Scholarly article on topic 'Engineering of novel Staphylococcal Protein A ligands to enable milder elution pH and high dynamic binding capacity'

Engineering of novel Staphylococcal Protein A ligands to enable milder elution pH and high dynamic binding capacity Academic research paper on "Chemical sciences"

CC BY-NC-SA
0
0
Share paper
Academic journal
Journal of Chromatography A
OECD Field of science
Keywords
{"Protein A" / "Z domain" / Antibody / "Affinity chromatography"}

Abstract of research paper on Chemical sciences, author of scientific article — Timothy M. Pabst, Ronnie Palmgren, Annika Forss, Jelena Vasic, Mariko Fonseca, et al.

Abstract We describe novel Staphylococcal Protein A ligands that enable milder elution pH for use in affinity chromatography. The change in elution pH is the result of point mutations to the protein sequence. Two novel ligands are investigated in this study. The first, designated Z(H18S)4, represents a histidine to serine substitution single mutation. The second, designated Z(H18S, N28A)4, is a double mutant comprising histidine to serine and asparagine to alanine mutations. Both are compared against the unmutated sequence, designated Z4, which is currently utilized in a commercially available Protein A stationary phase for the purification of molecules containing Fc domains. The ligands are coupled to a chromatography support matrix and tested against a panel of antibodies and an Fc fusion protein for elution pH, dynamic binding capacity, step-wise elution, and capture from clarified culture media. Results demonstrate that the novel ligands result in milder elution pH, on average >0.5 pH units, when tested in a pH gradient. For step-wise elution at pH 4.0, the Z(H18S, N28A)4 ligand showed on average a greater than 30% increase in yield compared to Z4. Importantly, for the antibodies tested the mutations did not result in a decrease in dynamic binding capacity or other desirable attributes such as selectivity. A potential application of the novel ligands is shown with a pH sensitive molecule prone to aggregation under acidic conditions.

Academic research paper on topic "Engineering of novel Staphylococcal Protein A ligands to enable milder elution pH and high dynamic binding capacity"

ELSEVIER

Contents lists available at ScienceDirect

Journal of Chromatography A

journal homepage www.elsevier.com/locate/chroma

Engineering of novel Staphylococcal Protein A ligands to enable milder elution pH and high dynamic binding capacity

Timothy M. Pabsta1, Ronnie Palmgrenb1, Annika Forssb, Jelena Vasicb, Mariko Fonsecaa, Christopher Thompson2, William K. Wanga, Xiangyang Wanga, Alan K. Huntera *

a MedImmune, Purification Process Sciences, One MedImmune Way, Gaithersburg, MD 20878, USA b GE Healthcare, Life Sciences R&D, Bjorkgatan 30, SE-751 84 Uppsala, Sweden

ABSTRACT

We describe novel Staphylococcal Protein A ligands that enable milder elution pH for use in affinity chromatography. The change in elution pH is the result of point mutations to the protein sequence. Two novel ligands are investigated in this study. The first, designated Z(H18S)4, represents a histidine to serine substitution single mutation. The second, designated Z(H18S, N28A)4, is a double mutant comprising histidine to serine and asparagine to alanine mutations. Both are compared against the unmutated sequence, designated Z4, which is currently utilized in a commercially available Protein A stationary phase for the purification of molecules containing Fc domains. The ligands are coupled to a chromatography support matrix and tested against a panel of antibodies and an Fc fusion protein for elution pH, dynamic binding capacity, step-wise elution, and capture from clarified culture media. Results demonstrate that the novel ligands result in milder elution pH, on average >0.5 pH units, when tested in a pH gradient. For step-wise elution at pH 4.0, the Z(H18S, N28A)4 ligand showed on average a greater than 30% increase in yield compared to Z4. Importantly, for the antibodies tested the mutations did not result in a decrease in dynamic binding capacity or other desirable attributes such as selectivity. A potential application of the novel ligands is shown with a pH sensitive molecule prone to aggregation under acidic conditions.

© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-SA

license (http://creativecommons.org/licenses/by-nc-sa/3.0/).

CrossMark

ARTICLE INFO

Article history: Received 16 December 2013 Received in revised form 9 June 2014 Accepted 14 August 2014 Available online 19 August 2014

Keywords: Protein A Z domain Antibody

Affinity chromatography

1. Introduction

Staphylococcal Protein A (SpA) is a 42 kDa single chain polypeptide localized to the outer surface of Staphylococcus aureus [1-4]. Native SpA comprises five highly homologous Fc binding domains designated E, D, A, B, and C, followed by a cell wall binding domain designated X [5-7]. The five Fc binding domains are organized in an anti-parallel a-helical arrangement [5]. The structure of an IgG Fc fragment bound to SpA has been solved and the amino acid residues implicated in binding have been identified [3,8].

