Integration of Ion Exchange Chromatography in Downstream ...
Transcript of Integration of Ion Exchange Chromatography in Downstream ...
1
1
Integration of Ion Exchange Chromatography in Downstream
Processing of Proteins
SCI Training Course, Cambridge, September 2012
Sylvio Bengio, Ph.D.Pall Life Sciences
2
DisclaimerThis presentation is the Confidential work product of Pall Corporation and no portion of this presentation may be copied, published, performed, or redistributed without the express written authority of a Pall corporate officer.
© 2012 Pall Corporation.
3
Pall Chromatography Toolbox
Lab Scale Methods & Screening Scale-Up & ProductionProcess Development
� Chromatography sorbents� Mustang XT140 and XT5000
capsules� Resolute® columns / Packing
support� Packing stations� PKP and PK systems
� Chromatography sorbents� Mustang XT140 and XT5000
capsules� Resolute® columns / Packing
support� Packing stations� PKP and PK systems
� Chromatography sorbents� Mustang XT5 capsules� Mustang XT Acrodisc� 5 mL prepacked PRC
columns� Empty laboratory LRC
glasscolumns
�AcroPrep™ and AcroWell™ filter plates,SELDI Services for sorbent screening
�1 mL prepacked PRC columns�Prepacked AcroSep™ columns�Mustang® XT Acrodisc or XT5
2
4
Defining Ion Exchange Chromatography
Ion exchange chromatography (�-än eks-ch�nj kr�-m�-tä-gr�-f�):�
A separation technique utilizing a stationary phase that contains either acidic groups for exchanging cations or basic groups for exchanging anions present in the mobile phase
under defined conditions of pH and ionic strength
“”
5
Proteins have different isoelectric points
5.6
6.99.7
9.287.49
8.309.59
7.14
4.74
4.06 6.71
6.02
5.22 8.30
Isoelectric point (pI)Calculated using pK values of amino acids as described in: Bjellqvist, et al, The focusing positions of polypeptides in immobilized pH gradients canbe predicted from their amino acid sequences.Electrophoresis 1993, 14, 1023-1031
6
Proteins have different hydrophobicities
-0.166
-0.586-0.314
-0.422-0.018
-0.261-0.876
-0.476
-0.553
-0.649 -0.197
-0.080
0.193 0.084
GRAVY (Grand Average of Hydrophobicity) scoreCalculated as the sum of hydrophobicity values of all amino acids, divided by the number of residues in the sequence.
3
7
R-NH4+R-COOH
Binds to Anion exchanger
Binds to Cation exchanger[H+] = 10-1 moles/liter
[H+] = 10-14 moles/liter
When the pH is higher than the pI, the net charge on a protein is negative
-
When the pH is lower than the pI, the net charge on a protein is positive
+
pH 1
pIIsoelectric Point (pI) Net charge is zero 00
pH 14
R-COO- R-NH3
pH = -log [H+]
8
Ion-Exchange Chromatography
� Routinely used in process-scale protein purification� Large number of products on the market� Anion and Cation exchangers� Applications : Mabs, polyclonal IgG, Recombinant
proteins, plasma derivatives, Vaccines …
9
Main Ligands Used for Ion Exchange� Anion exchange
� Strong
� Weak
� Cation exchange
� Strong
� Weak
R-CH2CH2NH+(CH2CH3)2Diethylaminoethyl (DEAE )
R-N+(CH3)3Quaternary ammonium (Q)
R-CH2COO-
Carboxymethyl (CM or C)
R-CH2SO3-
Sulfonic acid (S)
4
10
Formats: Beads or Membrane adsorbers
Sorbents Membranes
11
« Salt Tolerant » Ion Exchange Ligands
� Vast majority of proteins acidic. Will be negatively charged around neutral pH or higher, anion exchange (Q, Deae) is more commonly used than cation (S, CM) for capture steps.
