EP1044378A2

EP1044378A2 - Real-time monitoring of an analyte by chromatography using an on-line assay

Info

Publication number: EP1044378A2
Application number: EP98965982A
Authority: EP
Inventors: Gregory S. Blank; Robert Lee Fahrner; Martin Vanderlaan
Original assignee: Genentech Inc
Current assignee: Genentech Inc
Priority date: 1997-12-31
Filing date: 1998-12-09
Publication date: 2000-10-18
Also published as: AU2199099A; WO1999034220A2; WO1999034220A3

Abstract

The presence of a contaminating analyte in a protein mixture may obscure detection of a protein of interest. In addition, antibody breakthrough is obscured by flow-through of contaminant proteins when loading cell culture fluid onto affinity columns. An on-line chromatographic assay described herein allows real-time measurement of contaminating analyte during purification or antibody breakthrough during loading and collection of a protein of interest in preparative chromatography. Breakthrough curves are generated allowing the user to end loading of cell culture fluid and/or to end effluent collection for maximization of efficient purification and preventing waste of cell culture fluid.

Description

REAL-TIME MONITORING OF AN ANALYTE BY CHROMATOGRAPHY USING AN ON-LINE ASSAY

BACKGROUND Purification of a protein from a cell culture mixture is affected by several factors including, but not limited to, the amount of material loaded onto a chromatographic column as well as the types and levels of contaminating materials present in the protein mixture .

Immobilized protein A affinity chromatography is an important technique for industrial purification of pharmaceutical antibodies because it provides a method to extract antibodies from complex solutions (Hjelm, H et al , FEBS Letters ( 1972) 28( 1 ), Leser, E W et al , Journal of Chromatography ( 1992) 584 43) Protein

A binds to IgG of different subclasses, primarily to the F_c region but also in some cases to the F_ab (Surolia,

A et al Trends in Biochemical Sciences (1982) 7 74, Ey, P L et al , Immunochemistry (1978) 15 429, Reis, K J , J Clin Lab Immunol (1984) 13 75. Lindmark, R et al , Journal of Immunological Methods (1983) 62 1 Because of its specific interaction with antibodies, protein A affinity chromatography is often used as the first chromatographic recovery step for producing recombinant monoclonal antibodies, concentrating and highly purifying the antibodies in a single step

An important factor in protein A affinity chromatography is the dynamic capacity of the chromatography media Dynamic capacity is the amount of antibody that can be loaded before antibody breakthrough occurs, and it is affected by several factors including column media, column length, and flow rate (Horstmann, B J et al , Chem Eng Res Des (1989) 62 243, van Sommeren, A P G et al , Preparative

Biochemistry (1992) 22 135) Directly measuring breakthrough capacity involves looking for product breakthrough during loading, which can be done at absorbance at 280 nm for pure samples When protein A affinity chromatography is used as the first purification step for recombinant antibodies, antibody breakthrough is hidden by the presence of a significant amount of unbound protein and other impurities in the flow-through that obscure UV detection of product breakthrough, because for recombinant proteins of interest produced in cell culture, absorbance at 280 nm is dominated by protein and other impurities Such impurities also obscure elution of the protein of interest from the chromatographic column and/or co-elute with it, thereby limiting the purity of the desired product

SUMMARY OF THE INVENTION Rapid assay technology allows product analysis to be run on-line in real-time to find product breakthrough On-line means that the preparative column effluent is directly sampled by the assay, and real time means that a quantitative measurement of antibody breakthrough (such as assay peak area) is available as soon as the assay is run Assay technology such as high performance liquid chromatography, flow injection immunoassay, biosensors, and perfusion chromatography can produce protein analysis m 20 seconds to ten minutes (Chen, H et al , Journal of Chromatography A (1995) 705 3, Nilsson, M , et al ,

Journal of Chromatography (1992) 597 383. Mulchandani, A et al , Critical Reviews in Biotechnology, (1995) 15 105, Afeyan, N B et al , Sio/Technology (1990) 8 203, Fulton, S P et al Journal of Chromatography (1991) 547 452) Rapid chromatographic affinity assays have been used on-lme for real- time monitoring of chromatography (Chase, H.A., Biosensors (1986) 2:269; Hunt, A.J. et al, Journal of Chromatography A (1995) 708:61.

