), fibroblasts (IFN-β), and immune cells (IFN-γ). See, for example, Peskta, S. (1986) supra; Sen, G. C. and Lengyel, P. (1992) supra; and Pestka, S., ed., Interferons Part C in Meth. Enzymol. Vol. 119, Academic Press, Inc., New York, N.Y. (1986).

"IFN-β" refers to fibroblast IFN, which in man is derived from a single gene lacking introns. The DNA or polynucleotide sequence of human IFN-β is described in Taniguchi, T. et al. (1980a) supra and U.S. Pat. No. 5,326,859. The human IFN-β cDNA encodes a pro- polypeptide of 187 amino acids in length. A 21 amino acid signal sequence is cleaved off to form the mature, secreted IFN-β which consists of 166 amino acids.

"ΙΡΝ-β-la" refers to a recombinant human IFN-β, expressed in Chinese hamster ovary ("CHO") cells. As shown in FIG. 1, mature secreted ΙΡΝ-β-la is 166 amino acids in length, corresponding to native IFN-β. ΙΡΝ-β-la is N-linked glycosylated at the asparagine residue at position 80 (Asp80). See, for example, Innis, M. A. and McCormick, F. et al. (1986) supra and U.S. Pat. No. 4,966,843. "IFN-P-lb" refers to recombinant human IFN-β expressed in E. coli host cells and having a cysteine-to-serine amino acid substitution at position 17 (Serl7). When IFN-P-lb is processed in E. coli and the N-terminal methionine is removed, it is 165 amino acids in length with Ser2 at the N-terminus. IFN-P-lb is not glycosylated. See, for example, Mark, D. F. et al. (1984) supra and U.S. Pat. No. 4,588,585.

"Inclusion bodies" are aggregated proteins that form during overexpression of recombinant proteins in E. coli.

"Native conformation" of a protein, in the present context, refers to the secondary, tertiary and quaternary structures of a protein as it occurs in nature in a biologically active state.

"Native" or "naturally occurring" proteins or polypeptides refer to proteins or polypeptides recovered from a source occurring in nature. The term "native IFN-β" or "naturally occurring IFN-β" would include native or naturally occurring IFN-β and fragments thereof, and would include post-translational modifications of IFN-β and fragments thereof, including, but not limited to, acetylation, carboxylation, glycosylation, phosphorylation, lipidation, acylation, and cleavage.

"Non-glycosylated" means a protein that does not contain any covalently linked saccharides.

"Recombinant proteins or polypeptides" refer to proteins or polypeptides produced by recombinant DNA techniques, i.e., produced from cells, microbial or mammalian, transformed by an exogenous recombinant DNA expression construct encoding the desired protein or polypeptide. Proteins or polypeptides expressed in most bacterial cultures will typically be non-glycosylated. Proteins or polypeptides expressed in yeast may have a glycosylation pattern different from that expressed in mammalian cells.

"Refolding" (renaturing, naturing), in the present context, means that a fully or partially denatured protein adopts secondary, tertiary and quaternary structure like that of the cognate native molecule. A (properly) refolded protein has biological activity which is substantially higher than that of the non-refolded molecule. Where the native protein has disulfide bonds, oxidation to form native intramolecular disulfide bonds is a desired component of the refolding process. "Subject" is used to mean an animal, preferably a mammal, including a human or non- human. The terms patient and subject may be used interchangeably.

"Therapeutically effective amount" as used herein shall mean that drug dosage that provides the specific pharmacological response for which a drug is administered in a significant number of subjects in need of such treatment. It is emphasized that "therapeutically effective amount," administered to a particular subject in a particular instance will not always be effective in treating the conditions described herein, even though such dosage is deemed a "therapeutically effective amount" by those skilled in the art.

"Treating" or "treatment" as used herein covers the treatment of disease-state in a mammal, preferably a human. In a preferred embodiment, the disease-state to be treated is characterized by symptoms associated with multiple sclerosis (MS), such as weakness, numbness, tremor, loss of vision, pain, paralysis, loss of balance, bladder and bowel dysfunction, and cognitive changes (primary symptoms); repeated urinary tract infections, disuse weakness, poor postural alignment and trunk control, muscle imbalance, decreased bone density, shallow, inefficient breathing, and bedsores (secondary symptoms); and depression (tertiary symptoms), and includes:

(i) inhibiting the condition, i.e., arresting its development; or

(ii) relieving the condition, i.e., causing regression of the condition.

"Zwitterionic" refers to a compound that is electrically neutral but carries formal positive and negative charges on different atoms. Zwitterions are polar and usually have a high solubility in water.

"Zwittergent" refers to a detergent or surfactant that is zwitterionic.

The embodiments of the present invention described herein are not intended to be exhaustive or to limit the invention to the precise forms disclosed in the detailed description. Rather, the embodiments are chosen and described so that others skilled in the art may appreciate and understand the principles and practices of the present invention.

Aspects of the present invention relate to improved strategies and media to accomplish high pressure disaggregation and refolding beneficially applied to non-glycosylated interferons and variants thereof. Such a process results in significant increases in overall disaggregated protein yields and higher purity product. Desirably, non-glycosylated interferon proteins used in aspects of the present invention are non-glycosylated versions of the α, β, and/or γ human interferons or variants thereof. Most desirably, the interferon proteins are non-glycosylated versions of human interferon-β

("nongly-IFN-β") or variants thereof such as interferon beta- lb. IFN-β polypeptides may be glycosylated or non-glycosylated. It has been reported in the literature that both the glycosylated and non-glycosylated IFN^'s show qualitatively similar specific activities and that, therefore, the glycosyl moieties may not be involved in and contribute to the biological activity of IFN-β. Therefore, the non-glycosylated IFN^'s have substantially similar biological activity as otherwise identical IFN^'s that are naturally glycosylated.

The principles of the present invention preferably are applicable to human non- glycosylated IFN-β and variants thereof such as the protein according to SEQ ID No. 1

MSYNLLGFLQRSSNFQCQKLLWQLNGRLEYCLKDRMNFDIPEEIKQLQQFQKEDAALTIYEML QNIFAIFRQDSSSTGWNETIVENLLANVYHQINHLKTVLEEKLEKEDFTRGKLMSSLHLKRYY GRILHYLKAKEYSHCAWTIVRVEILRNFYFINRLTGYLRN (SEQ ID No.: 1)

Biologically active with respect to variants of IFN-β preferably means that the variants retain IFN-β activities, particularly the ability to bind to IFN-β receptors. In preferred embodiments the IFN-β variant retains at least about 25%, about 50%, about 75%, about 85%, about 90%, about 95%, about 98%, about 99% or more of the biological activity of the polypeptides whose amino acid sequences are given in SEQ ID NO: l with respect to the cytopathic effect (CPE) assay (Peska, Methods in Enzymology, vl l9, pg. 14-23, 1986). IFN-β variants whose activity is increased in comparison with the activity of the polypeptides shown in SEQ ID NO: l are also encompassed. The biological activity of IFN-β variants can be measured by any method known in the art. Examples of such assays can be found in Fellous et al. (1982) Proc. Natl. Acad. Sci USA 79:3082-3086; Czerniecki et al. (1984) J. Virol. 49(2):490-496; Mark et al. (1984) Proc. Natl Acad. Sci. USA 81:5662-5666; Branca et al. ( 981) Nature 277:221-223; Williams et al. (1979) Nature 282:582-586; Herberman et al. (1979) Nature 277:221-223; Anderson et al. (1982) J. Biol. Chem. 257(19): 11301-11304; and the IFN-β potency assay described herein (see Example 15).

Variants include biologically active fragments of the peptide shown in SEQ ID No. 1 or biologically active fragments of variants of the peptide shown in SEQ ID No. 1. These biologically active fragments or truncated forms of IFN-β or variants thereof are generated in any convenient manner such as by scission of the peptide itself or such as by removing amino acid residues from the full-length IFN-β amino acid sequence using chemical, enzymatic, or recombinant DNA techniques well known in the art.

Variants also include biologically active embodiments in which one or more amino acid residues not essential to biological activity are deleted, replaced, or added, including when such deletions, replacements, and additions enhance biological activity. One such variant includes a modification of the human IFN-β sequence shown in SEQ ID NO: l, wherein one or more cysteine residues that are not essential to biological activity have been deliberately deleted or replaced with other amino acids to eliminate sites for either intermolecular crosslinking or incorrect intramolecular disulfide bond formation. IFN-β variants of this type include those containing a glycine, valine, alanine, leucine, isoleucine, tyrosine, phenylalanine, histidine, tryptophan, serine, threonine, or methionine substituted for the cysteine found at amino acid 17 of the mature native amino acid sequence. Serine and threonine are the more preferred replacements because of their chemical analogy to cysteine. Serine substitutions are most preferred.

Another such variant includes a modification of the human IFN-β sequence shown in SEQ ID NO: l, wherein the amino-terminal methionine is removed. In recombinant synthesis, this is accomplished using widely known techniques by which the terminal methionine is removed by the E. coli enzyme methionine aminopeptidase so that the amino-terminal sequence is Ser-Tyr-Asn . One such variant is the variant according to SEQ ID No. 2:

SYNLLGFLQRSSNFQSQKLLWQLNGRLEYCLKDRMNFDIPEEIKQLQQFQKEDAALTIYEMLQ NIFAIFRQDSSSTGWNETIVENLLANVYHQINHLKTVLEEKLEKEDFTRGKLMSSLHLKRYYG RILHYLKAKEYSHCAWTIVRVEILRNFYFINRLTGYLRN (SEQ ID NO: 2) wherein the amino acid residue at position 17 is changed from cysteine to serine, and the terminal methionine in the native sequence is deleted. A common designation for this sequence is IFN-β C17S, indicating the replacement of the cysteine residue of position 17 with a serine. The calculated molecular weight of this variant is about 19880 and the theoretical pi is 9.02. The skilled artisan will appreciate that other biologically active variants can be derived by introducing additional additions, deletions, replacements or modifications to amino acids of the native human IFN-β sequence according to SEQ ID No. 1 or variants thereof. For example, in addition to position 17, other conservative amino acid substitutions may be made at one or more predicted, preferably nonessential amino acid residues. A "nonessential" amino acid residue is a residue that can be altered from the wild- type sequence of IFN-β without unduly altering its biological activity, whereas an "essential" amino acid residue is required for biological activity. A "conservative amino acid substitution" often is one in which the amino acid residue is replaced with an amino acid residue having a similar side chain. Families of amino acid residues having similar side chains have been defined in the art. These families include amino acids with basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), uncharged polar side chains (e.g., glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine), nonpolar side chains (e.g., alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), beta-branched side chains (e.g., threonine, valine, isoleucine), and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine).

Alternatively, variant IFN-β sequences can be made by introducing amino acid modifications, replacements, deletions, or additions randomly along all or part of the entire IFN-β peptide chain. These mutants could be used to generate fusion proteins, PEGylated proteins or others as known by one skilled in the art. The resultant mutants can be screened for IFN-β biological activity to identify mutants that retain activity. Following mutagenesis, the encoded protein can be expressed recombinantly, and the activity of the protein can be determined using standard bioassay techniques described herein. Biologically active variants of IFN-β will generally have at least 80%, more preferably about 90% to about 95% or more, and most preferably about 96% to about 99% or more amino acid sequence identity to the amino acid sequence of mature native IFN-β of SEQ ID No. 1, which serves as the basis for comparison. By "sequence identity" is intended the same amino acid residues are found within the variant polypeptide and the polypeptide molecule that serves as a reference when a specified, contiguous segment of the amino acid sequence of the variant is aligned and compared to the amino acid sequence of the reference molecule. For purposes of optimal alignment of the two sequences for the purposes of sequence identity determination, the contiguous segment of the amino acid sequence of the variant may have additional amino acid residues or deleted amino acid residues with respect to the amino acid sequence of the reference molecule. The contiguous segment used for comparison to the reference amino acid sequence will comprise at least 20 contiguous amino acid residues.

Corrections for increased sequence identity associated with inclusion of gaps in the variant's amino acid sequence can be made by assigning gap penalties. Methods of sequence alignment are well known in the art.