� Novel sorbents and membranes allow protein capture at moderate to high conductivities (e.g. >10 mS/cm) with high binding capacities.� For capture : direct load of feedstock� For polishing (flowthrough mode)
� Usually based on primary amine ligands
12����������
Typical Steps in Ion Exchange
Equilibration Load Feedstock Wash Column
Re-equilibration Clean in Place Elution
5
13
Ion Exchange Typical ChromatogramProteins are monitored by UV absorbance at 280 nm
14
UF/DF
BioreactorBioreactorMedia Filtration
UF/DF Virus
Filtration
Membrane Chrom: DNA PolishingFinal Filtration
Media
Cell Separation & Clarification
Chromatography in Process
ChromatographyCapture
ChromatographyIntermediate
15
Microfiltration (MF)
BioreactorBioreactor
Protein A
Ultrafiltration/Diafiltration (UF/DF) Virus
Filtration
Media
Example: Typical Mab Platform Process
Cation-exchange (CEX)
Anion-exchange (AEX)
Centrifugation Depth Filtration
6
16
Bind/Elute and Flowthrough Modes
Bind and EluteBind and EluteTarget bindsTarget binds
Contaminants passContaminants pass
Target Target & &
ContaminantsContaminants
FlowthroughFlowthrough ModeModeTarget passesTarget passes
Contaminants bindContaminants bind
17
Bind/Elute or Flowthrough mode ?
� Relative concentration of target� Relative pI of target vs main impurity� Feedstock volume � Subsequent downstream process� Target stability to process conditions� Production facilities and equipment
18
Selectivity (�) : ���������������� ���������� �����������������������������������������������������������
������ �� ��������� ������������� �� ��������������� �� Expressed as : V2 – V0 with V0 : void volume of the column
� = ------------- V1 : elution volume of peak 1 V1 – V0 V2 : elution volume of peak 2
W 1 W 2
V oV o lum e
V1 V 2 V t
Resolution : ���������������������������������� ������������� ����������������������������� � ��� ���� �� ���� ����������� �� ��������� ���� ���������.
Expressed as : 2(V2 – V1) with W1 : peak 1 width at half peak height Rs = -------------- W2 : peak 2 width at half peak height W1 + W2
Selectivity and Resolution
7
19
Chemical Composition of Sorbents
Composite
Polysaccharides
AgaroseDextranCellulose
Synthetic polymersAcrylic
MethacrylatePolystyrene-DVB
Mineral oxidesSilicaCeramic
Zirconium oxideHydroxyapatite
Sepharose®, Capto™, Ultrogel®
Sephadex®
HyperCel™, Cellufine™
Spherodex®, Spherosil®
Ceramic HyperD®
HyperZ®, ZirChrom®
CHT™, HA Ultrogel®,
Trisacryl®, UNOsphere™
Toyopearl® (GigaCap), Fractogel® EMD, Macro-Prep®
Mono®, Source™, Poros®
20
Sorbent Chemistry has an impact on Selectivity
Anion exchangers:� Q rigid cellulose sorbent � Rigid Q agarose sorbent� Q polymeric sorbent
Cation exchangers: � S rigid cellulose sorbent� Rigid S agarose sorbent� S polymeric sorbent
Protein mix eluted by Linear salt gradient
21
BioreactorBioreactorMedia Filtration
UF/DF Virus
Filtration
Membrane Chrom: DNA PolishingFinal Filtration
Media
Cell Separation & Clarification
Capture Step Chromatography
Capture
ChromatographyIntermediate
8
22
Key parameters for Capture
� Selectivity for the target protein� Dynamic Binding Capacity for the target� Volume and conductivity of feedstock
� Most ion exchangers loose capacity if crude (non-diluted) feedstock is loaded: need for additional unit operations: UF/DF, dilution
� Final column volume and equipment issues
23
Enhancing Target Protein Capture by IEX
� Check selectivity for your target� Check « real-life » Capacity for your target
� Vendor’s specs are only given for guidance, « real-life »capacities tend to be lower !