The potential of SpA to be used as an affinity ligand for protein purification has been recognized for decades. Early SpA affinity resins consisted of native Protein A coupled to a base matrix most often through covalent bonding to amines. Since then, dramatic improvements have been made in Protein A chromatography stationary phases. Among the most significant innovations in SpA

* Corresponding author. Tel.: +1 301 398 4142; fax: +1 301 398 9322.

E-mail address: hunterak@medimmune.com (A.K. Hunter). 1 These authors contributed equally to this work.

engineering is the Z domain, which represents a synthetic analogue of the native B domain developed for purification of Fc-fusion proteins [9,10]. A derivative of the Z domain engineered for greater alkaline stability has gained widespread use for capture of recombinant therapeutic proteins.

For industrial purification of antibodies and Fc fusion proteins, Protein A chromatography is routinely utilized as part of a platform approach [7,11-19]. In most instances, the Protein A column is placed first in the purification train to capture product from clarified cell culture broth [20-22]. This configuration provides an optimal balance of cost of goods for manufacturing, process similarity for different molecules, and process robustness. However, process models developed to predict facility capacity and cost of goods tend to be sensitive to Protein A capture column dynamic binding capacity (DBC) [23,24]. As a result, the last decade has seen introduction of multiple generations of Protein A stationary phases designed to achieve ever higher DBCs.

One disadvantage of Protein A for use as an affinity chromatogra-phy ligand is that low pH conditions, typically pH 3-4, are required for elution of bound proteins. It is generally understood that low pH has the potential to destabilize proteins and contribute to aggregate

http://dx.doi.org/10.1016/j.chroma.2014.08.046

0021-9673/© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-SA license (http://creativecommons.org/licenses/by-nc-sa/3.0/).

formation [17,19,25]. In particular, Fc-fusion proteins are often very sensitive to low pH conditions, greatly complicating the process development and manufacturing of these proteins. To overcome this challenge, successful strategies have included the use of aggregation suppressors such as arginine and urea [17,26], as well as SpA engineering to increase elution pH through destabilization of the ligand itself or the ligand-Fc interaction [27,28].

When designing a Protein A stationary phase, a balance must be struck between several often competing properties. To minimize aggregate formation, high elution pH is desirable; however, when considering large scale bioprocessing, changes that lead to higher elution pH must not be achieved at the expense of a substantial reduction in DBC. Other properties associated with modern Protein A media, such as high selectivity, low elution pool volume and high flow rates should also be maintained to the greatest extent possible.

In this work, we examine the chromatographic behavior ofnovel Z domain stationary phases engineered to achieve higher elution pH while maintaining high DBC. To study behavior under column chromatography conditions, the ligands are expressed and coupled to a chromatography support matrix. The effect of mutations on elution pH is tested using a panel of antibodies and an Fc-fusion protein under low loading conditions with a linear pH gradient. Similarly, the impact of mutations on DBC is investigated with breakthrough experiments. To test performance in a typical bioprocess scenario, stepwise elution under high loading conditions and product capture from clarified culture broth are investigated.

2. Materials and methods

2.1. Construction ofZ domain ligands and stationary phases

Three affinity ligands were used in this work, comprised of proprietary Z-domain derivatives. MabSelect SuRe (GE Healthcare) is a commercially available Protein A stationary phase that incorporates a ligand, designated Z4, comprising four repeats of the Z domain engineered to be alkaline resistant. A linker with a C terminal cysteine is added to the last repeat to facilitate coupling to a chromatography support matrix. The other two stationary phases used in this work incorporate novel ligands. The first, designated Z(H18S)4, is identical to the SuRe Z4 ligand with the exception of a histidine to serine mutation at position 18 included for each repeat. The final stationary phase incorporates a ligand designated Z(H18S, N28A)4. It is also identical to the SuRe Z4 ligand with the exceptions of a histidine to serine mutation at position 18 as well as an asparagine to alanine mutation at position 28 included for each repeat.

