WO2003057720A2

WO2003057720A2 - Recovery of biofilament proteins from biological fluids

Info

Publication number: WO2003057720A2
Application number: PCT/CA2003/000025
Authority: WO
Inventors: Costas N. Karatzas; Yue Huang; Carl Turcotte
Original assignee: Nexia Biotechnologies, Inc.
Priority date: 2002-01-11
Filing date: 2003-01-13
Publication date: 2003-07-17
Also published as: WO2003057720A3; GB0416692D0; GB2399820A; AU2003202330A1

Abstract

The present invention relates to methods of purifying and recovering biofilament proteins, such as spider silk proteins, from complex biological solutions, including, for example, cell culture media, plant extract, milk, urine, saliva, sweat, tears, seminal fluid, and blood. The process comprises steps of concentration and purification. Additional clarification steps and polishing procedures may also be added. The process employs precipitation, filtration, centrifugation and chromatography methods.

Description

RECOVERY OF BIOFILAMENT PROTEINS FROM BIOLOGICAL FLUIDS

This application is entitled to and claims priority benefit under 35 U.S.C. § 119(e) to U.S. Provisional Application No. 60/347,471, filed January 11, 2002, which is incorporated herein by reference in its entirety.

1. INTRODUCTION

The present invention relates to methods of purifying recombinant biofilament proteins from biological mixtures, in particular, spider silk proteins from complex biological fluids such as blood, urine, or milk derived from transgenic animals. The invention also encompasses recovery of these proteins from other sources such as cell culture media and plant extracts. Also contemplated are methods for producing aqueous solutions comprising such biofilament proteins, suitable for fiber spinning. The biofilament proteins recovered by the methods of the invention can be processed into filaments or fibers by spinning methodologies. These fibers are useful in a wide variety of medical applications and industrial materials.

2. BACKGROUND OF THE INVENTION

The evolutionary success of spiders has been tightly linked to the diversity, production, and use of silks. Silks play a central role in a spider's ability to capture prey, construct shelter, reproduce, and escape from predators. Orb-web spinning spiders have as many as seven sets of highly specialized glands, each producing silk with different amino acid compositions, mechanical properties, and functions. The physical properties of a silk fiber are influenced by a combination of parameters, including the silk's primary amino acid sequence, the spider's diet, spinning mechanism, and environmental conditions in which the silk was produced. Dragline silk, the strongest silk produced by a spider, is used as the safety line and as the frame thread for the spider's web. It has the ability to dissipate energy, and balances stiffness and strength in both extension and compression without evidence of failure. The protein core of dragline silk fibers is composed of a mixture of two soluble proteins, and is secreted from specialized columnar epithelial cells of the major ampullate (MA) gland of orb-weaver spinning spiders. These proteins are designated as ADF-3 and ADF-4 in A. diadematus and as MaSpI and MaSpU in Nephilia clavipes.

Dragline spider silks have a number of high performance mechanical properties that make them superior to the commercially available "superfilaments," such as Spectram

(Allied Signal) and Kevlar™ (Du Pont). For example, dragline silk from Nephilia clavipes is three times tougher than synthetic fibers such as aramid fibers {e.g., Kevlar™) and five times stronger than steel. Vollrath et al, Nature 410:541 (2001). In addition, the dragline silk of A. diadematus demonstrates high tensile strength (1.9 GPa; -15 gpd), greater than that of steel (1.3 GPa) and aramid fibers. In addition to this exceptional strength, dragline silk also exhibits substantial elasticity (about 35%). Gosline et al, Endeavor 10:37-43 (1986).

The unique mechanical properties of spider silk filaments and an inability to domesticate spiders have driven numerous attempts to artificially manufacture spider dragline silks for industrial and medical applications. As a first step towards this goal, partial cDNA clones encoding for the two protein components of dragline silk have been isolated and characterized from two species of orb-web weaving spiders {A. diadematus and N. clavipes). See Guerette, P.A., et al, Science 272 :112 (1996) ; Xu, M., et al, Proc. Natl. Acad. Sci. 87:7120 (1990); Hinman, M.B., et al, J. Biol. Chem. 267:19320 (1992). Genetic information for some of these spider silks has been studied and DΝA fragments coding for various silk proteins have been isolated. See, e.g., U.S. Patents Νos. 5,728,810; 5,994,099; 5,989,894; 5,756,677 and 5,733,771, hereby incorporated by reference.

Biological methods for the production of proteins have shown promise as synthetic alternatives, particularly for complex biochemicals, such as polypeptides and proteins. The simplest and most thoroughly investigated biological method for recombinant protein production employs microbiological systems. Primitive prokaryotic cells are easily engineered to produce a variety of biomolecules, including polypeptides and proteins. Of the bacterial expression systems, Escherichia coli is the most extensively characterized and easiest to manipulate. It has been used to produce a wide variety of polypeptides, in large part, due to its relatively simple genetic expression requirements, short doubling time, and robust growth under laboratory conditions.

In the case of recombinant spider silk proteins, however, the use of E. coli has enjoyed only limited success. E. coli does not readily secrete the desired polypeptides,

- 2 - SUBSTITUTΕ SHEET (RULE 26) adding to the time and expense required to isolate the desired compound from inclusion bodies. Yeast, a lower eukaryotic cell, has also been used to produce recombinant spider silk protein. Fahnestock, S.R. et al, Reviews in Molecular Biotechnology 74:105 (2000). However, yeast, like E. coli, has also failed to produce soluble recombinant spider silk protein in a form that can be processed into a dope solution for spinning into filaments.

Successful expression of silk protein in bacteria and yeast has been limited to relatively short polypeptides. This is due to the highly repetitive structure and unusual secondary structure of the spider silk mRNA, which leads to inefficient translation caused by pausing and premature termination. Moreover, spider silk genes are susceptible to recombination and rearrangement in the repetitive areas of the gene, further limiting successful expression in prokaryote systems and low eukaryote organisms. Spider silk proteins produced by such sources, e.g., bacteria and yeast, must be solubilized using harsh solvents, such as organic acids, di- and trihaloacetic acids and haloalcohols (e.g., hexafluoroisopropanol), as well as the use of chaotropic agents, such as lithium thiocyanate, guanidine thiocyanate, and urea to solubilize the biofilament proteins prior to spinning. In particular, the reported examples of production of spider silk proteins by bacteria and yeast sources all include the use of urea to solubilize spider silk proteins. Fahnestock, S.R. et al, PCT publication WO 94/29450 (1994); Lewis, R.V. et al, Protein Exp. Purif 7:400-406 (1996). The addition of harsh treatments, i.e., solvents and chaotropic agents may change the strength and elasticity of the resulting biofilaments.

Spider silk proteins can be more successfully produced using recombinant methods in a variety of biological fluids, including mammalian cell cultures (for example, MAC-T, BHK, and CHO cells), the milk, urine, saliva, seminal fluid of transgenic animals (cows, goats, sheep, and pigs), plant exudates, and extracts of transgenic plants. Although production of the desired proteins in these systems holds much commercial promise, the recovery of these proteins from the resulting, complex biological fluids can be prohibitively complex, time-consuming and expensive.

Purification of proteins by conventional chromatography is usually achieved using a combination of chromatographic methods including gel-filtration, ion-exchange, hydrophobic-interaction, dye-interaction, affinity and immunoaffinity chromatography.

However, it is rarely possible to purify a protein to homogeneity in a single chromatographic step. See Current Protocols in Molecular Biology, Vol. 2, 10.9.2, Ausubel et al, eds., 1993. Typically, it is necessary to utilize sequential chromatographic steps and

- 3 - SUBSTITUTΕ SHEET (RULE 26) to analyze the protein after each purification step to determine if the protein is homogeneous. In the case of spider silk proteins, chromatographic purification is an even greater challenge because of the unique, repetitive structure of these proteins. Accordingly, there is a need in the art for a cost-effective and scaleable process that can be used to recover and purify biofilament proteins, such as recombinant spider silks, in high yield and high purity, from the complex biological fluids in which they are produced. For the production of biofilaments to be commercially feasible, there is a need for methods that are low in labor-intensity, yet enable recovery and purification to homogeneity of these proteins in very few steps to yield biofilament proteins of sufficient purity and in preparative amounts, such that they can be used to spin fibers on an industrial scale.

3. SUMMARY OF THE INVENTION

The present inventors have discovered methods for efficient, cost-effective recovery of recombinant biofilament proteins from complex biological fluids which allow recovery and purification of these proteins in a minimal number of steps. The methods of the present invention are amenable to large scale production and purification. The methods of the invention yield biofilament proteins in water-soluble form, which may be spun into fibers. These biofilament proteins may be spun into fibers having a commercially useful tensile strength and elasticity. Natural silk proteins are almost completely insoluble when in the form of fibers. Thus, the methods of the invention provide a great advantage because the proteins are recovered in water-soluble form. This facilitates further purification of the recovered proteins, as well as use in manufacturing of silk-related valuable products. For example, the proteins may be spun into fibers, which are useful in cosmetic, medical, military, and industrial applications. The fibers can be used in the manufacture of medical devices such as surgical meshes, sutures, medical adhesive strips, replacement ligaments, and skin grafts, and in a wide range of industrial and commercial products, such as fishing line, netting, clothing fabric, bullet-proof vest lining, container fabric, backpacks, knapsacks, bag or purse straps, cable, rope, adhesive binding material, non-adhesive binding material, strapping material, tent fabric, tarpaulins, pool covers, vehicle covers, fencing material, sealant, construction material, weatherproofing material, flexible partition material, and sports equipment; and, in fact, in nearly any use of fiber or fabric for which high tensile strength and elasticity are desired characteristics. According to the methods of the present invention, recovery and purification of biofilament proteins are achieved by differential solubility {e.g., precipitation, partitioning), interaction with solid phases {e.g., chromatography), or behavior in magnetic or electrical fields. As such, the invention provides methods for recovering and purifying biofilament proteins from biological fluids, comprising various combinations of techniques which are useful for (a) clarifying the mixture containing the biofilament protein, (b) concentrating the biofilament protein in solution, and/or (c) purifying the biofilament from the concentrate. In particular, the methods of the invention encompass the use of one or more techniques, alone or in combination, including but not limited to: continuous centrifugation, tangential flow filtration, salt-induced precipitation, acid precipitation, EDTA-induced precipitation, and chromatography techniques such as anion exchange chromatography, cation exchange chromatography, affinity chromatography, and hydrophobic interaction chromatography, in particular, in an expanded bed absorption chromatography. Depending on the application, the source of the mixture containing the biofilament protein, and other parameters, it is contemplated that the invention encompasses methods whereby any one technique can be sufficient to recover a biofilament protein or produce an aqueous solution thereof suitable for fiber spinning. In additional embodiments, the resulting biofilament protein and aqueous solution thereof can be further prepared for fiber spinning as described in U.S. Patent Application Serial No. , entitled Methods and Apparatus for Spinning Silk Proteins, filed January 13, 2003 (attorney docket no. 9529-012), herein incorporated by reference in its entirety. Even where multiple techniques are used for recovering a biofilament protein or produce an aqueous solution thereof, the methods of the invention obviate the need for complicated, expensive purification processes involving sequential chromatographic steps to purify and recover biofilament proteins. The methods of the invention yield sufficiently pure biofilament proteins and aqueous solutions thereof which are suitable for spinning into fibers for use in medical applications and industrial applications, in particular in applications where high purity is critical. {E.g., U.S. Patent Application Serial No. , entitled Methods and Apparatus for Spinning Spider Silk Protein, filed January 13, 2003 (attorney docket no. 9529-012)) In one embodiment the present invention provides a method for recovering a biofilament protein from a biological fluid comprising subjecting the biological fluid to tangential flow filtration across a membrane of sufficient porosity to permit the biofilament protein to pass through the membrane, thereby producing a permeate comprising the biofilament protein. The biofilament protein is then precipitated from the permeate by adding a salt. Any salt capable of precipitating the biofilament protein may be used, preferably ammonium sulfate. In an alternate embodiment, the biofilament protein is precipitated from the permeate by adding a metal chelator, preferably a calcium chelator, more preferably EDTA (ethylene diamine tetraacetic acid). In a specific embodiment, the salt or EDTA precipitate is re-dissolved in an aqueous solution. In one such embodiment, the salt or EDTA precipitate is solubilized using a chaotropic agent, preferably guanadine hydrochloride, and an aqueous solution comprising the biofilament protein is prepared, e.g., by diafitration and or buffer exchange to remove the chaotropic agent. Preferably the aqueous solution is substantially free of chaotropic agents and organic solvents. Aqueous solutions of biofilament protein produced according to the invention may optionally be further subjected to one or more purification steps using various types of chromatography, either before precipitation or after re-dissolution. In such embodiments, preferably the biofilament protein is isolated at about 80% to about 99% purity. The invention also provides methods of recovering biofilament proteins and producing aqueous solutions thereof suitable for fiber spinning which comprise subjecting a biological fluid to more than one filtration step in order to clarify and concentrate the biological fluid comprising the biofilament protein. These methods comprise subjecting a biological fluid to a first tangential flow filtration step across a first membrane which is of sufficient porosity to permit the biofilament protein to pass through, thereby producing a first permeate comprising the biofilament protein, and a first retentate. The first permeate is then subjected to a second tangential flow filtration step across a second membrane having sufficient porosity to prevent the biofilament protein from passing through, thereby producing a second permeate and a second retentate, wherein the biofilament protein is concentrated in the second retentate. The invention further contemplates methods of recovery wherein the biofilament protein is subjected to a number of concentrating tangential flow filtration steps, wherein in such embodiments, it is the final membrane that has sufficient porosity to prevent the biofilament protein from passing through, thereby producing a final permeate and a final retentate, wherein the biofilament protein is concentrated in this final retentate.

According to the methods of the invention, the second and/or final retentate may be subjected to one or more chromatographic purification steps. In alternate embodiments, the biofilament protein is precipitated from the second and/or final retentate by adding a salt, followed solubilization of the salt precipitate using a solvent, preferably guanadine, and re- dissolution of the protein in an aqueous solution, preferably not containing any organic solvents or chaotropic agents. In an alternate embodiment, the biofilament protein is precipitated from the second and or final retentate by adding a metal chelator, preferably a calcium chelator, more preferably EDTA. Following tangential filtration and salt or EDTA precipitation steps, optionally, additional purification can be achieved using chromatography techniques or further precipitation and redissolution steps.