� Optimize loading, wash/elution pH …� Check elution column volumes� Optimize residence time (start with 4-5 min, decrease to
2 min. if positive)
24
Prepacked Columns are convenient tools to check both Selectivity and Capacity parameters
� Packing performances guaranteed by the supplier
� Easy to connect to AKTA™ or other system
� Columns can be connected in series to mimic a bed height typical of a pilot-scale column
� Use typical flow rate of 1 mL/min (300 cm/hr)
9
25
Elution Conditions
� Elution can be achieved as usual by salt gradient. � Elution may also be done by decreasing the pH. In this
case, it allows to recover the product at lowerconductivity allowing a direct load to an orthogonal ion exchanger or simplifying formulation
26
Ion Exchange for Capture : purification of HSA from crude, undiluted plasma
� Direct capture of HSA from undiluted feedstock: Purity > 99%, Yield ~90%� Elution can be prompted by lowering pH only, no need to add NaCl, this
allows a direct orthogonal load on S HyperCel cation exchange sorbent.
27
High Throughput Sorbent Screening
� HTS methods are well suited to quickly select the best sorbent for a specific step
� Saves time, sample and manpower� Needed at early stages of process development� Use 96 well plates combined with Design of
Experiments (DoE) software tools
Second international conference: Avignon, June 4-7, 2012
10
28
High Throughput Sorbent Screening Platform (Pall)
TECAN Freedom EVO® workstation allows flexible and precise fullyautomated high throughput screening of different chromatographysorbents in AcroPrep™ 96-well filter plates.
29
Capacity of Ion Exchange sorbents� Dynamic Binding Capacity (DBC) is a key parameter in
process chromatography.� To maximize productivity during capture steps, with
feedstream having high expression titers (e.g. Monoclonal antibodies 5-10 g/L)
� DBC varies according to the sorbent type, the linear flow rate or residence time
� pH and ionic strength of the feedstock also impact DBC.� New generation of « salt tolerant » sorbents
30
Dynamic Binding Capacity vs. pH and Conductivity(BSA model protein)
High dynamic binding capacity (DBC) over a wide range of pHs and conductivities at short residence time leads to flexibility and better productivity.
Sorbent: HyperCel STAR AX (Pall) Column: 0.5 cm I.D. x 5 cm bed height (~1 mL); Sample: 5 mg/mL BSA in equilibration buffer; Equilibration buffer: 25 mM Tris-HCl, pH 7.0 – 8.5; Conductivity3 – 20 mS/cm; Residence time: 1 – 4 min (0.25 – 1 mL/min).
11
31
Ion Exchange for Capture : purification of HSA from crude, undiluted plasma
HyperCel STAR AX89% Yield
DBC 30 mg/mLElution in 50 mM Na Acetate, pH 4.0
CEX95% Yield
DBC 60 mg/mLElution in 50 mM Na Phosphate,
0.15 M NaCl, pH 7.0
Cryo Poor PlasmapH 7.6, Conductivity 11 mS/cm
Rigid DEAE Agarose73% Yield
DBC 11 mg/mLElution in 50 mM Na Acetate, pH 4.0
CEX95% Yield
DBC 60 mg/mLElution in 50 mM Na Phosphate,
0.15 M NaCl, pH 7.0
32
Ion Exchange for Capture : purification of HSA from crude, undiluted plasma
Loading undiluted plasma (11 mS/cm) on a conventional DEAE sorbent results in loss of Dynamic Binding Capacity (would need 3-6 fold dilution)
33
Process Economics AnalysisBioSolve software
� Cost of goods (US$/g) reduced by almost 70%
� Significant savings on labour cost (>50%) and water usage (>55%)
� More than 20% higher throughput (kg/year)
12
34
Ion Exchange at Capture Step: Conclusions� Check selectivity for the target first� Choose sorbent that maximizes Dynamic Binding
Capacity with minimal feed pretreatment (UF/DF, dilution)
� Consider re-equilibration duration, column volumes needed for elution and cleaning in place regimes
� Consider column final volume at production scale and equipment costs
35
UF/DF
BioreactorBioreactorMedia Filtration
UF/DF Virus
Filtration
Membrane Chrom: DNA PolishingFinal Filtration
Media
Cell Separation & Clarification
Intermediate Step
ChromatographyCapture
ChromatographyIntermediate
36
Key parameters for Intermediate step
� Selectivity for the target protein vs main contaminants� e.g. Host Cell Proteins, aggregates, misfolds….