The novel constructs were expressed using recombinant Escherichia coli. After fermentation, the cells were subject to heat treatment to release the Z domain ligands into the media. The crude extract was then clarified by microfiltration with a 0.2 |im membrane. The membrane permeate was loaded onto IgG Sepharose (GE Healthcare). The resin was washed with phosphate buffered saline (PBS) and eluted by lowering to pH 3. The elution pool was adjusted to neutral pH, reduced by addition of dithiothreitol (DTT), and purified by anion exchange chromatography. The purified ligands were analyzed by LC-MS (Agilent) to confirm the molecular weight.

Following expression and purification, ligands were coupled to 85 | m diameter cross-linked agarose porous beads using standard techniques via a thioether linkage. Ligand density was determined by amino acid analysis based on the total amino acid content recovered from a stationary phase sample. Table 1 summarizes the Z domain stationary phases used in this work.

Table 1

Stationary phase properties.

Ligand name Base matrix dP (|im)a Normalized

ligand density

Z4 Cross-linked agarose 85 1.00

Z(H18S)4 Cross-linked agarose 85 1.05

Z(H18S, N28A)4 Cross-linked agarose 85 1.03

a Average bead diameter.

Table 2

Antibody and Fc-fusion protein properties.

Molecule Type p/a MWb (kDa) rc (nm)

mAbl IgGl 9.2 l47 5.7

mAb2 IgGl 8.9 l46 5.3

mAb3 IgGl 9.4 l45 5.4

mAb4 IgGl 8.4 l44 5.6

mAb5 IgG4 7.l l47 6.6

BsAb Bispecific 8.2 200 6.5

Fcl Fc-fusion (IgGl) 5.7 90 4.7

a pi of main peak by isoelectric focusing.

b Molecular weight by mass spectrometry, including glycosylation. c Hydrodynamic radius measured by DLS.

2.2. Buffer reagents and protein preparations

Chemicals used for buffer preparation were obtained from Sigma (St. Louis, MO, USA) and JT Baker (Phillipsburg, NJ, USA). Recombinant human serum albumin (HSA) was obtained from InVitra (Fort Collins, CO, USA). Antibodies and an Fc-fusion protein were expressed in Chinese hamster ovary (CHO) cells using standard cell culture techniques. To generate purified material, clarified cell culture broth was purified by Protein A chromatography and then by ion exchange chromatography. Table 2 summarizes the antibodies and Fc-fusion proteins used in this work.

2.3. Chromatography columns and instrumentation

Chromatography experiments were conducted using a GE Healthcare ÄKTA Explorer 100. Resins were packed to approximately 20 cm bed height in 0.5 cm diameter Tricorn columns (GE Healthcare) for all experiments except dynamic binding capacity experiments, which used 1.1 cm diameter Vantage L11 columns from Millipore (Billerica, MA, USA) packed to a 5 cm bed height.

2.4. Protein concentration determination by absorbance (A280)

Protein concentrations of purified samples were determined using a Nanodrop 2000c from Thermo Scientific (Wilmington, DE, USA) with the microvolume pedestal and measurement at a wavelength of 280 nm.

2.5. Antibody concentration determination by protein A HPLC (ProA-HPLC)

Concentration of mAbs and Fc-fusion proteins in clarified cell culture broth was determined by analytical high performance Protein A chromatography (ProA-HPLC) using a POROS A 20 (4.6 mm i.d. x 10 cm, 20 |im) column obtained from Life Technologies (Grand Island, NY, USA) with an Agilent 1100 HPLC system (Palo Alto, CA). The binding mobile phase buffer consisted of phosphate buffered saline (PBS), pH 7.2 and the elution buffer was PBS, pH 2.2; both at a flow rate of 3.5 mL/min. Each injection was eluted for a total of 2.5 min. Samples were applied to the column neat and the elution profile was monitored at 280 nm using the HPLC spectrophotometer. Elution peak area was converted to a

protein concentration using a standard curve generated with purified material.

2.6. pH measurement

Buffer and protein solution pH was measured offline using a SevenMulti pH meter from Mettler Toledo (Columbus, OH, USA) equipped with an InLab Expert Pro pH probe from Mettler Toledo. The meters were calibrated with pH 2, 4, 7, and 10 standards.

2.7. Dynamic binding capacity (DBC) experiments

For DBC experiments, defined as 10% breakthrough, columns were equilibrated with 25 mM Tris, pH 7.5 and then loaded until breakthrough was observed. Prior to column loading, purified material was diluted to 1 mg/mL with equilibration buffer to match the equilibration conditions. All experiments were operated at a residence time of 4 minutes. The contributions of column void volume and system delay volume were subtracted from the protein load volume to obtain DBC. The maximum absorbance of the load was measured with the column in bypass.