Thus, the determination of percent identity between any two sequences can be accomplished using a mathematical algorithm. One preferred non-limiting example of a mathematical algorithm utilized for the comparison of sequences is the algorithm of Myers and Miller (1988) Comput. Appl. Biosci. 4: 11-7. Such an algorithm is utilized in the ALIGN program (version 2.0), which is part of the GCG alignment software package. A PAM120 weight residue table, a gap length penalty of 12, and a gap penalty of 4 can be used with the ALIGN program when comparing amino acid sequences. Another preferred, non-limiting example of a mathematical algorithm for use in comparing two sequences is the algorithm of Karlin and Altschul (1990) Proc. Natl. Acad. Sci. USA 90:5 873-5877, modified as in Karlin and Altschul 7(1993) Proc. Natl. Acad. Sci USA 90:5873-5877. Such an algorithm is incorporated into the NBLAST and XBLAST programs of Altschul et al. (1990) J. Mol. Biol. 215:403-410. BLAST amino acid sequence searches can be performed with the XBLAST program, score=50, wordlength=3, to obtain amino acid sequence similar to the polypeptide of interest. To obtain gapped alignments for comparison purposes, gapped BLAST can be utilized as described in Altschul et al. (1997) Nucleic Acids Res. 25:3389-3402. Alternatively, PSI-BLAST can be used to perform an integrated search that detects distant relationships between molecules. See Altschul et al. (1997) supra. When utilizing BLAST, gapped BLAST, or PSI-BLAST programs, the default parameters can be used. See www.ncbi.nlm.nih.gov. Also see the ALIGN program (Dayhoff (1978) Atlas of Protein Sequence and Structure 5:Suppl. 3, National Biomedical Research Foundation, Washington, D.C.) and programs in the Wisconsin Sequence Analysis Package, Version 8 (available from Genetics Computer Group, Madison, Wis.), for example, the GAP program, where default parameters of the programs are utilized. When considering percentage of amino acid sequence identity, some amino acid residue positions may differ as a result of conservative amino acid substitutions, which do not affect properties of protein function. In these instances, percent sequence identity may be adjusted upwards to account for the similarity in conservatively substituted amino acids. Such adjustments are well known in the art. The non-glycosylated interferon used in the practice of aspects of the present invention can be obtained from natural or non-natural sources. Natural interferon can be sourced from humans, other primates, dogs, cats, rabbits, goats, sheep, bovines, equines, porcines, avians, and the like. Glycosylated forms of natural interferon proteins such as IFN-β harvested from natural sources can be de-glycosylated to form non- glycosylated interferons. Laura Runkel et al., "Structural and Functional Differences Between Glycosylated and Non-Glycosylated Forms of Human Interferon- β (IFN-β)" Pharmaceutical Research Vol. 15, No. 4, 1998, pages 641-649.

In preferred embodiments of the present invention, the IFN-β is recombinantly produced. By "recombinantly produced IFN-β" is intended IFN-β or variants that has comparable biological activity to mature native IFN-β and that has been prepared by recombinant DNA techniques. IFN-β can be produced by culturing a host cell transformed with an expression vector comprising a nucleotide sequence that encodes an IFN-β polypeptide or variants. The host cell is one that can transcribe the nucleotide sequence and produce the desired protein, and can be prokaryotic (for example, E. coli) or eukaryotic (for example a yeast, insect, plant, or mammalian cell). Examples of recombinant production of IFN-β are given in Mantei et al. (1982) Nature 297: 128; Ohno et al. (1982) Nucleic Acids Res. 10:967; Smith et al. (1983) Mol. Cell. Biol. 3:2156, and U.S. Pat. Nos. 4,462,940, 5,702,699, and 5,814,485; herein incorporated by reference. Human interferon genes have been cloned using recombinant DNA ("rDNA") technology and have been expressed in E. coli (Nagola et al. (1980) Nature 284:316; Goeddel et al. (1980) Nature 287:411; Yelverton et al. (1981) Nuc. Acid Res. 9:731; Streuli et al. (1981) Proc. Natl. Acad. Sci. U.S.A. 78:2848). Alternatively, IFN-β can be produced by a transgenic animal or plant that has been genetically engineered to express the IFN-β protein of interest in accordance with methods known in the art.

Proteins or polypeptides that exhibit native interferon- β -like properties may also be produced with rDNA technology by extracting poly-A-rich 12S messenger RNA from virally induced human cells, synthesizing double-stranded cDNA using the mRNA as a template, introducing the cDNA into an appropriate cloning vector, transforming suitable microorganisms with the vector, harvesting the microorganisms, and extracting the interferon- β therefrom.

Alternatively, IFN-β can be synthesized chemically, by any of several techniques that are known to those skilled in the peptide art. See, for example, Li et al. (1983) Proc. Natl. Acad. Sci. USA 80:2216-2220, Steward and Young (1984) Solid Phase Peptide Synthesis (Pierce Chemical Company, Rockford, 111.), and Baraney and Merrifield (1980) The Peptides: Analysis, Synthesis, Biology, ed. Gross and Meinhofer, Vol. 2 (Academic Press, New York, 1980), pp. 3-254, discussing solid-phase peptide synthesis techniques; and Bodansky (1984) Principles of Peptide Synthesis (Springer- Verlag, Berlin) and Gross and Meinhofer, eds.

(1980) The Peptides: Analysis, Synthesis, Biology, Vol. 1 (Academic Press, New York), discussing classical solution synthesis. IFN-β can also be chemically prepared by the method of simultaneous multiple peptide synthesis. See, for example, Houghten (1984) Proc. Natl. Acad. Sci. USA 82:5131-5135; and U.S. Pat. No. 4,631,211.

Note that recombinant IFN-β expressed in bacterial cells tends to be produced in non- glycosylated form. Thus, inclusion bodies of IFN-β are non-glycosylated as synthesized. Similarly, chemically synthesized IFN-β tends to be produced in non-glycosylated form unless affirmative reaction steps are carried out to accomplish glycosylation.

In a representative mode of practice, an aggregated mixture of non-glycosylated-IFN-β ("nongly-IFN-β") beneficially processed in the practice of the present invention is obtained by recovering inclusion bodies from E. coli pellets derived from E. coli host cells in which the nongly-IFN-β had been recombinantly synthesized. Inclusion bodies tend to be completely aggregated, with high levels of non-native intermolecular beta-sheet secondary structures.

The inclusion bodies (also referred to as refractile bodies) can be recovered from these pellets by any suitable technique such as those described, for example, in U.S. Pat. No.

4,652,630. High pressure homogenizers are commercially marketed to carry out cell lysis to recover inclusion bodies for this purpose. For example, the host cell can be disrupted by mechanical means such as a Manton-Gaulin homogenizer or French press. It is preferred that the disruption process be conducted so that cellular debris from the host organism is so disrupted that it fails to sediment from the homogenate solution under low speed centrifugation sufficient to sediment the inclusion bodies. The inclusion bodies are preferably resuspended, washed and centrifuged again. The supernatant is discarded yielding a substantially pure preparation of inclusion bodies. Although not critical to the practice of the present invention, it is preferred that the inclusion body preparation be homogenized again to ensure a freely dispersed preparation devoid of agglomerated inclusion bodies. The preparation may be homogenized in a Manton-Gaulin homogenizer at 3000-5000 psig. When using a high pressure homogenizer for cell lysis, the whole cells can be suspended in a 20 mM Tris, 2 mM EDTA buffer prior to processing. Chemical methods can also be used to disrupt the cells and recover the inclusion bodies. The Examples below include a representative chemical methodology for carrying out this kind of recovery.

Prior to high pressure refolding, inclusion bodies are first solubilized or are at least partially solubilized. Pre-solubilization, while not required for high pressure refolding of other proteins, is advantageous with IFN-β because it leads to increased yields. The low solubility of IFN-β aggregates was found to hinder proper refolding when it is treated with high pressure directly from inclusion body isolation. This is overcome by pre-solubilizing the inclusion bodies.

Solubility of inclusion bodies may be achieved using any method known in the art. With IFN-β, it was discovered that higher yields can be achieved in a low pH solubilizing media that includes the presence of one or more surfactants or detergents and in the absence of an organic solvent, such as, but not limited to, aliphatic alcohols. The pH of the solubilization media may be any suitable acidic pH to achieve the desired solubilization. The pH is not too acidic to risk acid-induced degradation of the protein material, and is not too basic to inhibit solubilization. In one embodiment, the pH is less than 7, often less than about 6. While it is desirable that the solubilizing media have a pH of about 2.8, the instant invention is not so limiting and the pH may be in the range from about 1 to about 4.

A wide variety of buffers or strong acids, may be used to control the pH of the solubilization media. In one embodiment, pH may be controlled by a strong acid, which may include any acid or combination of acids that provide the desired pH and improve solubility of the protein without affecting protein structure. Such strong acids may include, but are not limited to, hydrochloric acid, phosphoric acid, nitric acid, sulfuric acid, acetic acid, citric acid, and combinations thereof. In preferred embodiments, the strong acid is hydrochloric acid (HC1). Strong acid concentrations may be provided in any amount to achieve the desired pH, as exemplified herein or otherwise apparent to one of ordinary skill in the art.

Surfactants or detergents used in conjunction with the low pH solubilizing media may include any compound or combination of compounds that reduce the surface tension of the aqueous liquid carrier and similarly facilitate, at least partially, aggregate dissolution.

Surfactants may be anionic, cationic, nonionic, zwitterionic, or mixtures of these.

Representative examples include t-octylphenoxypolyethoxy-ethanol; polyoxyethylene sorbitan; sodium dodecyl sulfate; sodium lauroyl sarcosinate; cetylpyridinium chloride;

deoxycholate; sodium octyl sulfate; sodium tetradecyl sulfate; polyoxyethylene ether; sodium cholate; octylthioglucopyranoside; n-octylglucopyranoside; alkyltrimethylammonium bromides; alkyltrimethyl ammonium chlorides; sodium bis (2-ethylhexyl) sulfosuccinate; beta- oxtyl-glucopyraniside (BOG, a nonionic surfactant), 3-[(3-

Cholamidopropyl)dimethylammonio]-l-propanesulfonate (CHAPS, a zwitterionic surfactant), combinations of these, and the like. Representative examples of suitable surfactants are also commercially available under trade designations including Zwittergent SB 3-14; Brij-35;

Tween-20; Pluronic F-68; Tetronics, Tween-80, Triton X-100; and NDSB-201, but may also include variants thereof known in the art. By way of non-limiting example, zwittergent SB 3- 14 may be provided, as known, or with varying sized alkyl chains (e.g. 3-12 carbons) and achieve similar results.

It has been found that using zwitterionic surfactants (zwittergents), in particular, facilitates solubilization when used in conjunction with a strong acid. Indeed, use of a zwitterionic surfactant for disaggregation yielded the most effective precondition for solubilization and subsequent refolding of non-glycosylated IFN-β from inclusion bodies. In one preferred refolding admixture, refolding yields of over about 90% were obtained using a zwitterionic surfactant in a low pH solution. In further embodiments, refold yields

approximated 90-100% were achieved using a zwitterionic surfactant in a low pH solution for disaggregation. A particularly preferred, but non-limiting, zwitterionic surfactant is commercially available under the trade designation Zwittergent SB 3-14 which is a

zwitterionic detergent that has been shown to have a strong protein binding affinity and prevent aggregation in proteins in some applications, including the purification of IFN-β. The zwitterionic surfactant may be provided in any amount effective to facilitate solubilization. In certain aspects, such amounts are between about 0.1% to about 1.0%. In further non-limiting embodiment, the zwittergent may be provided in an amount of about 0.2 wt %.

In further embodiments, sodium dodecyl sulfate (SDS) is not used in the process for preparing a composition. To this end, the solubilization solution is substantially free of SDS. In one aspect, the term "substantially free" means the solution has less than about 3% SDS, less than about 2% SDS, less than about 1% SDS or no measureable amount of SDS. In further embodiments, the refolding mixture, precipitate, and final product is similarly substantially free from SDS.

In addition to surfactants or detergents, the solubilizing media may optionally include one or more chemical protein denaturants to further facilitate inclusion body disaggregation. Chemical denaturants or chaotropic agents include compound(s) that disrupt the noncovalent intermolecular bonding within the protein, permitting the amino acid chain to assume a substantially random conformation. In aspects of the instant invention, such chemical denaturants include, without limitation, guanidine hydrochloride (guanidinium chloride, GdmCl), sodium thiocyanate, urea, and/or a surfactant such as SDS. Because the chemical denaturants are provided in conjunction with a low pH environment and zwitterionic surfactant, they are not needed in high concentrations and may be provided solely to optimize solubilization. To this end, they may be used at concentrations lower, in certain instances 2 - 4 times lower, than that typically used in the art. By way of example, typical concentrations of urea for denaturing proteins are about 8 M. In embodiments of the instant invention the concentration of urea used in the solubilization of IFN-β may be much lower at about 2 M. The chemical denaturant may be provided in any amount effective to facilitate protein disaggregation and/or solubilization. In certain aspects, such amounts are between about 1M to about 4M. In further non-limiting embodiments, particularly when chemical denaturant is urea, it may be provided in an amount of about 2M.