� Consider Flowthrough mode (contaminants bind)� High resolution steps may help in some cases
� Smaller beads may also increase process costs
13
37
Ion Exchange in Flowthrough mode : Removal of contaminants in rec. Protein (CHO)
� Target in Flow through, contaminants bind with higher capacity vs conventional sorbent
� Capacity at 15 mS/cm for contaminants >2-fold higher than Q rigid agarose
� Other data (not shown) also demonstrates:Excellent target recovery (no protein loss)Good removal of protein contaminants and pigmentsEfficient regeneration with 0.1 M HClfollowed by 1 M NaOH
CHO Cell Culture Supernatant Courtesy of Cytheris, France
0
1
2
3
4
5
6
7
8
HyperCel STAR AX Q Rigid agarose
DB
C f
or c
onta
min
ant p
rote
ins
mg/
mL
15 mS/cm
38
Key parameters for Final Polishing Step
� High Resolution sometimes needed� Smaller beads may also increase process costs
� Consider Membrane adsorbers instead of packed columns � DNA and HCP removal� Disposale chromatography
Resolution in IEX is influenced by
� Buffer pH/Protein pI� Particle Size� Load� Gradient Slope� Flow Rate
14
40
Hepatitis B Virus Particle
(0.05um)
B. diminutaBacterium (0.2um)
Yeast Cell (3um)
E. coli Bacterium (0.5 X 1.5um)
Human Hair
(75um)
IgG (0.03um, 160KD)
•Albumin0.01um,
67KD
Drawn to show relative size comparison.Shapes are not accurate representations
Particle Sizes
41
50 µm100 µm
30 µm 20 µm
10 µm 5 µm Analytical:HPLC (>500 psi)
Semi-Preparative:MPLC, FPLC , Medium PressureLiquid chromatography(50 - 150 psi)
Preparative:Low Pressure Liquid chromatography (5 - 50 psi)
Bead Size and the Different Types of Chromatography
Small beads give high back pressure but great resolution, good for analytical work and demanding small scale preps
Big beads give low back pressure but less resolution, good for preparative work in large columns
Note that the selectivity is the same. All peaks elute in the same position, they only get broader as the particle size gets bigger.
Resolution and Particle size
15
43
Membrane versus Resin Chromatography
44
Benefits of Membrane Adsorbers vs Column Chromatography
� Disposable, plug and play, no time consuming equipment prep
� Less surface area but all of it is accessible to the product� Orders of magnitude faster flowrates� Use of a membrane adsorber will allow the polishing step
to be accomplished using a device of much reduced volume.
� Ideally suited for capture of large proteins, or polishing
45
Ion Exchange Mustang� Membranes
� Polyethersulphone (PES) membrane substrate
� Cross-linked polymers containing functional groups – Q, S, E
� Nodular surface - high surface area
� High voids - 0.8 µm pores
� Multiple layers (16) - controlled bed depth
Mustang Q membrane
Mustang Q XT Capsules
16
46
Mustang® Membrane Chromatography
5000 mL140 mL5 mL0.86 mL
XT5000XT140XT5XT Acrodisc®
10 mL
CLM05
60 mL
CL3
780 mL520 mL260 mL0.86 mL
NP8NP7NP6XT Acrodisc®
Single-Use Mustang Capsules : Q, S
Re-usable Mustang XT for Q chemistry
(standard) Acrodisc® Q & S : 0.18 mL not for scale-up Mustang E has different volumes : 0.12mL; 10; 40; 160; 320; 480 mL
������� ������ ��
��� ����� ������� ��
