AU707719B2 - Process for protein refolding by means of buffer exchange using a continuous stationary phase capable of separating proteins from salt - Google Patents

Description

WO 97/04003 PCT/US96/11581 PROCESS FOR PROTEIN REFOLDING BY MEANS OF BUFFER EXCHANGE USING A CONTIN- UOUS STATIONARY PHASE CAPABLE OF SEPARATING PROTEINS FROM SALT BACKGROUND OF THE INVENTION The present invention relates to a process for refolding a protein which utilizes rapid size exclusion chromatography to separate the reduced, denatured protein from the denaturant solution. In a preferred embodiment, the invention is related to a quick and efficient process utilizing cellulosic rolled stationary phase to promote high protein refold yields while significantly decreasing the volume needed to achieve protein refolding.

Recombinant DNA technology has permitted the expression of foreign (heterologous) proteins in microbial and other host cells. In many instances, high expression of the heterologous protein leads to formation of high molecular weight aggregates called "refractile bodies" or "inclusion bodies". Recovery of the desired protein which is in the form of such refractile bodies has presented a number of problems.

First of all, it can be difficult to separate the refractile proteins from other host cellular materials.

Second, it can be difficult to subsequently remove refractile body protein contaminants from the desired refractile body protein. Third, and most troublesome, the refractile body protein is often in the form which, while identifiable as the desired protein, is not biologically active.

In most instances, denaturants and detergents guanidine hydrochloride, urea, sodium dodecylsulfate (SDS), Triton X-100) have to be used to WO 97/04003 PCT/US96/11581 2 extract the protein. The resultant solution containing the denatured protein with the individual polypeptide chains unfolded is then treated to remove the denaturant or otherwise reverse the denaturing conditions and thereby permit renaturation of the protein and folding of the polypeptide chains in solution to give protein in native, biologically form. With proteins of pharmaceutical interest, use of these denaturants can create potential problems because it is difficult to remove the denaturants completely from the isolated proteins. Further, when strong denaturants like guanidine are employed, renaturation can be difficult, if not impossible. There have been several renaturation protocols described in the art; see U.S. 5,235,043 (Collins et al) and U.S. 4,734,362 (Hung et al) and references cited therein.

One example of a protein which can be expressed in microbial host cells is secretory leukocyte protease inhibitor (SLPI). SLPI is of therapeutic interest in the treatment of disease states that involve leukocyte-mediated proteolysis such as emphysema and cystic fibrosis. Full stability and activity of this protein requires proper refolding with 16 cysteine residues involved in 8 intramolecular disulfide bonds.

Recombinant SLPI (rSLPI) is currently renatured by diluting the product stream to a concentration of 0.2 mg/ml to bring the solubilizing salts and reductants to an acceptable concentration that allow the protein to refold; Harcum et al., Biotech.

Lett., 15(9), 943-948 (1993). This method, while yielding biologically active SLPI, requires a large dilution and results in unwieldly large process volumes when attempting to process large amounts of the protein.

Other recombinant proteins utilize similar renaturation/refolding methods; see Protein Refolding, Georgiou and Bernardez, Eds., ACS Symposium WO 97/04003 PCT/US96/I 1581 3 Series No. 470 (1991); Jaenicke and Rudolph, Chap. 9 in Protein Structure, T. Creighton, ed.(1988). The need exists for a process which reduces the refold process volumes associated with the refolding of these proteins and makes production of large amounts of these proteins more commercially practicable.

Buffer exchange is an important separation technique used in the production of biopharmaceuticals and common methods such as tangential flow filtration and size exclusion chromatography have been tested for their ability to separate denaturants/reductants from protein. Utilization of tangential flow filtration results in a protein gradient near the membrane, membrane fouling, and protein aggregation. Size exclusion chromatography using hydrophilic gels is slow and difficult, due to compressibility of these gels, and also results in substantial dilution of the protein.

Nonetheless, size exclusion chromatography may be preferred where protein aggregation or denaturation is an issue; Kurnik et al., Biotech. and Bioeng., 45(2), 149-157 (1995) A size exclusion chromatography method based on a new type of cellulosic stationary phase facilitates protein/salt separations at mobile phase velocities of 500 to 6000 cm/hour. Consequently, it is possible to remove salts from a denatured protein and collect the protein in a quick and efficient manner, in smaller volumes, and at higher protein concentrations, thereby making use of size exclusion chromatography in a refolding process practical. Specifically, this chromatography method is based on a continuous, woven stationary phase which is rolled into a cylindrical configuration and packed into a chromatography column.