Solubilization of the protein is conducted under sufficient reaction conditions and for a sufficient time to achieve desired yield. Incubation, for example, may occur at any suitable temperature, but is not so high as to risk thermal degradation of the protein material and is not so cool as to slow the solubilization. Using room temperature is suitable and convenient and is preferred, although not limiting, to the instant invention. The incubation period can vary and can be a function of the reaction temperature.

While non-limiting to the invention, incubation periods may be between about 30 seconds to about 48 hours. In further embodiments, incubation periods may be between about 10 minutes and about 12 hours. In even further embodiments, the incubation period is about 20 - 30 minutes at room temperature. In further embodiments, the incubation period may be any length of time to achieve aggregate solubilization of greater than about 10%. In further embodiments, the incubation period may be any length of time to achieve aggregate

solubilization of greater than about 25% or about 50%. In even further embodiments, the incubation period may be any length of time to achieve aggregate solubilization that approximates about 90 - about 100% or a relatively homogenous solution of protein.

Post-solubilization, the remaining aggregated protein, if any, in the solubilized protein solution may be optionally measured by any one method selected from the group consisting of analytical ultracentrifugation, size exclusion chromatography, field flow fractionation, light scattering, light obscuration, fluorescence spectroscopy, gel electrophoresis, GEMMA analysis, mass-flow imaging and nuclear magnetic resonance spectroscopy (that is, the percentage can be based on any one method of analysis, to the exclusion of other methods of analysis).

Alternatively, the amount of aggregated protein in the protein composition measured by at least one method selected from the group consisting of analytical ultracentrifugation, size exclusion chromatography, field flow fractionation, light scattering, light obscuration, fluorescence spectroscopy, gel electrophoresis, GEMMA analysis, mass-flow imaging and nuclear magnetic resonance spectroscopy (that is, the percentage can be based on any one method of analysis, without necessarily excluding other methods of analysis).

According to a preferred methodology, residual aggregate content is determined by Size-Exclusion High Performance Liquid Chromatography (SE-HPLC). SEC-HPLC analysis of protein fractions may be conducted on an Agilent 1100 HPLC system equipped with a TSK G2000 SWXL size exclusion column (Tosohaas). The HPLC parameters are as follows:

mobile phase of about 10 mM HC1 in water, flow rate of about 0.5 mL/min (isocratic), room temperature, an injection of about 25 μΐ with absorbance measured at both 215 and 280 nm. Another mobile phase of about 10 mM HC1 and about 0.01% polysrobate 20 has also been used successfully. Aggregate content is determined by calculating the percentage of monomeric material relative to higher order species. The equation used is as follows: Aggregate Content % = Peak Area of Aggregate/(Peak Area of Monomer + Peak Area of Aggregate)* 100

Based on the result of such measurements, the solubilization pre-treatment may be continued until the desired solubility is reached, or the process may proceed to the next step.

Several advantages stemmed from solubilizing the inclusion bodies prior to refolding.

First, it allows the aggregated interferon mixture to be refolded much more effectively under high pressure to provide processed protein material with reduced aggregate content. In preferred modes of practice described below, IFN-P-lb containing less than 10 percent, even less than 5 percent, and even less than 1 percent of aggregated content after refolding, with yields of about 30% or about 40%, or over 40% were achieved.

Once adequate solubility is obtained, the solubilized protein-containing media may be prepared for high pressure refolding in a refolding mixture. In one embodiment, this includes raising the pH of the media to a level desirable for high pressure refolding. The optimal pH for refolding of non-glycosylated IFN-β and its variants has been found to be slightly alkaline, typically between about 7 and about 11. Often the pH is less than about 10.5 and desirably less than about 10 to minimize degradation or deamidation of the protein material. A preferred, but non-limiting, pH range may be about 7.5 to about 9.5 with an ideal pH being about 8.7. One or a combination of a wide range of buffering agents can be used in the mixture to help adjust/maintain the pH at the desired value or range. Examples include, but are not limited to carbonate, Tris, MOPS, MES, HEPES, CAPS, CHES, and the like. The use of CAPS around pH 10 and CHES around pH 9 are preferred. The use of CHES is more preferred. Additional refolding additives may be provided to the mixture, including, but not limited to, amino acids, glycols, polyethylene glycols (PEG's) or the like.

The refolding of non-glycosylated interferon proteins such as nongly-IFN-β generally is accompanied by the formation of one or more disulfide bonds. Accordingly, the refolding admixture may also include oxidizing/reducing agents (e.g. disulfide chemistry) that help promote the formation of the disulfide bonds. These can include agents that work mainly through an oxidation/reduction mechanism, such as when using the combination of

iodosobenzoic acid oxidizing agent (-40 μΜ) and DTT reducing agent (~2 mM). These can also include suitable pairs of disulfide shuffling agents such as the cysteine/cystine pair, the reduced glutathione (GSH)/oxidized glutathione (GSSH) pair, the cysteamine/cyamine pair, combinations of these, and the like. The reducing agent cysteine can also be used alone without cystine, with dissolved oxygen serving as oxidizing agent. The disulfide shuffling agents may be used at any suitable concentration such as about 0.3 mM oxidized agent and about 1.3 mM reduced agent. The use of cysteine/cystine or cysteine alone is preferred, however, the type of disulfide chemistry used in the refolding mixture can impact the refolding yield significantly. Thus, the choice of refolding mixture may be a function of what solubilization conditions were used prior to high pressure treatment.

The concentration of the interferon in the buffer may vary over a wide range. If the concentration is too low, the high pressure refolding throughput may be less than desired. On the other hand, if the concentration of the interferon is too high, re-aggregation may ensue or refolding yields may decrease due to solubility limits. Balancing such concerns, using from about 0.05 mg/mL to about 5 mg/mL, more desirably about 0.1 to about 3 mg/mL, and most desirably about 0.5 g/mL to about 2 mg/mL of interferon is suitable.

While not limiting to the instant invention, one protocol for solubilizing and preparing the protein for refolding may be carried out as follows. Pellets comprising the inclusion bodies are suspended in purified water at a ratio of about 10 mL of water per 1 g of wet inclusion body. Inclusion bodies are then solubilized in a solution containing about 10 mM HCl, about 2 M urea, about 0.2% zwittergent SB 3-14 at a protein concentration of about 0.5 mg/mL. The solution is then mixed at room temperature until solubilization is complete, which typically takes about 20-30 min. After solubilization, the pH is adjusted to approximately 8.7 by the addition of 50 mM CHES and 0.13 mM cysteine is added as a disulfide shuffling agent.

Using the foregoing protocol, or any modified protocol in accordance with the teaching herein, aspects of the present invention use one or more improved high pressure techniques to accomplish further interferon disaggregation, if necessary, and to facilitate refolding. These improved techniques may be advantageously used singly or in combination to obtain improved refolding yields of interferon, particularly in view of the solubilization protocol described above.

Generally, the refolding sample is subjected to a pressure that is sufficiently high to cause at least a portion of aggregated interferon to disaggregate and refold. Although refolding samples may include small amounts of a chaotrope, as indicated above, aspects of the present invention are distinguished from conventional methods that rely mainly upon relatively high concentrations of a strong chaotrope to accomplish disaggregation and refolding. In representative embodiments, the pressure used may be in the range from about 1000 bars to about 5000 bars. Refolding yields tend to be greater with increasing pressure up to an optimum pressure range beyond which increasing pressure will inhibit refolding. Selecting a suitable pressure within this range will depend upon factors including the formulation of the refolding sample, whether the aggregated protein mixture or aggregated protein being treated was subjected to a solubilization and precipitation treatment, the pressure stability of the native, monomeric protein of interest, the temperature, and the like.

When the foregoing solubilization protocol is used, particularly in combination with a zwitterionic surfactant and/or strong acid, a much wider range of high pressures leads to at least some IFN-β disaggregation and refolding, although higher pressures are still preferred. In these embodiments, the pressure may be in the full range from about 1000 bars to about 5000 bars, but desirably in the range from about 2000 bars to about 4500 bars, more desirably about 2800 to about 3800 bars, e.g., about 3200 or about 3500 bars.

In the course of a pressure treatment it may be desirable to ramp the pressure up to the desired pressure(s) over a period of time to avoid undue generation of thermal energy and/or otherwise degrade the interferon, allow the sample to remain at the elevated pressure(s) for a period of time, and then ramp the pressure down over a suitable time period, also to avoid undue aggregation or other degradation if the pressure ramping down were too fast. For instance, the pressure may be increased from ambient to the desired incubation pressure in one or more stages occurring over a period ranging from about 3 min to about 48 hr, desirably about 10 min to about 8 hr. Similarly, the pressure may be decreased to ambient pressure in one or more stages, each occurring with similar time periods. In one mode of practice, linearly increasing the pressure up to about 3500 bars in a period of about 32 min is suitable. In another mode of practice, linearly increasing the pressure up to about 3200 bars in a period of about 32 min is suitable. In a further mode of practice, linearly decreasing the pressure from about 3500 bars or about 3200 bars to ambient pressure in a period of about 10 min is suitable. This process is scalable to large scale manufacturing, such as, but not limited to, 1,000 liter processing.

A wide range of high pressure holding periods may be used to accomplish at least some degree of disaggregation and refolding. Representative time periods range from about 0.5 sec to about 48 hr, desirably about 2 min to about 24 hr, more desirably about 10 min to about 24 hr, most preferably about 2 to about 4 hr. It is believed that shorter periods lead to lower yields due to the slow kinetics of aggregate dissociation. It is also believed that a maximum yield is reached due to thermodynamic equilibrium effects, inasmuch as thermodynamic equilibrium has been shown to be a factor in protein refolding. Seefeldt, M. B., C. Crouch, et al. (2006). "Specific volume and adiabatic compressibility measurements of native and aggregated recombinant human interleukin 1 - receptor antagonist: Density differences enable pressure- modulated refolding." Journal of Biotechnology and Bioengineering , v98, 476-485. In one embodiment, a 2-hour incubation period is used to treat non-glycosylated IFN-β inclusion bodies.

The pressure treatment may occur at a wide range of temperatures. If the temperature is too low, then the kinetics of aggregate dissociation and refolding can be slowed as well as the disruption of hydrogen bonds. Thermal degradation of the interferon may occur if the temperature is too high. Balancing such concerns, representative modes of practice may carry out high pressure treatment at a temperature in the range from about 0 to about 50°C, more desirably from about 0 to about 30°C. In many embodiments, a temperature of about 25°C is suitable. Carrying out the pressure treatment at cooler temperatures, e.g., about 4°C, has been observed to increase yield and purity in some embodiments. The aforementioned temperature is the water bath temperature in which the pressure vessel is held during the high pressure treatment.

As a consequence of using such high pressure treatment in combination with pre- solubilization, disaggregated and refolded interferon have been generated with yields of > about 90%.

Thus, in an embodiment of the present invention, a method of preparing a composition comprising non-glycosylated interferon to maximize the solubilization of inclusion bodies and aggregated non-glycosylated interferon, comprising the steps of:

a) solubilizing the aggregated interferon;

b) incorporating the solubilized interferon into a refolding mixture; and c) applying an amount of pressure to the refolding mixture rendering effective refold of at least a portion of the interferon to form a composition comprising non-glycosylated interferon; wherein the resulting composition comprises less than about 5 percent of protein as residual aggregate.

Another embodiment comprises the steps of:

a) solubilizing the aggregated interferon into low pH solubilizing media that contains at least one surfactant or detergent;

b) incorporating the solubilized interferon into a refolding buffer that contains at least one oxidizing/reducing agents that help promote the formation of the disulfide bonds; and

c) applying an amount of pressure to the refolding mixture incorporating the

interferon effective to refold at least a portion of the interferon; wherein the resulting composition comprises less than about 5 percent of protein as residual aggregate.

Another embodiment comprises the steps of:

a) solubilizing the aggregated interferon into low pH solubilizing media that contains at least one zwitterion detergent;

b) incorporating the solubilized interferon into a refolding buffer that contains at least one oxidizing/reducing agents that help promote the formation of disulfide bonds; and

c) applying an amount of pressure to the refolding mixture incorporating the

Another embodiment comprises the steps of:

a) solubilizing the aggregated interferon into low pH solubilizing media that contains at least one zwitterion detergent; and

b) incorporating the solubilized interferon into a refolding buffer that contains at least one oxidizing/reducing agents that help promote the formation of the disulfide bonds and refolding of the protein

wherein the resulting composition comprises less than about 5 percent of protein as residual aggregate.