In another embodiment, the biofilament protein is precipitated from a biological fluid by a salt, preferably ammonium sulfate. In a specific embodiment, the salt precipitate is re-dissolved in an aqueous solution. In one such embodiment, the salt precipitate is solubilized using a chaotropic agent, preferably guanadine hydrochloride, and an aqueous solution comprising the biofilament protein is prepared, e.g., by diafitration and or buffer exchange to remove the chaotropic agent. Preferably the aqueous solution is substantially free of chaotropic agents and organic solvents. Aqueous solutions of biofilament protein produced according to the invention may optionally be further subjected to one or more purification steps using one or more chromatography steps, either before precipitation or after re-dissolution. In such embodiments, preferably the biofilament protein is isolated at about 80% to about 99% purity.

In yet another embodiment, the biofilament protein is precipitated from a biological fluid by a metal chelator, preferably a calcium chelator, e.g., EDTA (ethylene diamine tetraacetic acid). In a specific embodiment, the EDTA precipitate is re-dissolved in an aqueous solution. In one such embodiment, the salt precipitate is solubilized using a chaotropic agent, preferably guanadine hydrochloride, and an aqueous solution comprising the biofilament protein is prepared. Preferably the aqueous solution is substantially free of chaotropic agents and organic solvents. Aqueous solutions of biofilament protein produced according to the invention may optionally be further subjected to one or more purification steps using various types of chromatography, either before precipitation or after re- dissolution. In such embodiments, preferably the biofilament protein is isolated at about 80% to about 99% purity. In yet additional embodiments, the biological fluid comprising a biofilament protein is treated with an acid to achieve a pH in the range of about 3.0 to about 5.5, thereby producing an acidified mixture. The acidified mixture is centrifuged to clarify the biological fluid. The clarified solution can be further subjected to salt and/or EDTA precipitation, followed optionally by further purification using various types of chromatography. In an alternate embodiment, the clarified solution is purified using various types of chromatography without undergoing salt and/or EDTA precipitation.

In another embodiment, the biological fluid comprising a biofilament protein are subjected to one or more chromatographic steps utilizing methods including but not limited to anion exchange chromatography, cation exchange chromatography, affinity chromatography, or hydrophobic interaction chromatography. Any such chromatographic methods are preferably used in an expanded bed absorption chromatography mode.

The methods of the invention are applicable to a wide variety of biological fluids which can be an aqueous solution produced by or derived from an animal, preferably a transgenic mammal, including, but not limited to, milk, urine, saliva, seminal fluid, blood, sweat, and tears, or extracted or derived from a plant. In specific embodiments, the methods of the invention are applicable to milk produced by a transgenic mammal engineered to express a recombinant biofilament protein in its milk, mammalian cell culture, urine produced by a transgenic animal, plant extracts, and plant extrudates, {e.g., as extruded into a hydroponic growth media.) Biofilament proteins recovered and aqueous solutions of biofilament protein produced according to the methods of the invention may be used directly to spin fibers, or they may be subjected to additional concentration and/or purification steps prior to spinning. In a specific embodiment, the biological fluid is milk and the milk is first subjected to filtration, preferably milk produced by a transgenic animal engineered to express the recombinant biological filament in its milk. Biofilament proteins may be recovered according to the invention using tangential flow filtration techniques to achieve clarification and/or concentration of the milk. In one embodiment, the tangential flow filtration uses a membrane of sufficient porosity to allow the biofilament protein to pass through the membrane but exclude casein, fat, micelles and particulate matter from the milk.

In another specific embodiment, the milk is first subjected to acidification and centrifuged in order to exclude casein, fat, micelles and particulate matter from the milk, producing a whey solution. In yet another specific embodiments, the milk is first defatted by continuous centrifugation, thereby resulting in skim milk.

In alternate specific embodiments, the biological fluid is cell culture, urine or defatted milk. In such embodiments, the fluid can be first subjected to expanded bed absorption chromatography. In other embodiments, the fluid is first subjected to salt precipitation. In yet other embodiments, the fluid is first subjected to EDTA precipitation. In yet other embodiments, the fluid is first subjected to EDTA and salt precipitation. In yet other embodiments, the fluid is first subjected to filtration. It is contemplated that each of the above embodiments can further comprises one or more steps in order to further purify the biofilament protein recovered from the biological fluid.

Accordingly, in addition to the methods described above, the invention encompasses the aqueous solutions of biofilament proteins recovered according to the above methods of the invention. The invention also includes methods for producing such aqueous solutions of biofilament proteins from biological fluids. The methods of the present invention can be used to recover proteins from a variety of mixtures, including aqueous solutions, dispersions, suspensions, biological fluids and plant extracts. The methods of the invention are applicable to recovery of all types of proteins, including naturally occurring proteins, transgenically produced proteins, and proteins which have been made using by classical chemical synthesis. The methods are particularly useful to purify biofilament proteins from a biological fluid produced by a transgenic organism expressing a transgene encoding the biofilament protein. The transgenic organism can be a plant or an animal. Preferably, the methods of the invention are used to recover a biofilament protein from a biological fluid.

3.1. Definitions

The term "clarify" as used herein means to remove insoluble materials from the mixture containing the biofilament protein to be recovered. Such materials include, but are not limited to, suspended particles, insoluble particulate matter, cells, cell debris, micelles, plant structures, fats, lipids, fatty acids, nucleic acids, casein, and insoluble non-biofilament proteins. Clarification encompasses removing these materials by filtration, centrifugation, addition of enzymes to hydrolyze and/or solubilize the particulate matter, addition of flocculating agents, and various types of precipitation, including acid precipitation and precipitation by the addition of salts. Clarification can be accomplished by a single one of these techniques or by a combination of any two or more thereof. Clarification means both the complete and partial removal of these materials from a solution. According to the invention, clarification encompasses the removal of 35%, 50%, 60%, 70%, 80%, 90%, 95%, 99%), up to 100%) of these materials from a solution. The term "concentrating" as used herein means increasing the amount, on a weight per volume basis, of the biofilament protein in a solution. The concentration process will result in a solution that is, at least two-fold, three-fold, four-fold, six-fold, eight-fold, tenfold, twenty- fold, fifty- fold, 75-fold, 100-fold or greater concentration in the biofilament protein compared to the raw biological fluid or other starting fluid containing the biofilament protein. According to the invention, the concentration step will result in the biofilament protein accounting for at least 0.1%, 1.0%, 5%, 10%, 20%, 30%, 40%, or even 50%) w/v in the resulting solution. In certain embodiments, the concentration step will result in the biofilament protein accounting for as much as 60%, 70%, 80%, or 90% w/v in the resulting solution. Preferably, the biofilament protein will be concentrated to a greater extent than the contaminants. The mixtures used in the methods of the present invention can be concentrated by any one or more of a variety of techniques known to those skilled in the art. For example, concentration is accomplished by precipitating the biofilament protein from solution by addition of a salt, followed by re-dissolving the precipitated biofilament protein in an aqueous solution, preferably an aqueous buffer. Alternatively, concentration may be achieved by a filtration method as described herein.

The term "biological fluid" as used herein means an aqueous solution produced by an animal, preferably mammal, including, for example, milk, urine, saliva, seminal fluid, blood, sweat, or tears. Also included is cell culture media that has been conditioned by native or transgenic cells. Preferably, the cells are of mammalian origin and secrete a biofilament protein, although plant and fungal cell culture media are also encompassed by the invention. Further included are plant extracts which include aqueous or organic extractions of any plant structure including shoots, leaves, roots, stems and seeds. Plant extracts can also be derived from exudates or guttation fluids. The phrase "solution of a biofilament protein suitable for spinning into fibers" means any liquid containing a biofilament protein that is amenable to extrusion and/or spinning fibers. Preferably, the solution comprises a substantially pure biofilament protein, e.g., having about 80-99% purity. Preferably the solution has a pH about 11. Preferably the solution is aqueous and does not contain any organic solvents or chaotropic agents. Preferably, the solution contains a biofilament protein which is a recombinant spider silk protein at about 2-40% (w/v) spider silk protein, although more concentrated solutions may be used {e.g., 40-80%). Aqueous buffers that promote a liquid crystal structure are most preferable. By "purify" is meant a process of removing or separating contaminating biological molecules from the biofilament to be isolated. Contaminating biological molecules to be removed include, for example, non-biofilament proteins (and even biofilament proteins other than those desired to be purified), lipids, fatty acids, nucleic acids, cells, cellular debris, and particulate matter. The process of purification removes, at least 50%, but preferably 75%, more preferably 85%, even more preferably 95%, or most preferably 99%, of the contaminating biological molecules present in the raw fluid. Accordingly, a solution which is "substantially pure" for biofilament proteins is one in which biofilament proteins account for 50%, 60%, 70%, 75%, 80%, 85%, 90%, 95%, or even 99% and up to 100% of the soluble protein present in solution.

By "biofilament" is meant a fibrous protein that is normally produced and secreted by any one of a variety of insects and arachnids. Biofilaments are composed of alternating crystalline and amorphous regions. Exemplary biofilaments include spider silk as well as other externally spun fibrous protein secretions found in a variety of insects {e.g., Bombyx mori). Preferable biofilaments include those that when secreted are subjected to shear forces and mechanical extension, have a poly-alanine segment, forming an α-sheet transition, forming a β-crystal that stabilizes its structure. Preferably, the amorphous domain of a biofilament forms a n-pleated sheet with inter-n-sheet spacings between about 3 angstroms and 8 angstroms in size, and more preferably, between 3.5 angstroms and 7.5 angstroms in size.

By "substantially identical" is meant a polypeptide or nucleic acid exhibiting at least about 50%, 70%, 85%, 90%, 95%, or even 99% identity to a reference amino acid or nucleic acid sequence. Unless otherwise specified for polypeptides, the length of comparison sequences will generally be at least 20 amino acids, preferably at least 30 amino acids, more preferably 40 amino acids and most preferably 50 amino acids. Unless otherwise specified for nucleic acids, the length of comparison sequences will generally be at least 60 nucleotides, preferably at least 90 nucleotides, and more preferably at least 120 nucleotides.

4. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a flowchart depicting various exemplary purification techniques described in detail herein and the combinations thereof which may be used to purify biofilament proteins from biological fluids according to methods of the invention. These biological fluids include, for example but not limited to cell culture media, plant extracts, urine from transgenic animals or milk from transgenic animals.

FIG. 2 is a plot of electrical charge versus pH data for recombinantly produced MaSpII, ADF-3, MaSpI, MiSpI and flagelliform spider silk proteins. Charge versus pH data for bovine serum albumin (BSA) is included for comparison purposes. The data was generated by inputting representative amino acid sequences for these proteins into the "DNAMAN" computer program.

FIG. 3 is a schematic representation of a tangential flow filtration system which may be used for clarification and concentration of biological fluids according to the methods of the invention. The system may be used in the clarification and concentration of cell culture media, as described in Example 4.

FIG. 4 is a schematic representation of a tangential flow filtration system which can be used for both clarification and concentration of biological fluid according to the methods of the invention. The system may be used in the clarification and concentration of milk produced by transgenic animals, as described in Example 5 and Example 15.

FIG. 5 is the amino acid sequence of a representative MaSpI protein which may be recovered according to the methods of the invention, arranged so that the amino acid repeat motifs can be observed.

FIG. 6 is the amino acid sequence of a representative MaSpTJ protein which may be recovered according to the methods of the invention.

FIG. 7 is the amino sequence of a representative ADF-3 protein which may be recovered according to the methods of the invention.

5. DETAILED DESCRIPTION OF THE INVENTION

The present invention provides methods for recovering and isolating biofilament proteins from a wide variety of mixtures and biological fluids using various combinations of one or more methods which clarify or concentrate the biological fluid and/or purify the biofilament protein. Preferably, the methods of the invention recover and isolate such biofilament proteins in an aqueous solution appropriate for spinning. In certain embodiments, the aqueous solution is substantially free of organic solvents or chaotropic molecules or concentrated solutions of organic acids. Described below are detailed procedures for recovering these proteins according to the principal steps of the invention, using various techniques, including precipitation, filtration and chromatographic methods.

The inventors have found that biofilament proteins produced recombinantly in eukaryotic cells, in particular mammalian cells, preferably mammalian epithelial cells, and precipitated according to the invention can be readily re-dissolved and then diafiltered into aqueous buffer solutions which are substantially free of harsh solubilizing agents such as organic solvents or chaotropic agents. More preferably, the cells are mammary epithelial cells. Recombinant spider silk proteins previously produced in E. coli or yeast and precipitated reportedly could only be dissolved and maintained in solution in strong denaturing solvents, such as hexafluoroisopropanol, guanidine hydrochloride, urea and/or organic acids. Not to be bound by any particular mechanism, but this difference in solubility maybe attributed to the presence of the carboxy- terminus in spider silk proteins produced in epithelial cells, suggesting that the more hydrophilic carboxy-terminus of about 100 amino acids, which is absent in the E. coli and yeast-produced proteins, increases the solubility of the secreted silks.