Frequently, contaminants in biopharmaceuticals are monitored off-line and, if possible, separated from the protein of interest. Previously disclosed methods of assaying contaminants in a protein mixture include an immunoassay for cellular proteins (EP 03798887). Assays for the presence of DNA and proteins in biopharmaceuticals are reviewed by Briggs, J. and Panfili, P.R. (Briggs, J. and Panfili, P.R. (1991) Anal. Chem. 63:850-859). A review of assay development for host cell contaminant protein in recombinant biopharmaceuticals is presented by Eaton, L.C. (Eaton, L.C. (1995) J. Chromatography A 705: 105-114).

Measuring analyte levels on-line in real-time leads to a rapid evaluation of chromatographic conditions that affect capacity. Because of its specificity, immobilized protein A chromatography is useful for assaying breakthrough of antibodies, providing robust assays; combining this with perfusion chromatography allows very rapid assays.

In one aspect of the invention, a method is provided herein to analyze effluent composition by using the on-line assay to quickly measure analyte breakthrough or elution of impurities. The information gained from analyzing the effluent is used to control loading of a cell culture mixture onto a first chromatographic column (such as a preparative column) and/or to control collection of effluent.

In another aspect of the invention, a method of separating a protein of interest from a contaminant is disclosed. The method comprises a) loading a protein mixture onto a first chromatography column, wherein the mixture comprises an analyte including, but not limited to, a protein of interest and at least one contaminant, and b) monitoring the amount of analyte in effluent from the first chromatography column as the analyte elutes from the first column by deflecting column effluent to a separate on-line assay apparatus for a period of time sufficient to provide enough analyte to be monitored in the separate assay apparatus, wherein the monitoring requires less than twenty percent of the total elution time of the analyte on the chromatography column and wherein the monitoring measures the amount of analyte in the effluent from the chromatography column. Preferably, the first chromatography column is a preparative chromatography column. Preferably, the separate on-line assay apparatus comprises a second chromatography column.

In an embodiment of the invention the method further includes controlling a chromatographic device when the amount of analyte in the effluent reaches a predetermined amount.

In another embodiment of the invention, the chromatographic device to be controlled is a loading device capable of directing a protein mixture onto the column or away from the column, wherein the loading device is controlled based on information from the monitoring step. The controlling may be by manual, mechanical or electronic means, or a combination of means.

In another embodiment of the invention, the chromatographic device to be controlled is an effluent collection device capable of directing column effluent into a collection reservoir or away from the collection reservoir, wherein the device is controlled based on information from the monitoring step. The controlling may be by manual, mechanical or electronic means, or a combination of means.

In still another embodiment of the invention, the chromatographic device to be controlled comprises a loading device and an effluent collection device. In yet another embodiment of the invention, the monitoring requires less than ten percent of the total elution time of analyte on the chromatography column, preferably less than five percent of the total elution time of the protein of interest from the chromatography column.

In another embodiment of the invention, the protein of interest is a fusion protein wherein a protein is fused covalently to a portion of an IgG molecule. Preferably the protein of interest includes, but is not limited to, C2B8, rhuMAb-HER2 (recombinant human monoclonal antibody to HER2), an anti-IgE antibody (for example, E25 and E26), a TNF receptor-IgG fusion (for example, TRY), human DNase (hDNase), and IGF-1.

In still another embodiment of the invention, the contaminant is a protein including, but not limited to, an aggregate of IGF proteins, and host cell proteins (HCPs) such as CHOP (Chinese hamster ovary cell proteins), ECP (E. coli proteins), and insect cell proteins.

In another aspect of the invention, an isolated protein of interest is provided by a method comprising: a) loading a protein mixture onto a first chromatography column, wherein the mixture comprises an analyte including, but not limited to, the protein of interest and at least one contaminant, and b) monitoring the amount of analyte in effluent from the first chromatography column as the analyte elutes from the first column by deflecting column effluent to a separate on-line assay apparatus for a period of time sufficient to provide enough analyte to be monitored in the separate assay apparatus, wherein the monitoring requires less than twenty percent of the total elution time of the analyte on the chromatography column and wherein the monitoring measures the amount of analyte in the effluent from the chromatography column. Preferably, the separate on-line assay apparatus comprises a second chromatography column. Preferably, the purified protein of interest is C2B8, rhuMAb-HER2, IGF-1, or hDNase. Preferably, the isolated protein of interest is present in the final product at a level of at least 80%, more preferably at least 90%, and most preferably at least 95% of the total protein in the product at the level of detection used for standard assay of the product.

In an embodiment of the invention the purified protein of interest is provided by further including controlling a chromatographic device when the amount of analyte in the effluent reaches a predetermined amount.

