Today, monoclonal antibodies (mAbs) represent the largest and most progressive biopharmaceutical market. The mAbs used for clinical applications are subject to very high purity requirements and there is thus a demand for more efficient and economical purification processes. Chromatography using protein A resins results in mAbs with high yield and purity and is therefore a key step in most mAb purification processes. The mAb-containing cell supernatant is often loaded directly onto the protein A column without prepurification, causing extensive bioburden on the resin. To increase the lifetime of the protein A resin and to improve the final purity of the mAb, we have included an upstream multimodal AIEX column (WorkBeads™ 40 TREN) in flow-through mode. This procedure removed 95% of the host cell proteins and 99% of the host cell DNA. The extensive removal of host cell impurities protected the protein A resin and resulted in a mAb product of improved purity.
Monoclonal antibodies
Monoclonal antibodies (mAbs) are genetically identical antibody copies used primarily in therapy and diagnostics. The main therapy areas for mAbs include cancer, inflammatory and autoimmune diseases, and viral infections. The global mAbs market accounted for approximately USD 100 billion in 2017 with an expected continued growth over the coming years. Most of the therapeutic antibodies are produced in vivo, i.e., they are overexpressed in a biological system. This puts a high demand on the downstream process. A challenging aspect of treatment with mAb drugs is the common requirement for frequent and high doses (up to grams of mAbs per year per patient) which in turn lead to corresponding levels of drug-related impurities. Even small traces of impurities may cause big cumulative effects. The purity requirements are therefore extremely stringent.
mAb purifications
Today the most widely used purification technique is affinity chromatography using protein A resins. This technique gives high yield of antibodies with a purity of more than 95%. For therapeutic use however, downstream chromatographic polishing steps are necessary to further enhance the purity by reducing process-related impurities such as host cell proteins (HCP), host cell DNA (HCD) and viruses to the very low levels required. The polishing steps should also remove any protein A leached from the protein A resin (ligand leakage) and aggregates which can be toxic in a pharmaceutical product.
WorkBeads affimAb
WorkBeads affimAb is an alkali-stable protein A resin from Bio-Works designed for mAb purification (Table 1). This high-capacity resin allows a higher purity of eluted mAbs from cell supernatants than the market leading resin. The extraordinary high binding capacity, even at high flow rates (short residence times), allows high productivity in downstream bioprocesses.
Bioburden on protein A resins
Purification of antibodies or Fc-fusion proteins from mammalian host cells, such as Chinese Hamster Ovary (CHO), which today is the primary mAb expression system, results in extensive bioburden on the protein A column in the form of nucleic acids and protein impurities. The host cell nucleic acids together with host cell proteins in general, cause damage to the protein A resin. Regular cleaning-in-place (CIP) is mandatory in the purification process, but accumulative fouling of the packed column will still occur. Maximized lifetime of the protein A resin is thus an important requirement during the purification process development. It would be desirable to add a purification step before the protein A purification step in order to reduce the impurities that constitute the worst threat to the resin, and also in the process improve the final purity of the target antibody. Here, we apply upstream WorkBeads 40 TREN for pretreatment to protect the protein A resin from the heavy bioburden caused by the feed.Table 1. Properties of WorkBeads affimAb
WorkBeads 40 TREN
WorkBeads 40 TREN is a high salt tolerant anion exchange chromatography (AIEX) resin with a ligand based on Tris(2-aminoethyl)amine, TAEA, that is positively charged below pH 9. This ligand introduces both ionic and hydrophobic interactions giving the resin a unique selectivity. The resin is effective in removal of most host cell impurities is thus an excellent choice for pretreatment in mAb purifications.