5.1. Salt-Induced Precipitation

According to the methods of the invention, the technique of salt-induced precipitation may be used to provide a biofilament protein solution of sufficient purity that the solution may be used directly for spinning fibers for use in industrial applications, without the need for subsequent purification steps. Alternatively, if the biofilament protein is to be used in medical applications, such as for sutures or skin graft substitutes, it may be necessary to further purify the biofilament protein solution by additional chromatography steps prior to spinning. According to the methods of the invention, where the biological fluid is cell culture media, urine, or a plant extract, salt-induced precipitation may be applied directly to the raw biological fluid as it is received, without the need for preliminary purification or treatment. Determination of the treatment of other biological fluids is within the skill of the art. Alternatively, when the biological fluid is milk produced by a transgenic animal, salt- induced precipitation is applied after the milk has been clarified. Any salt capable of effecting precipitation may be used, including but not limited to ammonium sulfate, sodium sulfate, sodium chloride, sodium acetate, potassium chloride, potassium sulfate, and potassium phosphate. In a preferred embodiment, an aqueous solution of ammonium sulfate

- 13 - SUBSTITUTΕ SHEET (RULE 26) may be added to the mixture, causing the biofilament protein to precipitate. The salt is added as an aqueous solution, preferably at a concentration of about 12-50 grams per 100 ml of solution {i.e., 10-70% of the saturation point of the salt at 4° C). In a particularly preferred embodiment, the salt is added as an aqueous solution wherein the concentration is 10-20% of the saturation point of the salt at 4°C. The mixture is incubated at a temperature of about

1-15 °C, preferably at about 4°C. The mixture is then centrifuged, preferably at about 20,000xg at about 4°C for one hour to pellet the precipitated biofilament protein. According to the methods of the invention, the conditions for precipitating and pelleting the biofilament protein will vary depending on the machine, volume, and/or scale and such conditions can be determined by the skilled artisan. The supernatant is removed and the biofilament protein is then re-dissolved in an appropriate buffered aqueous medium, such as phosphate buffered saline. Examples of buffer solutions which may be used include, but are not limited to, 20 mmol glycine with pH 10 and 20 mmol phosphate, pH 7.4. Centrifugation can be repeated to remove any undissolved solids. Preferably, the re-dissolved biofilament protein solution is more concentrated than the original biological fluid; however, this is not required. In preferred embodiments, the biofilament protein is solubilized by a solvent or chaotropic agents such as guanidine HCl or urea, followed by buffer exchange or diafiltration into an aqueous solution that does not contain the solvent or chaotropic agent. The above-described procedures for salt-induced precipitation followed by redissolution of the biofilament protein are particularly useful when the biological fluid is cell culture media, urine from a transgenic animal, or plant extracts or exudates. According to the invention, when salt-induced precipitation is used, a separate clarification step is not necessary because centrifugation causes the insoluble material, which is typically lipophilic, to rise to the top of the solution, where it can be easily removed by routine procedures. Thus, clarification of the raw biological fluid and concentration of the biofilament solution can be achieved in the same step, while simultaneously removing contaminating, non- biofilament proteins. The technique of salt-induced protein precipitation may also be applied to a solution that has already been clarified, to provide a concentrated solution suitable for spinning or further purification steps.

5.2. EDTA-Induced Precipitation

According to the methods of the invention, EDTA precipitation (or metal chelator, preferably calcium chelator induced precipiation) may be used to provide a biofilament protein solution of sufficient purity that the solution may be used directly for spinning fibers for use in industrial applications, without the need for subsequent purification steps, particularly in the recovery of biofilament protein from defatted milk. Alternatively, if the biofilament protein is to be used in medical applications, such as for sutures or skin graft substitutes, it may be necessary to further purify the biofilament protein solution by additional chromatography steps prior to spinning.

According to the methods of the invention, where the biological fluid is cell culture media, urine, or a plant extract, EDTA precipitation may be applied directly to the raw biological fluid as it is received, without the need for preliminary purification or treatment. Determination of the treatment of other biological fluids is within the skill of the art.

Alternatively, when the biological fluid is milk produced by a transgenic animal, EDTA- induced precipitation is applied after the milk has been clarified. The methods also encompass use of any metal chelator, preferably a calcium chelator to precipitate the biofilament protein. According to the methods of the invention, the conditions for precipitating and pelleting the biofilament protein will vary depending on the machine, volume, and/or scale and such conditions can be determined by the skilled artisan. The supernatant is removed and the biofilament protein is then re-dissolved in an appropriate buffered aqueous medium, such as phosphate buffered saline. Examples of buffer solutions which may be used include, but are not limited to, 20 mmol glycine with pH 10 and 20 mmol phosphate, pH 7.4. Centrifugation can be repeated to remove any undissolved solids.

Preferably, the re-dissolved biofilament protein solution is more concentrated than the original biological fluid; however, this is not required. In preferred embodiments, the biofilament protein is solubilized by a solvent or chaotropic agents such as guanidine HCl or urea, followed by buffer exchange or diafiltration into an aqueous solution that does not contain the solvent or chaotropic agent.

The above-described procedures for EDTA precipitation followed by redissolution of the biofilament protein are particularly useful when the biological fluid is cell culture media, urine from a transgenic animal, or plant extracts or exudates. According to the invention, when EDTA precipitation is used, a separate clarification step is not necessary because centrifugation causes the insoluble material, which is typically lipophilic, to rise to the top of the solution, where it can be easily removed by routine procedures. Thus, clarification of the raw biological fluid and concentration of the biofilament solution can be achieved in the same step, while simultaneously removing contaminating, non-biofilament proteins. The technique of EDTA protein precipitation may also be applied to a solution that has already been clarified, such as defatted milk to provide a concentrated solution suitable for spinning or further purification steps.

5.3 Acid Precipitation

According to the methods of the present invention, acid precipitation may be used as a clarification step in the recovery of biofilament proteins from biological fluids. Acid precipitation is particularly useful to remove insoluble materials such as fats and casein from milk prior to isolation of the protein. According to the technique of acid precipitation, the pH of the milk is adjusted by treatment with an acid until a pH of about 2.0-6.5 is achieved. The pH achieved can be 2.0- 3.0, 3.0-4.0, 4.0-5.0, 5.0-6.0, preferably less than 4.0, more preferably 2.0-4.0, most preferably 3.0-4.0. Preferred acids include, but are not limited to acetic acid, hydrochloric acid and phosphoric acid; however, any acid capable of bringing the pH to the desired range may be used. Generally, the acid precipitation step is carried out at a temperature of about

1-25 °C, but preferably at about 4 °C. In one embodiment, the temperature at which the acid precipitation step is carried out can be 1-25 °C, preferably 2-15 °C, more preferably 3- 10 °C, most preferably 4-5 °C. After adding the acid, the acidified mixture is centrifuged at approximately 200,000xg, at 4-5 °C, 1-25 °C, preferably 2-15 °C, more preferably 3-10 °C, most preferably 4-5 °C, causing the casein to collect at the bottom of the mixture and the fat and other lipophilic material to rise to the top. In an embodiment, the acidified mixture is centrifuged at least 30 minutes after adding the acid, at least one hour after adding the acid, at least 2 hours after adding the acid, at least 6 hours after adding the acid, at least 12 hours after adding the acid, up to 24 hours after adding the acid. In preferred embodiments, the acidified mixture is stirred prior to centrifugation. The resulting clear solution, known as the "whey" fraction, contains the desired biofilament protein and various whey proteins. If necessary, solid phase whey proteins can be further fractionated by additional centrifugation. The layer of fat on top of the solution is easily removed, and the whey fraction is either filtered or simply decanted from the solid casein. The biofilament protein may then be isolated directly from the whey fraction by salt-induced precipitation or by chromatography techniques as described hereinbelow in Section 5.4.

Clarification of milk may also be accomplished by first centrifuging the raw mixture, preferably in a continuous manner, to remove fats and lipids, resulting in "skim milk." Preferably, the milk is centrifuged 10,000xg at 4°C for 30 minutes. The casein is then removed from the skim milk by a filtration method or by acid precipitation. In preferred embodiments, the casein is removed from the skim milk by EDTA precipitation. The above-described clarification procedures may be repeated one or more times to obtain the desired level of clarification.

5.4 Tangential Flow Filtration

According to the methods of the invention, various filtration methods may be employed to achieve concentration, clarification, or both, of the biological mixture containing the protein to be recovered. Any one or more of a wide variety of filtration methods known in the art may be used, depending on the type of raw mixture or biological fluid being purified, and the production system employed. Preferably, the filtration method used is tangential flow filtration ("TFF").

The technique of TFF uses a filtration system that is operated in a "cross flow" or a "tangential flow" configuration. In contrast to conventional filtration techniques, TFF involves recirculation of the feed stream pumped across, i.e., tangent to, the filtration membrane surface. This tangent flow sweeps the membrane surface and prevents rapid plugging of the filtration device. In operation, as the feed stream is pumped through the filtration cartridge, the retentate, comprising species excluded by the membrane pores, continues through the circulation loop while the permeate, including solvent and solutes transported through the membrane, is collected on the other side of the membrane.

A basic TFF system comprises a pump, feed reservoir, permeate reservoir, pressure gauges, valving and a filtration membrane cartridge. The TFF system may use a hollow fiber polymeric, plate and frame, tubular ceramic, or spiral wound polymeric configuration. Examples of TFF systems suitable for use in the present invention are commercially available from Amersham Biosciences.

5.4.1. Clarification by TFF

TFF may be the primary means of clarification of a biological fluid; however, centrifugation can be substituted for, or used in conjunction with, a filtration protocol. In alternate embodiments, acid precipitation can also be used to clarify a biological fluid, preferably in conjunction with centrifugation. Whether centrifugation should be used prior to, subsequent to, or in place of filtration of the biological fluid will depend on fluid composition, the types of filtration processes available, and other technical considerations. For biological fluids containing large amounts of insoluble material, including, for example, cell culture media, centrifugation prior to filtration is optimal. In this case, a continuous centrifugation protocol is preferable. For clarification of cell culture media, a micro filtration membrane, preferably having a pore size in the range of about 0.05 μm to about 1.0 μm, is used in the TFF system; however, other pore sizes can be used for specific applications. Pore size is about 0.05 to 0.5, e.g., 0.05 to 0.1 μm, 0.1 to 0.2 μm, 0.2 to 0.3 μm, 0.3 to 0.4 μm, or 0.4 to 0.5 μm. Centrifugation, preferably continuous centrifugation, may be used in addition to, or as an alternative to TFF, to remove cell debris from cell culture media. Clarification of biological fluids, such as milk, may be achieved using either an ultrafiltration membrane with a molecular weight cutoff of 500 kDa to 1000 kDa or a microfiltration membrane with a pore size of 0.1 to 0.2 μm. Preferably the molecular weight cutoff is 750 kDa. Preferably, the process is operated at room temperature; however any temperature in the range of about 4 to 60°C may be used. The process can be operated at a temperature of 4 to 60°C, preferably

10 to 40°C, more preferably 15 to 30 °C, most preferably about 20 to 25 °C.

For clarification, the main criterion for selection of the membrane is that the soluble biofilament protein must be able to pass through the membrane to the permeate while cellular debris, particulate matter, lipids, fats, casein, and other flocculate are retained in the circulation stream, also referred to herein as the retentate. The selection of the membrane pore size will depend on the relative size of the biofilament protein being recovered. For example, recombinant spider silk proteins are typically about 10-300 kD in size. A typical recombinant spider silk protein is about 60 kD. When recovering recombinant spider silk proteins from a biological fluid such as milk, an ultrafiltration membrane with a pore size of about 750 kD is effective.

The technique of diafiltration can be used in conjunction with a TFF clarification process to increase the amount of biofilament protein recovered. In this technique, an additional volume of water or other liquid is added to replace the volume of permeate that has already come through the filter. Alternatively, a buffer solution comprising, for example, about 10-200 mmol of arginine may be used for adding additional volume. In an embodiment, the buffer solution comprises, 10-200 mmol of arginine, preferably 20-100 mmol of arginine, more preferably 30-70 mmol, most preferably 40-60 mmol of arginine. In a particularly preferred embodiment the buffer solution comprises 50 mmol arginine. Moreover, the methods of the invention contemplate use of any buffer solution with a pH of about 6.8. The pH of the buffer can be 5.0 to 8.0, preferably 5.5-7.5, more preferably 6.0 to 7.0. In preferred embodiments, the pH of the buffer is the pH of milk. Other buffer solutions contemplated comprise citrate or phosphate. Such citrate or phosphate buffers comprise 10- 200 mmol, preferably 20-100 mmol, more preferably 30-70, most preferably 40-60 mmol of citrate or phosphate, respectively. In particularly preferred embodiments, the buffer comprises 50 mmol of citrate or phosphate. Additional volumes, e.g., 1-20 such volumes may be added. Typically, addition of about 6-10 additional volumes is sufficient to achieve transport of more than 90% of the biofilament protein from the raw biological fluid through the membrane into the permeate.

5.4.2. Concentration bv TFF

Preferably, TFF is used to concentrate a mixture that has already been clarified, either by filtration or by acid precipitation. TFF can also be used to concentrate a mixture that has been salt or EDTA or salt and EDTA precipitated. Concentration may be achieved using tangential flow ultrafiltration as described above, using an ultrafiltration membrane with a cutoff size in the range of about 5-100 kD at an operating temperature of about 1- 60°C. In the methods of the invention, the cutoff size is in the range of about 5-10 kD, 10- 20 kD, 20-30 kD, 30-40 kD, 40-50 kD, 50-60 kD, 60-70 kD, 70-80 kD, 80-90 kD. The operating temperature can be 1-5 °C, 5-10°C, 10-15 °C, 15-20°C, 20-30°C, 30-40°C, 40-

50°C, or 50-60°C, preferably less than 25 °C, more preferably less than 10°C, most preferably less than 5°C but no lower than 1 °C. In a preferred embodiment, the cutoff size is 10-30 kD and the operating temperature is 4°C. In the TFF concentration step, the biofilament protein remains in the circulation stream (retentate), while low molecular weight components and impurities pass into the permeate. Preferably, the process results in an increase in protein concentration of at least about two-fold.

For example, for a sample of milk produced by a transgenic animal which has about 1 gram per liter biofilament protein, and a total protein content of about 60 g protein per liter, clarification can be expected to reduce the total protein content to about 12 g per liter. In this case, the TFF concentration step can achieve an increase in protein concentration of about five-fold. With cell culture media, the concentration step can generally achieve an increase in protein concentration of about ten- fold to about 100-fold. Diafiltration can be used in conjunction with TFF, as described above, to increase the amount of impurities removed in the concentration process. In a preferred embodiment, the clarification and concentration processes are performed in a tandem TFF diafiltration unit, which eliminates the need for additional diafiltration buffer. For example, simultaneous clarification and concentration can be achieved using a TFF system configured with two constant volume diafiltration units, with permeate from the clarification unit being used as the diafiltration buffer for the concentration unit.