Preferably, the first chromatography column is a protein A affinity column when the protein of interest is an IgG-containing protein, such as an IgG fusion protein. However, any chromatographic material useful for the separation of the protein of interest from a contaminant may be used, and the detection method is UV absorption. Preferably, the assay used in the separate on-line assay apparatus is a chromatographic assay comprising a second chromatographic column. The second chromatographic column may contain the same material as the first (preparative) chromatographic column or it may be different according to what analyte is being monitored by the on-line assay. The first and second chromatography columns and the detection are chosen to be appropriate for the purification of the protein of interest. Preferably, where the analyte is HCP, the separate on-line assay comprises the second chromatography column, which column comprises anti-HCP antibodies immobilized onto a solid support matrix. In a preferred embodiment of the invention, the HCP is CHOP and the second chromatography column comprises anti-CHOP antibodies immobilized onto a solid support matrix. Preferably, detection of analyte is by UV absorbence. Other preferred detection methods include, but are not limited to, absorption in the visible or infrared spectrum, NMR, biosensor, and the like.

These and other objects, advantages and features of the present invention will become apparent to those persons skilled in the art upon reading the details of the methods as more fully set forth below. DESCRIPTION OF THE EMBODIMENTS

Before the present methods are described, it is to be understood that this invention is not limited to the particular processes described as such methods may, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting since the scope of the present invention will be limited only by the appended claims. It must be noted that as used in this specification and the appended claims, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a protein" includes mixtures of proteins and so forth. Definitions

The term "breakthrough" refers to a rapid increase in analyte concentration in the column effluent indicating that the analyte is no longer being retarded by the column bed material. Where the analyte is the protein of interest, breakthrough is monitored for the purpose of determining loading parameters for the first chromatographic column and/or to control the amount of protein of interest loaded onto the column.

The term "monitor" refers to periodic analysis of chromatographic column effluent such that the concentration of a chosen an analyte is detected by any means appropriate for such detection. Such detection may include, but is not limited to, UV absorption, antibody interaction, and visible light detection.

The term "loading" refers to the amount of a cell culture mixture comprising a protein of interest applied to a chromatography column. Preferably loading is ceased as breakthrough of a chosen analyte (such as the protein of interest) begins.

The term "protein of interest" refers to a protein that may be an analyte and is a protein that the user wishes to isolate from contaminants in a cell culture mixture.

The term "contaminant" or "impurity" refers to a component in the cell culture mixture comprising the protein of interest, or to a component introduced during the purification of the protein of interest, which component is undesirable in the final product- The contaminant or impurity may include, but is not limited to, a protein, nucleic acid, host cell, host cell debri, serum, peptone, degraded protein of interest, detergent, antifoam material, pluronic polyol, buffer component, retrovirus and the like.

The term "analyte" refers to either a protein of interest or a contaminant or impurity, the detection of which is monitored in the column effluent.

The term "column effluent" refers to the solution and proteins contained therein flowing from the chromatography column. "Elution time" refers to the time from loading of a protein on the column to the time that the protein emerges from the column.

The term "control" refers generally to the manipulation of a chromatographic device, which manipulation is performed when the amount of an analyte reaches a predetermined level in the first chromatographic column effluent. Manipulation includes, but is not limited to manual, mechanical or electronic means. A chromatographic device includes, but is not limited to, a loading device such as a loading loop, a collection device such as a fraction collector or product pool collector, a detection device such as a UV detector, a computer for the calculation of data from the monitoring step or for subsequent control of other devices.

Symbols used herein are defined in Table I.

TABLE I r Antibody concentration in the column effluent (g/L) Antibody concentration in the load (g/L) o

F Volumetric flow rate (L/hr)

L Column length (cm)

Q Breakthrough capacity (g/L) t Time (hr) u Linear flow rate (cm/hr)

Y Column volume (L)

' c

The instant invention is shown and described herein in what is considered to be the most practical, and the preferred embodiments- It is recognized, however, that departures may be made therefrom which are within the scope of the invention, and that obvious modifications will occur to one skilled in the art upon reading this disclosure.

All references provided herein are hereby incorporated by reference in their entirety.

BRIEF DESCRIPTION OF THE DRAWINGS Figs. 1A and IB. Fig. 1A is a chromatogram of an on-line assay for monitoring antibody breakthrough. The eluted peak is antibody. Fig. IB is a chromatogram from a preparative protein A affinity column. Load was stopped when C C₀ = 0.05.

Figs. 2A - 2D are graphs generated using the on-line assay to monitor complete breakthrough curves for (Fig. 2A) Sepharose A, (Fig. 2B) Poros 50 A, (Fig. 2C) Prosep A at five different flow rates, and (Fig. 2D) Prosep A at six column lengths.