2.8. Linear pH gradient elution chromatography

For linear pH gradient elution experiments, columns were equilibrated with 25 mM Tris, 150 mM NaCl, pH 7.5 and then purified protein (diluted to 4 mg/mL with equilibration buffer) was loaded on the column to a load challenge of 5 g/L resin. The column was re-equilibrated, washed with 25 mM citrate, pH 6.7, and then eluted in a linear gradient to 25 mM citrate, pH 2.7 over 10 column volumes. Experiments were performed at a flow rate of 300 cm/h with online UV absorbance monitored at 280 nm. Elution pH at the peak maximum was calculated through linear interpolation between the measured pH values of the two buffers used to form the gradient, taking into account column void volume and the system delay volume.

2.9. Stepwise elution chromatography

For stepwise elution experiments, columns were equilibrated with 25 mM Tris, 150 mM NaCl, pH 7.5 and then purified protein (diluted to 4 mg/mL with equilibration buffer) was loaded on the column to a load challenge of 30 g/L resin and 10 g/L resin for antibodies and the Fc-fusion protein, respectively. The column was re-equilibrated and then eluted in stepwise fashion to 25 mM acetate, pH 4.0. Product pools were collected over 2 column volumes to mimic typical large-scale bioprocess conditions. Experiments were performed at a flow rate of 300 cm/hr with online UV absorbance monitored at 280 nm. Step yield was determined using mass of product in the load and pool (as determined by A280).

2.10. Capture from clarified cell culture broth

For capture of clarified CHO cell culture broth, columns were equilibrated with 25 mM Tris, 150 mM NaCl, pH 7.5 and then clarified cell culture broth was loaded on the column to a load challenge of 30 g/L resin and 10 g/L resin for antibodies and the Fc-fusion protein, respectively. The column was re-equilibrated and then washed with 25 mM Tris, 1 M NaCl, pH 7.5 to remove HCP. The column was re-equilibrated with 25 mM Tris, 150 mM NaCl, pH 7.5 and then eluted in stepwise fashion to 25 mM acetate, pH 4.0. Product pools were collected over 2 column volumes. Experiments were performed at a flow rate of 300 cm/hr with online UV absorbance monitored at 280 nm. Step yield was determined using mass of

0 5 10 15 20 25

Column volumes (CV)

Fig. 1. Elution of mAb 2 in a linear pH gradient from 6.7 to 2.7 on Protein A columns loaded to 5 g/L. The pH gradient is shown as a dotted line.

product in the load (as determined by ProA-HPLC) and pool (as determined by A280).

2.11. High performance size exclusion chromatography (HP-SEC)

Analytical high performance size exclusion chromatography (HP-SEC) was performed using a TSKgel G3000SWXL (7.8 mm i.d. x 30 cm, 5 |m) column obtained from Tosoh Biosciences (King of Prussia, PA, USA) with an Agilent 1100 HPLC system (Palo Alto, CA). The mobile phase buffer consisted of 100 mM sodium phosphate, 100 mm sodium sulfate, pH 6.8 at a flow rate of 1 mL/min. Each injection was eluted for a total of 20 min. The elution profile was monitored at 280 nm using the HPLC spectrophotometer. Samples were applied to the column neat and the injection volume was adjusted to load approximately 250 |g of total protein on the column. The column was calibrated using gel filtration standards from Bio-Rad (Hercules, CA, USA).

2.12. Host cell protein measurements

Host cell protein (HCP) concentrations were measured using the bioaffy sandwich immunoassay on the Gyrolab xP workstation from Gyros AB (Uppsala, Sweden). Capture and detection antibodies were in-house reagents raised against HCP from the cell line used to produce the antibodies and Fc-fusion proteins used in this work.

2.13. Residual Protein A measurements

Residual Protein A in elution pools was measured with an in-house enzyme-linked immunosorbent assay (ELISA) using commercially available anti-Protein A capture antibodies.

2.14. Dynamic light scattering (DLS)

Dynamic light scattering measurements were made with a DynaPro Plate Reader II from Wyatt Technology (Santa Barbara, CA, USA) with purified protein samples diluted to 5 g/L with 25 mM Tris, 150 mM NaCl, pH 7.5. Hydrodynamic radius was estimated using Dynamics (v.7) software.