Such a stationary phase is termed "rolled stationary phase" because of its configuration.

Rolled stationary phases are either derivatized or underivatized, preconditioned, cylindrically wound, woven textiles and the fibers that make up the woven matrix are capable of withstanding flow velocities of 6000 cm/hr without bed compression, while maintaining constant plate height.

In one preferred embodiment the present invention utilizes this rapid size exclusion chromatography, followed by protein refolding, to provide a process which significantly reduces process volumes associated with the refolding of proteins like SLPI, that normally require large dilutions in order to create the proper refolding environment. In preferred embodiments the process of the present invention significantly minimizes tankage requirements, water requirements and process time, increases protein concentration, and improves the viability of downstream processing of large amounts of proteins. The process of the present invention makes scale-up production of large volumes of proteins practical and provides a valuable tool to those preparing proteins, particularly recombinantly produced proteins, for therapeutic use.

SUMMARY OF THE INVENTION The present invention is directed to a process for refolding a protein, preferably a recombinantly expressed protein, which utilizes rapid size exclusion chromatography to •t •separate the reduced, denatured protein from the denaturant solution.

According to a first aspect the present invention consists of a process for 20 refolding and renaturing proteins, wherein said proteins attain their active form, said process consisting of: passing a mixture of denatured protein and denaturing agents over a separation device, wherein said separation device is selected from the group consisting :of continuous stationary phase, particulate stationary phase or a membrane capable of carrying out separation based on differences in size between proteins and denaturing i .°.agents, and wherein said device is capable of rapidly fractionating said protein from said denaturing agents; and allowing said protein to refold without an additional buffer exchange step.

According to a second aspect the present invention consists in a process for refolding and renaturing a protein, wherein said protein attains its active form, said ,4 process comprising: 4a passing a mixture of denatured protein and denaturing agents over a rolled stationary phase; allowing said protein to refold.

According to a third aspect the present invention consists in a process for refolding and renaturing rSLPI, wherein said rSLPI attains its active form, said process comprising: passing a mixture of denatured rSLPI and denaturing agents over a rolled stationary phase, wherein said rolled stationary phase comprises DEAE derivatized polyester; and allowing said rSLPI to refold.

Unless the context clearly requires otherwise, throughout the description and the claims, the words 'comprise', 'comprising', and the like are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense; that is to say, in the sense of "including, but not limited to".

In a preferred embodiment, the invention is directed to a process which utilizes rolled stationary phase technology (RSPTTM) to separate denaturants and reducing agents from proteins in large volumes of fermentation product stream. Surprisingly, separations ;carried out in this manner took only a few minutes, resulted in high yields of recovery and, when followed by a short incubation o *o* WO 97/04003 PCT/US96/11581 5 period, gave a high yield of properly refolded protein at higher protein concentrations. More importantly, the separation technique substantially reduces refold process volumes, demonstrating that a combination of size exclusion chromatography followed by protein refolding will significantly enhance process throughput.

In a preferred embodiment, a fermentation product stream comprising a mixture of denatured protein and denaturing agents is passed rapidly over a separation device to partially fractionate protein from denaturants/reductants, followed by capture of protein to allow for protein refolding. Preferably, the denatured protein is a recombinantly expressed protein and the separation device is a rolled stationary phase capable of carrying out separation based on differences in size between protein and salt. Most preferably, the denatured protein is recombinant SLPI and the separation device is a RSPT' column consisting of a woven matrix of cotton/40% polyester, wherein the cellulose is a DEAE derivatized material.

BRIEF DESCRIPTION OF THE DRAWINGS FIGURE 1 is a schematic diagram of the Superperformance® column packed with rolled stationary phase.

FIGURE 2 depicts a porosity distribution curve. Eluent was deionized water, with 100 Al sample injection of 10 mg/mL PEG, dextrans, and other molecular weight probes dissolved in water. Flow rate was mL/minute.