Another embodiment comprises the steps of: a) solubilizing the aggregated interferon into low pH solubilizing media that contains at least one zwitterion detergent;

b) incorporating the precipitated interferon into a refolding mixture comprising a disulfide shuffling reagent; and

c) applying an amount of pressure to the refolding mixture incorporating the

interferon effective to refold at least a portion of the interferon; wherein the resulting composition comprises less than about 5 percent of protein as residual aggregation.

Another embodiment comprises the steps of:

a) solubilizing the aggregated interferon;

b) incorporating the solubilized interferon into a refolding mixture; and c) applying over 3000 bars of pressure to the refolding mixture incorporating the interferon in a manner effective to refold at least a portion of the interferon; wherein the resulting composition comprises less than about 5 percent of protein as residual aggregation.

Another embodiment comprises the steps of:

a) solubilizing at least a portion of the aggregated interferon in an acidic aqueous medium optionally in the presence of a surfactant or a detergent; b) incorporating the solubilized interferon into a refolding mixture; and c) applying an amount of pressure to the refolding mixture incorporating the

It will be appreciated that all combinations of these methods are specifically

contemplated, including, for example, the use of specific surfactants and/or disulfide bond shuffling agents optionally in combination with specific pressure conditions as described herein. Further, pharmaceutical compositions comprising a therapeutically effective amount of an interferon, said interferon composition being made according to each of these possible methods as described herein are specifically contemplated.

After high pressure treatment, the interferon product mixture is desirably purified to remove various impurities. These impurities include E. coli host cell proteins, aggregated interferon, higher molecular weight IFN-β species that appears to lack a desired disulfide bond as observed by SDS-PAGE and RP-HPLC analyses, endotoxins, fermentation process-related impurities, and the like.

As a part of purification, it is desirable, though non-limiting, to include a salt or pH- induced precipitation step. Such precipitation may be accomplished using any methods known to one of skill in the art. In one embodiment, the pressure-treated solution is brought to a pH in the range of about 5.0 to about 8.0 using any buffer or, preferably, an acid. In further embodiments, the pH of the solution is brought to about 5.5 to about 7.5 and in further embodiments to about 6.0. While many acids, such as, but not limited to, acetic acid, citric acid, succinic acid, sulfuric acid, nitric acid, may be used to achieve this pH, in certain embodiments, the acid is phosphoric acid. The acid may be provided in any concentrations to achieve the desired pH, as exemplified herein or otherwise apparent to one of ordinary skill in the art. Other means of selective precipitation of proteins known to one skilled in the art, such as using sodium chloride, ammonium sulfate, magnesium sulfate, and the likes, to selectively or preferentially "salt out" either the protein of interest or impurities can also be used as a part of the purification scheme.

Precipitation of the protein of interest or impurities is conducted under sufficient reaction conditions and for a sufficient time to achieve desired yield. Incubation, for example, may occur at any suitable temperature, but is not so high as to risk thermal degradation of the protein material and is not so low as to render the reaction too slow. Using room temperature is suitable and convenient and is preferred, although not limiting, to the instant invention.

The incubation period can further vary and can be a function of the reaction

temperature. While non-limiting to the invention, incubation periods may be between about 30 seconds to about 48 hours. In further embodiments, incubation periods may be between about 10 minutes and about 12 hours. In even further embodiments, the incubation period is about 10 - 20 minutes at room temperature. In further embodiments, the incubation period may be any length of time to achieve protein precipitation of greater than about 10%. In further embodiments, the incubation period may be any length of time to achieve protein precipitation of greater than about 25% or about 50%. In even further embodiments, the incubation period may be any length of time to achieve protein precipitation that approximates 90 - 100%. The resulting precipitate may be centrifuged or filtered so as to either collect or remove the precipitate. The precipitate may be centrifuged at any speed and for any length of time to separate or pellet the precipitated protein from the remaining supernatant. In one non-limiting embodiment the precipitate is centrifuged at about 9,000-10,000 rpm for approximately 5-15 minutes. The resulting supernatant is collected and the impurity-containing pellet is discarded. Alternatively, the solution after pH- or salt-induced precipitation may be filtered e.g. with a depth filter, to either collect or remove the precipitated material. The material may be further purified directly or frozen and stored for later purification.

Additional purification methods may include any methods known in the art. In one embodiment, purification may be achieved using chromatography or a series of

chromatographic methods. Chromatography using a series of three columns is an example of one illustrative purification technique. According to such a technique, Blue Sepharose, Cu IMAC, and SP or CM Sepharose columns are used. These columns can be used in any order. The Blue Sepharose column can be used to capture the interferon out of the refolding solution. The interferon is then eluted from the column. The other two columns help to reduce impurities such as host cell proteins and aggregates. The Cu-IMAC column functions in one regard as a metal binding column to help remove metal impurities bound to the interferon by chelation or the like. The SP or CM Sepharose column functions via cationic exchange to separate the interferon from impurities such as endotoxins based upon charge. The Cu IMAC column helps remove host cell proteins as well as residual endotoxins. Additionally, aggregates and E. coli host cell proteins elute prior to the main IFN-β peak. Other columns such as Zinc IMAC columns could also be used. CM and SP Sepharose columns seem to offer similar performance. The SP Sepharose column offers an opportunity to add a final polishing step to upgrade the purity of the interferon. Endotoxins and E. coli host cell proteins can be further removed. Representative conditions for carrying out purification using these columns are described in the examples below.

The Blue Sepharose and the Cu-IMAC columns have the potential of leaching column components into the final protein pool. Also, it would be desirable to use buffers during chromatographic purification with components that are generally recognized as safe, such as those on the US FDA GRAS list. Accordingly, alternative modes of practice of purification involve a tertiary column purification that avoids the use of Blue Sepharose and IMAC-Cu²⁺ columns. The examples below show how this procedure is carried out with respect to this column.

In further embodiments, Applicants have surprisingly found that, with the Blue Sepharose, altering the ionic strength prior to elution improved the protein purity in the column eluent. More specifically, prior to loading, the Blue column was equilibrated with a low ionic strength loading buffer. In one embodiment, the loading buffer is phosphate buffer at a concentration of about 20 mM. In further embodiments, approximately 5 column volumes (CVs) of the loading buffer were used to pre-wash the column. This step removes impurities, particularly E coli. host cell proteins, that elute with the addition of propylene glycol at low ionic strength, resulting in a greater purity IFN-β.

Following chromatography, the purified interferon can be incorporated into any one or more desired pharmaceutical compositions. The pharmaceutical compositions may be filtered such as by 0.22 μιη membrane filtration and may be stored in containers such as plastic bags or bottles. The pharmaceutical compositions are used to administer the interferon in

therapeutically effective amounts. By "therapeutically effective amount" is intended an amount that is useful in the treatment, prevention, or diagnosis of a disease or condition, or symptoms thereof. Typical routes of administration include, but are not limited to, oral administration, nasal delivery, pulmonary delivery, and parenteral administration, including transdermal, intravenous, intramuscular, subcutaneous, intraarterial, and intraperitoneal injection or infusion. In one such embodiment, the administration is by injection, preferably subcutaneous injection. Injectable forms of the compositions of the invention include, but are not limited to, solutions, suspensions, and emulsions. Typically, a therapeutically effective amount of IFN-β comprises about 0.01 μg/kg to about 5 mg/kg of the composition, preferably about 0.05 μg/kg to about 1 mg/kg, more preferably about 0.1 μg/kg to about 500 μg/kg, even more preferably still about 0.5 μg/kg to about 30 μg/kg per dose and/or on a daily basis.

The pharmaceutical compositions, particularly those including nongly-IFN-β or variants thereof, are useful in the diagnosis, prevention, and treatment (local or systemic) of clinical indications responsive to therapy with this protein. Such clinical indications include, for example, disorders or diseases of the central nervous system (CNS), brain, and/or spinal cord, including Alzheimer's disease, Parkinson's disease, Lewy body dementia, multiple sclerosis, epilepsy, cerebellar ataxia, progressive supranuclear palsy, amyotrophic lateral sclerosis, affective disorders, anxiety disorders, obsessive compulsive disorders, personality disorders, attention deficit disorder, attention deficit hyperactivity disorder, Tourette

Syndrome, Tay Sachs, Nieman Pick, and schizophrenia; nerve damage from cerebrovascular disorders such as stroke in the brain or spinal cord, from CNS infections including meningitis and HIV, from tumors of the brain and spinal cord, or from a prion disease; autoimmune diseases, including acquired immune deficiency, rheumatoid arthritis, psoriasis, Crohn's disease, Sjogren's syndrome, amyotropic lateral sclerosis, and lupus; and cancers, including breast, prostate, bladder, kidney and colon cancers. Administration of IFN-β or its muteins to humans or animals may be delivered orally, intraperitoneally, intramuscularly, subcutaneously, intravenously, intranasally, or by pulmonary delivery as deemed appropriate by the physician.

According to one formulation option, the purified interferon is incorporated into pharmaceutical compositions suitable for subcutaneous injection such as a solution, suspension, or emulsion. Such formulations generally comprise a pharmaceutically acceptable liquid carrier. By "pharmaceutically acceptable liquid carrier" is intended a carrier that is conventionally used in the art to facilitate the storage, administration, and/or the healing effect of the therapeutic ingredients. A carrier may also reduce any undesirable side effects of the IFN-β with low or no toxicity to the patient. A suitable carrier should be stable, i.e., substantially incapable of reacting with other ingredients in the formulation. It should not produce significant local or systemic adverse effects in recipients at the dosages and concentrations employed for treatment. Such carriers are generally known in the art. Water with various degrees of purity such as water for injection (WFI) and purified water as defined in pharmaceutical pharmacopeias are exemplary liquid carrier.

The formulations for subcutaneous injection of IFN-β desirably have a pH of about 3.0 to about 5.0, about 3.0 to about 4.5, about 3.0 to about 4.0, about 3.5 to about 4.0, or about 4.0. At this pH, the IFN-β is soluble, stable, and resistant to aggregate formation. Such formulations may be provided with or in the substantial absence of sodium dodecyl sulfate (SDS) and/or Human Serum Albumin (HSA). In certain aspects, the formulations of the present invention are prepared and administered with the substantial absence of SDS and/or HSA. In one aspect, the term "substantially free" or "substantial absence" means the formulation has less than about 3% SDS or HSA, less than about 2% SDS or HSA, less than about 1% SDS or HSA or no measureable amount of SDS or HSA. The composition also desirably comprises a buffer in an amount that is sufficient to maintain the pH of the pharmaceutical composition within plus or minus 0.5 units of a specified pH, and which desirably is present at a concentration no greater than about 60 mM, preferably about 10 mM to about 50 mM. Suitable buffers that can be used to prepare the composition in which the IFN-β is solubilized include, but are not limited to, glycine, aspartic acid, succinate salt, citrate salt, formate salt, acetate salt, glutamic acid, histidine, imidazole, and phosphate. Sodium salts of succinate, citrate, formate, and acetate are preferred.

The formulations may also comprise a tonicity modifying agent in an amount sufficient to render the compositions isotonic with body fluids. Tonicity is a measure of effective osmolality or effective osmolality. Tonicity is a property of a solution in reference to a particular membrane, and is equal to the sum of the concentrations of the solutes which have the capacity to exert an osmotic force across that membrane. If a cell is placed in a hypotonic solution (one of lower tonicity than the cell contents), the water concentration is greater outside the cell and so osmosis produces a net movement of water into the cell. If the medium is isotonic, the water concentration is the same on either side of the cell membrane, and there is no net movement of water. If the medium is hypertonic, the water concentration inside the cell is greater. This leads to net movement of water out of the cell. Under the condition, animal cells shrivel up and plant cells become plasmolysed (the cell membrane pulls away from the cell wall in places as the cytoplasm shrinks).

For composition of IFN-β, nonionic tonicity modifying agents are preferred. The compositions can be made isotonic with a number of tonicity modifying agents ordinarily known to those in the art. These may be carbohydrates of various classifications (see, for example, Voet and Voet (1990) Biochemistry (John Wiley & Sons, New York).

Monosaccharides classified as aldoses such as glucose, mannose, arabinose, and ribose, as well as those classified as ketoses such as fructose, sorbose, and xylulose can be used as nonionic tonicifying agents in aspects of the present invention. Disaccharides such a sucrose, maltose, trehalose, and lactose can also be used. In addition, alditols (acyclic polyhydroxy alcohols) such as glycerol, mannitol, xylitol, and sorbitol are nonionic tonicifying agents useful in aspects of the present invention. The most preferred nonionic tonicifying agents are trehalose, sucrose, and mannitol, or a combination thereof. The nonionic tonicifying agent is added in an amount sufficient to render the formulation isotonic with body fluids. When incorporated into the pharmaceutical compositions, the nonionic tonicifying agent is present at a concentration of about 1% to about 15%, depending upon the agent used. Thus, in one embodiment, the nonionic tonicifying agent is trehalose or sucrose at a concentration of about 5% to about 15%, or about 9% by weight per volume. In certain aspects, the tonicifying agent is trehalose and is provided at this concentration. In another embodiment, the nonionic tonicifying agent is sorbitol at a concentration of about 4% to about 6%, preferably about 5% by weight per volume.