Figs. 3A and 3B are graphs showing breakthrough capacity determined from the breakthrough curves of Fig. 2. Fig. 3 A is a graph showing the effect of flow rate on breakthrough capacity for three column media. Fig. 3B is a graph showing the effect of column length on capacity.

Figs. 4A and 4B. Fig. 4A is a chromatogram of an on-line assay for monitoring IGF-I purity. Fig. 4B is a chromatogram resulting from monitoring the preparative column.

Fig. 5 is a bar graph showing the results from 5 runs pooling IGF-I from 85% to 85% purity. Percent for IGF and aggregate is percent of total peak area by the off-line HPLC analysis for both the load an pool. Yield is percent yield by the off-line HPLC analysis. Error bars are one standard deviation for five runs.

Figs. 6A and 6B. Fig. 6A is a chromatogram of the on-line assay for monitoring CHOP. Fig. 6B is a chromatogram of preemptive purification of C2B8, with results from the on-line CHOP assay overlaid.

All references cited in this application are herein incorporated by reference in their entirety. EXAMPLES The following examples are set forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how to use the method of the invention and are not intended to limit the scope of what the inventors regard as their invention. Efforts have been made to insure accuracy with respect to numbers used (e.g. amounts, temperature, etc.) but some experimental errors and deviation should be accounted for. Example 1: Materials Used for Real-Time Monitoring of Analyte Breakthrough.

Poros A/M columns, Poros 50 A media, and BioCAD chromatography instruments were from

PerSeptive Biosystems (Framingham, MA). Prosep A chromatography media was from Bioprocessing (Consett, England) and Sepharose A was from Pharmacia (Uppsala, Sweden). Harvested cell culture fluid

(HCCF, where cells and cell debris are removed by tangential flow filtration) used as load material was obtained from Genentech, Inc. (South San Francisco, CA). The HCCF contained a humanized monoclonal antibody (with a murine F_v and a human F_c, produced in Chinese hamster ovary cells) at a concentration of about 0.5 g/L. The humanized monoclonal antibodies analyzed herein are rhuMAb-HER2 and C2B8 produced in separate cell cultures.

Example 2: On-line Assay Chromatography

The following on-line assay parameters were used for monitoring the breakthrough of rhuMAb-HER2 or C2B8 during protein A affinity chromatographic purification of these proteins of interest. On-line analysis was performed at room temperature by a perfusion affinity assay. The column was a 30 mm length X 2.1 mm diameter Poros A/M with a 20 μm particle size. Buffer A was 10 mM sodium phosphate, 150 mM sodium chloride, pH 7.2, and buffer B was 10 mM sodium phosphate, 150 mM sodium chloride, pH 2.2. Two flow rates were used: 20 ml/min for purge, 3.5 ml/min for run. Injection volume was 20 μl from the process stream using an in-line sampling T to directly sample the preparative column effluent. Detection was at 280 nm. The method was: purge 3.5 ml A, run 15 column volumes (CV) A, inject, run 12 CV A, purge 3.5 ml B, run 20 CV B. Chromatography was run on a BioCAD/RPM. On the BioCAD instrument, purging takes the column out of the flow path, allowing increased flow rate to flow path to rapidly fill with buffer; the flow rate is decreased and the column is then placed back in line. This purge/run method decreases assay time. The BioCAD/RPM software automatically integrates the antibody peak and displays the result immediately after the assay is finished. Example 3: Preparative Chromatography for Protein A Affinity Chromatography

Preparative protein A affinity chromatography used 0.66 cm diameter columns. Columns were equilibrated with at least 5 CV of 25 mM Tris, 25 mM NaCl, 5 mM EDTA, pH 7.1 and eluted with at least 3 CV of 0.1 M acetic acid, pH 3.5. All buffers and load material were filtered through a 0.22 μm filter prior to use. Chromatography was run on a BioCAD. Example 4: Purification of a Humanized Monoclonal Antibody by On-line Protein Breakthrough Assay

The following purification method was applied to purification of rhuMAb-HER2 or C2B8. The effluent from the preparative protein A affinity column was monitored for C2B8 product breakthrough using an immobilized protein A perfusion chromatography assay, shown in Fig. 1A, where convective mass transport allows the assay to be run at 6000 cm/hr, resulting in a 2.3 minute assay. The assay provides a fast, accurate, and reliable method for real-time analysis of antibody breakthrough. The on-line assay measures protein breakthrough in real-time during loading, where antibody breakthrough is obscured at 280 nm by flow-through of contaminant proteins. The preparative assay for the protein of interest (C2B8 in this example) is shown in Fig. IB. The assay must be fast enough to run about ten assays during product breakthrough for accurate breakthrough measurement. For this study, a 2.3 minute assay is sufficient, but the assay time could be reduced by increasing the flow rate on the assay column or by decreasing the volume of equilibration.