3. Results and discussion

3.1. Linear pH gradient elution behavior

Fig. 1 shows results of linear pH gradient elution of mAb 2 on Z4, Z(H18S)4, and Z(H18S, N28A)4 stationary phases. As is seen in the

Table 3

Summary of linear pH gradient elution.

Molecule

pH at elution peak maximum

Z4 Z(H18S)4 Z(H18S, N28A)4

mAbl 3.6 3.8 4.0

mAb2 3.7 4.2 4.5

mAb3 3.7 4.1 4.4

mAb4 3.7 4.0 4.2

mAb5 3.6 3.9 4.0

BsAb 4.0 4.2 4.6

Fcl 3.7 5.4 -

figure, Z(H18S, N28A)4 elutes at higher pH than Z(H18S)4, which in turn elutes at higher pH than Z4. The elution profile, as determined by peak height and shape, on all three resins was similar. Moreover, the peaks all appeared fairly symmetric, showing slight fronting during the early part of elution.

The pattern shown in Fig. 1 was consistent across all antibodies tested in this work. Table 3 summarizes results of linear pH gradient elution experiments. For a single resin, differences were observed in pH at elution peak maximum for different molecules. However, for a single molecule, the trend was always the same where Z(H18S, N28A)4 elutes at higher pH than Z(H18S)4 which elutes at higher pH than Z4. The Fc1 molecule showed the greatest difference in elution pH among the different resins. A value was not reported for the Z(H18S, N28A)4 resin as the protein began to elute isocrat-ically prior to the start of the gradient at pH 6.7. Moreover, the Fc1 molecule showed much broader elution peaks on the Z(H18S, N28A)4 and Z(H18S)4 resins compared to the Z4 resins as shown in Fig. 2. This behavior, the reason for which is unknown, was not observed for any antibodies.

3.2. Dynamic binding capacity

Results of DBC experiments for mAb 5 are shown in Fig. 3. The profiles showed the characteristic sigmoidal shape expected for breakthrough curves. DBC results for all molecules tested are shown in Table 4. Unlike the trend observed for linear pH gradient elution experiments, as summarized in Table 3, for DBC of antibodies the highest capacity was observed for the Z(H18S)4 resin, the second highest capacity was seen with Z(H18S, N28A)4, and the lowest capacity was seen for Z4. However, the difference was in our view minor and given the trend in DBC is similar to what one would expect based on the ligand density measurements, it cannot be concluded that the mutations alone result in higher DBC. This could occur, for example, through higher binding stoichiometry. Nonetheless, the fact that the mutations did not result in an

Fig. 3. Breakthrough curves for mAb 5 on Protein A resins.

apparent decrease in DBC compared to Z4 ligand, which is used in a modern commercial Protein A resin, is a significant finding.

While mAbs showed very similar capacities when comparing ligands, the behavior of the Fc-fusion protein was markedly different. For this molecule, the capacity of all three resins was much lower compared to the antibodies and the novel ligands showed lower capacity than the Z4 ligand. Again, as was the case for the gradient elution experiments, the reason for the behavior is unknown. Based on prior work in this area we can speculate on possible causes. In their detailed work, Ghose, Hubbard and Cramer noted that binding stoichiometry for a panel of antibodies and Fc fusion proteins ranged from 2.4 to 3.1, showing not all available Protein A binding domains could be fully utilized [12]. This effect was attributed to intra-ligand steric hindrance. In the same work, it was noted that inter and intra-ligand steric hindrance strongly impacted Protein A stationary phase binding capacity. The dynamic light scattering results shown in Table 2 demonstrate the Fc1 molecule has the smallest hydrodynamic radius of any of the proteins used in this work. Thus, inter-ligand steric hindrance would seem to be an unlikely cause. As a result, the most likely explanation is that features of the Fc1 molecule must exist which result in severe intra-ligand hindrance. The novel mutations appear to exacerbate this effect, leading to lower capacity than seen for Z4. As each Fc-fusion protein is unique, this data cannot be seen as predictive for the behavior of other fusion constructs.

These results demonstrate that higher elution pH can be achieved without sacrificing DBC through novel point mutations to the Z domain. However, other important properties such as the impact of ligand mutations on selectivity under conditions typically encountered in modern bioprocessing remain to be addressed. The sections below are devoted to these topics.

Table 4

Summary of dynamic binding capacities for antibodies and an Fc-fusion protein on Protein A resins.

Fig. 2. Elution of Fc1 in a linear pH gradient from 6.7 to 2.7 on Protein A columns loaded to 5 g/L. The pH gradient is shown as a dotted line.