Compositions encompassed by aspects of the invention may have as little as about 0.01 mg/mL IFN-β and as much as about 20.0 mg/mL IFN-β. In various embodiments, the IFN-β is present at a concentration of about 0.01 mg/mL to about 20.0 mg/mL, about 0.015 mg/mL to about 12.5 mg/mL, about 0.025 mg/ml to about 10.0 mg/ml, about 0.05 mg/ml to about 8.0 mg/ml, about 0.075 mg/ml to about 6.0 mg/ml, about 0.1 mg/ml to about 4.0 mg/ml, about 0.125 mg/ml to about 2.0 mg/ml, about 0.175 mg/ml to about 1.0 mg/ml, about 0.2 mg/ml to about 0.5 mg/ml, about 0.225 mg/mL to about 0.3 mg/mL, or about 0.25 mg/mL.

The pharmaceutical composition may additionally comprise a solubilizing agent or solubility enhancer that contributes to the protein's solubility. Compounds containing a guanidinium group, most preferably arginine, are suitable solubility enhancers for IFN-β. Examples of such solubility enhancers include the amino acid arginine, as well as amino acid analogues of arginine that retain the ability to enhance solubility of IFN-β. Such analogues include, without limitation, dipeptides and tripeptides that contain arginine. Additional suitable solubilizing agents are discussed in U.S. Pat. Nos. 4,816,440; 4,894,330; 5,004,605; 5,183,746; 5,643,566; and in Wang et al. (1980) J. Parenteral Drug Assoc. 34:452-462; herein incorporated by reference.

In addition to those agents disclosed above, other stabilizing agents, such as

ethylenediaminetetracetic acid (EDTA) or one of its salts such as disodium EDTA, can be added to further enhance the stability of the liquid pharmaceutical compositions. The EDTA acts as a scavenger of metal ions known to catalyze many oxidation reactions, thus providing an additional stabilizing effect. Other suitable stabilizing agents include nonionic surfactants, including polyoxyethylene sorbitol esters such as polysorbate 80 (Tween 80) and polysorbate 20 (Tween 20); polyoxypropylene-polyoxyethylene esters such as Pluronic F68 and Pluronic F127; polyoxyethylene alcohols such as Brij 35; simethicone; polyethylene glycol such as PEG400; lysophosphatidylcholine; and polyoxyethylene-p-t-octylphenol such as Triton X-100. Classic stabilization of biopharmaceuticals by surfactants is described, for example, in Levine et al.(1991) J. Parenteral Sci. Technol. 45(3): 160-165, herein incorporated by reference. Such agents may be provided in any amount so as to exhibit the foregoing stabilizing effects. In one aspect, the stabilizing agent include nonionic surfactants, such as but not limited to polysorbate 20, which is provided in a concentration between about 0.005 wt and 0.05 wt , about 0.01 wt to about 0.05 wt , or about 0.02 wt .

In certain aspects the pharmaceutical composition is aggregate free. As used herein, the term aggregate free means the presence of less than about 5% of protein aggregates

(particularly interferon aggregate) in the composition, less than about 4% of protein aggregates in the composition, less than about 3% of protein aggregates in the composition; less than about 2% of protein aggregates in the composition; less than about 1% of protein aggregates in the composition; or the composition is entirely free from protein aggregates.

For subcutaneous formulation, it is desirable that the pooled fractions from

chromatographic purification are dialyzed or diafiltered into the formulation buffer. However, attempts to dialyze or diafilter nongly-IFN-β directly into formulation buffers tend to cause significant precipitation of the protein. To avoid this, dialysis or diafiltration of the interferon into 10 mM HC1 containing 9 weight percent trehalose has successfully provided soluble protein. Concentrated stock of buffer or other excipients can be added to yield the desired final formulation. In other embodiments of the invention, the pharmaceutical compositions of the invention can be prepared in a form that is suitable for pulmonary delivery and administering the preparation to the subject via pulmonary inhalation. By "pulmonary inhalation" is intended that the pharmaceutical composition is directly administered to the lung by delivering the composition in an aerosol or other suitable preparation from a delivery device into the oral or nasal cavity of the subject as the subject inhales through the oral or nasal cavity. By "aerosol" is intended a suspension of solid or liquid particles in flowing air or other physiologically acceptable gas stream. Other suitable preparations include, but are not limited to, mist, vapor, or spray preparations. Pulmonary inhalation could also be accomplished by other suitable methods known to those skilled in the art. These may include liquid instillation using a suitable device or other such methods. Pulmonary inhalation results in deposition of the inhaled protein composition in the alveoli of the subject's lungs. Once deposited, the protein may be absorbed, passively or actively, across the alveoli epithelium and capillary epithelium layers into the bloodstream for subsequent systemic distribution.

Pulmonary administration of a polypeptide or protein such as IFN-β requires dispensing of the biologically active substance from a delivery device into a subject's oral or nasal cavity during inhalation. For purposes of aspects of the present invention, pharmaceutical compositions comprising IFN-β or variants thereof are administered via inhalation of an aerosol or other suitable preparation that is obtained from an aqueous or nonaqueous solution or suspension form, or a solid or dry powder form of the pharmaceutical composition, depending upon the delivery device used. Such delivery devices are well known in the art and include, but are not limited to, nebulizers, metered-dose inhalers, and dry powder inhalers, or any other appropriate delivery mechanisms that allow for dispensing of a pharmaceutical composition as an aqueous or nonaqueous solution or suspension or as a solid or dry powder form.

Thus, in certain aspects, the compositions of the invention for pulmonary delivery encompass liquid compositions and dried forms thereof. For purposes of aspects of the present invention, the term "liquid" with regard to pharmaceutical compositions or formulations is intended to include the term "aqueous", and includes liquid formulations that are frozen. By "dried form" is intended the liquid pharmaceutical composition or formulation is dried by techniques including freeze drying (i.e., lyophilization; see, for example, Williams and Polli (1984) J. Parenteral Sci. Technol. 38:48-59), spray drying (see Masters (1991) in Spray- Drying Handbook (5th ed; Longman Scientific and Technical, Essez, U.K.), pp. 491-676; Broadhead et al. (1992) Drug Devel. Ind. Pharm. 18: 1169-1206; and Mumenthaler et al.

(1994) Pharm. Res. 11: 12-20), or air drying (Carpenter and Crowe (1988) Cryobiology 25:459- 470; and Roser (1991) Biopharm. 4:47-53). The term "lyophilize" with regard to IFN-β pharmaceutical formulations is intended to refer to rapid freeze drying under reduced pressure of a plurality of vials, each containing a unit dose of the interferon formulation therein.

Lyophilizers, which perform the above described lyophilization, are commercially available and readily operable by those skilled in the art. In one embodiment of the present invention, the liquid composition is prepared as a lyophilized composition.

As used herein, the terms "solid" and "dry powder" may be used interchangeably with reference to the pharmaceutical compositions suitable for pulmonary delivery. By "solid" or "dry powder" form of a pharmaceutical composition is intended the composition has been dried to a finely divided powder having a moisture content below about 10% by weight, usually below about 5% by weight, and preferably below about 3% by weight. Preferred particle sizes are less than about 10.0 μιη mean diameter, more preferably less than about 7.0 μιη, even more preferably about less than about 6.0 μιη, even more preferably in the range of about 0.1 to about 5.0 μιη, most preferably in the range of about 1.0 to about 5.0 μιη mean diameter.

Where the liquid pharmaceutical composition is lyophilized prior to use in pulmonary delivery, the lyophilized composition desirably is milled to obtain the finely divided dry powder of particles within the desired size range noted above. Where spray-drying is used to obtain a dry powder form of the liquid pharmaceutical composition, the process is carried out under conditions that result in a substantially amorphous, finely divided dry powder of particles within the desired size range noted above. Similarly, if the starting pharmaceutical composition is already in a lyophilized form, the composition can be milled to obtain the dry powder form for subsequent preparation as an aerosol or other preparation suitable for pulmonary inhalation. Where the starting pharmaceutical composition is in its spray-dried form, the composition has preferably been prepared such that it is already in a dry powder form having the appropriate particle size for dispensing as an aqueous or nonaqueous solution or suspension or dry powder form in accordance with pulmonary administration. For methods of preparing dry powder forms of pharmaceutical compositions, see, for example, WO 96/32149, WO 97/41833, WO 98/29096, and U.S. Pat. Nos. 5,976,574, 5,985,248, and 6,001,336.

The resulting dry powder form of the composition is then placed within an appropriate delivery device for subsequent preparation as an aerosol or other suitable preparation that is delivered to the subject via pulmonary inhalation. Where the dry powder form of the pharmaceutical composition is to be prepared and dispensed as an aqueous or nonaqueous solution or suspension, a metered-dose inhaler, or other appropriate delivery device is used. A pharmaceutically effective amount of the dry powder form of the composition is administered in an aerosol or other preparation suitable for pulmonary inhalation. The amount of dry powder form of the composition placed within the delivery device is sufficient to allow for delivery of a pharmaceutically effective amount of the composition to the subject by inhalation. Thus, the amount of dry powder form to be placed in the delivery device will compensate for possible losses to the device during storage and delivery of the dry powder form of the composition.

Following placement of the dry powder form within a delivery device, the properly sized particles as noted above are suspended in an aerosol propellant. The pressurized nonaqueous suspension is then released from the delivery device into the air passage of the subject while inhaling. The delivery device delivers, in a single or multiple fractional dose, by pulmonary inhalation a pharmaceutically effective amount of the composition to the subject's lungs. The aerosol propellant may be any conventional material employed for this purpose, such as a chlorofluorocarbon, a hydrochloro-fluorocarbon, a hydrofluorocarbon, or a hydrocarbon, including trichlorofluoromethane, dichlorodifluro-methane,

dichlorotetrafluoromethane, dichlorodifluoro-methane, dichlorotetrafluoroethanol, and 1,1,1,2- tetra-fluoroethane, or combinations thereof. A surfactant may be added to the pharmaceutical composition to reduce adhesion of the protein-containing dry powder to the walls of the delivery device from which the aerosol is dispensed. Suitable surfactants for this intended use include, but are not limited to, sorbitan trioleate, soya lecithin, and oleic acid. Devices suitable for pulmonary delivery of a dry powder form of a protein composition as a nonaqueous suspension are commercially available. Examples of such devices include the Ventolin metered-dose inhaler (Glaxo Inc., Research Triangle Park, N.C.) and the Intal Inhaler (Fisons, Corp., Bedford, Mass.). See also the aerosol delivery devices described in U.S. Pat. Nos.

5,522,378, 5,775,320, 5,934,272 and 5,960,792.

Where the solid or dry powder form of the HSA-free IFN-β pharmaceutical composition is to be delivered as a dry powder form, a dry powder inhaler or other appropriate delivery device is preferably used. The dry powder form of the pharmaceutical composition is preferably prepared as a dry powder aerosol by dispersion in a flowing air or other

physiologically acceptable gas stream in a conventional manner. Examples of commercially available dry powder inhalers suitable for use in accordance with the methods herein include the Spinhaler powder inhaler (Fisons Corp., Bedford, Mass.) and the Ventolin Rotahaler (Glaxo, Inc., Research Triangle Park, N.C). See also the dry powder delivery devices described in WO 93/00951, WO 96/09085, WO 96/32152, and U.S. Pat. Nos. 5,458,135, 5,785,049, and 5,993,783, herein incorporated by reference. The dry powder form of the HSA-free pharmaceutical composition comprising IFN-β or biologically active variant thereof can be reconstituted to an aqueous solution for subsequent delivery as an aqueous solution aerosol using a nebulizer, a metered dose inhaler, or other suitable delivery device. In the case of a nebulizer, the aqueous solution held within a fluid reservoir is converted into an aqueous spray, only a small portion of which leaves the nebulizer for delivery to the subject at any given time. The remaining spray drains back into a fluid reservoir within the nebulizer, where it is aerosolized again into an aqueous spray. This process is repeated until the fluid reservoir is completely dispensed or until administration of the aerosolized spray is terminated. Such nebulizers are commercially available and include, for example, the Ultravent nebulizer (Mallinckrodt Inc., St. Louis, Mo.) and the Acorn II nebulizer (Marquest Medical Products, Englewood, Colo.). See also the nebulizer described in WO 93/00951, and the device for delivering aerosolized aqueous formulations described in U.S. Pat. No. 5,544,646.

