Application Note
USD 2870
Contaminant Removal by Anion Exchange on
Salt Tolerant HyperCel™ STAR AX Sorbent
Prior to Downstream Purification of
Recombinant Interleukin 7
1.
2.
Summary
u
This study describes the use of HyperCel STAR AX “salt tolerant” anion exchange sorbent to remove
protein contaminants prior to other chromatography steps in the purification process of recombinant
human interleukin 7 (rhIL-7) in Chinese Hamster Ovary (CHO) cell culture supernatant (CCS).
u
The performance of HyperCel STAR AX sorbent was compared with another commercial anion
exchanger.
u
The ability to capture proteins at different sample conductivities using different feedstock dilutions was
measured for both sorbents.
u
A process economics analysis was performed to evaluate the benefit of the elimination of feedstock
dilution at process scale.
Introduction
“Negative mode” or “Flowthrough (FT) Mode” chromatography is commonly used to remove contaminant
proteins from high expression cell culture supernatants. In this approach, contaminants are retained while
the target protein is unbound.
Conventional anion exchangers can be used to selectively separate contaminants from the target protein
but they require CCS dilution to lower ionic strength or diafiltration to achieve sufficient capacity. These
units of operation, however, negatively impact process economics by increasing buffer consumption and
processing time and by limiting throughput.
HyperCel STAR AX “salt tolerant” anion exchange sorbent was used to allow the direct removal of
contaminant proteins from undiluted feed while rhIL-7 was not captured. Data shown in this study
evaluates the process economics impact of using “salt tolerant” ion exchange instead of conventional
anion exchange.
Table 1
Properties of HyperCel STAR AX Sorbent
Average particle size
Ion exchange ligand
Dynamic binding capacity1 at conductivity 15 mS/cm
Recommended operating range of feedstock conductivity
Recommended cleaning conditions2
1
2
3.
80 μm
Primary amine
>100 mg BSA/mL within pH range 7.5 – 8.0
2 – 15 mS/cm
1 M NaOH
Determined using 5 mg/mL BSA in 25 mM Tris-HCl , 0.14 M NaCl at 2 minute residence time.
Injection of 5 column volumes (CV) of 0.5 – 1 M NaOH, 1 hour contact time.
Materials and Methods
3.1.
CHO Feedstock and Analytical Methods (CHO CCS kindly provided by Cytheris SA, France)
The rhIL-7 concentration in the feedstock was estimated at 0.1 mg/mL using hIL-7 ELISA assay
(Cell Sciences). The total protein concentration was 0.39 mg/mL using a Bradford protein assay
(Pierce Thermo Scientific).
The CHO supernatant was used undiluted as well as 2- and 4-times diluted by double distilled
water. The feedstock was filtered through a Pall 0.2 μm Supor® 200 sterilizing filter prior to
application on chromatography sorbent.
2
3.2.
Chromatography Runs
All the chromatography runs were performed using ÄKTAexploreru 100 system (GE Healthcare).
HyperCel STAR AX and rigid Q agarose sorbents used in the study were packed in 1 mL columns
(0.5 x 250 mm, YMC Europe GmbH).
The chromatography protocol used is summarized in Table 2. Prior to use, each sorbent was
washed with 20 column volumes (CV) of buffer B. A total protein load of 40 mg per mL of sorbent
was applied on each sorbent. This corresponded to:
– 100 mL of undiluted filtered CHO CCS
– 200 mL of 2-fold diluted (1 volume of CCS + 1 volume of water for injection) filtered CHO CCS
– 400 mL of 4-fold diluted (1 volume of CCS + 3 volumes of water for injection) filtered CHO CCS.
Flowthrough fractions were collected and analyzed for total protein and rhIL-7 content to determine
the dynamic binding capacity (DBC) for contaminant proteins and the yield of recovery of rhIL-7.
Table 2
Chromatography Run Sequence (1 minute Residence Time)
Step
Equilibration
Load
Wash
Elution
Strip
CIP
1
3.3.
CV
20
(Not applicable)
10
20
20
20
Solution Used
Buffer A1
Equivalent to 40 mg of protein per mL of sorbent
Buffer A1
Step elution Buffer B1
0.1 M HCl
1 M NaOH
Buffer A: 25 mM Tris-HCl, pH 8; Buffer B: 25 mM Tris-HCl, 1 M NaCl, pH 8.
Process Economics Analysis
The process economics analysis was conducted using the BioSolve Cost of Goods (CoG) analysis
software from Biopharm Services Ltd.
Three different dilutions of the CCS were considered for the process economics analysis (see
Figure 1):
– Scenario 1: No feedstock dilution - Conductivity 16 mS/cm.
– Scenario 2: 2-fold dilution - Conductivity 8 mS/cm.
– Scenario 3: 4-fold dilution - Conductivity 4 mS/cm.