FIGURE 3 is the experimental protein elution profile for a 2 mL sample of BSA (2 mg/mL) in 50 mM Tris WO 97/04003 PCTUS96/11581 6 buffer (pH 8.0) containing 500 mM NaCl. The flow rate was 2 mL/minute. Trace at UV 280 nm, at 0.5 AU.

FIGURE 4 is the experimental protein elution profile for a 2 mL injection of Sample 1. The flow rate was 2 mL/minute. Trace at UV 280 nm, at 0.5 AU.

FIGURE 5 is the experimental protein elution profile for a 2 mL injection of Sample 1 injected onto two columns connected in a series. The flow rate was 1 mL/minute. Trace at UV 280 nm, at 0.5 AU.

DETAILED DESCRIPTION The processes by which size exclusion chromatography can be used to facilitate protein refolding processes are described in more detail in the discussion below and are illustrated by the example provided below. The example shows various aspects of the invention and includes results of use of RSPTM to separate recombinant SLPI from denaturants and reducing agents. The results were surprising in that utilization of RSPT M for the separation resulted in high protein refolding yields, with substantially reduced process volumes, and higher protein concentrations.

Included in the processes of the present invention are any proteins, preferably proteins expressed by DNA technology in any host microorganism, wherein said proteins must be isolated from denaturants and renatured/refolded. In particular, those proteins which require refolding environments involving low concentrations of denaturants, and therefore large volumes, yet can refold in relatively high concentrations of protein, without forming aggregates.

Exemplary proteins which may be useful in the present invention include SLPI, brain-derived growth factor WO 97/04003 PCT/US96/11581 7 (BDNF), glial-derived growth factor (GDNF), nerve growth factor (NGF), and neurotropic factor-3 (NT-3).

In general, SLPI useful in the present invention has the sequence of human SLPI, or closely related analogues thereof. The SLPI may be produced by mammalian cells outside the body, or it may be isolated from natural sources. Preferably, the SLPI is recombinant SLPI (rSLPI) produced as described by Seely and Young, Chap. 16 in ACS Symposium Series No. 470, Protein Refolding; Georgiou and Bernardez, Ed.;ACS: Wash 206-216 (1991). While the procedures of Seely and Young are the preferred method for producing rSLPI, modifications and changes could be made to that process as known in the art.

Separation devices contemplated for use in the present invention include the use of any continuous stationary phase, particulate stationary phase, or membrane capable of: rapidly fractionating denatured protein from denaturants/reductants; and promoting protein renaturation/refolding to a biologically active form. As used herein, "rapidly fractionating" is defined as sufficient separation of denatured protein from denaturants/reductants to allow the protein to refold, but quickly enough such that the protein is removed from the separation device before refolding of denatured protein exceeds Rapid size exclusion chromatography contemplated for use in the present invention is preferably carried out over rolled stationary phase technology (RSPTF) columns. These columns were developed based upon the discovery that cellulosic solid sorbent materials, especially continuous stationary phases (which had been shown previously to demonstrate excellent and rapid separations at eluent linear velocities in excess of 5000 cm/hr (Yang et al., Adv.

Biochem. Eng., 49, 148-160, 1993)), could be treated WO 97/04003 PCT/US96/11581 8 with cellulase enzymes to significantly improve the protein adsorption capacity of the solid sorbent material.

The cellulase enzymes are produced by and can be obtained from suitable microorganisms such as fungi, using conventional techniques, or can be obtained from commercial sources. It is preferred that the cellulase enzyme employed have a molecular weight of about 20,000 to about 100,000, more preferably about 50,000 or more.

The preferred sorbent material is cellulose based and may be particulate, fibrous, or preferably, a continuous phase comprising a woven or non-woven fabric.

Moreover, the sorbent material can be derivatized to introduce ionic or nonionic functional groups as well known and used in the art of chromatography to introduce cation exchange, anion exchange and/or affinity character to the sorbent. The derivatized sorbent material is preferably an amino-functionalized material such as a dialklaminoalkyl cellulose, DEAE cellulose, although celluloses containing other functional groups such as sulfate, alkylsulfate, carboxymethyl, phosphate, quaternary salt or other beneficial groups can also be prepared in accordance with the invention. Alkyl groups in these functional groups typically contain 1 to about 5 carbon atoms. As one example, to prepare a preferred DEAE cellulose material, a cotton fabric can be immersed into a mixture of NaOH and DEAE for a period of several hours, for example about 6 to 10 hours. In such a process, the fabric to liquid ratio is preferably in the range of about 1:25 to about 1:50 W/V, and the concentration of DEAE is preferably up to about 1M.