Table 2. Properties of WorkBeads 40 TREN
When WorkBeads 40 TREN is applied upstream WorkBeads affimAb in the process it is used in flow through mode to adsorb undesirable impurities while allowing mAbs to pass through the column without binding, since the majority of mAbs are basic, and mainly positively charged at neutral or low pH. The heavy sample load of impurities on the WorkBeads affimAb column will thereby be reduced and the purity of the eluted mAb increased, since host cell nucleic acids and a major portion of the host cell proteins will be adsorbed on WorkBeads 40 TREN.
Experiments
CHO cell supernatant with overexpressed mAbs was loaded onto a WorkBeads affimAb column, see Figure 1. The UV trace shows a large amount of impurities in the flow through (Figure 1, see also highlighted area in Figure 4). To study the positive effects of WorkBeads 40 TREN as a pretreatment column, a prepacked GoBio Mini TREN 5 mL column was placed upstream of the WorkBeads affimAb column. The GoBio Mini TREN column was positioned in a separate valve on the chromatographic system to allow automatic inline or offline flow upstream of the WorkBeads affimAb column. This allows for placing of both columns inline during sample loading and bypass of the GoBio Mini TREN column during mAb elution, and finally also bypass of the WorkBeads affimAb column during regeneration of the GoBio Mini TREN column. Figure 2 shows the hardware setup.
.webp?width=526&height=532&name=f1%20(1).webp)
Figure 1. A purification run with only WorkBeads affimAb to demonstrate purification from crude sample by loading a CHO cell supernatant with overexpressed mAbs. Abs280 (blue line) and concentration (%) of elution buffer (green line) are shown in the chromatogram. Injection of sample occurs at time point 0. Cleaning-in-place (CIP) with 0.5 M NaOH occurs after the elution step to regenerate the resin, followed by a re-equilibration step.
In Figure 3 the mAb purification profile is shown at the same sample load and under the same experimental conditions as described in Figure 1. The mAbs were eluted in a single peak and a major part of sample impurities was eluted from WorkBeads 40 TREN in the regeneration steps, which consist of a salt elution step followed by cleaning-in-place (CIP) with 1 M NaOH.
Figure 3. A mAb purification where 20 mL clarified CHO cell supernatant was loaded onto WorkBeads 40 TREN + WorkBeads affimAb connected in series. Impurities in the sample will bind to WorkBeads 40 TREN. Sample loading and washing steps were performed with WorkBeads 40 TREN in-line, but during the elution and CIP steps WorkBeads 40 TREN was bypassed. Regeneration steps were run with the WorkBeads affimAb column in bypass position. Abs280 (blue line), concentration (%) of elution buffer (green line) and regeneration of WorkBeads 40 TREN (grey box) are shown in the chromatogram. Injection of sample occurs at time point 0.
Figure 2. Flow chart of the purification method in which WorkBeads 40 TREN is acting as a pretreatment resin before WorkBeads affimAb. The GoBio Mini TREN column is positioned in Valve INV-907, position 2, and WorkBeads affimAb is positioned between valves 2 and 3. The top flow chart illustrates the flow during sample loading (red arrows) and the bottom flow chart illustrates the flow during the elution of mAbs (red arrows). Other steps in the process are not shown.
Experimental conditions
To validate the advantage of WorkBeads 40 TREN as a pretreatment resin, a full experiment setup was designed including two protein A resins, WorkBeads affimAb and MabSelect SuRe™ (Cytiva), with two different sample loading conditions (low and high sample load). MabSelect SuRe resin was included in our study to compare the performance of WorkBeads affimAb with a market leading protein A resin. Thus, in total eight mAb purifications were performed, where 20 mL or 100 mL clarified CHO cell supernatant were loaded onto the two protein A columns with and without WorkBeads 40 TREN as a pretreatment resin.
In process scale a high percentage of the dynamic binding capacity (DBC) of the resin is loaded to maximize each purification run. Here, the low sample load corresponds to 15% of the DBC for WorkBeads affimAb, whereas the high sample load corresponds to 70% of the DBC for WorkBeads affimAb.