All references, including patents, patent applications, internet citations, and technical articles cited herein are incorporated herein by reference in their respective entireties for all purposes.

Embodiments of the present invention will now be further described in the context of the following illustrative examples. EXAMPLES

Example 1 - Expression of C17S IFN-beta-lb in E. coli.

A. Construction of CI 7S IFN-β gene sequence

The IFN-β in the C17S sequence (according to SEQ ID No. 2 and shown in Figure 1) was expressed in E. coli.

The gene for IFN-β was isolated by PCR amplification of human genomic DNA (Cat. #

636401, Clontech, CA), using primers BARO 1

(5 ' C ACGTGCATATG AGCTAC AACTTGCTTGGATTC) (SEQ ID NO: 4) and BARO 4 (5 ' CGG A ATTCTTAGTTTCGG AGGT A ACCTGTA AG) (SEQ ID NO: 5). The resulting fragment was digested with restriction enzymes Ndel and EcoRI, and cloned into similarly digested and calf intestine alkaline phosphatase (ClP)-treated pUC19. Several clones were isolated and sequenced, and one clone with the correct sequence was subjected to PCR-based mutagenesis using forward primer BARO 8

(5 ' AGC AGCAATTTTCAGTCTCAGAAGCTTCTGTGGCAATTG) (SEQ ID NO: 6) and reverse primer BARO 9 (5 ' CAATTGCCAC AGAAGCTTCTGAGACTGAAAATTGCTGCT) (SEQ ID NO: 7) as described in Higuchi R. (PCR Protocols; M.A. Innis et al, eds.,1990, Academic Press), which changed the codon TGT for Cysteine at position 17 (C17) to TCT for Serine ( C17S). After mutagenesis, the gene was cloned into pUC19, and sequenced to confirm the presence of the C17S mutation. The final sequence of the IFN-β C17S gene is according to SEQ ID No. 3 as shown in Figure 2. In this Figure, IFN-β C17S is encoded by DNA sequence shown in uppercase. The codon TCT for Ser-17 is underlined, as are recognition sites for Ndel, Hindlll, and EcoRI.

B. Construction of the Preliminary Expression Plasmid and Expression in recA+ strains

The IFN-β C17S gene was isolated as an Ndel-EcoRI fragment and cloned into similarly digested and CIP-treated pET21a+ (Novagen, WI). The vector pET21a+ carries the phage T7 promoter and lac operator for regulated expression, and confers ampicillin resistance. The resulting plasmid was transformed into BL21 (DE3) and Rosetta2 (DE3), two strains that express the T7 RNA polymerase under the control to the lac promoter. The two strains are the same except for that Rosetta2 (DE3) contains the plasmid pRARE2 which expresses tRNAs for the rare E. coli codons AUA (He), AGG, AGA, CGG (Arg), CUA (Leu), CCC (Pro), and GGA (Gly), and confers chloramphenicol (Cam) resistance. Strain A [IFN-β C17S/pET21a+/BL21 (DE3)] and Strain B [IFN-β C17S/pET21a+/Rosetta (DE3)] were grown at 37°C in LB medium containing the appropriate antibiotics (Strain A, ampicillin; Strain B, ampicillin +

chloramphenicol) to an optical density at 600 nm (OD-600) of 0.5. Expression of IFN-β C17S was induced by addition of 0.75 mM IPTG. Samples were taken for analysis by SDS-PAGE shortly before induction, and one, two, three and four hours after induction. We found that Strain A expressed no observable IFN-β C17S whereas Strain B expressed IFN-β C17S at the level of -20% of total cell protein at two and three hours after induction. C. Construction of the Final Expression Plasmid The IFN-β C17S gene was then transferred to pET24a+, a T7 promoter vector that carries the kanamycin resistance gene. pBarl2 was used to transform BL21 (DE3) and Rosetta2 (DE3) (EMD4 Biosciences), and the resulting strains, Strain C and Bar23, were subjected to growth (with LB medium containing the appropriate antibiotics), IPTG induction and expression analysis as described above. We found that Strain C expressed no detectable IFN-β C17S whereas Bar23 expressed between 10% and 15% of total cell protein. We assume that both Strain B (above) and Bar23 express IFN-β C17S because these strains contain pRARE2, which carries tRNA genes for rare E. coli codons. Such codons are present in the IFN-β C17S gene.

D. Development of the Final Expression Strain

Strains BLR (DE3) and HMS 174 (DE3) (Novagen) were transformed with pRARE2 to produce strains D and E. These strains, in addition to containing the pRARE2 plasmid, are recA-. Strains D and E were transformed with pBAR12 to give Bar24 and Bar25, respectively, which were evaluated for IFN-β C17S expression as described above, except that the 4 hour time point after induction was eliminated. These strains produced substantial levels of IFN-β C17S, with the HMS 174-based strain (a K12 strain) expressing approximately twice as much of the protein as the BLR-based (B) strain. This result and the data from additional

experiments comparing the two strains using other media (not shown) led us to choose Bar 25 as the final IFN-β production strain, which is designated as the Bar 25 protein herein. We then confirmed the sequence of the IFN-β C17S gene in Bar25 using the T7 primer

(5 ' TAATACGACTC ACTATAGGG) (SEQ ID NO: 8) and the T7 terminator primer

(5 ' GCTAGTTATTGCTC AGCGG) (SEQ ID NO: 9).

E. Media Development

Table 1 shows results of experiments comparing a variety of media, which were conducted to look for improved growth characteristics and IFN-β C17S expression in shake flasks, relative to LB medium. Based on these data, our optimal medium for expression is 4% yeast extract, 0.1 M MES, pH 6.5, 1% NaCl. Additionally, the presence of glycerol in the fermentation media improved cell density. Although the expression method provided in this example differs from what has been published previously ( U.S. Pat. No. 4,450,103), the expression system has been shown to provide the proper amino acid sequence according to SEQ ID No. 2 in Fig. 1.

Table 1.

Young

Medium O.D. (600 nm) O.D. (600 nm) Units

No. YE Tryptone Buffer and pH NaCl Glycerol Induction Harvest g/OD.L

1 3%D 1%T 0.1 M MES 6.5 1% 0 1.5 2.75 0.10

2 4%D 0 0.1 M MES 6.5 1% 0 1.52 2.97 0.10

3 2.4%D 1.2%T 0.09 M P0₄ 7.5 0 0.50% 5.23 7.29 0.05

4 2%D 3.5%T 0.09 M P0₄ 7.5 0.50% 0 3.77 5.5 0.05

5 4%D 0 0.1 M MES 6.5 1% 0 1.83 2.29 0.10

6 4%D 0 0.1 M MES 6.5 1% 1% 2.85 4.45 0.05

7 4%D 0 0.1 M MES 6.5 1% 0 2.97 3.72 0.03

8 4%D 0 0.09 M P0₄ 7.5 1% 0 2.48 3.39 0.05

9 4%D 0 0.1 M MES 6.5 1% 0 1.58 3.22 0.10

10 4%D 0 0.1 M MES 6.5 1% 1% 3.33 5.78 0.08

11 4%D 0 0.1 M MES 6.5 1% 0 3.03 4.76 0.10

12 4%M 0 0.1 M MES 6.5 1% 0 2.59 4.32 0.08

13 4%M 1%T 0.1 M MES 6.5 1% 0 2.72 4.4 0.10

14 4%M 1%S 0.1 M MES 6.5 1% 0 2.92 4.7 0.08

15 4%M 0 0.1 M MES 6.5 1% 0 3.01 4.65 0.08

Cells were grown at 37°C and induced with 0.75 mM IPTG. Media are as shown except No. 7 contains 0.3 mM ZnCl₂ at the time of induction. YE=yeast extract; D=Difco YE; M=Marcor YE; T=Tryptone; S=Soytone.

Example 2 - Examination of the Effect of "traditional" Refolding Conditions on the Refolding of IFN-β from Inclusion Bodies.

US patents #7,064,192 and #6,489,450 teach the skilled artisan the general methods for refolding protein aggregates using high pressure (Randolph, Carpenter et al. 1999). Of these methods, 2000 bar pressure treatment at a temperature of 25°C, for sixteen hours, at a refolding pH of 8.0, in the presence of 4 mM reduced glutathione (GSH) and 2 mM oxidized glutathione (GSSG) has been used for the refolding of proteins that contain disulfide bonds such as hen egg white lysozyme, placental bikunin, and malaria pfs48 (St. John, Carpenter et al. 2002; Seefeldt, Ouyang et al. 2004; Seefeldt 2005; Lee, Carpenter et al. 2006). In line with those disclosures and teachings, inclusion bodies of IFN-β were obtained (Example 2A) and pressure treated at 2000 bar for sixteen hours at 25°C in aqueous solutions containing 50 mM Tris, pH 8.0, 4 mM GSH, 2 mM GSSG (Example 2B). After depressurization, the samples were tested for refolding by RP-HPLC (Example 2C) and found to have no refolded protein (yield of 0%). The addition of arginine, a commonly used refolding agent that has been taught extensively in the prior art (Arakawa and Tsumoto 2003; Tsumoto, Umetsu et al. 2004; Seefeldt 2005), did not improve the refolding yields.

A. Generation of Washed Inclusion Bodies

An E. coli whole cell pellet (~5 g) is thawed and suspended in 100 mL BPER Reagent (Pierce Chemical). The suspension is stirred for 20 minutes at room temperature. One mL of lysozyme stock solution (10 mg/2 mL H₂0, made fresh) is added to lyse the cells. The mixture is stirred an additional 10 min at room temperature. 150 μΐ of DNAse solution (2 mg/mL, frozen stock) is added to break down DNA and decrease the viscosity of the mixture. After an additional 1 hour of stirring, the lysed E. coli suspension is centrifuged in a JA-14 rotor at 8000 rpm for 10 min. The supernatant is decanted. The resulting pellet of insoluble material is suspended in 200 mL of distilled water and re-centrifuged. The final IB pellet can be frozen at -20°C for storage or subjected to additional processing steps.

B. High Pressure Refolding Experiments

High pressure refolding experiments were conducted by creating stock solutions of IFN-β inclusion bodies, IFN-β methanol precipitate, 500 mM CAPS buffer (pH 10), 5% zwittergent SB 3-14, and 100 mM redox components. The stock solutions were used to create 500 μΐ_^ of refolding solution and mixed in Eppendorf tubes. The mixed samples were then placed in sealed syringes and pressure treated as described previously (Seefeldt, Ouyang et al. 2004). Care was taken to ensure that insoluble protein aggregate was properly suspended and pipetted correctly without filtering or loss in the Eppendorf tubes. The concentration of protein present in the inclusion body precipitate was determined by an RP-HPLC method. The protein concentration was obtained by taking an absorbance reading at 280 nm prior to methanol precipitation and calculating the protein concentration using an extinction coefficient of 1.5 cm^-mg^-ml.

C. Reverse-Phase High Performance Liquid Chromatography (RP-HPLC) The conditions for analyzing refolds are as follows: HPLC System: Agilent HP 1100, Column: C3 Zorbax with guard column 300SB 3.5 μιη 4.6 x 150 mm, Solvents: A: Water, 0.1% TFA, B: Acetonitrile, 0.1 %TFA, Flow Rate: 1 mL/min, Temp: 30°C, Injection: 20 μΐ, Detection: Absorbance at 215 nm and 280 nm

Gradient: Time (min) %B

0 27

1 45

36 56

37 100

40 27

D. Sodium dodecyl sulfate - polyacrylamide gel electrophoresis (SDS-PAGE)

Reducing and non-reducing SDS-PAGE was used to examine the purity and disulfide content of pressure treated IFN-β. Approximately 5 μg of protein from the processed supernatant was added to the 2X SDS-PAGE sample buffer (Invitrogen, Carlsbad, CA). The SDS-protein mixture was heated for 5 minutes at 100°C. 4-20% Novex Tris-glycine precast gels were used (Invitrogen, Carlsbad, CA), with 400 ml of diluted 10X Tris-glycine running buffer (BioRad, Hercules, CA). Gels were run for fifty minutes, and staining was conducted with methanol/acetic acid-free Coomassie blue for total protein analysis (BioRad, Hercules, CA).