Molecule DBCio% (g/L resin)

Z4 Z(H18S)4 Z(H18S, N28A)4

mAbl 43.8 50.7 49.3

mAb2 39.0 44.5 43.9

mAb3 42.5 47.3 43.1

mAb4 41.2 47.4 46.8

mAb5 46.2 51.5 51.9

BsAb 48.3 52.9 49.8

Fcl 21.7 17.7 14.0

Table 5

Summary of pH 4.0 stepwise elution experiments.

Molecule Yield (%)

Z4 Z(H18S)4 Z(H18S, N28A)4

mAbl 60.8 72.4 76.4

mAb2 71.2 95.4 97.9

mAb3 69.5 94.0 98.2

mAb4 62.6 85.6 90.8

mAb5 48.2 80.1 88.5

BsAb 57.0 66.4 69.7

Fcl 11.1 89.1 93.1

3.3. Stepwise elution behavior

Table 6

Host cell protein clearance from clarified cell broth.

HCP (ng/mg)

mAb1 mAb2 mAb5 Fc1

Load (CB) 34,305 398,406 30,769 66,445

Pool-pH 3.5 elution

Z4 497 1088 268 671

Z(H18S)4 559 323 216 522

Z(H18S, N28A)4 231 315 659 457

Pool-pH 4.0 elution

Z(H18S)4 727 135 236 431

Z(H18S, N28A)4 372 95 176 352

Table 5 summarizes results of stepwise elution experiments at pH 4.0. In most cases, there is a marked increase in yield when comparing the Z(H18S)4 and Z(H18S, N28A)4 resins to the Z4 resin. This observation further corroborates the linear gradient elution data. It also demonstrates a benefit can be obtained from the Z(H18S)4 and Z(H18S, N28A)4 resins under conditions of high loading typically encountered in bioprocessing.

The difference in yield was most dramatic for the Fc fusion protein, where it increased from 11% on Z4 to 89% and 93% on the Z(H18S)4 and Z(H18S, N28A)4 resins, respectively. For the antibodies the average yield increased 21% when comparing Z(H18S)4 to Z4 and 25% when comparing Z(H18S, N28A)4 to Z4.

3.4. Purification of proteins from clarified cell culture broth and nonspecific binding

Fig. 4 shows a representative chromatogram obtained for purification of the Fc1 molecule from clarified cell culture broth (CB) on the Z(H18S)4 resin with pH 4.0 elution. Purification from CB on both the Z(H18S)4 and Z(H18S, N28A)4 was similar to that of Z4, resulting in product elution peaks having volumes of less than 2CV.

Table 6 summarizes HCP clearance for three antibodies and the Fc1 molecule. No trends were observed with respect to removal of HCP across the different resins and molecules tested. This result is consistent with prior work suggesting HCP levels post Protein A are largely dependent on interactions between the antibody and the HCP, with little contribution due to interactions between the resin and the HCP [29,30].

As an additional test for nonspecific binding, recombinant HSA was loaded on each column to 30 g/L resin and the column was washed with 1 M NaCl, eluted, stripped, and sanitized as normal.

— Absorbancc

r — 1 — Conductivity _

AfL 1 1 ..... 1 1 1 1 1 1 1 • pH

\__J 1 1 1 1 1 .........; -

14 12 10

8 6 4 2

¡Z) £

0 20 40 60 80 100

Volume (mL)

Fig. 4. Representative chromatogram showing purification of Fc1 clarified cell culture broth on Z(H18S)4. The column was loaded to 10g/L resin and eluted at pH 4.0.

No nonspecific binding was observed during the run as determined by the chromatography workstation UV trace. In addition, a blank run was performed after the HSA run, no carry over was observed based on the UV trace.

3.5. Impact of elution condition on a pH sensitive molecule

Table 7 summarizes results of monomer purity obtained in elution pools for purification of Fc1 on the Z4, Z(H18S)4 and Z(H18S, N28A)4 resins after a 24 h hold at room temperature without neutralization. These conditions represent a realistic bioprocess scenario, where product pools must frequently be held for planned or unplanned reasons such as equipment malfunctions or process deviations. For this molecule, the ability to elute at higher pH translated into 45% reduction in aggregate levels. The higher monomer purity and reduced aggregate burden has obvious benefits, as separation of protein aggregates often represents a significant challenge. Thus, the ability to achieve reduction in aggregate levels can be expected to lead to higher manufacturing facility throughput and lower cost of goods.