SDS-PAGE analysis of the purifications showed a significant reduction of host cell protein impurities in the mAb eluates when WorkBeads 40 TREN was included upstream. A major portion of HCP was adsorbed to WorkBeads 40 TREN and thus removed from the sample before it was loaded onto the WorkBeads affimAb column (see flow through lanes +/- WorkBeads 40 TREN in Figure 4). The same pattern was seen for the two protein A resins at both low and high sample loads. These analyses also demonstrated the non-binding of mAbs on WorkBeads 40 TREN at pH 7.4 resulting in no loss of recovery (see mAb elute fractions +/- WorkBeads 40 TREN in Figure 4).
The HCP and HCD results of the mAb purifications are shown below in Table 3. The values are presented in ppm levels (ng HCP or ng HCP/mg eluted IgG) since impurities have been shown to mainly coelute with the monoclonal antibodies (this is especially true for agarose-based beads).
Table 3. Analyses of eluates collected from eight combination runs on WorkBeads affimAb and MabSelect SuRe with or without WorkBeads 40 TREN.
Figure 4. SDS-PAGE analyses of the flow through fractions and the collected mAb eluates. M: marker, CHO: clarified CHO cell supernatant, FT: flow through, IgG: eluted mAb. The highlighted area (red box) illustrates the HCP removal by WorkBeads 40 TREN resin.
Low levels of HCP are one of the key quality attributes during downstream process purification development for biopharmaceuticals. HCP analysis using an enzyme-linked immunosorbent assay (ELISA) for measurement of CHO host cell proteins (see Table 3, third column and Figure 5) showed a low level of HCP in the mAb eluates from WorkBeads affimAb at a low sample load. This value was further reduced with WorkBeads 40 TREN employed upstream of WorkBeads affimAb. The eluted mAbs from MabSelect SuRe contained more than 10-fold higher levels of HCP (~8000 ppm) compared to WorkBeads affimAb eluates (~670 ppm). When WorkBeads 40 TREN was added upstream of MabSelect SuRe, the HCP level was reduced to a 10-fold lower level, indicating its ability to remove protein impurities.
At the high sample load (100 mL CHO cell supernatant) the same trend was seen with a 5-fold difference between MabSelect SuRe eluates and WorkBeads affimAb eluates. The absolute HCP levels were increased for all samples, with 1400–2000 ppm for mAb eluates using WorkBeads affimAb and 7300–11000 ppm for mAb eluates using MabSelect SuRe. By including an upstream WorkBeads 40 TREN, all mAb eluates contained a reduced HCP level, but the reductions were more significant at lower sample loads.
An HCP analysis of the flow through fractions of the sample with and without WorkBeads TREN demonstrated an HCP removal of 95%, resulting in a significant reduction of the bioburden on the protein A resins (Figure 5). Any remaining HCP impurities and tentative mAb aggregates can be removed by applying downstream ion exchange polishing steps to reduce the HCP level below the recommended pharmaceutical requirements.
The HCD (host cell DNA) analysis using the PiccoGreen assay also showed superior results for WorkBeads affimAb eluates compared to eluates from MabSelect SuRe, with a more than a 20-fold difference, 3.9 ppm vs. 114 ppm at low sample load (see Table 3, fourth column, and Figure 6). However, adding an upstream WorkBeads 40 TREN column decreased the DNA content to approximately 1 ppm for both eluates, demonstrating the ability of WorkBeads 40 TREN to remove host cell DNA.
At high sample load the absolute HCD levels were higher (~10 ppm for WorkBeads affimAb eluates) and the same trend was observed as for the HCP impurities, i.e., addition of a WorkBeads 40 TREN purification step reduced the HCD levels. There were approximately 5-fold higher HCD impurity levels in the eluates from MabSelect SuRe. Analyses of the different flow through fractions showed that about 99% of the HCD was removed from the sample when a WorkBeads 40 TREN column was connected upstream of the protein A column.