������������������ �������������������������������������� ���������
�������������������� ���� ���� �������!� "������� ��#� $
��
%�����������
&!�'���������������������!���( ������� '��������������� ����)���)%*� +'��������� � , ��))��)����-#�� ���!�'��������.�)�'� ���( ���������)��!�� �/��� ')����� �����������*'��������0�1���)���)���������'')���� ������)�����2���)�*���)����������)����/�����'!���%�)�����)����)�-
17
�3
, ��)���
%- 4.�)��������� �2��/� ����!��� ���*��'!����'�)��!��*���'����!���� �.�)�� #��**��*����-
��� �� ���� �������� �� �� ���������� ������������������������ ���
����������������� ����� ��� ��������� �����������������
%%%- 5�� �.�)��� ������*�!�� �� ������!��� ���*��'!����'-
%6- ��)��/���)��������.���)�)�������-
0�
� �� ��%*� +���� �, ��)����7 ����������
&!���������!��� ���*��'!����'�-
&( �.������')��������'�8)�( '� ������ ������������)�������-
��)��!��*"�9#���� ���� �.�)$/ &( ����!��:���( �����.����*����8
� 4�� �'�� �); � �2��/� ����!��� ���*��'!�����**��*������ �.�)-
� �������������� ������� � ����������������� ��������� ��
�� ��� �������������!�"���
CaptureProtein A
IntermediateCIEX
PolishingMEP AIEX
Pure Antibody Batch
Cell free supernatant
0+
������)��� ��!���
� ������#�
%*� +( ���2'������������� /���� � , ��))��)����"���/����!$-
� �����#
#))�!��� ���*��'!����'�( ���'������ ������*<=&#; ������ �"� 4� ��)�! ���$- 4�� �'�� �); ���������� �����*> ? � �� ��������������( �����'')�������))-
��������#
� %*� + '�����8��@�42�)����� !��� ���*��'!��5������)�9#8����*���� #����"%�.����*��$��:���������.�� 5-� 5������)� �8*������ ���� � , /� �/41%�#" �*��� &��!��)�*���$-� 6�����'����* �����8�2�����)���.��� '��.����-
18
0
��)��!��* ��� ��� �.�)���������)�9#
5������)�9#������) #�'�������!��)������!�'*/���*�"A+��'*B����$-
5���)�����������)�9#������������ %4C���'
�����*? � �� �������������� ( �������������������)�9#��� �.�)8� �������)���8 �����*#������� ? ��-+�� )���.�)�� �� %����� ���������)�8 �����*? ��'��)��+�� )���.�)�� �� �������������)�8 �����*? ��'��)��D�� )���.�)�� �
+-E*0�� )-�*B� )$����� ��!���������� %&
-FE*00�� )�-�*B� )$����� ��!���������� %�
'���������$�����!����
(��������� )* � ��$������
0F
%� ')�� �������� ���!� �����*? ���'
�9#�'����*�2'���� ����( ���'������ ����.������!�������*���9 #��#%4C- ���������( ��� �!�� �'')��� ���'������ �������-
+ ����������������������������������8
1���8 %4C�)���� �� ������������� " #$8 #�������> �����*?5�����*������8 �� 9�/�!��'!���'��-���- 9� )�9#�)�����������8 �� 9�/�!��'!���'��-��+ 9� )
������������������) * � ���,�����������
�����������������������������
D�� �B��
�� �B��B++� #G "���� $
0�� �B��
�0� �B��BD� #G "���� $
+ ������������������)* � ���,������8
1���8 0-0E* �)�&!�� ���9#��������*������ �� ������������� " #$8 #�������> �����*?5�����*������8 �� 9�/�!��'!���'��-���- 9� )�9#�)�����������8 �� 9�/�!��'!���'��-��+ 9� )
'�����,���������������- ��� �������) * � ��������
�������- �,�����������������������./�01��2
'����������������- ����������������!�����3/4 �������
5'���������� � ������2�
$�����������������������)* � ����!����������������
������2�'������������� � �- ����������������������
�����������������2