According to the present invention, the solid sorbent material may include fibers of two different materials. For example, the sorbent may include a fabric comprising derivatized cellulose fibers, combined WO 97/04003 PCT/US96/11581 9 with another type of fiber designed to reinforce and improve the overall mechanical properties of the stationary phase. For example, derivatized cellulose and synthetic fibers such as polyester nylon or Kevlar® aromatic polyamide fibers can be blended to achieve an advantageous stationary phase. The stationary phase may also include fibers of cellulose which have been separately derivatized with differing derivatizing agents, e.g. DEAE- and sulfate-derivatized cellulose fibers which have been blended together in a fabric.

As indicated above, the invention contemplates the hydrolysis of a cellulose based sorbent material with a cellulase enzyme for a duration sufficient to form the modified sorbent material with an increased protein adsorption capacity. According to one mode for preparing the material, the sorbent material is treated with the cellulose enzyme for up to about 6 hours at a pH of about 3 to about 8, more preferably a pH of about 4 to about 6. Temperatures during these treatments may vary so long as the temperature employed does not denature or otherwise inactivate the enzyme.

Temperature of about 4'C to about 80"C are typical, and more preferably fall within the range of about 20° to about 60'C. A preferred hydrolysis protocol in work to date has included exposing the cellulosic material to the cellulase enzyme for about 1 hour at a pH of about to about 6 and at a temperature of about The enzyme concentrations may also vary widely in treating the cellulosic material, for example ranging up to about 50 GCU/mL or more. More preferred cellulase enzyme concentrations are in the range of about 2 to about 10 GCU/mL. In this regard, one GCU is defined as one Genecor unit, which is equivalent to 1 FPU, a standardized level of enzyme activity based upon the rates at which strips of filter paper are hydrolyzed by cellulytic enzymes.

WO 97/04003 PCTIUS96/11581 10 After the enzyme treatment, the enzyme is deactivated, for example by immersing the stationary phase in hot water to denature the enzyme. In this regard, when carrying out methods of the present invention, it is important that the cellulase-mediated hydrolysis be terminated prior to complete breakdown or fragmentation of the cellulose phase material, as this will provide materials having poor mechanical properties and/or which will lead to the collection of fines which deleteriously affect column performance. Preferred methods will be carried out so as to achieve stationary phases have breaking strengths of at least about 5 lbf.

Preferred methods of the present invention also include a cellulose conditioning step which includes swelling the fabric or other cellulosic material in water or a solution of a swelling agent such as an organic or inorganic base, e.g. ammonia, ethylene diamine, or caustic. Sodium hydroxide (NaOH) solutions are preferred for these purposes. Pretreatment with swelling agents such as sodium hydroxide increases reactivity with respect to enzyme hydrolysis. This is believed to result in an increased internal porosity and surface area accessible to protein either directly (through swelling) or indirectly (by facilitating enzyme attack) Optionally, the cellulose conditioning step may also include a prederivatization step. Cellulose prederivatization may be accomplished for example, by immersing a cellulose based material in a mixture of NaOH and a derivatizing agent such as 2-(diethylamino) ethyl chloride (DEAE-CI). After conditioning and/or prederivatization, the fabric can be washed, for example with deionized water, prior to further treatment with the cellulase enzyme.

Once prepared, the stationary phases of the invention can be packed into metal, plastic, glass or -Ilother columns suitable for use in liquid or other chromatographic techniques. For example, to pack a modified, rolled continuous phase of the invention, an aperture can be punched or drilled in the end of the phase, and a cord made from a material having a high tensile strength, e.g. Kevlar® aromatic polyamide fiber, can be threaded through the aperture. The cord can then be threaded through the column and used to pull the phase into the column.

Preferred columns will have packing densities of at least about 0.5 g/cc, usually in the range of 0.5 to 0.6 g/cc. As well, preferred columns will have void fractions as low as about 0.4 and even ranging to about 0.3 or lower.