Biofilament concentration can be achieved by alternative methods appropriate to the characteristics of each biological fluid. Recombinant spider silk proteins in solutions that are low in contaminant proteins and insoluble materials may be recovered directly by salt- induced precipitation, followed by filtration or centrifugation. Diafiltration may then be used to remove the precipitating buffers. This method is particularly useful for the recovery of silk proteins from the urine of transgenic animals with kidney or bladder specific compression of the protein. Urine typically has little if any particulate matter, is acidic in character and low in protein and lipid content, making it ideal for direct application of the concentration step, without the need for a separate clarification step. Clarification by TFF followed by concentration by TFF are also preferred for the recovery and purification of biofilament proteins from urine.

5.5. Purification by Chromatographic Methods

Salt-induced precipitation, and the clarification and concentration methods described herein typically result in biofilament solutions within which the biofilament protein is about 80-95% pure. Similar purity levels can be similarly obtained using EDTA and acid precipitation methodology. Solutions of biofilaments having purity in this range are suitable for spinning fibers that may be used in a wide variety of industrial products, such as fishing lines, cords, or bullet-proof vests. However, for biofilaments to be used in medical applications, higher purity may be required. Accordingly, the biofilament solutions of the invention may be further purified using various chromatographic techniques described herein. According to the invention, the resulting purified, concentrated solution is substantially free of non-biofilament proteins, and has a biofilament purity in excess of

95%. More preferably, the biofilament protein is 98-99% pure. Biofilaments will account for at least 60%>, 70%, 80%, 90%, 95%, or even 99% of the soluble protein in solution, on a weight basis. Procedures according to this step can be repeated or performed in series until the required level of purity is achieved.

The biofilament protein can be purified using one or more of various chromatographic techniques known to those skilled in the art, or may be recovered by simply precipitating the protein from solution. Chromatographic methods are optimized for each biofilament protein, based on its physico-chemical properties. The most useful chromatographic methods include, but are not limited to, anion exchange, cation exchange, size exclusion, affinity, and hydrophobic interaction chromatography. Fast performance, reverse phase, or normal phase high-performance liquid chromatographic methods can also be used to recover biofilament proteins according to the invention.

The resins used in the chromatographic steps of the invention may be bonded to suitable solid supports including glass beads, silica-based resins, cellulosic resins, agarose beads, cross-linked agarose beads, polystyrene beads, cross-linked polyacrylamide resins and other supports that are insoluble under proper operating conditions. These supports can be modified with reactive groups that allow attachment of proteins by amino groups, carboxyl groups, sulfhydryl groups, hydroxyl groups and/or carbohydrate moieties. Examples of coupling chemistries include cyanogen bromide activation, N- hydroxysuccinimide activation, epoxide activation, sulfhydryl activation, hydrazide activation, and carboxyl and amino derivatives for carbodimide coupling chemistries. These and other solid media are well-known and widely used in the art and are available from commercial suppliers. Selection of a particular method is a matter of routine design and is determined in part by the properties of the chosen support. These exchangers and ligands can also be cross-linked to a membrane and used as membrane chromatography media.

The present inventors have discovered methods for ion exchange chromatographic purification of recombinantly produced biofilament proteins, based on electrical charge versus pH data for these proteins. To obtain this data, a representative amino acid sequence of a recombinantly produced MaSpII protein was input into the "DNAMAN" computer program. For every 0.2 pH unit, the program calculated the corresponding electrical charge of the protein at that pH. The same calculations were also performed for representative amino acid sequences for recombinantly produced MaSpI, ADF-3, MiSpI and flagelliform spider silks.

Figure 2 is a graph of the charge versus pH data for these proteins. Charge versus pH data for bovine serum albumin (BSA) is included for comparison purposes. The figure shows that recombinantly produced MaSpII and ADF-3 proteins have a charge of zero or very close to zero over a wide pH range (4 to 8). In contrast, commonly occurring proteins such as albumin rapidly acquire a significant charge as the buffer pH moves away from the isoelectric point {i.e., the pH at which the protein has zero charge). At pH<4, the MaSpII and ADF-3 proteins have only a slight positive charge. The proteins remain essentially uncharged until about pH 9 when a negative charge is acquired. Thus, the only ion exchange chromatography method useful for purifying ADF-3 and MaSpπ is anion exchange chromatography at pH>9.

In contrast to the MaSpII and ADF-3 proteins, MaSpI is positively charged over a wide range, from pH 2 to its isoelectric point of about pH 10. In addition, MaSpI can accommodate a negative charge at pH above its isoelectric point. As MaSpI can accommodate both positive and negative charges, in theory, either cation or anion exchange chromatography could be used to purify MaSpI proteins. However, in practice, cation exchange chromatography is preferred for MaSpI purification at pH 7.5 to 8, because most contaminant proteins are negatively charged. Thus, the contaminating proteins will pass through the column at this pH, while the MaSpI protein is captured on the column.

The MiSpI protein, like the MaSpI protein, exhibits a positive charge over a wide pH range, but has a slightly lower isoelectric point of about pH 9, above which the protein takes on a negative charge. Because MiSpI has a charge profile similar to that of MaSpI, cation exchange chromatography techniques suitable for purification of MaSpI should prove suitable for MiSpI as well.

The recombinantly produced flagelliform protein has a similar charge profile to that of the MaSpII and ADF-3 proteins, except at pH less than about 5, where the flagelliform protein accommodates a positive charge while the charge of the MaSpII and ADF-3 proteins is close to zero. Thus, it is expected that both anion and cation exchange chromatography techniques will be suitable for purifying flagelliform proteins. (a) Anion Exchange Chromatography

As discussed above, anion exchange chromatography is preferred for the purification of ADF-3 and MaSpII proteins. During sample loading, the biofilament protein is negatively charged and retained on a positively charged resin in the column. Suitable anion exchange media include, for example, derivatized dextrans, agarose, cellulose, polyacrylamide, specialty silicas, and the like. These materials are also suitable for cation exchange chromatography. Suitable anion exchange resins include, for example, diethylaminoethyl (DEAE), quarternary aminoethyl (QAE), and quartemary ammonium (Q). The solution containing the biofilament protein is loaded onto an anion exchange chromatography column pre-equilibrated with a low ionic strength buffer. The pH of equilibration buffer is typically in the range of about pH 5 to about pH 12, preferably about pH 10. The biofilament proteins bind to the anion exchange resin while most contaminant proteins pass into the eluant. The biofilament protein is then selectively eluted from the column, thereby separating it from contaminant proteins remaining on the column. For example, the biofilament protein may be eluted using a high salt buffer {e.g., 150 mmol NaCl). Alternatively, the pH of the buffer can be decreased to elute the biofilament protein. Isolation by anion exchange chromatography typically yields substantially pure biofilament proteins. Such material is suitable for use in a wide variety of industrial applications. Material of 99% or greater purity suitable for use in medical and pharmaceutical applications can be obtained by performing one or more additional chromatography steps after the isolation step. Such steps can include, for example, additional anion exchange chromatography, affinity chromatography, size exclusion chromatography, hydrophobic interaction chromatography, and/or cation exchange chromatography. These chromatographic methods can be performed in addition to, or instead of anion exchange chromatography (if appropriate). Following one or more chromatographic separations, the buffer may be substituted or modified to facilitate filament spinning. Ultrafiltration or dialysis can be used for buffer exchange and/or further concentration of the purified protein.

(b) Cation Exchange Chromatography

Cation exchange chromatography is particularly useful for the purification and isolation of MaSpI proteins. In cation exchange chromatography, the biofilament protein is positively charged and retained on a negatively charged resin in the column. Suitable cation exchange resins may be any media comprising carboxymethyl (CM), sulfopropyl (SP), methyl sulfonate. The solution containing the biofilament protein to be recovered is loaded onto a cation exchange chromatography column pre-equilibrated using a low ionic strength buffer {e.g., 10 mmol phosphate buffer or an acidic buffer), typically in the range of about pH 3 to about pH 9, preferably about pH 7.5-8. The biofilament protein binds to the column while most of the contaminant proteins flow through the column. The biofilament protein is then selectively eluted using a high salt buffer {e.g., 500 mmol NaCl). Alternatively, the biofilament protein can be selectively eluted using a buffer with increased pH. (c) Affinity Chromatography

Affinity chromatography is also a preferred technique for purification and isolation of biofilament proteins. The affinity ligands used in this technique can be any molecule which specifically binds the biofilament protein, including monoclonal antibodies, peptides, dye-based molecules, and other natural or chemically-modified molecules. Polyclonal antibodies such as those discussed in Section 6.10 herein are also useful for purification and isolation of biofilament proteins. The affinity chromatography column is equilibrated with a buffer having a pH from about 2 to about 11. After loading the solution containing the biofilament protein to be recovered, the column is washed with a buffer containing a salt and/or detergents to remove non-specific binding proteins. The biofilament protein is then selectively eluted using an appropriate buffer.

For example, in the case of conditioned cell culture media containing a biofilament protein with a histidine tag, the media may be purified using a Nickel column {e.g., Ni-NTA column, Qiagen), in which the nickel specifically interacts with the histidine. The cell culture media is adjusted to contain 6M urea, and then loaded onto a Ni-NTA column and processed as described by the manufacturer. The bound proteins may be eluted using a wash buffer containing lOOmM imidazole. Dye-based molecules {e.g., PIKSI-M test column kit, Prometic) and antibodies {e.g., rabbit, see Example I) may also be used as affinity columns to purify biofilament proteins according to the invention. (d) Hydrophobic Interaction Chromatography

Hydrophobic interaction chromatography (HIC) is also useful in the methods of the invention, both as a primary means of isolation or as an additional step. With HIC, the biofilament proteins are separated from contaminant proteins based on interactions with different hydrophobic groups on their surfaces. Exemplary HIC media include those media derivatized with phenyl, butyl, or octyl groups, such as Phenyl-Sepharose FF (Pharmacia),

Toyopearl butyl 650 (Toso Haas, Montgomeryville, PA), Octyl-Sepharose (Pharmacia) and similar other media; or polyacrylic resins, such as Amberchrom CG 71 (Toso Haas) and related media. The HIC column is pre-equilibrated with a high ionic strength buffer having a pH from about 2 to about 11. The pH of the buffer can vary depending on the charge of the protein being purified and can be determined using the chart depicted in FIG. 2 herein.

When the solution containing the biofilament protein {i.e., a hydrophobic protein) is loaded on the column, the biofilament protein and other hydrophobic molecules bind to the resin, while most of the contaminant proteins pass through the column. The biofilament protein is then selectively eluted using an appropriate buffer, preferably water. In specific embodiments, the buffer is a low conductivity solvent, preferably a no-salt buffer. Other solvents contemplated are guanadine HCl or urea.

Any of the chromatographic media described above can be cross-linked to a membrane and used in a membrane chromatography configuration to improve the efficiency of purification. In order to achieve greater purity, a finishing step may be performed in which any of the above-mentioned chromatography methods are used in sequence or repeatedly until the desired purity is achieved. In addition, size exclusion chromatography (SEC), which separates the biofilament protein from contaminant proteins based on molecular size difference, can be used as a finishing step. The recovered biofilament protein can also be subjected to additional ultrafiltration steps to improve purity. Preferably, the finishing step removes 50%> of the remaining contaminating proteins, more preferably 75%, even more preferably 85%, most preferably 95%, or even 99% of the remaining contaminating proteins removed. (e) Expanded Bed Absorption (EBA) Chromatography

Any of the above-described chromatography methods may also be used in an expanded bed absorption (EBA) mode for recovering biofilament proteins. EBA chromatography is particularly useful for isolation of proteins from cell culture media and from defatted (skim) milk produced by transgenic animals. In an EBA process, the biofilament protein is captured when the protein solution is pumped into the column in an upward direction, thereby expanding the packed column bed. Subsequently, the column is washed with an appropriate solution, which also flows in an upward direction, while the biofilament protein is retained on the solid support. Elution of the biofilament protein occurs when the eluant is pumped in the downward direction, which packs the column. In more traditional packed-bed methods, in which the resin is confined between the bottom of the column and the flow adaptor, clogging occurs when particulate matter and cell debris are unable to flow around the closely packed resin beads. In contrast, EBA columns are fed from below, and the adaptor is held away from the packed-resin level, giving the resin room to expand, thus creating spaces between the beads. As buffer is injected from below, the resin becomes fluidized, and the beads form a stable concentration gradient when their sedimentation velocity equals the upward liquid flow velocity. The goal is to achieve a flow rate in which the terminal velocity of the feedlot particulates is not exceeded by that of the resin beads. To achieve this differential, resin beads which have been modified to include inert quartz or metal alloy cores may be used depending on the application.

Unlike traditional resins, in which the beads are relatively uniform in size, the beads of EBA resins are variable, typically ranging in size from 50 to 400 mm. Thus, the larger particles populate the lower portion of the fluidized bed while the smaller particles populate the upper portion. As the sample feed is injected, the articulates and cell debris move freely around the resin beads and eventually leave through the top of the column. The biofilament protein of interest is retained on the column. The column is generally washed, and then allowed to repack. The flow is then reversed, and the desired protein is eluted from the column using additional methods .

The use of EBA chromatography eliminates the need for a separate clarification step. In the case of cell culture media, the raw solution may be applied directly to the EBA chromatography system. The cell debris and other soluble materials are retained in the column but do not clog the column as would be the case with standard chromatographic methods. In the case of milk produced by a transgenic animal, it is preferred that the fat is removed from the milk to produce skim milk prior to subjecting the milk to the EBA step. Defatting the milk may be achieved by standard methods well known to those skilled in the art.

5.6. Spider Silk Proteins Recovered According to the Methods of the

Invention

Spider silk proteins are designated according to the gland or organ of the spider in which they are produced. Spider silks known to exist include major ampullate (MaSp), minor ampullate (MiSp), flagelliform (Flag), tubuhform, aggregate, aciniform, and pyriform spider silk proteins. Spider silk proteins derived from each organ are generally distinguishable from those derived from other synthetic organs by virtue of their physical and chemical properties. For example, major ampullate silk, or dragline silk, is extremely tough. Minor ampullate silk, used in web construction, has high tensile strength. An orb- web's capture spiral, in part composed of flagelliform silk, is elastic and can triple in length before breaking. Gosline, J., et al, J. Exp. Biol. 202:3295 (1999). Tubuhform silk is used in the outer layers of egg-sacs, whereas aciniform silk is involved in wrapping prey, and pyriform silk is laid down as the attachment disk. The biofilament proteins produced and recovered according to the methods of the present invention may be any recombinantly produced spider silk protein, including recombinantly produced major ampullate, minor ampullate, flagelliform, tubuhform, aggregate, aciniform, and pyriform proteins. Similarly, the proteins recovered according to the methods of the present invention may be any type of biofilament proteins such as those produced by a variety of arachnids, including, but not limited to Nephilia clavipes, Arhaneus ssp., and A. diadematus. Also suitable for use in the invention are proteins produced by insects such as Bombyx mori. Dragline silk produced by the major ampullate gland of Nephilia clavipes occurs naturally as a mixture of at least two proteins, designated as MaSpI and MaSpII. Similarly, dragline silk produced by A. diadematus is also composed of a mixture of two proteins, designated ADF-3 and ADF-4.