During loading of protein A affinity columns, no antibody is present in the column effluent until the available dynamic capacity of the column is exhausted and antibody begins flowing through the column without binding (Arnold, F.H. et al., The Chemical Engineering Journal (1985) 30:B9). Breakthrough is often expressed as C;/C₀, the fraction of loaded antibody that is in the column effluent. During breakthrough, C C₀ starts at zero when all loaded antibody is bound to the column and ends at one when the column is saturated and all loaded antibody is flowing through the column unretained (Johnston, A. et al, Journal of Chromatography (1991) 548:127). The shape of the breakthrough curve for affinity absorption results from the ability of the antibody to transfer onto the absorbate surface, which is determined by three mass transfer mechanisms (Cowan, G.H. et al.. Journal of Chromatography (1986) 363:37): film diffusion (mass transfer from the mobile phase to the surface of the particle), pore diffusion (diffusion into the particle), and the kinetics of the interaction between the antibody and protein A. With no mass transfer effects (ideal behavior), C_j/C₀ would increase to one the instant that saturation capacity was reached, resulting in a very sharp breakthrough curve. Diffusion changes the shape of the breakthrough curve because with diffusion antibody starts breaking through before the column is saturated (Mao, Q.M. et al, Journal of Chromatography (1995) 691 :273).

Breakthrough curves were plotted as C_j/C₀ vs. antibody loaded, where the antibody loaded is F-C₀ t V_c

(see Figs. 2A - 2D). Using this plot method, the column capacity (in grams of antibody per liter of column volume) can be directly read from the plot, and it corrects for differences in flow rate or column length used to generate the breakthrough curves.

Because flow rate, column length, and column media affect diffusion throughout the column, these factors affect the shape of the breakthrough curve (Chase, H.A., Journal of Chromatography (1984) 297:179). Several breakthrough curves generated by the on-line assay are shown in Fig. 2. Figs. 2 A - 2D show breakthrough curves for five flow rates on three different column media.

For Sepharose A, at higher flow rates the shape of the breakthrough curve degenerates, and much antibody has to be loaded before saturation is reached because many of the binding sites are difficult to reach. At high flow rates, only the most accessible binding sites are loaded before breakthrough occurs.

Poros 50 A has symmetric breakthrough curves, indicating short diffusion path lengths. The breakthrough curves for Prosep A have a good shape when C_j/C₀ < 0.5, but degenerate when C_j/C₀ > 0.5 at higher flow rates, indicating that some binding sites are deep in the media. Affinity columns are often loaded to breakthrough capacity Q_b, measured as the amount of protein of interest (such as an antibody) loaded at C,/C₀ = 0 05, the smallest breakthrough that can be accurately measured with the on-line assay When it is desired that column loading of the protein of interest be controlled and stopped at the pomt when breakthrough of the protein of interest begins, a column loading device is manipulated such that loading is stopped when C./C₀ is approximately 0 05, or when the amount of protein of interest unretained by the first chromatography column (such as a preparative column) reaches an unacceptable level with regard to reduced purification efficiency

In the range studied, breakthrough capacity is linearly related to flow rate, and column length has a logarithmic relationship to capacity, which approximates the breakthrough curve approachmg ideal behavior at sufficiently long column lengths (Figs 3 A and 3B) Sepharose A has a high Q_b at low flow rates, but Q_b drops off quickly as flow rate increases because only the most accessible binding sites are filled Poros 50 A and Prosep A both have shallower slopes than Sepharose A, and Poros 50 A has a higher capacity than Prosep A at all flow rates studied

The on-line assay measures antibody breakthrough during loading of preparative protein A chromatography, leading to a fast evaluation of breakthrough capacity The assay can be applied to other antibody processes where real-time measurement of antibody concentration is needed By immobilizing different affinity hgands on the assay media, this assay could be modified to measure breakthrough of other proteins in real-time It is useful for monitoring processes where breakthrough is obscured by flow through of contaminant proteins The on-line assay described herein may be applied using any affinity column specific for the protein of interest, where the protein of interest may be a whole protein, a fragment thereof, or a fusion protein and includes, for example, antibodies or fusion proteins containing antibody fragments Preferably, the protein of interest includes without limitation such proteins as IgG fusion proteins, C2B8, rhuMAb-HER2, E25, E26, TRY, hDNase, and IGF-l Where the protein of interest is IGF-1, the contaminant includes without limitation an aggregate of IGF proteins Where the protein of interest is produced CHO cell culture, the contaminant includes without limitation CHOP Where the protein of interest is produced in E colt cell culture, the contaminant includes without limitation ECP