The novel combination of chromatography and refolding described in the present invention allow for external control of such parameters as temperature, protein concentration, time, and concentration of denaturants/reductants. It is well known by those skilled in the art that these parameters affect refold efficiencies.

Temperature can be controlled by jacketing the chromatography column and by 15 controlling the temperature of the eluting buffer. Protein concentration (and *o concentration of denaturants/reductants) can be controlled by adjusting sample size, column length, and initial concentration of the sample to be injected into the separation device. The unique flow properties of the rolled stationary phases, feature of a constant plate height regardless of eluent velocity, allows flow rate to be changed as *20 necessary without affecting the column's efficiency, and without 20 necessary without affecting the column's efficiency, and without WO 97/04003 PCT/US96/11581 12 encountering pressure drop associated with other types of gel permeation and size exclusion media. The present invention combines these factors into a novel process where the operational parameters of sample size, stationary phase characteristics, and column length are used in place of a diluting buffer to achieve protein refolding in greatly reduced process volumes.

Although the invention has been described and illustrated with respect to certain protein refolding processes which utilize rapid size exclusion chromatography to separate denaturants/reductants from protein prior to refolding, it will be apparent to one of ordinary skill that additional embodiments may exist without departing from the scope of the invention.

The following examples will illustrate in more detail the various aspects of the present invention.

EXAMPLE 1 This example describes the preparation of the RSPTM columns and protein sample preparations used in the experiments.

RSPT

M Column Preparation In the present invention, the RSPT M column consisted of a woven matrix of 60% cotton/40% polyester, rolled into a cylinder, and inserted into a 10 mm i.d.

glass Superperformance® column Merck, Darmstadt, Germany). The 60/40 cotton/polyester fabric blend was supplied by Cotton, Inc (Raleigh, and derivatized as follows: the fabric is stored in a 67-73 F and 60-70% relative humidity conditional room for at least 3 days before it is cut and weighed for treatment; the WO 97/04003 PCT/US96/11581 13 fabric is pretreated with 0.5 M DEAE-C1 (Sigma Chemical Corp., St. Louis, MO) in 18% NaOH at 22 0 C for 6 hours, and then repeatedly rinsed with deionized water; the water is squeezed out of the pretreated fabric by hand and the fabric immersed in an enzyme solution (preheated to 50 0 C and with a total volume of enzyme solution to weight of fabric is 30:1 containing 9 GCU cellulase (solution made by mixing 9 parts of 50 mM citrate buffer (pH 4.8) with 1 part Cytolase T M 123 (Genencor, Inc.)) for one hour, and then rinsed repeatedly with deionized water; the fabric is placed in boiling water for 5 minutes to deactivate the enzyme and then repeatedly rinsed in deionized water at room temperature; step is repeated.

The column apparatus is illustrated in Figure 1. The porosity of this type of column was characterized using D 2 0, glucose, polyethylene glycol, and dextran probes dissolved in deionized water at concentrations of 2 mg/ml. Eluent was deionized water.

The probes had molecular weights ranging from 20 X 106 to 2 X 106. The resulting porosity distribution curve is given in Figure 2.

Two RSPTr columns were used in the experiments. Column 1 consisted of fabric rolled to fit the 10 mm i.d. glass column and had a bed height of 105 mm and a packed bed volume of 8.4 mL. Column 2 was a mm i.d column with a bed height of 114 mm and a packed bed volume of 9.0 mL.

Sample Preparation The purified rSLPI protein solution used for these experiments was prepared as described by Seely and Young. supra. Protein concentration was 2 mg/mL.

Sample 1. Solid guanidine-HCl was added to the protein solution to obtain a final concentration of WO 97/04003 PCT/US96/11581 14 3M guanidine-HCl. After 30 minutes, dithiothreitol (DTT) was added to give a 5 mM solution. After minutes the pH was adjusted to 8.0, and NaCl added to 500 mM. The protein concentration was 2 mg/ml.

EXAMPLE 2 This example describes the size exclusion chromatography separations where RSPTm was used to separate rSLPI from denaturants and reducing agents, and refolding analysis of the fractionated protein preparations.

Size exclusion chromatoaraohv The objective was to separate rSLPI from the other constituents and obtain rSLPI containing 0.3M guanidine, as this concentration of guanidine had been used in previously published methods.