Example 3 - Solubilization, High Pressure Refolding, Acid precipitation, and Clarification A. Materials

IFN-β inclusion body suspension (31 mg/mL)

MQ Water

IFN-β inclusion body suspension (31 mg/mL)

CHES (Sigma C2885 or equivalent)

Zwittergent SB 3-14 (3-(N,N-Dimethylmyristylammonio)propanesulfonate) (Sigma

T7763 or equivalent)

L-Cysteine (Sigma C7352 or equivalent)

Urea (JT Baker 4206 or equivalent)

Concentrated HC1 (JT Baker 9535 or equivalent) Sodium Chloride ACS Grade (Baker 3624-05 or equivalent)

Concentrated Phosphoric Acid (Fisher A260 or equivalent)

B. Reagents

4 M Urea, prepared fresh daily, 20 mM Cysteine, prepared fresh daily

1 M HC1

1 M Phosphoric Acid

500 mM CHES, pH 9.2

2% Zwittergent SB 3-14

5 M NaCl

C. Solubilization

High pressure refolding cannot currently employ mixing, due to the high pressures employed and current equipment limitations. The inclusion bodies are dense and, if used directly in high pressure refolding, can settle quickly and result in a high protein concentration at the bottom of the pressure treatment vessel. For many proteins, this is not critical if the solubility of the protein is sufficient. However, for IFN-β, the solubility is very low under alkaline pH and thus the settling of the inclusion bodies in the pressure vessel can significantly decrease the process yield. Consequently, steps were taken to examine solubilization of the interferon inclusion bodies prior to pressure treatment.

It was discovered that although zwittergent and HC1 individually could not solubilize the inclusion bodies, but when combined, 100% solubilization was obtained, as indicated above. Phosphoric and sulfuric acid were also tested, but did not provide as effective solubilization at 10-20 mM concentrations. A solution of 10 mM HC1 and 0.2% zwittergent SB 3-14 was effective at solubilization of the interferon inclusion bodies within 10-20 minutes. Longer incubation times had neither positive nor negative effects on yield.

D. Pressure Refolding

The high pressure refold is conducted by first solubilizing inclusion bodies in a solution containing 10 mM HC1, 2 M urea, 0.2% zwittergent SB 3-14 at a protein concentration of 0.5 mg/mL (as analyzed by SDS-PAGE). The pH of this solution is approximately 2.8. In the final solution, HC1 is provided at 5 to 20 mM; urea is provided at 1 to 4 M; and the zwittergent is provided at 0.1 to 0.4 wt. %. The IB suspension is vortexed to ensure that it is homogenous prior to pipetting. The IFN-β inclusion bodies are then pipetted into the above described solution and mixed. This solution is then gently mixed and incubated at ambient temperature until completely clear and free of visible particles (approx. 10-20 min).

After solubilization is completed, the pH is adjusted to approximately 8.7 by the addition of CHES to 50 mM (500 mM CHES stock solution at pH 9.2) and 0.13 mM cysteine is added to enable disulfide bond formation. While mixing, the entire volume of CHES and cysteine is added to the solubilized inclusion body solution. The solution may become slightly cloudy. The low pH solubilization step is employed to enhance dissolution of particulate material prior to high pressure refolding.

The refold solution is transferred to a syringe or bottle to initiate refolding. Refolding is initiated by pressure treating at 3200 bar for two hours, with a 32 minute pressurization and 10 minute depressurization time.

In certain instances, refolding yields slightly greater than 100% were observed. This is due to the discrepancies between inclusion body protein concentrations as determined by SDS- PAGE relative to the RP-HPLC method used to determine the refolded protein when calculating yield. The SDS -PAGE test was used to determine the interferon concentration of inclusion bodies because of high levels of protein impurities present.

E. Acid Precipitation and Clarification

In order to remove E. coli. host cell proteins (HCPs) and to improve the longevity of the Blue Sepharose resin used subsequently in protein purification, an acid precipitation step was added prior to the Blue column. After refolding, the refold is warmed to 25°C and 1 M phosphoric acid is added while mixing in small increments until a final pH of 6.0 + 0.1 is reached. Once this pH is reached stirring is ceased and the solution is incubated at ambient temperature for 10-20 min.

The solution at pH 6.0 is clarified by centrifugation (10,000 X g) for about 5-10 minutes prior to analysis or column loading. Depth filters are also capable of removing the precipitate, such as using a Pall K900P or BIO 20. Although both filters remove insoluble material, the BIO 20 does not exhibit non-specific product binding, resulting in a 90% step yield. NaCl is then added to the clarified solution at a final concentration of 765 mM to improve purification on the subsequent Blue column. At this point, the refold can be held or stored at 4°C for approximately 5 days with a slight loss of yield due to deamidation and aggregation. Figure 4 shows a representative SDS-PAGE gel of the refold pre-clarification (Lane 2) and post-clarification (lane 3). The results showed significant removal of impurity proteins by the acid precipitation process.

Example 4 - Surfactant Screening

The zwitterionic surfactant zwittergent SB 3-14 was found effective in solubilizing interferon inclusion bodies and in subsequent pressure refolding of the protein. The concentration of zwittergent was optimized by a central composite statistical experimental design (SED) (Example 5 and Figure 3). It was found that to ideal, but not limiting, improvement in the refold was observed when the amount of the zwittergent 3-14

concentration was increased from 0.05% to 0.2%, as it increased the solubility of IFN-β at alkaline pH. Such a concentration may be provided in any amount, however, between about 0.1 to about 1.0 wt %.

Example 5 - Optimization of Zwittergent Concentration, Protein Concentration and Time with Statistical Experimental Design (SED)

Studies were conducted to evaluate the interactions between zwittergent concentration

(0.05 - 0.4%), protein concentration (0.5-1 mg/ml), and refolding time (0.5 - 2 hr.) on the refolding yield. Over the broad parameters studied, optimum yields were obtained at 0.5 mg/ml IFN-β concentration, 0.2% zwittergent and refolding times of 2 hr (time data not shown). The experimental design space is shown in Figure 3. The strong effect of zwittergent 3-14 concentration on the refolding yield is apparent, more than doubling the yield from -30% to near 100%. Higher protein concentrations (1 mg/ml) can be used in the refold with somewhat lower refolding yield but increasing throughput. For the purposes of optimization, a concentration of 0.5 mg/ml was used to maximize yield, but can be increased to improve throughput. Higher protein concentrations could confound downstream processing. Refolding time had a small effect on yield, being slightly higher at 2 hrs. Longer times were found not beneficial. Example 6 - Optimizing Precipitation Step

E. coli host cell protein (HCP) precipitation was examined as a function of propylene glycol concentration (0-30%) and pH (4-8). Analysis by SDS-PAGE demonstrated that dilution of the refold 5X into 75 mM citrate (pH 4.0), 25% propylene glycol (PG), and 0.5% zwittergent 3-14 was effective at precipitating a majority of HCPs, with less than 20% loss in IFN-β. HCP precipitation was examined coming directly out of the refold, without propylene glycol addition. Adjustment to pH 6.0 using 1 M phosphoric acid was determined to be effective. Note that if propylene glycol is present during this step, losses of IFN-β may occur. This may be due to the strong change in solubility as a function of pH and temperature. The solubility appears to be lowest at pH 6 and the protein is more prone to aggregate. The solubility increases again at pH 3. The interferon solubility decreases with lower temperature and thus the solution could be warmed to 25°C prior to pH 6 adjustment.

Ionic strength was also examined to determine its effects on both HCP impurity removal and IFN-β losses. The addition of NaCl to a final concentration of 765 mM removes a significant amount of HCP during loading on the Blue by maintaining the HCP in the column flow through. The salt is added after the solution has been clarified to prevent increased ionic strength from resolubilizing the precipitated HCPs. The addition of 5% propylene glycol prior to loading on the Blue also increases yields. Subsequent studies examined Blue loading conditions in more detail.

Example 7 - GE Blue Affinity Chromatography Purification

Blue Chromatography was conducted by using GE Blue Sepharose resin with an

AKTA Explorer. Two column sizes were tested; a 5 mL Blue HP column with column dimensions of 2.5 cm in length and 1.6 cm in diameter and a 50 mL Blue FF column with an identical length but a diameter of 5 cm. All conditions described below were used, with the exception that at the 50 mL step scale, FF resin was used and a flow rate of 12 column volume

(CV)/hr was used as a result of equipment limitations. Figure 9 provides an overview of the steps provided in this example.

A. Reagents

Blue Loading Buffer (50 mM Na-Phosphate, pH 7.1, 1 M NaCl)

Blue Wash Buffer (20 mM Na-Phosphate, pH 7.1) Blue Elution Buffer (20 mM Na-Phosphate, pH 7.1, 50% (v/v) Propylene Glycol)

B. Purification Method

The Blue column is equilibrated with 3-5 CV's of loading buffer. 5% (v/v) propylene glycol (PG) is added to the clarified refold, which is then loaded to a maximum level of 4.3 mg

IFN-p/mL resin. The protein is loaded onto the column and the column is connected to the chromatograph. Once the protein is loaded, the column is washed using 20 mM phosphate (pH 7.0) until a stable baseline for both A₂8o and conductivity is observed.

Next, the column is washed with Wash Buffer until a stable baseline for both A₂₈₀ and conductivity is observed. After the wash, the protein is eluted in the Elution Buffer.

Representative chromato grams are shown in Figure 5. The Blue column removes both HCPs and product aggregates. Lower purity material elutes on the leading edge of the peak.

As shown, overall purity was improved significantly by the washing of the column in low ionic strength buffer (20 mM phosphate) prior to equilibration.

Example 8 - Copper Immobilized Metal Affinity Chromatography Purification

Copper Immobilized Metal Affinity Chromatography was optimized and conducted at the 5 mL scale using GE HP EVIAC resin on an AKTA Explorer. Two column sizes were tested; a 5 mL column with column dimensions of 2.5 cm length and 1.6 cm D and a 38 mL with a length of 7.2 cm and a diameter of 2.6 cm. Figure 10 provides an overview of the steps provided in this example.

A. Reagents

Cu Sulfate Solution (10 mM CuS0₄)

2% Zwittergent 3-14

Elution Buffer (20 mM Na-Phosphate - USP grade, pH 7.1, 0.2 M NaCl, 0.05%

Zwittergent 3-14, 10% Ethylene glycol) B. Purification Method

The maximum protein load is 2.0 mg/ml resin. Based on the RP-HPLC assay, the volume of pooled blue fractions containing up to 10 mg of IFN-β is measured. Next,

Zwittergent 3-14 is added to make the pooled blue fractions 0.05% Zwittergent 3-14. Prior to loading, the column is equilibrated. Columns are equilibrated by washing with 3-5 CV of Elution Buffer. The 1 ml column is removed and set it aside until needed. The 5 ml column is charged with 2.5 column volumes of 10 mM CuS04 solution and washed with 5 column volumes Elution Buffer.

The protein is loaded onto the column, collected and retained in the flow through.

Next, the 1 mL column is connected to the end of the 5 mL column. Both columns are connected to the chromatographic system and begin collecting data, including A₂8o and conductivity.

The column is then washed with Elution Buffer until a stable baseline for both A₂go and conductivity is observed. The protein was eluted with a 0-100% gradient of 50 mM imidazole. Lower purity material elutes on the leading edge of the peak. Fractions were measured having >15% of the maximum peak height absorbance. Accordingly, the Cu column pool should be initiated when the absorbance is 15% of the maximum peak height, and the pool will stop once the absorbance drops below 15% of the maximum peak height, where it is then measured. The Cu IMAC column at the 5 ml scale had yields of 58% +/- 2% (n=3). At the 38 ml step scale (50 ml Blue column), yields of the Cu IMAC column were 69% +/- 15% (n=3), demonstrating repeatable column performance. A representative chromatogram of the IMAC step is shown in Figure 6, the IMAC column is effective at removing impurities that elute at fractions 15 and 16.

Example 9 - Cation Exchange Chromatography Purification

The HP/FF-Sepharose Cation Exchange Chromatography was conducted using GE SP Sepharose FF resin on an AKTA Explorer. All runs were operated at the 5 ml scale with HP resin for process development. For the 50 ml scale production, multiple cycles were performed using a 5 ml FF resin. Figure 11 provides an overview of the steps provided in this example.

A. Reagents

Load Buffer (50 mM Na-Phosphate - USP Grade, pH 6.7, 0.05% Zwittergent 3-14,

50% Propylene Glycol - USP Grade, 0.01% Tween-20)

Elution Buffer (20 mM Na-Phosphate - USP Grade, pH 6.7, 50% Propylene Glycol

USP Grade, 0.01% Tween-20) B. Purification Method

While mixing, the IMAC pool is diluted with 5X the volume of Load Buffer. A 6X dilution of the IMAC pool is conducted (1 part IMAC Pool to 5 parts Elution Buffer).