The biofilament proteins recovered according to the invention may be monomeric proteins, fragments thereof or dimers, trimers, tetramers or other multimers of a monomeric protein. The biofilament proteins are encoded by nucleic acids, which can be joined to a variety of expression control elements, including tissue-specific animal or plant promoters, enhancers, secretory signal sequences and terminators. These expression control sequences, in addition to being adaptable to the expression of a variety of gene products, afford a level of control over the timing and extent of production.

Spider silk proteins are dominated by iterations of four simple amino acid motifs (1) polyalanine (Ala,,); (2) alternating glycine and alanine (GlyAla)_n; (3) GlyGlyXaa; and (4)

GlyProGly(Xaa)_π, where Xaa represents a small subset of amino acids, including Ala, Tyr, Leu, and Gin (for example, in the case of the GlyProGlyXaaXaa motif, GlyProGlyGlnGln is the major form). Hayashi, et al, J. Mol. Biol. 275:773, 1998; Hinman, et al, Trends in Biotech. 18:374-379, 2000. Spider silk proteins may also contain spacers or linker regions comprising charged groups or other motifs, which separate the iterated peptide motifs into clusters or modules.

Modules of the GlyProGly(Xaa)_n motif are believed to form a β-turn spiral structure which imparts elasticity to the protein. Major ampullate and flagelliform silks both have a GlyProGlyXaaXaa motif and are the only silks which have elasticity greater than 5-10%. Major ampullate silk, which has an elasticity of about 35%, contains an average of about five β-turns in a row, while flagelliform silk, which has an elasticity of greater than 200%, has this same module repeated about 50 times. The polyalanine (Ala and (GlyAla)_n motifs form a crystalline β sheet structure which provides strength to the proteins. The major ampullate and minor ampullate silks are both very strong, and at least one protein in each of these silks contains a (Ala / (GlyAla). module. The GlyGlyXaa motif is associated with a helical structure having three amino acids per turn (3₁₀ helix), and is found in most spider silks. The GlyGlyXaa motif may provide additional elastic properties to the silk.

The methods of the present invention are applicable to purification and recovery of biofilament proteins which comprise the above-mentioned motifs. In particular, the methods of the invention encompass recovery of biofilament proteins having a sequence that is substantially about 50% identical to a sequence selected from the group consisting of:

AlaAlaAlaAlaAla (SEQ ID NO: 3)

GlyAlaGlyAla (SEQ ID NO: 4)

GlyAlaGlyAlaGlyAla (SEQ ID NO: 5)

GlyAlaGlyAlaGlyAlaGlyAla (SEQ ID NO: 6)

GlyAlaGlyAlaGlyAlaGlyAlaGlyAla (SEQ ID NO: 7)

GlyAlaGlyAlaGlyAlaGlyAlaGlyAlaGlyAla (SEQ ID NO: 8)

GlyAlaGlyAlaGlyAlaGlyAlaGlyAlaGlyAlaGlyAla (SEQ ID NO: 9)

GlyGlyTyrGlyGlnGlyTyr (SEQ ID NO: 10)

AlaAlaAlaAlaAlaAlaAlaAla (SEQ ID NO: 11)

GlyGlyAlaGlyGlnGlyGlyTyr (SEQ ID NO: 12)

GlyGlyGlnGlyGlyGlnGlyGlyTyrGlyGlyLeuGlySerGlnGlyAla (SEQ ID NO: 13)

AlaSerAlaAlaAlaAlaAlaAla (SEQ ID NO: 14)

GlyProGlyGlnGln (SEQIDNO: 15)

(GlyProGlyGlnGln)₂ (SEQ ID NO: 16)

(GlyProGlyGlnGln)₃ (SEQIDNO: 17)

(GlyProGlyGlnGln)₄ (SEQIDNO: 18)

(GlyProGlyGlnGln)₅ (SEQIDNO: 19)

(GlyProGlyGlnGln)₆ (SEQ ID NO: 20)

(GlyProGlyGlnGln)₇ (SEQ ID NO: 21)

(GlyProGlyGlnGln)₈ (SEQ ID NO: 22)

GlyProGlyGlyGlnGlyGlyProTyrGlyProGly (SEQ ID NO: 23)

SerSerAlaAlaAlaAlaAlaAlaAlaAla (SEQ ID NO: 24)

GlyProGlySerGlnGlyProSer (SEQ ID NO: 25) GlyProGlyGlyTyr (SEQ ID NO: 26)

Preferably, the biofilament protein has a C-terminal portion with an amino acid sequence repeat motif which is from about 20 to about 40 amino acids in length, more preferably 314 amino acids in length, and a consensus sequence which is from about 35 to about 55 amino acids in length, more preferably 47 amino acids in length. Also, preferably, the biofilament protein has an amino acid repeat motif (creating both an amorphous domain and a crystal-forming domain) having a sequence that is substantially identical to: Ala Gly Gin Gly Gly Tyr Gly Gly Leu Gly Ser Gin Gly Ala Gly Arg Gly Gly Leu Gly Gly Gin Gly Ala Gly Ala Ala Ala Ala Ala Ala Ala Gly Gly (SEQ ID NO: 1 ), which is found in Nephilia spidroin (MaSpI) proteins. In another embodiment, it is preferred that the biofilament protein has a consensus structure that is at least substantially identical to: Cys Pro Gly Gly Tyr Gly Pro Gly Gin Gin Cys Pro Gly Gly Tyr Gly Pro Gly Gin Gin Cys Pro Gly Gly Tyr Gly Pro Gly Gin Gin Gly Pro Ser Gly Pro Gly Ser Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala (SEQ ID NO:2), of the Nephilia spidroin 2 (MaSpπ) proteins. Preferably, the biofilament protein, when subjected to shear forces and mechanical extension, has a polyalanine segment that undergoes a helix to β-sheet transition, where the transition forms a β-sheet that stabilizes the structure of the protein. It is also preferred that the biofilament has an amorphous domain that forms a β-pleated sheet such that the inter-β sheet spacings are between 3 and 8 angstroms; preferably between 3.5 and 7.5 angstroms.

The biofilament proteins produced and recovered according to the methods of the present invention include recombinantly produced MaSpI and MaSpII proteins, as described in U.S. Patent Nos. 5,989,894 and 5,728,810, which are hereby incorporated by reference. These patents disclose partial cDNA clones and amino acid sequences of spider silk proteins MaSpI and MaSpII. The present invention encompasses MaSpI and MaSpII proteins that have been recombinantly produced using the cDNA clones of the '894 and '810 patents, as well as proteins which comprise the amino acid sequences disclosed therein.

The methods of the invention are also applicable to minor ampullate spider silk proteins, such as those disclosed in U.S. Patent Nos. 5,756,677 and 5,733,771, and to flagelliform silks, such as those disclosed in U.S. Patent No. 5,994,099, also hereby incorporated by reference. The methods of the invention are also applicable to spider silk proteins described in U.S. Provisional Application Serial No. 60/315,529 (incorporated by reference).

The MaSpI and MaSpπ spider silk or fragment or variant thereof usually has a molecular weight of at least about 16,000 daltons, preferably 16,000 to 100,000 daltons, more preferably 50,000 to 80,000 daltons for fragments, and greater than 100,000 but less than 300,000 daltons, most preferably 120,000 to 300,000 daltons for the full-length protein.

The sequences of the spider silk proteins disclosed herein may have additional amino acid residues or amino acid sequences inserted within or at the terminal ends thereof so long as the protein possesses the desired physical characteristics. Likewise, some of the amino acid residues or amino acid sequences may be deleted from the protein so long as the protein possesses the desired physical characteristics. Amino acid substitutions may also be made in the sequences so long as the protein possesses the desired physical characteristics. Examples of recombinantly produced MaSpI and MaSpII proteins which may be recovered according to the methods of the invention are depicted in Figures 5 and 6, respectively. Figure 5 shows the sequence of a representative MaSpI protein arranged so that the amino acid repeat motifs can be seen. Figure 6 shows the sequence of a representative MaSpII protein arranged so that the amino acid repeat motifs can be seen. The methods of the invention may also be used to recover recombinantly produced ADF-1, ADF-2, ADF-3 and ADF-4 proteins from biological fluids. These proteins are produced naturally by the Araneus diadematus species of spider. The ADF-1 generally comprises 68% poly(Ala)₅or (GlyAla)₂-₇, and 32% GGYGQGY (SEQ ID NO: 10). The ADF-2 protein generally comprises 19% ρoly(A)₈, and 81% GGAGQGGY (SEQ ID NO: 12) and GGQGGQGGYGGLGSQGA (SEQ ID NO: 13). The ADF-3 protein generally comprises 21% ASAAAAAA (SEQ ID NO: 14) and 79% (GPGQQ)_n, where n = 1-8. The

ADF-4 protein comprises 27% SSAAAAAAAA (SEQ ID NO: 24) and 73% GPGSQGPS (SEQ ID NO: 25) and GPGGY (SEQ ID NO: 26). An example of a recombinantly produced ADF-3 protein which maybe recovered according to the methods of the invention is depicted in Figure 7, which shows the sequence of a representative ADF-3 protein, arranged so that the amino acid repeat motifs can be seen.

It is contemplated that the methods of the invention may be used to recover mixtures of biofilament proteins, comprising at least two biofilament proteins or fragments such as those described above. In preferred embodiments, the recovered mixture comprises one or more spider silk proteins or fragments thereof.

Abbreviations for amino acids used herein are conventionally defined as described herein below unless otherwise indicated.

Amino Acid Three-letter One-letter abbreviation symbol

Alanine Ala A

Arginine Arg R

Asparagine Asn N

Aspartic Acid Asp D

Asparagine or aspartic acid Asx B

Cysteine Cys C

Glutamine Gin Q

Glutamic acid Glu E

Glutamine or glutamic acid Glx Z

Glycine Gly G

Histidine His H

Leucine Leu L

Lysine Lys K

Methionine Met M

Phenylalanine Phe F

Proline Pro P

Serine Ser s

Threonine Thr T

Tryptophan Trp w

Tyrosine Tyr Y

Valine Val V

5.6.1. Transgenic Animals According to the methods of the invention, biofilament proteins are extracted from mixtures comprising biological fluids produced by transgenic animals, preferably transgenic mammals. Transgenic animals useful in the invention are animals that have been genetically modified to secrete a target biofilament in, for example, milk or urine. The methods of the invention are applicable to biological fluids from any transgenic animal capable of producing a recombinant biofilament protein. Preferably, the biological fluid is milk, urine, saliva, seminal fluid, or blood derived from a transgenic mammal. Preferred mammals are rodents, such as rats and mice, ruminants including, for example, goats, cows, sheep, and pigs. In one particularly preferred embodiment, the animal is a goat. See U.S. Patent No. 5,907,080, hereby incorporated by reference. The transgenic animals useful in the invention may be produced as described in WO 99/47661 and U.S. Patent No. 20010042255, also hereby incorporated by reference.

5.6.2. Cell Culture Media The methods of the present invention are also applicable to conditioned media recovered from mammalian cell cultures which have been engineered to produce the desired biofilaments as secreted proteins. Mammalian cell lines capable of producing the subject proteins can be obtained by cDNA cloning, or by the cloning of genomic DNA, or a fragment thereof, from a desired cell, as described by Sambrook, J., et al, Molecular Cloning: A Laboratory Manual, 3d edition, Cold Spring Harbor Laboratory Press (2001).

Examples of mammalian cell lines include, but are not limited to, BHK (baby hamster kidney cells), CHO (Chinese hamster ovary cells), and MAC-T (mammary epithelial cells from cows).

5.6.3 Plant Sources

The methods of the invention can also be applied to plant extracts. Several methods are well known in the art by which to engineer plant cells to produce and secrete a variety of heterologous polypeptides. See, e.g., Esaka et al, Phytochem. 28:2655-2658 (1989); Esaka et al, Physiologia Plantarum 92:90-96 (1994); Esaka et al, Plant Cell Physiol 36:441-446 (1995) and Li et al, Plant Physiol. 114:1103-1111. Transgenic plants have also been generated to produce spider silk proteins. Scheller, K-H et al, Nature Biotech. 19:573 (2001); see also WO 01/94393 A2, incorporated by reference. Exudates produced by whole plants or plant parts may be used in the methods of the present invention. The plant portions for use in the invention are intact and living plant structures. These plants materials may be distinct plant structures, such as shoots, roots or leaves. Alternatively, the plant portions may be part or all of a plant organ or tissue, provided the material contains the biofilament protein to be recovered.

Having been externalized by the plant or the plant portion, exudates are readily obtained by any conventional method, including intermittent or continuous bathing of the plant or plant portion (whether isolated or as part of an intact plant) with fluids. Preferably, exudates are obtained by contacting the plant or portion with an aqueous solution such as a growth medium or water. The fluid-exudate admixture may then be subjected to the purification methods of the present invention to obtain the desired biofilament protein. The proteins may be recovered directly from a collected exudate, preferably guttation fluid, or a plant or a portion thereof.