All process chromatography was carried out using a Pharmacia Biopilot Unit LCC-500 Plus (Piscataway, with a UV detector (ISCO Type 6 Optical Unit, Model 228). Protein was detected at 280 nm, while a conductivity detector (Cole Parmer, VWR Module 1052) was used to detect emergence of the salt peak. The UV monitor also detected DTT and thereby gave an indication of resolution of DTT from the protein.

The signal from the detectors was recorded simultaneously on a Linseis Chart Recorder (Type 7045A).

A full scale deflection of the conductivity detector corresponded to 12.9 mmhos. This was determined by injection of standard solutions of guanidine-HCl in elution buffer directly into the detector.

Samples were injected onto the column using a Rheodyne injector fitted with 100 .il, 2.0 mL, 3.0 mL or mL sample loops. The start of the chromatogram was WO 97/04003 PCT/US96/11581 15 measured as the time at which the sample first entered the column.

The mobile phase consisted of 50mM Tris buffer (pH with 500 mM NaCI. The combination of the buffer and salt was chosen based on prior experiments in which 500 mM NaCl was shown to suppress BSA binding onto the DEAE cellulose. Hence, the mobile phase as well as the sample of rSLPI contained 500 mM NaCI.

The columns were calibrated with 2 mg/mL bovine serum albumin (BSA) protein (Sigma Chemical Corp., St. Louis, MO) dissolved in 50mM Tris buffer (pH containing 500 mM NaCI. The protein elution profile for a 2 mL sample is depicted in Figure 3.

Refoldina Analysis Analytical chromatography was performed on a SynChropak RP-8 column (Synchrom, Inc., Linden NJ). A 100 .l sample at 0.2 mg/mL is injected. Buffer A is water w/0.1% trifluoroacetic acid (TFA) and Buffer B is acetonitrile (100%) w/0.1% TFA. The column is run at room temperature and a gradient of 19% to 34% acetonitrile at l%/minute is used to resolve correctly folded rSLPI (elutes at 13.8 minutes) from unfolded form (elutes at 17.9 minutes). Recovery of protein was calculated by dividing the area of the peak at 13.8 minutes by the total area of all peaks.

Experimental Runs Run 1. 2 mL of Sample 1 as prepared as described in Example 1 was loaded onto Column 1. This represented 25% of the total bed volume, or 83% of the void fraction accessible to the rSLPI. A flow rate of 2 mL/minute was used. The protein peak which eluted between 2.5 mL and 5 mL elution volume was collected.

WO 97/04003 PCT/US96/11581 16 This collection was based on the BSA elution profile.

The protein elution profile for Run 1 is depicted in Figure 4. The Figure 4 data demonstrates that RSPT M can be used to effectively separate rSLPI from the other constituents. The collected fraction was quickly mixed then allowed to incubate for 4 hours at 200°C, followed by refolding analysis. The results of Run 1 are summarized in Table 1.

Run 2 This run was similar to Run 1 except that the protein peak which eluted between 2.5 mL and mL elution volume was collected. The collected fraction was then allowed to incubate for 4 hours at 200°C, followed by refolding analysis. The results of Run 2 are summarized in Table 1.

_Run 3. This run was similar to Run 1 except that the protein peak which eluted between 2.5 mL and 6 mL elution volume was collected. The collected fraction was then allowed to incubate for 4 hours at followed by refolding analysis. The results of Run 3 are summarized in Table 1.

Runs 4-5. 3 mL of Sample 1 as prepared as described in Example 1 was loaded onto Column 1. This represented 37.5% of the total bed volume, or 125% of the void fraction accessible to the rSLPI. A flow rate of 2 mL/minute was used. The protein peak which eluted between 2.5 mL and 5.5 mL elution volume was collected.

The collected fraction was then allowed to incubate for 4 hours at 200°C, followed by refolding analysis. The results of Runs 4-5 are summarized in Table 1.

Run 6-8. These runs are the same as Runs except that collection of the protein peak was between 2.5-6.0, 2.5-6.6 and 2.5-6.1 mL, respectively. The collected fractions were then allowed to incubate for 4 hours at 200°C, followed by refolding analysis. The results of Runs 6-8 are summarized in Table 1.

WO 97/04003 PCT/US96/11581 17 Eun_.. In this run, Column 1 and Column 2 were connected in series, and 2 mL of Sample 1 injected.