The detection method may be any method suitable for the protein of interest and suitable for on-line, real-time analysis

Table II shows the results from loading to breakthrough (determined by the on-line assay), compared to loading by volume for rhuMAb-HER2 and C2B8 produced CHO cell culture In each case, yield and purity (determined by CHOP ELISA) was equivalent TABLE II

L (cm) -.₀ (cm/hr) Product Loading Method Media Yield (%) CHOP (μg/ml)

/ 14 500 C2B8 13 g/L by volume Prosep 94 ± 1 13 ± 1

2 14 500 C2B8 Prosep 90 ± 1 16 ± 1

C 0.05 by on-line assay

4 20 800 rhuMAb-HER2 20 g/L by volume Prosep 91 ± 3 28 ± 4

5 20 800 rhuMAb-HER2 Prosep 90 ± 2 27 ± 4

C 0.05 by on-line assay

6 11 1500 rhuMAb-HER2 9 g/L by volume Prosep 88 ± 4 22 ± 3

7 11 1500 rhuMAb-HER2 Prosep 86 ± 5 24 ± 1

C 0.05 by on-line assay

8 10 1100 rhuMAb-HER2 13 g/L by volume Poros 107 ± 4 42 ± 5

9 10 1 100 rhuMAb-HER2 Poros 106 ± 3 40 ± 3

C 0.05 by on-line assay

Example 5: Purification of IGF- 1 in the presence of IGF Aggregates

During purification of IGF-I, IGF-I is separated from aggregated IGF-I. Purified IGF-I is difficult to reliably collect because the preparative chromatogram does not indicate the elution positions of IGF-I and aggregated IGF-I. An on-iine reversed-phase assay can assay for IGF-I and aggregated IGF-I every 4 minutes and enable real-time control of product pooling.

Preparative IGF-I purification was performed on a 1 X 25 cm (20 ml) Bakerbond C4 40 μm 275 A column (Fig. 4B). Buffer A was 50 mM acetic acid 50 mM citric acid pH 3.0 and Buffer B was 50 mM acetic acid 20 mM citric acid pH 3.0 with 50% hexylene glycol. The load was 12.5 mg IGF/ml CV of folded pool. The column was run at a temperature of 30 C. A flow rate of 20 column volumes (CV)/hr was used for the Equilibrate/Load/Wash, and a flow rate of 9 CV/hr was used for the Gradient/Regenerate. The method was: Equilibrate 3 CV 100% Buffer A, Load, Wash 2 CV 100% Buffer A, Gradient 0-50% Buffer B/ 15 CV, Regenerate 2 CV 100% Buffer B. Detection was at 280 nm. Off-line analysis was performed by an HPLC assay. The column was a 4.6 X 25 mm Vydac C18 5 μm

300A. Buffer A was 0.12% trifluoroacetic acid (TFA) in water and Buffer B was 0.1% TFA in acetonitrile. The flow rate was 2 ml/min. The temperature was 50° C. A volume of 100 μl was injected, containing about 10 μg of IGF-I. The method was: 27.5-28.5% B / 9 min, 28.5-40% B / 4 min. 40-90% B / 2 min, hold 90% B for 1 min, 27.5% B for 4 min. Detection was at 214 nm. On-line analysis was performed by a rapid reversed-phase assay (Fig. 4A). The column was a 2.1 X

30 mm Poros R2/M column run at 4 ml/min (20 ml/min purge) at room temperature. Buffer A was 0.1% TFA in 10% acetonitrile, Buffer B was 0.1% TFA in 27%, and Buffer C was 0.1% TFA in 60% acetonitrile. 5 μl of the preparative column effluent, sampled from a sampling T, was injected. Detection was at 214 nm.

Five runs were collected from 85% to 85% purity (calculated as the peak area of IGF-I divided by the sum of the peak areas of IGF-I and aggregated IGF-I), determined in real time by the on-line assay. The purified pools were analyzed by the off-line HPLC assay. The results of this analysis indicate that a purified

IGF-I pool can be reliably collected (Fig. 5).