Extracts useful in the invention may be derived from any transgenic plant capable of producing a recombinant biofilament protein. Preferred for use in the methods of the present invention are plant species representing different plant families including, but not limited to, monocots such as ryegrass, alfalfa, turfgrass, eelgrass, duckweed and wilgeon grass; dicots such as tobacco, tomato, rapeseed, azolla, floating rice, water hyacinth, and any of the flowering plants. Other preferred plants are aquatic plants capable of vegetative multiplication such as Lemna and other duckweeds that grow submerged in water, such as eelgrass and wilgeon grass. Water-based cultivation methods such as hydroponics or aeroponics are useful for growing the transgenic plants of interest, especially when the silk protein is secreted from the plant roots into the hydroponic medium from which the protein is recovered. The plant used in the present invention may be a mature plant, an immature plant such as a seedling, or a plant germinating from a seed. According to the methods of the invention, the recombinant polypeptide is recovered from an exudate of the plant, which may be a root exudate, guttation fluid oozing from the plant via leaf hydathodes, or other sources of exudate, independent of xylem pressure. The proteins may be exited or oozed out of a plant as a result of xylem pressure, diffusion, or facilitated transport {i.e., secretion).

5.7. Uses of Spider Silk Fibers The biofilament proteins recovered using the methods of the invention may be spun and can be utilized in a vast and diverse array of medical, military, industrial and commercial applications. The fibers can be used in the manufacture of medical devices such as but not limited to surgical mesh, sutures, medical adhesives, replacement ligaments, and skin grafts, and in a wide range of industrial and commercial products, such as but not limited to cable, rope, netting, fishing line, clothing fabric, bullet-proof vest lining, container fabric, backpacks, knapsacks, bag or purse straps, adhesive binding material, non-adhesive binding material, strapping material, tent fabric, tarpaulins, pool covers, vehicle covers, fencing material, sealant, construction material, weatherproofmg material, flexible partition material, or sports equipment; and, in fact, in nearly any use of fiber or fabric for which high tensile strength and elasticity are desired characteristics.

6. EXAMPLES

The following examples are meant to illustrate the principles and advantages of the present invention. They are not intended to be limiting in any way.

6.1. Example 1: Expression of ADF-3 in BHK or MAC-T Cells

An expression vector was constructed to express, in mammalian cell culture, the ADF-3 protein, which is found in nature in the dragline silk of Arhaneus spp. The expression vector consisted of: (1) a cytomegalovirus ("CMV") promoter; (2) the coding region of ADF-3 cDNA; (3) a DNA fragment coding for the immunoglobulin kappa ("IgK") signal peptide sequence, attached at the 5' end of the coding region; and (4) a hygromycin resistance gene. The purpose of the IgK signal peptide was to direct the cells to secrete the ADF-3 protein produced by the cells into the extracellular medium. The presence of the hygromycin resistance gene enabled identification of the cells that incorporated the expression vector.

All molecular manipulations were carried out following standard procedures. See, e.g., Sambrook, J., et al, Molecular Cloning: A Laboratory Manual, 3d ed. (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2001). All DNA cloning manipulations were performed using E. coli STBIJ competent cells Canadian Life Science,

Burlington, Canada). Restriction and modifying enzymes were purchased from New England BioLabs (Mississauga, ON, Canada) unless otherwise specified. Construct integrity was verified using DNA sequencing analysis provided by Queens University (Kingston, ON, Canada) or McMaster University (Hamilton, ON, Canada). Primers were synthesized by Dalton Chemical, Inc. (North York, ON, Canada). PCR was performed using Ready-To- Go PCR beads (Pharmacia Biotech, Baie d'Urfe, PQ, Canada) or a Dynazyme kit (MJ Research, MA). Construction of ADF-3His Vector.

The ADF-3 gene sequence was PCR amplified from the plasmid Bluescript (BLSK)- ADF-3 (provided by Dr. Gosline). Two primers (primer 1: 5'-CGT ACG AAG CTT ATG CAC GAG CCG GAT CTG-3' (SEQ ID NO: 30); primer 2: 5'-ATT AAC TCG AGC AAG GGC TTG AGC TAC AGA-3' (SEQ ID NO: 31)) were designed according to ADF-3 sequences. Guerette, P.A., et al, Science 272:112 (1996). Primer 1 contains a Hindlll site and primer 2 was designed to incorporate a Xhol site. The PCR product was digested with Hindlll and Xhol, purified using a QiexIJ matrix (Qiagen, Chatsworth, CA) and cloned into the pSecTag-C vector (Invitrogen, CA) between the HindHI and Xhol sites. Construction of ADF-3 vectors. The ADF-3His construct was modified in order to remove the myc, His sequences, and 15 amino acid non-silk sequences present at the N-terminal. A linker containing a Xhol overhang (linker 1: 5'-TCG AGC TTG ATG TTT-3' (SEQ ID NO: 32)) was cloned into the ADF-3His expression cassette between the Xhol and Pmel sites. The 15 amino acid non- silk sequence at the 5' end of the vector was removed by inserting a linker (linker 2: 5'-CAG GAT CTG GAC AAC AAG GAC CCG GAC AAC AAG GAC CCG GAG AAC AAG

GAC CG GAG AAC AAG GAC CAT ATG GAG CCG GTG CAT CGG CCG GAG AG CAG CCG CTG GAG GTT ATG GAC GCG GAT GTG GAC AAC AAG AC CCA GCC AAC AAG GAG CTG G-3' (SEQ ID NO: 33)) into the above vector between the Sfil and Mscl sites. Concatamerization of ADF-3 coding region.

The ADF-3 coding region was released (Mscl and PvuIJ: 1.4kb) and subcloned into the same vector between the Mscl and PvuIJ sites. Using the same procedure, three copies of the ADF-3 coding region were inserted into the vector. Transfection and Selection of Stable Cell Lines. MAC-T cells are mammary epithelial cells. Huynh, H.T. et al, Exp. Cell Res.

197:191(1991). MAC-T cells were selected to produce spider silk proteins primarily for two reasons: (a) they are secretory epithelial cells, similar to the cell type that expresses silk in the spider glands (see Lucas, F., Discovery 25:20 (1964)); and (b) as they mimic ruminant lactation, they can provide preliminary information regarding the capacity of mammary epithelial cells to efficiently secrete soluble spider silks in the milk of transgenic animals. BHK or MAC-T cells were transfected with Lipofectamine (Canadian Life Science) as per the manufacturer's recommendations, using 10 μg of the plasmid DNA diluted into 0.25ml of DMEM and mixed with an equal volume of Lipofectamine (20 μg of lipid in 0.25 ml

DMEM). Stable transformants were selected in DMEM containing 10% fetal calf serum and 100 μg/ml hygromycin B (Canadian Life Science). Colonies surviving selection were picked after 7-8 days following transfection and expanded further. In general, under the cultured conditions used, BHK cells transfected with the spider silk constructs expressed higher amounts of the ADF-3 protein than the MAC-T cells. The selected transfectant colonies were cultivated in a hollow fiber cell culture system (Unisyn Cell-Pharm® System 2500™. The ADF-3 protein was produced in amounts averaging about 25-50 mg/L (about 20 μg/10⁵ cells per day) for periods of up to 3 months. A correlation was observed between the age of the reactor and the appearance of lower molecular weight spider silk proteins. The appearance of the protein "ladder" was probably due to termination errors of protein synthesis or internal recombination events.

Generation of Polyclonal Antibodies Against Dragline Spider Silk Proteins. Antibodies were raised in rabbits against both purified recombinant spider silk proteins (BHK-derived material) and synthetic peptides designed and based on sequences of N. clavipes and A. diadematus. Peptide synthesis, conjugation, immunization, bleeding, and serum preparations were carried out by Strategic BioSolutions (Ramona, CA). The immunizing peptide sequences were anti-MaSpU, GLGSQGAGRGGQGAGA-ΝH2 (SEQ ID NO: 34), anti-ADF-3, RAGSGQQGPGQQGPG-NH₂ (SEQ ID NO: 35).

Detection of Recombinant Spider Silk in Media and Purified Fractions. Quantification of recombinant spider silk proteins in conditioned media involved SDS-

PAGE and immunological evaluation (Western blotting analysis). See Coligan, L.E., et al, Current Protocols in Protein Science (John Wiley and Sons, Inc., NY, 2000.) Serum-free conditioned media was harvested from cells at 70-80% confluency at 24 hours. An aliquot of 20 μl was loaded onto 8-16% Tris-Glycine gels (Novex, Invitrogen), electrophoresed, and transferred by electroblotting into nitrocellulose membrane. Recombinant spider silk immunoreacting proteins on the membrane were detected using rabbit polyclonal antibodies raised against ADF-3 or MaSpI (1:5000 dilution) and goat anti-horseradish peroxidase conjugated second antibody. Detection was performed according to the manufacturer's protocol, using enhanced chemilluminescence (ECL) detection (Amersham/Pharmacia). For silver stain analysis, gels were stained using the GelCode SilverSNAP (Pierce, IL) kit, as directed by the manufacturer. Samples were prepared by adding 10M urea to a final concentration of 6M, loading buffer containing β-mercaptoethanol and heating for 5 minutes at 95 °C prior to loading. In the absence of urea, aberrant migration of recombinant spider silk was observed.

6.2. Example 2: Precipitation of ADF-3 from Cell Culture Media Using

Ammonium Sulfate A 13.5 liter sample of culture medium produced as described in Example 1 was analyzed by ADF-3-specific ELISA and determined to contain 1185 mg ADF-3. The total protein content of the sample was 25,461 mg, as determined using a commercially available protein assay kit (Biorad). Nine hundred grams of ammonium sulfate were slowly added to the 13.5 liter sample and dissolved with vigorous stirring. The mixture was incubated overnight at 4°C. ADF-3 was recovered in the precipitate by centrifugation at 20,000 xg for one hour. Analysis of the precipitate indicated a total protein content of 1466 mg, of which 1136 mg was determined to be ADF-3, with a purity of 77.6%. The yield of ADF-3 from this precipitation step was 95.5%.

6.3. Example 3: Purification of ADF-3 from Cell Culture Media Using Anion

Exchange Chromatography

The protein precipitates of Example 2 were resuspended in 135 mL of buffer A (20 mM glycine, pH 10), followed by centrifugation at 20,000 xg for 30 minutes to remove any insoluble material. The ADF-3 -containing supernatant was adjusted to pH 10 using concentrated NaOH (IO N).

An anion exchange column (5 cm x 11 cm) was prepared with POROS HQ50 resin (Perseptive Biosystems, MA) and equilibrated with 10 column columns of buffer A. The ADF-3 solution was loaded on the column at a flow rate of 100 ml/hour. The column was washed with 5 column volumes of buffer A. The ADF-3 protein was eluted with 260 ml of buffer B (buffer A with 0.15 M NaCl). The column was regenerated with 10 column volumes of buffer C (buffer A with 2 M NaCl).

The purity of the recovered ADF-3 was analyzed using silver staining, RP-HPLC, amino acid composition, SDS-PAGE and UV absorption. A total of 868 mg of ADF-3 was recovered at a purity of 94% (total protein content of the sample was 927 mg), with a yield of 76.4%. The peak containing the ADF-3 protein on RP-HPLC was identified by Western Blot analysis. Purity was estimated using peak area integration. Amino acid composition was performed as previously described. Heinrikson, R.L., et al, Anal. Biochem. 136:65 (1984). Purified material was quantified using the extinction coefficient method (at 280 nm). Gill, C. et al, Anal. Biochem. 182:319 (1989). The ADF-3 protein was further concentrated and successfully spun into a biofilament with mechanical properties similar to that of native dragline silk.

6.4. Example 4: Clarification of ADF-3 -Media by Tangential Flow Filtration

A tangential flow filtration system was assembled as illustrated schematically in Figure 3. A hollow fiber membrane cartridge with a pore size of 0.2 μm (CFP-2-E-4A, A G Technology Corp., Needham, MA) was equilibrated with a solution containing buffer A (20 mM Glycine, pH 10). The inlet pressure was adjusted to 2.5 psi and outlet pressure was adjusted to 0 psi. The system was operated at 4°C. A volume of 2.5 L cell culture media containing ADF-3, produced by the method of Example 1, was placed in the sample tank and introduced into the feed tank of the system. The sample was circulated through the system, with the permeate containing ADF-3 being collected in the permeate tank, and the retentate was fed back into the feed tank and recirculated. The system was run until 400 ml of media remained in the feed tank, at which time the diafiltration process was initiated. The buffer tank, containing 2.5 L of buffer A, was connected to the feed tank. The buffer was introduced gradually and circulated through the system for a period of approximately 2 hours, or until the final retention volume remaining in the feed tank was reduced to 200 ml. The final volume of clarified media containing ADF-3 collected in the permeate tank was expected to be 4.8 L.

6.5. Example 5: Concentration of ADF-3- Media Using Tangential Flow

Filtration.

The tangential flow filtration system used in Example 4 (Figure 3) was used for the concentration step, except that the 0.2 μm filter was replaced with a hollow fiber membrane cartridge having a 10 kDa molecular weight cutoff (UFP- 10B-A, A/G Technology Corp, Needham, MA). The system was equilibrated with a solution containing buffer A (20 mM Glycine, pH 10), the inlet pressure was adjusted to 15 psi and outlet pressure adjusted to 10 psi. The system was operated at 4°C. One liter of clarified ADF-3 media, prepared according to Example 4, was placed in the feed tank. The sample was circulated through the system, with the permeate being collected in the permeate tank and the concentrated media containing ADF-3 (retentate) being retained in the Feed Tank. The process was continued until the retentate volume was reduced to 100 ml. The entire process of concentration by tangential flow filtration took approximately 45 minutes.

6.6. Example 6: Purification of ADF-3 from Cell Culture Media Using Hydrophobic Interaction Chromatography (HIC) A Toyopearl butyl 650 chromatography column (0.7x10 cm; Toso Haas,

Montgomeryville, PA) was equilibrated with 30 ml of buffer A (10 mM NaPO4, 2 M NaCl, pH 7.5). A sample of 20 ml of concentrated ADF-3-media (prepared by the method of Example 5, above) was loaded on the column at a flow rate of 50 ml/h. The column was subsequently washed with 20 ml of buffer A and 20 ml buffer B (10 mM NaPO₄, 1.5 M NaCI, pH 8.0). The ADF-3 protein was eluted with 20 ml of buffer C (20 mM Glycine, pH

10). The column was regenerated with ddH₂O. The elution fractions were analyzed using SDS-PAGE, Silver staining and Western blot analysis. ADF-3 purity was about 50%. Elution fractions containing ADF-3 could be further purified using an anion exchange (POSROS HQ50) column.