A flow rate of 1 mL/minute was used. The protein elution profile for Run 9 is depicted in Figure 5. The Figure 5 data demonstrates that higher resolution can be obtained by connecting the columns in series. Protein peak collection started at 6 mL elution volume and stopped at 10.5 mL. The collected fraction was then allowed to incubate for 4 hours at 20 0 C, followed by refolding analysis. The results of Run 4 are summarized in Table 1.

Table 1 protein recovery 87.5 96.0 100 80.0 78.0 88.0 103 96.0 106 protein l (ma/mL) 1.43 1.28 1.34 1.60 1.55 1.52 1.54 1.64 0.94 45.2 45.7 37.8 25.1 24.1 0.2 0.2 41.0 As shown in Table 1, the highest yield of refolded protein was obtained for a 3 mL fraction collected 3 minutes after the 2 mL sample had been injected onto Column 1. This fraction contained 96% of the initial rSLPI injected at a concentration of 1.28 mg/mL. This represents a 6.4-fold increase in protein concentration over a previously published method, Seely and Young, sras., where refolding is carried out by diluting the WO 97/04003 PCT/US96/11581 18 mixture of protein, denaturants, and reducing agents by a factor of 10. After a four hour incubation, 46% of the protein had refolded to its active form. This yield is in line with refold yields reported for the previously published method.

And, although one can obtain separations having higher resolution by doubling the column length, the refolding results did not improve since the concentration of the reductants and denaturants were suboptimal. Run 9 still provides, however, a 4.7-fold improvement over the previous published method.

In summary, the results depicted in Table 1 demonstrate that separation of rSLPI from denaturants and reducing agents, using RSPTM, promotes refolding of rSLPI to an active form. The separations take less than minutes, give excellent yields, and significantly minimize the volume of material to be further processed.

Therefore, the data demonstrate that the combination of rapid size exclusion chromatography combined with appropriate protein pooling and renaturation conditions, significantly improve process throughput in a protein refolding process.

Claims (11)

1. 2. A process according to claim 1 wherein said separation device comprises a continuous stationary phase capable of carrying out separation based on differences in size between protein and salts. A process according to claim 2 wherein said continuous stationary phase is a S1Is rolled stationary phase.
2. 4. A process according to claim 3 wherein said rolled stationary phase comprises a *S.S cellulose based sorbent material. A process according to claim 4 wherein said cellulose based sorbent material is derivatised. a
3. 6. A process according to claim 3 wherein said rolled stationary phase comprises DEAE derivatised 60% cotton/40% polyester.
4. 7. A process according to claim I wherein said separation device comprises a particulate stationary i 20 phase capable of carrying out separation based on differences in size between protein and salts.
5. 8. A process according to Claim 1 wherein said separation device comprises a membrane capable of carrying out separation based on retention of protein based on size.
6. 9. A process for refolding and renaturing a protein, wherein said protein attains its active form, said process comprising: passing a mixture of denatured protein and denaturing agents over a rolled stationary phase; allowing said protein to refold. 10 10. A process according to Claim 1 or 9 wherein said protein is selected from the group consisting of rSLPI, BDNF, GDNF, NGF and NT-3.
7. 11. A process according to Claim 10 wherein said protein is rSLPI.
8. 12. A process for refolding and renaturing rSLPI, wherein said rSLPI attains its active 99 form, said process comprising: 15 passing a mixture of denatured rSLPI and denaturing agents over a rolled stationary phase, wherein said rolled stationary phase comprises DEAE derivatized cotton/40% polyester; and allowing said rSLPI to refold.
9. 13. A protein when prepared by the process of any one of Claims 1 to 11.
10. 14. rSLPI when prepared by the process of any one of Claims 1 to 12. A process for refolding and renaturing proteins, substantially as herein described with reference to any one of the Examples.
11. 20366-00.DOC -21- 16. A process for refolding and renaturing rSLPI, substantially as herein described with reference to any one of the Examples. DATED this 24th Day of February 1998 AMGEN, INC. Attorney: IAN T. ERNST Fellow Institute of Patent Attorneys of Australia of SHELSTON WATERS a. a a a a a a. a a a a. I 20366-00 DOC 20366-00 DOC