Each sorbent was tested using the above conditions. Process economics analysis was conducted
using scenario 1 (no dilution) only for HyperCel STAR AX sorbent.
The evaluation was performed using data obtained at the laboratory scale and transferred to
process scale, taking into account that 500 L batches and 10 batches are produced per year.
An ultrafiltration/diafiltration (UF/DF) step was added in the analysis, after the first chromatography
step, to mimic a process where the second chromatography step would require a low conductivity.
The analysis showed that UF/DF was not significantly impacting the whole process cost and is not
discussed further here (data not shown).
www.pall.com/biopharm
3
Figure 1
Process Scenarios for Process Economics Analysis
AEX Chromatography
Scenario 1: No Dilution
Initial Feedstock
Volume (L)
500
Total protein content
in feedstock (g/L)
0.4
rhIL-7 (g/L)
0.2
Conductivity (mS/cm)
16
Scenario 2: Dilution X2
VCF
Conductivity (mS/cm)
16
DV
2
Average flux (LMH)
75
500
Volume (L)
1000
Total protein content
in feedstock (g/L)
0.4
Total protein content
in feedstock (g/L)
0.20
rhIL-7 (g/L)
0.2
rhIL-7 (g/L)
0.1
Conductivity (mS/cm)
16
Conductivity (mS/cm)
8
8.5
Process time (H)
4
Rigid Q agarose
3
Initial volume (L)
500
Final volume (L)
500
Conductivity (mS/cm)
4
Ultrafiltration
Volume (L)
1000
VCF
Conductivity (mS/cm)
8
DV
1
Average flux (LMH)
85
DBC for contaminant proteins(1):
Rigid Q agarose
6
AEX Chromatography
Dilution 4x in DI Water
1
HyperCel STAR AX
AEX Chromatography
Dilution 2x in DI Water
Volume (L)
Initial Feedstock
Scenario 3: Dilution X4
500
DBC for contaminant proteins(1):
Initial Feedstock
2
Process time (H)
4
Initial volume (L)
1000
Final volume (L)
500
Conductivity (mS/cm)
4
Ultrafiltration
Volume (L)
500
Volume (L)
2000
Volume (L)
2000
VCF
Total protein content
in feedstock (g/L)
0.4
Total protein content
in feedstock (g/L)
0.10
Conductivity (mS/cm)
4
DV
0
Average flux (LMH)
100
rhIL-7 (g/L)
0.2
rhIL-7 (g/L)
0.05
Conductivity (mS/cm)
16
Conductivity (mS/cm)
4
DBC for contaminant proteins(1):
VCF: Volumetric Concentration Factor; DV: Diafiltration Volume
4.
Ultrafiltration
Volume (L)
Rigid Q agarose
(1)
11
4
Process time (H)
4
Initial volume (L)
2000
Final volume (L)
500
Conductivity (mS/cm)
4
DBC in mg/mL
Results and Discussion
4.1.
Impact of Feedstock Dilution on Dynamic Binding Capacity for Contaminant Proteins
and Yield of Recovery for rhIL-7
A breakthrough (BT) curve, using HyperCel STAR AX or rigid Q agarose sorbents, was determined
for the different feedstock dilutions using the Bradford protein assay.
After the first BT (unbound proteins flowing through the column), a second BT showed the sorbent
saturation and this point was used to determine the DBC for contaminant proteins. The DBC was
determined for the three samples used (no dilution, conductivity 15.8 mS/cm; 2-fold dilution,
8.4 mS/cm and 4-fold dilution, 4.2 mS/cm).
The results are presented in Figure 2. The DBC for contaminant proteins when using the crude
feedstock was 8.5 mg/mL on HyperCel STAR AX sorbent, while it was below 3 mg/mL on rigid
Q agarose sorbent. At the lower conductivity, the DBC of HyperCel STAR AX sorbent was
decreased, this behavior being specific to this sorbent. In contrast, the DBC of rigid Q agarose
was increased to 6 and 10 mg/mL respectively when conductivity was decreased to 8.4 (2-fold
dilution) and 4.2 mS/cm (4-fold dilution).
Therefore, a 4-fold dilution is required on the standard ion exchange sorbent to achieve a DBC
similar to that of HyperCel STAR AX sorbent without dilution. No significant difference in the yield
of recovery for rhIL-7 (>90%) was observed depending on sorbent or dilution.
4
Figure 2
DBC for Contaminant Proteins vs. CCS Dilution on HyperCel STAR AX and Rigid Q Agarose Sorbents.
10
10
8
6
8
6
4
4
2
2
0
0
15.8
8.4
4.2
Conductivity (mS/cm)
15.8
8.4
4.2
Conductivity (mS/cm)
Process Economics Analysis
The annual CoG was first used to obtain an overview of the whole process cost (Figure 3). Then,
other detailed information such as capital, labor and consumable cost was considered. Finally,
the usage of water was estimated to evaluate the impact of dilutions at process scale.