��������.���
G 6����
6 ($
)* �
)* �
��������.���
G 6����
0�0�
����)������� �)�( !������)����* �� ������������� '�)��!��* ���' ��� �!� '��������� ���)�
5�* �7�89 7:
����� �))����
�+ # �)����* /0�
�88���*)�G������������
�))5�� �.�)
)0�#�#��������!��� ���*��'!�8�������# �'����
)0&#�6����������.��������'#�������'� ������ ��� +!��'� F-0
)0;#������������������� �������
)0<#��������������� �� ������������ ������������ ���
)0/#�G)���/B�����)������� ��������)�( ��)�������8 H , F����
)0=#�6������� �.�)��������)������� "�)���.� �9$
%2&�>���������
9 $0'?���
)+ @ * 0'?���
19
0000
%*� +����.��� ����� �����*? ��)��!��*
+-+�
+-F�
+-��
�B�
$�����7�����������1�� �� !����
��+0�
0D�3
F�D��
+�+�
$����� A������7
0�1� ����
0�15��
0�15��
���������� ����
?��
+��-I+-�+-0� �B��D�� 1 �'��)�
3�-3I-�++-�� �B��+�� 1 �'��)�
3D-+I+-�+-3� �B��D�� 1 �'��)�
3�-0I+-�++-F� �B��+�� 1 �'��)�
� � ���!���
��� ������������ �9 A&B%�� (�����!��� �������� ����
�����!�( �� !�*!%*� +����.��� "J3DI$����'���������!� ���)� �'')���-
G���8 ���)���'������������#8 �� 9�/�!��'!���'� �-��������"�'�����)����)����*$8 �� 9�/�!��'!���'� �-��+ 9� ) ����������*��)���8 +8��)��������!� %4C�)���� ( ��!������#
&!� +�� ) �����*? ��'��)� ( ����)����� ��� '��������� ���!� 0�1���)������!�D�� )��'��)� ��� 0�1-
0D0D
������� ���������� ���!� � �� ����� �������� ���'���������
�����*? ��'��)� 68D�� 1
�)�( ����8FD1B!"+� 6B� ��$K
�����*? ��'��)� 68+�� 1
�)�( ����8F�� )B� ��KK"F 6B� ��$K
�����*? ��'��)� 68D�� 1
�)�( ����8FD1B!"+� 6B� ��$K
$�������������
F�1��101$������!����
0�10�10�10����
� � $'� &'� �
K 68� �� ����� .�)�� � ��������* ��'')��� �'�����������KK8��� �����!����)������� ��'������� � �2�� �� ( ������!��
�����*? D�� 1 �'��)�
0�
%� '�������85�� �.�)���������)�9#
#���� %4C�!��9 #�������( ���)��������!�)�( �* ���*�-&!�����*���� #�������)�����'')���������)������#%4C��� ')������� ')��( ������)�@���2�����))�����*�:��)�����:� 5 � ��!��"������������.���A+'*$-
5���)�������� �))�9#��')�����"����*���� #����$
F-�-F-��-#�������!��� ���*��'!� ���' " %4C$
�--D-D+3#����+���!��� ���*��'!� "�������#$
�B��0F��B�030+ �))��)���� ��'��������
�������� *B� )
�9#�*B� )
�������� *B� )
�9#�*B� )
������� ���'
)�!������� ?���&���%�7�0����
)�!������� ?�������%�7�0����
A+D'*B� )KA+�'*B� )KA��'*B� )K����)��)�
� � $�C &/%�7'� C /%�7'� " &/%�70�����1$����� ����
5���)�������� �))�9#���������� ������:� 5
K8��)�( )�( �� )�� �� ��:������������� ���!� ����� "11, ? $
&!�� �� ������������� ���'�!�( ���������.���� �.�)���������)�9#��'������������)�-
20
0�
���)������������ � ���
#� �� ������������� ���'( �����������))��� ')�� ����������*)�/���'�)��!��*���'������ �.�)���������)�9#���%*� +'�����������'��������'������������)�-
��)���������!�� �� ������������� ���� ��"D�� 1 �����*? ��'��)����0�1��)����$����������������������������"'�������)���)�*���).�����������!������������� '�������$-
4������.� �16 .������� �.�)����* �����*? �������� ( ���!�( �������)�/��( �� ���)( !�)� 6 .������� �.�)����������!���'��� �@�����")�����������.���$-
, .���)).���� �)������� ��'����� ( ���������������� �����������*����������*�������������)�����)����)�-
59
Thank You ! [email protected]