Prior to loading, the column is equilibrated by washing with at least 5 CV of Elution Buffer A using a peristaltic pump. The protein is loaded onto the column, collected and retained in the flow through. The column is connected to an AKTA and A₂8o and conductivity are monitored. The column is then washed until a stable baseline for both A₂go and

conductivity is observed. The protein is then eluted with a sodium chloride gradient in Elution Buffer. A representative chromatogram is shown in Figure 7. Yield across this column was good. Yields using the 5 ml column were approximately 67%. Using a 5 ml FF column loaded at a protein concentration of 3 mg IFN-β/ιηΙ resin (for the 50 ml Blue Step scale) a yield of 77% +/- 3% was obtained.

Example 10 - Process Optimization - 5 ml Process Overview

Process development and optimization was conducted primarily at the 5 ml scale, using 5 ml GE HiTrap columns. Process yield analysis for the 5 ml scale is shown in Table 2. Key elements to the new process include refolding at a protein concentration of 0.5 mg/ml after HCl solubilization and removal of solvent extraction. This process change has significantly improved refolding yields to approximately 100%. (Refolding yields appear to be greater than 100% due to the discrepancy between SDS-PAGE and RP-HPLC in measuring concentrations of IFN-β in inclusion bodies). The refolded solution is then clarified after adjustment to pH 6.0 to prevent column fouling and aid in protein purification. The Blue chromatography was modified by eluting with NaCl after first washing the column with propylene glycol at reduced ionic strength. The IMAC and SP columns are operated similarly to the previous process but the dilution schemes have been modified to reduce processing volumes in the system. After implementing these changes, the process was run three times at the 5 ml scale with the following yields and process purity: Table 2

Average Average Yields Lot # 093009 - Lot # 093009 - R2 Lot # 093009 - R3 Ste Yield 95% CI Cum. Yield

* Purification process yield assumes a 80% yield across the TFF, as reported by Avecia in the Phase I process

The purity of the material generated through TFF at the 5 ml scale is shown in Table 3.

Table 3

As seen, the interferon beta- lb produced through the process showed greater than 95% purity by RP-HPLC test and greater than 99% monomeric species by SE-HPLC.

Figure 12 shows a highly developed silver SDS-PAGE representative of the six process steps (Pressure Refold, Acid Precipitation and Clarification, Blue Column, IMAC Column, SP Column and Tangential Flow Filtration). The dark band between 14.4K and 21.5K molecular weight markers corresponds to interferon -pib. The clarification and Blue column steps remove a majority of the impurities, with subsequent column steps removing additional impurities and preparing the protein for final formulation.

Example 11 - 50 ml Process Overview for Preliminary Scale Analysis

To evaluate the process at a larger scale, the process was scaled from the 5 ml scale to the 50 ml scale (Blue column) and yields were analyzed after three subsequent runs (Table 4). This process was implemented before volume minimization occurred, however all columns were eluted and with the same conditions identified in Examples 7-9. Due to availability, the Blue and Cu IMAC resin at the 5 ml scale was HP grade, whereas the 50 ml scale was FF grade. The SP column was operated using a FF resin in cycles at the 5 ml scale and 50 ml Blue Column step scale, with a column loading of 3 mg/ml of resin. A summary of the yields for the process is shown in Table 4. The outlier in lot 081008 for the refold was a result inclusion body pipetting differences between operators. For lot 080508, the TFF step failed due to unknown causes.

Table 4

Average Average Yields Lot# 080508 Lot # 081008 Lot# 081108 Ste Yield 95% CI Cum. Yield

The purity of material produced at the 50 ml scale met all SEC, RP, CEX, HCP, Silver SDS-PAGE, and endotoxin assay specifications (Table 5). The purity (aggregate content, deamidation content, endotoxin and HCP content) for material generated through the proposed process was improved relative to the existing IFN-β process.

Table 5

Given the foregoing, the proposed process steps include elimination of the solvent result in an increase in overall refolding yields (from -30 % to 115% at 0.5 mg/ml for the original and revised process, respectively) over previously known methods, and an overall process yield improvement of approximately 10-20X, while maintaining process volumes that will facilitate operation of the process at the anticipated commercial scale. Example 12-Interferon Beta-lb Product Analyses

Interferon beta- lb produced by the methods described herein was analyzed for purity, presence of aggregates, and biological activity (potency).

A. Inclusion Body Solubilization and Pressure Treatment and Acid Precipitation

Inclusion body solubilization, pressure treatment, and acid precipitation were carried out as described under Example 3.

B. Column Chromatography Purification

Three column chromatography purification steps were carried out as described under Examples 7-9. C. Analytical Testing Methods

The RP-HPLC and SDS-PAGE analytical procedures are as described under Example 2C and 2D.

The SE-HPLC method using a TOSOH Bioscience TSK-gel G2000SWXL analytical column which has an exclusion limit between 5000 and 15000 Daltons for typical globular proteins. The mobile phase is 10 mM HCl, 0.01% (w/v) polysorbate 20 and absorbance at 215 nm is monitored.

The assay for interferon biological activity (potency) is performed in accordance with the European Pharmacopoeia Monograph 5.6, Assay of Interferons. Briefly, A549 cells are treated with dilutions of the test article or the NIBSC rhIFN beta- lb reference standard and then challenged with encephalomyocarditis virus. Interferon beta protects the cells from viral infection. The cell viability is determined after viral challenge and the potency is calculated in MlU/mg from the standard curve. The NIBSC reference standard 00/574 (Interferon Beta SER17 Mutein Human, rDNA, E. coli-derived, non-glycosylated) is used as the primary standard. An in-house reference standard is established for routine use through calibrating against the NIBSC reference standard. D. Interferon Beta- lb Analytical Testing Results

Figure 12 shows a SDS-PAGE results of interferon beta- lb produced by the methods described herein. The protein at the end of the series of purification methods appears as a single band reflecting its high purity.

Figure 13 shows a RP-HPLC chromatogram of interferon beta-lb. The main peak comprises >97% of the total protein.

Figure 14 shows a SE-HPLC chromatogram of purified interferon beta- lb. The protein elute mostly as a single peak reflecting it is >99% monomeric.

The biological activity of the interferon beta- lb sample are provided in Table 6. As seen, a sample of the interferon beta- lb produced by the methods described herein has a potency of 36.2 MlU/mg, comparable to that of the NIBSC reference standard (56.9 MlU/mg).

Table 6

Example 13 - Formulation Stability Data

A. Formulation of interferon beta- lb

10 mM sodium acetate buffer, pH 4.0 with 9% (w/v) trehalose and 0.02% (w/v) polysorbate 20

B. Stability testing results of interferon beta- lb prepared as described in the patent filing, in the formulation described

The foregoing formulation was stored at 5°C and 25 °C for up to two months. As indicated in the results below, the protein prepared with the process and formulation described herein was pure, soluble, monomeric, and biologically active at the time of preparation and throughout the 2 months of storage. Table 7: Stability of Interferon Beta- lb at 5 + 3°C

Table 8: Stability of Interferon Beta-lb at 25 + 3°C

Claims

What is claimed is:

1) A method of preparing a non-glycosylated interferon composition comprising the steps of:

a) solubilizing aggregated interferon in a solubilizing media having a pH less than about 7;

b) incorporating the solubilized interferon into a refolding admixture; c) applying an amount of pressure to the refolding admixture; and d) precipitating impurities from the refolding admixture using a salt or acid precipitation,

wherein less than about 5 percent of the non-glycosylated interferon are aggregated and non- monomeric.

2) The method of claim 1 wherein the pH of the solubilizing media is between about 1 to about 4.

3) The method of claim 1 wherein the solubilizing media comprises at least one surfactant that is not sodium dodecyl sulfate.

4) The method of claim 3 wherein the surfactant is a zwitterionic surfactant.

5) The method of claim 1 wherein the solubilizing media is substantially free from an organic solvent and sodium dodecyl sulfate.

6) The method of claim 5 wherein the organic solvent comprises an aliphatic alcohol.

7) The method of claim 1 wherein the solubilizing media further includes at least one chemical chaotropic protein denaturant.

8) The method of claim 7 wherein the chemical chaotropic protein denaturant comprises urea.

9) The method of claim 1 wherein the refolding mixture is provided at a pH between about 7 and about 11.

10) The method of claim 1 wherein the refolding mixture includes at least one

oxidizing/reducing agent that is adapted to promote the formation of the disulfide bonds. 11) The method of claim 10 wherein the oxidizing/reducing agent includes a cysteine or a cysteine/cystine pair.

12) The method of claim 1, wherein the amount of pressure applied to the refolding

admixture is at least about 3000 bars of pressure.

13) The method of claim 1 wherein impurities are precipitated from the refolding admixture using acid precipitation.

14) The method of claim 13 wherein the refolding admixture is brought to a pH between about 5.0 and about 8 to precipitate the impurities.

15) The method of claim 13 wherein the pH of the refolding mixture is reduced using one or more acids.

16) The method of claim 15 wherein the acid is selected from the group consisting of

phosphoric acid, acetic acid, citric acid, succinic acid, sulfuric acid, nitric acid, and combinations thereof.

17) The method of claim 15 wherein the acid comprises phosphoric acid.

18) A pharmaceutical composition comprising an interferon composition being made

according to a method of claim 1.

19) A method of preparing a non-glycosylated interferon composition comprising the steps of:

a) solubilizing aggregated interferon in a solubilizing media having a pH below about 7 and comprising at least one surfactant;

b) incorporating the solubilized interferon into a refolding admixture having a pH between about 7 and about 11 and including at least one oxidizing/reducing agent;

c) applying an amount of pressure to the refolding admixture; and d) precipitating impurities from the refolding admixture by lowering the pH of the refolding admixture,

wherein greater than about 95 percent of the non-glycosylated interferon is disaggregated and monomeric.

20) A method of preparing a non-glycosylated interferon composition comprising the steps of: a) solubilizing aggregated interferon in a solubilizing media having an acidic pH and being substantially free from an organic solvent and SDS;

b) incorporating the solubilized interferon into a refolding admixture;

c) applying an amount of pressure to the refolding admixture;

d) precipitating impurities within the admixture of step c) using a salt or acid precipitation;

e) removing said precipitate via centrifugation and/or filtration; and f) further purifying the non-precipitated protein;

wherein less than about 5 weight percent of the non-glycosylated interferon does not elute at the same time as native IFN-β.

21) The method of claim 20 wherein the step of purifying the precipitated protein

comprises elution through one or a series of chromato graph columns.

22) The method of claim 20 wherein the chromatograph columns are selected from the group consisting of Blue Sepharose, Cu IMAC, SP Sepharose, CM Sepharose and combinations thereof.

23) A pharmaceutical composition comprising:

a nonglycosylated interferon, said interferon comprising less than about 5 weight percent of aggregated protein;

one or more buffers at a concentration from about 10 mM to about 50 mM and providing a pH of about 3.0 to about 5.0; and

a nonionic tonicity modifying agent at a concentration of about 1% to about 15%.

24) The pharmaceutical composition of claim 23, wherein the buffer is selected from the group consisting of acetate, succinate, citrate, formate, and combinations thereof.

25) The pharmaceutical composition of claim 23, wherein the buffer comprises acetate.

26) The pharmaceutical composition of claim 23, wherein the nonionic tonicity modifying agent comprises trehalose.

27) The pharmaceutical composition of claim 23, further comprising a stabilizing agent selected from the group consisting of EDTA, a polyoxyethylene sorbitol ester, a polyoxypropylene-polyoxyethylene ester, a polyoxytheylene alcohol, simethicone, polyethylene glycol, lysophosphatidylcholine, polyoxyethylene-p-t-octylphenol, and combinations thereof, wherein the stabilizing agent is provided in an amount between about 0.005 wt % and about 0.05 wt %.

The pharmaceutical composition of claim 23, further comprising polysorbate 20 or polysorbate 80 in an amount between about 0.005 wt % and about 0.05 wt %.

The pharmaceutical composition of claim 23, wherein said interferon is selected from the group consisting of human IFN-a, IFN-B, and IFN-γ or variants thereof.

The pharmaceutical composition of claim 23, wherein said interferon is human IFN-B or variants thereof, said composition being aggregate free and having reduced immunogenicity when compared to a human IFN-B comprising aggregates.

The pharmaceutical composition of claim 23, wherein said composition is substantially free of SDS.

The pharmaceutical composition of claim 23, wherein said composition is substantially free of HSA.

The pharmaceutical composition of claim 23, wherein biological activity of the interferon is maintained at greater than 90% when stored at about 5°C or about 25 °C for about two months.

A method for treating an autoimmune disorder comprising;

Administering to a patient a therapeutically effective amount of interferon that is produced in accordance with claim 1.

The method of claim 34 wherein the autoimmune disorder is multiple sclerosis.