3.6. Resin lifetime and ligand stability

Protein A resin lifetime is considered critically important to the success of modern bioprocessing. As part of this work, greater than 30 use cycles were performed on all three resins. Each cycle consisted of protein loading, elution, low pH strip with 0.1 M acetic acid, and sanitization with 0.1N NaOH. Although not done as part of a formal resin lifetime study, and therefore conditions varied from run to run, no deterioration in performance was observed over the course of this study. Process performance and product quality after 30 cycles was similar to that obtained for early cycles. Consistent with what would be expected based on structural similarity, this suggests the novel ligands are likely to have stability profiles similar to the Z4 ligand used in MabSelect SuRe.

To further investigate ligand stability, levels of leached Protein A ligand were measured by ELISA in the product pool for all three resins operated under typical bind and elute conditions. As is shown in Table 8, levels of leached Protein A ligand were low, less than 1 ppm in each case, and were similar for all three resins. This result

Table 7

HP-SEC analysis of Fc1 in Protein A elution pools after 24 h hold at room temperature.

Elution pool HP-SEC (% aggregate)

pH 3.5 elution

Z4 34.0

Z(H18S)4 33.2

Z(H18S, N28A)4 32.5

pH 4.0 elution

Z(H18S)4 18.0

Z(H18S, N28A)4 18.3

Table 8

Summary of leached Protein A ligand in elution pools as measured by ELISA.

Molecule Leached Protein A (ng/mg mAb) [9]

Z4 Z(H18S)4 Z(H18S, N28A)4

mAb1 0.56 0.33 0.33 [10]

confirms that the stability of the novel ligands is comparable to the Z4 ligand used in MabSelect SuRe, and thus acceptable for industrial use.

4. Conclusions

This work demonstrates novel Staphylococcal Protein A mutations that result in milder elution pH for use in affinity chromatography. With one exception, this was accomplished without sacrificing dynamic binding capacity or other desirable attributes such as selectivity. This work highlights the great potential Protein A still holds for future advances. As has happened in the past, improvements to this important bioprocessing platform technology will help drive the next generation of protein therapeutics.

Acknowledgements

We would like to thank the Immunoassay group at MedImmune for providing host cell protein data and the Reagent Preparation group at MedImmune for generating clEF data and for providing samples of mAbs to make this work possible. We are also grateful to Kripa Ram for his review of this manuscript.

References

[1] H.K. Kim, V. Thammavongsa, O. Schneewind, D. Missiakas, Recurrent infections and immune evasion strategies of Staphylococcus aureus, Curr. Opin. Microbiol. 15(2012) 92-99.

[2] A. Surolia, D. Pain, M.I. Khan, Protein A: nature's universal anti-antibody, Trends Biochem. Sci. 7 (1982) 74-76.

[3] H. Gouda, M. Shiraishi, H. Takahashi, K. Kato, H. Torigoe, Y. Arata, I. Shimada, NMR study of the interaction between the B domain of Staphylococcal Protein Aandthe Fc portion of immunoglobulin G, Biochemistry 37 (1998) 129-136.

[4] G. Silverman, B cell superantigens: possible roles in immunodeficiency and autoimmunity, Semin. Immunol. 10 (1998) 43-55.

[5] M.A. Starovasnik, M.P. O'Connell, W.J. Fairbrother, R.F. Kelley, Antibody variable region binding by Staphylococcal Protein A: thermodynamic analysis and location of the Fv binding site on E-domain, Protein Sci. 8 (1999) 1423-1431.

[6] M. Graille, E.A. Stura, A.L. Corper, B.J. Sutton, M.J. Taussig, J.-B. Charbonnier, G.J. Silverman, Crystal structure of a Staphylococcus aureus Protein A domain complexed with the Fab fragment of a human IgM antibody: structural basis for recognition of B-cell receptors and superantigen activity, PNAS 97 (2000) 5399-5404.

[7] S. Hober, K. Nord, M. Linhult, Protein A chromatography for antibody purification, J. Chromatogr. B 848 (2007) 40-47.

J. Deisenhofer, Crystallographic refinement and atomic models of a human Fc fragment and its complex with fragment B of Protein A from Staphylococcus aureus at 2.9- and 2.8-Ä resolution Biochemistry 20 (1981) 2361-2370. M. Tashiro, R. Tejero, D.E. Zimmerman, B. Celda, B. Nilsson, G.T. Montelione, High-resolution solution NMR structure of the Z domain of Staphylococcal Protein A, J. Mol. Biol. 272 (1997) 573-590.