6.7. Example 7: Capturing ADF-3 Using HIC in Expanded Bed Absorption (EBA) mode

A STREAMLINE 25 (18-1110-51, Amersham/Pharmacia, Piscataway, New Jersey) column was prepared with 100 ml of STREAMLINE phenyl resin (17-5121-01, Amersham Pharmacia, Piscataway, New Jersey). The column was equilibrated with buffer

A (10 mM NaPO₄, 2 M NaCI, pH 7.5) and loaded with 5L of cell culture media produced by the method of Example 1 above. Both were fed upwardly through the column at 100 ml/h. The column was washed with buffer A and buffer B (10 mM NaPO₄, 1M NaCl, pH 8.0) using upward feeding. The spider silk protein was eluted using buffer C (10 mM NaPO₄, pH 8.0), flowing downwards at 50 cm/h. The column was regenerated with ddH₂O. The elution actions were analyzed using SDS-PAGE, Silver staining and Western blot analysis. The ADF-3 purity was about 50%. The elution fractions containing ADF-3 were further purified using an anion exchange (POSROS HQ50) column. 6.8. Example 8: Isolation of ADF-3 from Cell Culture Media Using His-tag Affinity Chromatography

A plasmid DNA (ADF-3His) was constructed as described in Example 1 to express ADF-3 with a histidine tag in mammalian cell culture. The ADF-3 cDNA coding region was modified by a 3 ' fusion of a DNA fragment encoding six His residues, and a 5 ' fusion of a fragment encoding the κ-casein signal peptide sequence. The signal peptide directs the secretion of ADF-3 from cells to extracellular medium.

MAC-T cells were co-transfected with the ADF-3 construct and another plasmid containing the kanamycin (Km) gene (see also Example 1). Stable cell clones having the ADF-3 construct were selected using Km resistance. A high expressor of ADF-3 was identified through immunoassay. This clone was cultivated in a hollow fiber fermentor as described in Example 1 and culture media containing ADF-3His was harvested regularly.

A 2x10 cm column was prepared with Ni-NTA agarose resin (Qiagen Inc., Chatsworth, USA) and equilibrated with 200 ml of buffer A (0.05 M Tris, 0.3M NaCl, 0.01M imidazole, pH 8.0). Imidazole was added to 225 ml of ADF-3His media, the pH was adjusted to 8.0 with 1 M NaOH, and the media was loaded in the column at a flow rate of 35 ml/h. The column was washed with 100 ml of buffer A followed by 100 ml of buffer B (0.05 M Tris, 0.3M NaCl, 0.02M Imidazole, pH 8.0). ADF-3His was eluted using 100 ml of buffer C (0.05 M Tris, 0.3M NaCI, 0.25M imidazole, pH 8.0). The eluant was analyzed as described in Example 1 using SDS-PAGE, Silver staining and Western blot analysis. A major band at 60 kD was observed following both Western blotting and Silver staining. Minor contaminants were observed on the silver stained gel. ADF-3His purity was estimated approximately at 90%. The ADF-3His protein was further concentrated and successfully spun into a biofilament with mechanical properties similar to that of native dragline silk.

6.9. Example 9: Clarification of Transgenic MaSpI-Milk by Acid Precipitation

Transgenic goats were constructed to secrete the N. clavipes dragline silk (MaSpI) in their milk. After inducing lactation in reproductively mature animals, milk was collected using standard procedures. A 250 ml volume of the milk (containing about 2.5 mg MaSpI,

1% of the total protein content) was acidified to pH 4.5 using hydrochloric acid, and incubated at 4°C for one hour. The casein precipitate was pelleted by centrifugation at 20000 xg for one hour and 205 ml of clarified MaSpI-containing whey was decanted. The pH was adjusted to 7.5 with sodium hydroxide and MaSpI-whey was further centrifuged at 20000 xg for 30 minutes to remove additional insoluble materials.

6.10. Example 10: Isolation ofMaSpIfrom Transgenic Milk Using Immuno- Affinity Chromatography

A polyclonal antibody against an MaSpI peptide was raised in rabbits (by Strategic Biosolutions, Ramona, CA). The anti-sera were purified using a Hi-Trap Protein G column (17-0404-03, Amersham/Pharmacia, Piscataway, NJ). Purified MaSpI IgG was immobilized on CNBr-activated Sepharose 4 Fast Flow resin (17-0981-01, Amersham/Pharmacia, Piscataway, New Jersey) and an affinity column (1.5 cm x 20 cm) was prepared.

The affinity column was equilibrated with PBS (10 mM NaPO₄, 150 mM NaCl, pH 7.5) and loaded (30 ml/h) with 200 ml of clarified MaSpI-containing-whey prepared according to Example 9 above. The column was washed with 200 ml of PBS, 200 ml of buffer B (PBS with 500mM NaCl) and an additional 100 ml of PBS. The MaSpI protein was eluted with 50 ml of buffer C (50 mM sodium acetate, 6 M urea, pH 3). The eluant was loaded on a Hi-Trap SP column (17-1551-01, Amersham Pharmacia, Piscataway, NJ) and, after washing with 10 ml of buffer D (10 mM NaPO₄, pH 8.0), the MaSpI protein was eluted with 10 ml of buffer E (buffer D with 1M NaCI).

The elution fractions were analyzed using RP-HPLC and a UV absorbance detector. Approximately 2 mg of MaSpI was recovered, with a purity of about 50%. The results were confirmed by SDS-PAGE/Silver staining and Western blot analysis.

6.11. Example 11: MaSpI Purification Using Preparative Reversed Phase HPLC The MaSpI protein isolated from Example 10 was further purified using preparative

RP-HPLC. The separation was performed on a Perseptive Bio-CAD 700E system with a 4.6mm x 100 mm R2 column (Perseptive Biosystems, MA). Mobile phase A consisted of aqueous 0.1 %> TFA and mobile phase B was acetonitrile with 0.085% TFA. The flow rate was 5 ml/min. The column was equilibrated with 15% of phase B prior to sample injection. A sample containing about 1 mg of total proteins was injected into the column. A linear gradient of 15-50% mobile phase B was run for 5 minutes followed by 50-100%> mobile phase B for 2 minutes. The MaSpI protein eluted at 4.9 min, as identified by Western blot analysis. In the preparative chromatography, the 4.9 min peak was collected and the sample was lyophilized to powder form. Subsequent analyses using SDS-PAGE/Silver staining and Western blot analysis confirmed the identity of the MaSpI protein. The purity was estimated to be greater than 90%.

6.12. Example 12: Purification of MaSpII from Transgenic Milk using Filtration. Ammonium Sulfate Precipitation, and Chromatography

Transgenic goats were developed to secrete the N. clavipes dragline silk MaSpJJ in their milk. The MaSpII protein was recovered from transgenic goat milk as described below.

A tangential flow filtration system was constructed as illustrated schematically in Figure 4. A volume of 500 ml of transgenic milk (containing approximately 500 mg of MaSpII) was placed in the sample tank. The buffer tank was charged with 500 ml of buffer A (200 mM arginine, pH 6.8) and connected to the feed tank. To start the clarification process, 500 ml of buffer A was introduced into the feed tank. Pump A was used to drive the clarification unit. A hollow fiber membrane cartridge of 750 kD cutoff (UFP-750-E- 4x2A, A G Technology Corp., Needham, MA) was equilibrated with buffer A. The inlet pressure was adjusted to 5 psi and the outlet pressure adjusted to 0 psi. The 500 ml transgenic milk containing MaSpII was then introduced into the feed tank. The sample was circulated through the clarification system, with the clarified permeate containing MaSpII being collected in the whey tank (permeate flux was 20 ml/minute) and the retentate being circulated back through the feed tank.

When the permeate volume collected in the whey tank reached 500 ml, the concentration process was initiated and run simultaneously with the clarification process.

Pump B was used to drive the concentration unit. A hollow fiber cartridge of 30 kD cutoff (UrPF-30-E-4x2C, A/G Technology Corp., Needham, MA) was used to concentrate the clarified whey. In the concentration unit, the inlet pressure was adjusted to 15 psi and outlet pressure adjusted to 10 psi. Pump C was used to maintain the equilibrium of flow rates between the clarification and concentration units. The clarification process was run for a total of 200 minutes, during which 8 feed volumes were circulated through the clarification system. The concentration process was continued until the final volume of retentate collected in the whey tank was reduced to 250 ml. Analysis of the whey concentration by ELISA indicated that a total 460 mg of MaSpII were recovered.

The whey concentrate containing the 460 mg of MaSpII was then subjected to ammonium sulfate precipitation. Ammonium sulfate (32g) was added slowly and with vigorous stirring to the 250 ml of whey concentrate. The mixture was incubated at 4°C overnight and the insoluble precipitate was recovered by centrifugation at 20000 xg for one hour.

The precipitate was resuspended in 100 ml of buffer B (20 mM glycine, pH 10) followed by centrifugation to remove insoluble material. The sample was then adjusted to pH 10 using NaOH, and conductivity was adjusted to 1.0 mS by adding 1000 ml of buffer

B. An anion exchange column (2.5 cm x 10 cm) was prepared with POROS HQ50 resin (Perseptive Biosystems, MA) and equilibrated with 10 column volumes of buffer B. The resuspended sample was loaded on the column at a flow rate of 100 ml/h. The column was washed with 5 column volumes of buffer B and MaSpII was eluted using 5 column volumes of buffer C (buffer B with 150 mM NaCI). The column was regenerated with 10 column volumes of buffer D (buffer B with 2M NaCl). The elution fractions were analyzed using RP-HPLC and a UV absorbance detector. A total of 400 mg of MaSpII protein was recovered with a purity greater than 90%. The results were confirmed by SDS-PAGE/Silver staining and Western blot analysis.

6.13. Example 13: Purification of MaSpII from Milk of Transgenic Goats using EDTA-induced Precipitation

Transgenic goats were constructed to secrete the N clavipes dragline silk MaSpII in milk. The MaSpII protein was purified from the transgenic goat milk as described below. 10 ml of transgenic goat milk containing approximately 5 mg of MaSpII was subjected to centrifugation at 10000 xg for 30 minutes at 4°C. The cream was separated from the milk after centrifugation. The cream layer was then punctured using a sharp pipette tip and skim milk was siphoned off. A quantity of 9 ml skim milk was obtained.

4.5 ml of 0.4 EDTA, pH 7, was added to the skim milk. The EDTA solution deconstructed the casein micelles by complexing its calcium core and resulted in the formation of a clarified milk serum and the precipitation of MaSpII. The mixture was incubated at 4 °C for 4 hours prior to centrifugation. Centrifugation was performed at 20000 xg for one hour at 4°C and insoluble materials were recovered by removing the liquid phase.

The precipitate was washed by homogenous resuspension in 10 ml of 0.15 M EDTA, pH 7, followed by centrifugation at 20000 xg for 30 minutes. This washing step was then repeated one time.

The resulting pellet was resolubilized in 2 ml of 6M guanidine and analyzed using UV absorbance, RP-HPLC, SDS-PAGE/Silver staining and Western blot. A total of 2.5 mg of MaSpπ protein was recovered with a purity greater than 90%. The results were confirmed by SDS-PAGE/Silver staining and Western blot analysis.

6.14. Example 14: Purification of Recombinant MaSpII Spider Silk Protein from Transgenic Goat Milk

A tangential flow filtration system was constructed as illustrated schematically in FIG. 4. A volume of 3180 ml of milk produced by transgenic goats (containing approximately 3000 mg of MaSpU) was placed in the Sample Tank. The Buffer Tank was charged with 3180 ml of Buffer A (50 mM Arginine, pH 6.8) and connected to the Feed Tank. To start the clarification process, 3180 ml of Buffer A was introduced into the Feed Tank. Pump A was used to drive the clarification unit. A hollow fiber membrane cartridge of 750kD cutoff (UFP-750-E-6 A, A/G Technology Corp, Needham, MA) was equilibrated with Buffer A. The inlet pressure was adjusted to 5 psi and outlet pressure to 0 psi. The sample of 3180 ml transgenic milk containing MaSpII was then introduced into the Feed Tank. The sample was circulated through the clarification system, with the clarified permeate containing MaSpU being collected in the Whey Tank (permeate flux was 100 ml/minute) and the retentate being circulated back through the Feed Tank. When the permeate volume collected in the Whey Tank reached 3180 ml, the concentration process was initiated and run simultaneously with the clarification process. Pump B was used to drive the concentration unit. A hollow fiber cartridge of 30kD cutoff (UPF-30-E-6C, A/G Technology Corp., Needham, MA) was used to concentrate the clarified whey. In the concentration unit, the inlet pressure was adjusted to 15 psi and outlet pressure to 10 psi. Pump C was used to maintain the equilibrium of flow rates between the clarification and concentration units. The clarification process was run for a total of 260 minutes, during which eight feed volumes were circulated through the clarification system. The concentration process was continued until the final volume of retentate collected in the Whey Tank was reduced to 1815 ml. Analysis of the whey concentration by Western blot indicated approximately 2700 mg of MaSpII recovered.

The whey concentrate containing 2700 mg of MaSpπ was then subjected to ammonium sulfate precipitation. Precisely 740 ml of 3.8M ammonium sulfate solution were added slowly to the 1815 ml of whey concentrate, with moderate stirring, to obtain a final concentration of ammonium sulfate of 1.1 M. The mixture was incubated at 4°C overnight and the insoluble precipitate was recovered by centrifugation at 20000 xg for one hour.

The precipitate was washed twice by homogeneous resuspension in 200 ml of 1.1M ammonium sulfate solution followed by centrifugation at 2000 xg for one hour. Three samples of 500 μl each were taken before the final centrifugation for analysis. Quantitative analysis of the samples was performed by UV absorbance spectroscopy at 280 nm, and qualitative analysis was performed by reverse phase HPLC. A total of 2112 mg of MaSpII protein in the form of a pellet was recovered with purity greater than 90%. The results were confirmed by SDS-PAGE/Silver staining and Western blot analysis.