Figure 5. (A) HCP analyses of the collected mAb eluates. A: WorkBeads affimAb, M: MabSelect SuRe. The values are presented in ppm levels (ng HCP or ng HCP/mg eluted IgG). (B) HCP level measurements of flow through CHO cell supernatant, with and without WorkBeads 40 TREN placed upstream.
Determination of dynamic binding capacity
To obtain an estimate of the capacity of WorkBeads 40 TREN, a dynamic binding capacity measurement for impurities (DBCimp) was developed. The DBCimp for the specific CHO cell supernatant used in this study was determined by frontal analysis at 10% breakthrough of the impurities at different flow rates/residence times in a GoBio Mini TREN 1 mL column. The level of HCD was used as an indicator of breakthrough since there seems to be a clear correlation between HCP and HCD impurities bound to the resin.
A DBCimp of 15 mL CHO cell supernatant was obtained at 4 minutes residence time in a GoBio Mini TREN 1 mL column. DBCimp correlated with residence times, as expected (see Figure 7). DBCimp values may differ depending on the cell line and expression conditions used.
Figure 6. (A) HCD analyses of the collected mAb eluates. A: WorkBeads affimAb, M: MabSelect SuRe. The values are presented in ppm levels (ng HCP or ng HCP/mg eluted IgG). (B) HCD level measurements of flow through of CHO cell supernatant, with and without WorkBeads 40 TREN.
Figure 7. (A) Frontal analysis determination of dynamic binding capacity. (B) Measured DBCimp at a residence time of 1, 4 and 9 minutes in a 7 × 28 mm column (GoBio Mini TREN 1 mL).
Conclusion
WorkBeads 40 TREN is an important tool for improving process purification of monoclonal antibodies. Clarified cell supernatants can easily be run through the resin to remove a major portion of impurities such as host cell nucleic acids (DNA and RNA) and host cell proteins. Early removal of these impurities eliminates bioburden on the protein A resin and should thus extend its lifetime. Reduction of impurities early in the purification process further enhances the final purity of the product. This is essential for pharmaceuticals and diagnostic mAb-based products.
In this study, WorkBeads 40 TREN was used for pretreatment to protect the downstream protein A resin. This resulted in 99% removal of host cell DNA and 95% removal of host cell proteins in the sample load, and additionally also resulted in higher purity of the mAb eluates. WorkBeads 40 TREN can also be placed downstream of the protein A resin as a polishing column in flow-through mode to collect impurities that coelute with the mAb from the protein A column at low pH. This scheme will however not protect the protein A resin.
Due to its orthogonal functionality, regardless of where in the mAb purification process WorkBeads 40 TREN is used, it will significantly improve the purity of the final mAb. We have also showed that WorkBeads affimAb is a high-performance protein A resin that enables the achievement of extremely high purity of mAbs with respect to both HCD and HCP impurities when compared to the market leading resin MabSelect SuRe.
Figure 8. An illustration of reduction of impurities by using WorkBeads 40 TREN for pre-treatment upstream of the WorkBeads affimAb column at a heavy sample load. (A) Direct load of the CHO sample onto the protein A column, (B) Loading via pretreatment. Red: mAbs, blue: HCP, and green: HCD.
Ordering information
Visit www.bio-works.com for information regarding all WorkBeads resins.
Orders: sales@bio-works.com or contact your local distributor.
For more information about local distributors and products visit www.bio-works.com or contact us at info@bio-works.com.
bio-works.com
Bio-Works, WorkBeads and GoBio are trademarks of Bio-Works Technologies.
ÄKTA is a trademark of Cytiva. All third-party trademarks are the property of their respective owners.
© Bio-Works. All goods and services are sold subject to Bio-Works terms and conditions of sale.
Contact your local Bio-Works representative for the most current information.
Bio-Works, Virdings allé 18, 754 50 Uppsala, Sweden. For local office contact information, visit bio-works.com/contact.
AN 40 603 001 BA