B. Nilsson, T. Moks, B. Jansson, L. Abrahmsén, A. Elmblad, E. Holmgren, C. Henrichson, T.A. Jones, M. Uthlén, A synthetic IgG-binding domain based on Staphylococcal Protein A, Protein Eng. Des. Sel. 1 (1987) 107-113. S. Ghose, M. Allen, B. Hubbard, C. Brooks, S.M. Cramer, Antibody variable region interactions with Protein A: implications forthe development ofgeneric purification processes, Biotech. Bioeng. 92 (2005) 665-673.

S. Ghose, B. Hubbard, S.M. Cramer, Binding capacity differences for antibodies and Fc-fusion proteins on Protein A chromatographic materials, Biotech. Bioeng. 96 (2007) 768-779.

S. Ghose, T. McNerney, B. Hubbard, Protein A affinity chromatography for capture and purification of monoclonal antibodies and Fc-fusion proteins: practical considerations for process development, in: A.A. Shukla, M.R. Etzel, S. Gadam (Eds.), Process Scale Bioseparations for the Biopharmaceutical Industry, CRC Press, Boca Raton, FL, 2006, pp. 463-490.

R. Hahn, R. Schlegel, A. Jungbauer, Comparison of Protein A affinity sorbents, J. Chromatogr. B 790 (2003) 35-51.

R. Hahn, P. Bauerhansl, K. Shimahara, C. Wizniewski, A. Tscheliessnig, A. Jungbauer, Comparison of Protein A affinity sorbents: II. Mass transfer properties, J.Chromatogr. A 1093 (2005) 98-110.

R. Hahn, K. Shimahara, F. Steindl, A. Jungbauer, Comparison of Protein A affinity sorbents: III. Lifetime study,J. Chromatogr. A 1102 (2006) 224-231.

A.A. Shukla, P. Gupta, X. Han, Protein aggregation kinetics during Protein A chromatography: case study for an Fc fusion protein, J. Chromatogr. A 1171 (2007) 22-28.

S. Vunnum, G. Vedantham, B. Hubbard, Protein A-based affinity chromatography, in: U. Gottschalk (Ed.), Process Scale Purification of Antibodies, John Wiley & Sons, Hoboken, NJ, 2009, pp. 79-102.

P. Gagnon, Technology trends in antibody purification, J. Chromatogr. A 1221 (2012)57-70.

U. Gottschalk, Bioseparation in antibody manufacturing: the good, the bad, and the ugly, Biotechnol. Prog. 24 (2008) 496-503.

B. Kelley, Industrialization of mAb production technology: the bioprocessing industry at a crossroads, MAbs 1 (2009) 443-452.

H.F. Liu, J. Ma, C. Winter, R. Bayer, Recovery and purification process development for monoclonal antibody production, MAbs 2 (2010) 1-20. B. Kelley, Very large scale monoclonal antibody purification: the case for conventional unit operations, Biotechnol. Prog. 23 (2007) 995-1008. M. Trexler-Schmidt, S. Sze-Khoo, A.R. Cothran, B.Q. Thai, S. Sargis, B. Lebreton, B. Kelley, G.S. Blank, Purification strategies to process 5 g/L titers of monoclonal antibodies, Biopharm Int. (Suppl. March) (2009).

M. Vázquez-Rey, D.A. Lang, Aggregates in monoclonal antibody manufacturing processes, Biotech. Bioeng. 108 (2011) 1494-1508.

T. Arakawa, J.S. Philo, K. Tsumoto, R. Yumioka, D. Ejima, Elution of antibodies from a Protein-A column by aqueous arginine solutions, Protein Exp. Purif. 36 (2004) 244-248.

S. Gülich, M. Ulhén, S. Hober, Protein engineering of an IgG-binding domain allows milder elution conditions during affinity chromatography, J. Biotechnol. 76 (2000) 233-244.

N.L. Brown, S.P. Bottomley, M.D. Scawen, M.G. Gore, A study of the interactions between an IgG-binding domain based on the B domain of Staphylococcal Protein A and rabbit IgG, Mol. Biotechnol. 10 (1998) 9-16.

A.A. Shukla, P. Hinckley, Host cell protein clearance during Protein A chromatography: development of an improved column wash step, Biotechnol. Prog. 24(2008) 1115-1121.

B. Nogal, K. Chhiba, J.C. Emery, Select host cell proteins coelute with monoclonal antibodies in Protein A chromatography, Biotechnol. Prog. 28 (2012) 454-458.