When compared to the rigid Q agarose sorbent, HyperCel STAR AX sorbent constantly provided
significant cost reduction and water saving, whatever the chosen scenario. The largest savings on
CoG (23%) was obtained when comparing the use of rigid Q agarose sorbent in Scenario 1 (no
dilution) to the use of HyperCel STAR AX sorbent.
Figure 3
Cost of Goods (CoG) Scenarios on Conventional vs. Salt Tolerant Anion Exchange
Annual CoG
(US$)
Capital
Labor
Consumables
Water Usage
(m3 per batch)
0
Savings with HyperCel STAR AX Sorbent (%)
4.2.
Rigid Q Agarose
12
DBC (mg/mL)
DBC (mg/mL)
HyperCel STAR AX
12
-10
-20
-30
-40
-50
Scenario 1 (no dilution - 16 mS/cm)
Scenario 2 (2-fold dilution - 8 mS/cm)
Scenario 3 (4-fold dilution - 4 mS/cm)
-60
The resulting savings obtained when using HyperCel STAR AX sorbent were still significant even for
Scenarios 2 (2-fold dilution) and 3 (4-fold dilution).
u
Labor: The labor cost difference between scenarios using rigid Q agarose and HyperCel STAR AX
sorbents was not significant.
u
Capital Cost: The savings on CoG using HyperCel STAR AX sorbent came mostly from savings
on capital equipment (25% savings with HyperCel STAR AX sorbent in Scenario 1). In Scenario 1,
the lower capacity for contaminant proteins on rigid Q agarose sorbent led to the use of a larger
column diameter (120 cm vs. 44 cm for HyperCel STAR AX sorbent) and a subsequently larger
www.pall.com/biopharm
5
pumping system needed for the chromatography skid (28 L/min vs. 4 L/min for HyperCel STAR
AX sorbent). Larger size vessels were also required for buffer preparation. In Scenarios 2 and 3,
the difference in capital cost was reduced to 11 and 7% respectively, as the capacity for
contaminant proteins was increased. The column diameter, still slightly higher (63 cm vs. 44 cm
for HyperCel STAR AX sorbent), required for Scenario 2 and the cost of the mixing equipment
required for the dilution step in Scenario 3 mainly made the difference in capital cost.
u
5.
Consumable Costs: They were also significantly reduced when using HyperCel STAR AX
sorbent and the difference remained stable through the various scenarios (around 25%). The
consumable cost was higher for Scenario 1 using rigid Q agarose (more resin required due to
lower capacity). As the capacity increased in Scenarios 2 and 3, there was no difference any
more in resin cost compared to the process using HyperCel STAR AX sorbent, but the consumables required for the dilution step made the difference in total consumable cost.
Conclusion
u
HyperCel STAR AX “salt-tolerant” anion exchange sorbent can capture contaminant proteins from
crude CHO feedstock early in the downstream process and contributes to improve the efficiency of
the further purification of rhIL-7.
u
A conventional rigid Q agarose sorbent used as a comparison required at least a 4-fold dilution of the
feedstock to obtain similar performance.
u
The process economics analysis showed that the elimination of the dilution with the use of HyperCel
STAR AX sorbent provided strong economical benefits (reduced capital and consumable cost,
reduced water usage) compared to a standard ion exchange sorbent.
Visit us on the Web at www.pall.com/biopharm
E-mail us at biopharm@pall.com
Corporate Headquarters
Port Washington, NY, USA
+1 800 717 7255 toll free (USA)
+1 516 484 5400 phone
biopharm@pall.com e-mail
European Headquarters
Fribourg, Switzerland
+41 (0)26 350 53 00 phone
LifeSciences.EU@pall.com e-mail
Asia-Pacific Headquarters
Singapore
+65 6389 6500 phone
sgcustomerservice@pall.com e-mail
International Offices
Pall Corporation has offices and plants throughout the world in locations such as:
Argentina, Australia, Austria, Belgium, Brazil, Canada, China, France, Germany,
India, Indonesia, Ireland, Italy, Japan, Korea, Malaysia, Mexico, the Netherlands,
New Zealand, Norway, Poland, Puerto Rico, Russia, Singapore, South Africa, Spain,
Sweden, Switzerland, Taiwan, Thailand, the United Kingdom, the United States, and
Venezuela. Distributors in all major industrial areas of the world. To locate the Pall
office or distributor nearest you, visit www.pall.com/contact.
The information provided in this literature was reviewed for accuracy at the time of
publication. Product data may be subject to change without notice. For current
information consult your local Pall distributor or contact Pall directly.
© 2012, Pall Corporation. Pall,
, HyperCel and Supor are trademarks of Pall
Corporation. ÄKTAexplorer is a trademark of GE Healthcare.
Filtration.Separation.Solution is a service mark of Pall Corporation. ® indicates a
trademark registered in the USA and TM indicates a common law trademark.
1/13, PDF, GN12.8112
USD 2870