6.15. Example 15 Preparation of Dope Solution of MaSpII Protein

6.15.1. Solubilization of the Spider Silk Protein Using Guanidine-HCl

Approximately 0.5 ml of guanidine-HCl (6 M) was added to 413 mg of the MaSpU pellet obtained as described in Example 1. The pellet was carefully ground with a glass rod to obtain a homogeneous mixture. Another 80 ml of guanidine-HCl (6 M) was added to the mixture and then incubated at 60°C in a water bath for 30 minutes. The suspension was briefly vortexed every 10 minutes during the 30 minute incubation period. Insoluble materials were removed from the MaSpU solution by decanting the supernatant following a one hour centrifugation at 30000 xg (4°C).

6.15.2. Buffer Exchange: Removal of Guanidine-HCl

Buffer exchange chromatography was performed using a Bio-Rad Biologic LP system (Bio-Rad Laboratories, Hercules, CA, USA). A 5x25 cm Sephadex G-25 medium resin column (Amersham, Piscataway, NJ, USA) was prepared and equilibrated using 2.0 L of 50 mM glycine buffer (pH 11), at a flow rate of 10 ml/min. The MaSpU supernatant prepared in the previous section was loaded on the column and the column was flushed with the 50 mM glycine buffer (pH 11). Under these conditions the MaSpU protein eluted while the guanidine-HCl remained bound to the column. Chromatography was monitored using UV absorption spectroscopy and conductivity measurements of the effluent. A 200 ml fraction of MaSpU solution (-2.0 mg/ml) was collected.

6.15.3. Concentration of the MaSpII Solution

The MaSPU solution recovered in the above section was concentrated using a 400 ml Stirred Cell system (Millipore, Jaffrey, NH, USA) equipped with a 10 kD cutoff YM 10 membrane (Millipore). The device was assembled according to manufacturer's instructions. The MaSpU solution (200 ml) was carefully added to the system and forced through the membrane at 55 psi. The MaSpU protein was retained in the retentate and the volume of

MaSpU solution was reduced from 200 ml to 10 ml. The retentate was recovered and the concentration of MaSpU, measured by UV absorbance, was 40 mg/ml.

The MaSpU solution was further concentrated by centrifugal filtration. An Ultrafree-15 Centrifugal Filter Unit equipped with a Biomax-10 membrane (lOkDa cutoff) (Millipore) was used to concentrate 7.5 ml of the MaSpU solution by centrifugation at 2000 xg for 20 minutes (4°C). The retentate was gently mixed in the centrifugal device and re-centrifuged five times for 20 minutes until the volume was reduced to 1.4 ml. The final concentration of MaSpU solution, determined by UV absorption spectroscopy, was 19.8% (w/v). Solutions thus prepared were subsequently used as the dope solution as described in U.S. Application Serial No. , entitled Methods and Apparatus for Spinning Spider

Silk Protein, filed January 13, 2003 (attorney docket no. 9529-012), herein incorporated by reference in its entirety.

6.16. Example 16: Production and Recovery of Recombinant Spider Silk

Proteins from Transgenic Tobacco Plants

Transgenic plants can be constructed to express spider silk proteins. See Scheller, K- H et al, Nature Biotech. 19:573 (2001); PCT publication WO 01/94393 A2. Following root formation, hydroponic cultivation can be initiated by placing the transgenic plants into sterile liquid MS medium containing 15 g/L of sucrose, 500 mg/L cefotaxime and 100 mg/L of kanamycin. Wild-type control plants {e.g., untransformed plants) can be cultivated in a similar medium without antibiotics. Plants can be positioned in a synthetic stopper to permit the roots to contact the medium under sterile conditions, while the remainder of the plants (e.g., plant regions above the hypocotyl, under normal gravimetric culture conditions) are kept in an open and non-sterile environment.

Approximately 2-3 weeks after rooting, exudates can be sampled by aspiration of the medium. The protein content of intercellular fluids can be tested, (intercellular fluids can be isolated, e.g., as described by Parent et al, Can. J. Bot. 52:564 (1984)). Intercellular proteins may be concentrated by ultrafiltration (e.g., using Microcon 10 membranes having a 10 kD cutoff, Amicon, Inc.). Samples can be incubated at 65 °C for 30 minutes to denature thermolabile proteins, rapidly cooled on ice, and stored at -20 °C.

All publications and patent applications cited in this specification are herein incorporated by reference as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference. The present invention is not to be limited in scope by the specific embodiments described herein. Indeed, as and although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be readily apparent to those of ordinary skill in the art, in light of the teachings of this invention via the foregoing description and accompanying figures, that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims. Such modifications are intended to fall within the scope of the claims of the invention.

Claims

What is claims is:

1. A method for recovering a biofilament protein from a biological fluid, said method comprising:

(a) subjecting said biological fluid to tangential flow filtration across a membrane of sufficient porosity to permit said biofilament protein to pass through the membrane, thereby producing a permeate comprising said biofilament protein; and

(b) precipitating said biofilament protein from said permeate by adding salt to said permeate sufficient to precipitate said biofilament protein,

thereby recovering said biofilament protein.

2. The method of claim 1, said method further comprising the steps of:

(a) solubilizing said biofilament protein; and

(b) preparing an aqueous solution comprising said biofilament protein,

wherein said aqueous solution is substantially free of chaotropic agents and organic solvents, and wherein said aqueous solution is suitable for spinning into fibers.

3. The method of claim 1, wherein said biological fluid is milk produced by a transgenic mammal.

4. The method of claim 3, wherein said transgenic mammal is a goat.

5. The method of claim 1 , wherein said biological fluid is cell culture medium.

6. The method of claim 1, wherein said biological fluid is urine.

7. The method of claim 1, wherein said method further comprises the step of purifying said biofilament protein using a chromatographic technique.

8. The method of claim 1, wherein said biofilament protein is a spider silk protein.

9. The biological filament protein recovered using the method of claiml .

10. A method for recovering a biofilament protein from a biological fluid, said method comprising:

(a) subjecting said biological fluid to one or more tangential flow filtration steps across a membrane of sufficient porosity to permit said biofilament protein to pass through the membrane, thereby producing a permeate comprising said biofilament protein;

(b) subject said permeate to a final tangential flow filtration step across a final membrane, wherein said final membrane is of sufficient porosity to prevent said biofilament protein from passing through said final membrane, thereby producing a retentate; and

(c) precipitating said biofilament protein from said permeate by adding salt to said retentate sufficient to precipitate said biofilament protein,

thereby recovering said biofilament protein.

11. The method of claim 10, said method further comprising the steps of:

(a) solubilizing said biofilament protein; and

(b) preparing an aqueous solution comprising said biofilament protein,

wherein said aqueous solution is substantially free of chaotropic agents and organic solvents, wherein said aqueous solution is suitable for spinning into fibers.

12. The method of claim 10, wherein said biological fluid is milk produced by a transgenic mammal.

13. The method of claim 12, wherein said transgenic mammal is a goat.

14. The method of claim 10, wherein said biological fluid is cell culture medium.

15. The method of claim 10, wherein said biological fluid is urine.

16. The method of claim 10, wherein said method further comprises the step of purifying said biofilament protein using a chromatographic technique.

17. The method of claim 10, wherein said biofilament protein is a spider silk protein.

18. The biological filament protein recovered using the method of claim 10.

19. A method for recovering a biofilament protein from a biological fluid, said method comprising precipitating said biofilament protein from said biological fluid by adding salt to said biological fluid sufficient to precipitate said biofilament protein, thereby recovering said biofilament protein.

20. The method of claim 19, said method further comprising the steps of:

(a) solubilizing said biofilament protein; and

(b) preparing an aqueous solution comprising said biofilament protein,

21. The method of claim 19, wherein said biological fluid is cell culture medium.

22. The method of claim 1, wherein said biological fluid is urine.

23. The method of claim 19, wherein said method further comprises the step of purifying said biofilament protein using a chromatographic technique.

24. The method of claim 19, wherein said biofilament protein is a spider silk protein.

25. The biological filament protein recovered using the method of claiml9.

26. A method for recovering a biofilament protein from milk produced by a transgenic mammal engineered to express said biofilament protein in its milk, said method comprising:

(a) treating said milk with an acid to achieve a pH in the range of about 3.0 to about 5.5, thereby producing an acidified mixture;

(b) centrifuging said acidified mixture to produce a clarified solution comprising said biofilament protein, which is free of casein, fat, micelles, cells and particulate matter from said milk; and

(c) precipitating said biofilament protein from said clarified solution by adding a salt sufficient to precipitate said biofilament protein,

thereby recovering said biofilament protein.

27. The method of claim 26, said method further comprising the steps of:

(a) solubilizing said biofilament protein; and

(b) preparing an aqueous solution comprising said biofilament protein,

28. The method of claim 26, wherein said method further comprises the step of purifying said biofilament protein using a chromatographic technique.

29. The method of claim 26, wherein said biofilament protein is a spider silk protein.

30. The biological filament protein recovered using the method of claim 26.

31. A method for recovering a biofilament protein from milk produced by a transgenic mammal engineered to express said biofilament protein in its milk, said method comprising:

(a) defatting said milk to produce skim milk;

(b) treating said skim milk with an acid to achieve a pH in the range of about 3.0 to about 5.5, thereby producing an acidified mixture;

(c) centrifuging said acidified mixture to produce a clarified solution comprising said biofilament protein, which is free of casein, fat, micelles, cells and particulate matter from said milk; and

(d) purifying said clarified solution using a chromatography technique,

thereby recovering said biofilament protein.

32. The method of claim 31 , wherein said transgenic mammal is a goat.

33. A method for purifying a biofilament protein from milk produced by a transgenic mammal, said method comprising:

(a) defatting said milk to produce skim milk;

(b) precipitating said biofilament protein from said skim milk by adding EDTA sufficient to precipitate said biofilament protein; thereby recovering said biofilament protein.

34. The method of claim 33, said method further comprising the steps of:

(a) solubilizing said biofilament protein; and

(b) preparing an aqueous solution comprising said biofilament protein,

35. The method of claim 33, wherein said transgenic mammal is a goat.

36. The method of claim 33, wherein said method further comprises the step of purifying said biofilament protein using a chromatographic technique.

37. The method of claim 33, wherein said biofilament protein is a spider silk protein.

38. The biological filament protein recovered using the method of claim 33.

39. A method for purifying a biofilament protein from a biological fluid, said method comprising subjecting said biological fluid to expanded bed absorption chromatography.

40. The method of claim 39, wherein said biological fluid is cell culture medium.

41. The method of claim 39, wherein said biological fluid is urine.

42. The method of claim 39,said method further comprising the step of purifying said biofilament protein using a chromatographic technique

43. A method for purifying a biofilament protein from milk produced by a transgenic mammal, said method comprising: (a) defatting said milk to produce skim milk;

(b) purifying said skim milk using expanded bed absorption chromatography.

44. The method of claim 43, said method further comprising the step of purifying said biofilament protein using a chromatographic technique

45. A method for purifying a biofilament protein from a biological fluid, said method comprising:

(a) subjecting said biological fluid to tangential flow filtration across a membrane of sufficient porosity to permit said biofilament protein to pass through the membrane, thereby producing a permeate comprising said biofilament protein;

(b) precipitating said biofilament protein from said permeate by adding EDTA to said permeate sufficient to precipitate said biofilament protein;

thereby recovering said biofilament protein.

46. The method of claim 45, said method further comprising the steps of:

(a) solubilizing said biofilament protein; and

(b) preparing an aqueous solution comprising said biofilament protein,

47. The method of claim 45, wherein said biological fluid is milk produced by a transgenic mammal.

48. The method of claim 47, wherein said transgenic mammal is a goat.

49. The method of claim 45, wherein said biological fluid is cell culture medium.

50. The method of claim 45, wherein said biological fluid is urine.

51. The method of claim 45, wherein said method further comprises the step of purifying said biofilament protein using a chromatographic technique.

52. The method of claim 45, wherein said biofilament protein is a spider silk protein.

53. The biological filament protein recovered using the method of claim45.

54. A method for purifying a biofilament protein from a biological fluid, said method comprising:

(b) subject said permeate to a final tangential flow filtration step across a final membrane, wherein said final membrane is of sufficient porosity to prevent said biofilament protein from passing through said final membrane, thereby producing a retentate;

(c) precipitating said biofilament protein from said permeate by adding EDTA to said retentate sufficient to precipitate said biofilament protein;

thereby recovering said biofilament protein.

55. The method of claim 54, said method further comprising the steps of:

(a) solubilizing said biofilament protein; and (b) preparing an aqueous solution comprising said biofilament protein,

56. The method of claim 54, wherein said biological fluid is milk produced by a transgenic mammal.

57. The method of claim 56, wherein said transgenic mammal is a goat.

58. The method of claim 54, wherein said biological fluid is cell culture medium.

59. The method of claim 54, wherein said biological fluid is urine.

60. The method of claim 54, wherein said method further comprises the step of purifying said biofilament protein using a chromatographic technique.

61. The method of claim 54, wherein said biofilament protein is a spider silk protein.

62. The biological filament protein recovered using the method of claim 54.

63. A method for purifying a biofilament protein from a biological fluid, said method comprising:

(a) precipitating said biofilament protein from said biological fluid by adding

EDTA and salt to said biological fluid sufficient to precipitate said biofilament protein;

thereby recovering said biofilament protein.

64. The method of claim 1 , said method further comprising the steps of:

65. The method of claim 63, wherein said biological fluid is cell culture medium.

66. The method of claim 63, wherein said biological fluid is urine.

67. The method of claim63, wherein said method further comprises the step of purifying said biofilament protein using a chromatographic technique.

68. The method of claim 63, wherein said biofilament protein is a spider silk protein.

69. The biological filament protein recovered using the method of claim63.

70. A method for purifying a biofilament protein from milk produced by a transgenic mammal engineered to express said biofilament protein from its milk, said method comprising:

(c) precipitating said biofilament protein from said clarified solution by adding a EDTA and salt sufficient to precipitate said biofilament protein,

thereby recovering said biofilament protein.

71. The method of claim 70, said method further comprising the steps of: (a) solubilizing said biofilament protein; and

(b) preparing an aqueous solution comprising said biofilament protein,

72. The method of claim 70, wherein said transgenic mammal is a goat.

73. The method of claim 70, wherein said method further comprises the step of purifying said biofilament protein using a chromatographic technique.

74 The method of claim 70, wherein said biofilament protein is a spider silk protein.

75. The biological filament protein recovered using the method of claim 70.