Example 6: Improved Purification of a Protein of Interest by Monitoring Contaminant in the Column Effluent Under conditions in which a contaminant elutes before, after or at approximately the same elution volume as the protein of interest, the method of the invention disclosed herein may be adapted to monitor the concentration of the contaminant(s) such that the purity of the protein of interest in the collected material is monitored and/or enhanced. Collection of the protein of interest as it elutes from the chromatography is controlled such that the amount of contaminating protein is kept to a relatively low acceptable level in the collected material. The methods of the invention may be applied together to the purification of a protein of interest such that both loading and collection are monitored and controlled, thereby limiting the amount of contaminating proteins significantly.

Disclosed herein is an ELISA of host cell proteins as trace contaminants in a preparation of a protein of interest (such as recombinant protein) produced in CHO cell cultures (CHO proteins or CHOPs). Process validation involves documenting the removal of host cell impurities by the purification process (Chen, A.B.

(1996) J. Biotech. Healthcare 3:70-80). CHOPs analysis is by ELISA using polyclonal goat antibodies raised to the complex mixture of CHOPs present in a blank run of the same cell line used in making the recombinant protein (for a general procedure, see, for example, Chen, A.B. (1996), supra). Immobilization of anti-CHOP polyclonal antibodies onto POROS AL was performed according to the manufacturer's instructions (PerSeptive Biosystems, Framingham, MA; product bulletin "POROS AL Media: Immobilizing Ligand for Perfusion

Immunoassays (1996)).

The separation, monitoring and control method disclosed herein is applied to, but not limited to, C2B8 purification from host cell culture contaminants such as CHOPs and to the control of a chromatographic device based on monitoring of contaminants in the chromatographic column effluent. The ELISA used herein has been qualified for products produced in different CHO cell subclones, and grown under different culture conditions. The ELISA showed that the quantitative comparison of blank cell run

CHOPs produced by these various processes were nominally identical in immunoreactivity. ELISA immunoreactivity was quantitatively equivalent independent of cell culture size (10 L vs. 12 kL), or use of bovine or porcine peptones in the culture media. The same anti-CHOPs antiserum can be used for a range of products, obviating the need for tailoring anti-CHOPs ELISAs to specific products, or requiring blank runs to be performed to scale. However, not all immunizations result in reagents suitable for a "generic" ELISA.

Anti-CHOPs antibodies were immobilized on an affinity column and used to develop a Real-time

Process Monitoring (RPM, PerSeptive Biosystems, Framingham, MA) application. As disclosed herein, the assay was used to monitor the separation of CHOPs from a C2B8. The level of CHOPs was below the level seen by A280 monitoring of an ion exchange column, but well within the range of the RPM, and correlated well with the ELISA assay on fractions collected from the run.

Anti-CHOPs antibodies were prepared by immunizing goats with CHOPs produced from the culture of a non-producing CHO clone. The CHOPs were also immobilized on a solid support matrix and used to affinity purify anti-CHOPs antibodies from the immune sera. These anti-CHOP antibodies were used to prepare both a "generic CHOPs ELISA" and immobilized on a solid support matrix to make an anti-CHOPs column for an online chromatographic assay.

Real-time Process Monitoring is an HPLC software system that allows the monitoring of a purification process for a specific analyte. The separation was performed for purification of C2B8 on an ion exchange column (Fig. 6B), while the presence of CHOP in the effluent was monitored on an anti-CHOP antibody on-line chromatographic assay (Fig. 6A). The preparative column was a 1.6 X 17.5 cm Poros 50 HS. The flow rate was 100 cm/hr, the load was 17 g/L of C2B8 protein A pool. The buffer was 20 mM MES pH 5.5. The procedure included the following steps: equilibrate with 2 CV 60 mM NaCl, Load, wash 1 CV 60 mM NaCl, gradient 60 - 500 mM NaCl over 5 C V, regenerate 4 C V 500 mM NaCl.

On-line analysis of CHOP used a 4.6 X 100 20 μm Poros with immobilized CHOP antibodies (Fig. 4A). The flow rate was 15 ml/min. The method was: equilibrate with PBS pH 7.2 and elute with PBS pH 2.0. The injection volume was 5 ml from the preparative column effluent. The anti-CHOPs affinity column assay (Fig. 4A) was a capture/elution assay taking 4 min. per cycle. Integration of the eluted peak provides a quantitative estimate of the CHOPs in the sample.

The purification of C2B8 involved separating an already highly purified protein from trace CHOPs. In fact, the CHOPs levels are too low to be seen by A280 detection. Fractions were also collected and submitted to the generic CHOPs ELISA, with quantitatively similar results to the RPM estimation of CHOPs. Monitoring of CHOPs levels in the final product is useful for the evaluation of product purity. Eluent collection control may be performed when monitored levels of CHOP in the eluent reach a predetermined level in order to maximize purification of the protein of interest.

Although the invention has been described with reference to the presently preferred embodiment, it should be understood that various modifications can be made without departing from the spirit of the invention. Accordingly, the invention is limited only by the following claims. What is claimed is:

Claims

1. A method of separating a protein of interest from a contaminant, the method comprising: a) loading a protein mixture onto a first chromatography column, wherein the mixture comprises the protein of interest and at least one contaminant as analytes, b) monitoring the amount of an analyte in effluent from the chromatography column as it elutes from the column, by deflecting column effluent to a separate on-line assay apparatus for a period of time sufficient to provide enough analyte to be tested in the on-line assay apparatus, wherein the monitoring requires less than twenty percent of the total elution time of the analyte on the chromatography column and wherein the monitoring measures the amount of analyte in the effluent from the chromatography column, and c) controlling a chromatographic device when the amount of analyte in the effluent reaches a predetermined amount.

2. The method of claim 1 , wherein the device is a loading device directing a protein mixture onto the column or away from the column, wherein the device is controlled in step c.

3. The method of claim 2, wherein the loading device is control such that loading is stopped when C_j/C₀ is approximately 0.05.

4. The method of claim 1, wherein the device is an effluent collection device directing column effluent into a collection reservoir or away from the collection reservoir, wherein the device is controlled in step c.

5. The method of 1, wherein the controlling of step c controls a loading device and an effluent collection device.

6. The method of claim 1, wherein the monitoring requires less than ten percent of the total elution time of the protein of interest on the first chromatography column.

7. The method of claim 1, wherein the monitoring requires less than five percent of the total elution time of the protein of interest on the chromatography column.

8. The method of claim 1, wherein the protein of interest is a fusion protein wherein a protein is fused covalently to a portion of an IgG molecule.

9. The method of claim 1, wherein the protein of interest is selected from the group consisting of

C2B8, rhuMAb-HER2, t-PA, TRY, hDNase, E25, and E26.

10. The method of claim 1 , wherein the protein of interest is IGF- 1.

1 1. The method of claim 10, wherein the contaminating protein is an aggregate of IGF proteins.

12. The method of claim 1, wherein the separate on-line assay apparatus comprises a chromatography column.

13. The method of claim 12, wherein the separate on-line assay chromatography column is selected from the group consisting of a reverse phase column, a protein A affinity column, an anti-HCP affinity column, an anti-CHOP affinity column, and an anti-ECP affinity column.

14. The method of claim 1 , wherein the detection method is UV absorption.

15. The method of claim 1, wherein the controlling of step c comprises electronic control of the chromatographic device.

16. A method of monitoring the separation of a protein of interest from a contaminant, the method comprising: a) loading a protein mixture onto a first chromatography column, wherein the mixture comprises the protein of interest and at least one contaminant, b) monitoring the amount of a contaminant in effluent from the chromatography column as it elutes from the column, by deflecting column effluent to a separate on-line assay apparatus for a period of time sufficient to provide enough contaminant to be detected, wherein the monitoring requires less than twenty percent of the total elution time of the analyte and contaminant on the chromatography column and wherein the monitoring measures the amount of contaminant in the effluent from the chromatography column.

17. The method of claim 16, wherein the separate on-line assay apparatus comprises a chromatography column.

18. The method of claim 17, wherein the chromatography column of the separate on-line assay apparatus is selected from the group consisting of a reverse phase column, a protein A affinity column, an anti- HCP affinity column, an anti-CHOP affinity column, and an anti-ECP affinity column.

19. An isolated protein of interest separated from a contaminant, the protein of interest produced by a method comprising: a) loading a protein mixture onto a first chromatography column, wherein the mixture comprises the protein of interest and at least one contaminant as analytes, b) monitoring the amount of an analyte in effluent from the chromatography column as it elutes from the column, by deflecting column effluent to a separate on-line assay apparatus for a period of time sufficient to provide enough analyte to be tested in the on-line assay apparatus, wherein the monitoring requires less than twenty percent of the total elution time of the protein of interest on the chromatography column and wherein the monitoring measures the amount of analyte in the effluent from the chromatography column, and c) controlling a chromatographic device when the amount of analyte in the effluent reaches a predetermined amount.

20. The isolated protein of interest of claim 19, wherein the protein of interest is selected from the group consisting of C2B8, rhuMAb-HER2, IGF-1, and hDNase.