Sartobind Membrane Chromatography. Dr. Stefan Fischer-Frühholz, February 15, 2012 Senior Product Manager Membrane Chromatography

Transcription

1 Sartobind Membrane Chromatography Dr. Stefan Fischer-Frühholz, February 15, 2012 Senior Product Manager Membrane Chromatography

2 Agenda 1. Membranes 2. Formats 3. Applications February 15, 2012 Page 2

3 Sartobind Membrane Adsorbers are formatted already for use Chromatographic adsorptive membranes Made from regenerated cellulose Surface modified (e.g. primary amine IEX) Grafted (e.g. quaternary ammonium IEX) Conventional ligands OH CH 2 O OH * n * O CH 2 O O CH 2 O OH O OH OH O OH OH OH * n * OH CH 2 OH OH hydrophilic X-linker OH base stable cellulose (Hydrosart ) February 15, 2012 Page 3

4 Diffusion limited gels (time) versus convection limited (flow rate) Average pore size nm Average pore size 3-5 μm February 15, 2012 Page 4

5 Dynamic binding capacity./. Flow rate -> Contaminant removal Dynamic binding capacity Membrane chromatography Gel chromatography Flow rate 40 x Bed volume/min February 15, 2012 Page 5

6 Dynamic binding capacity./. Size -> Capturing large molecules Size exclusion for proteins >100 kda when using gel matrix of microporous nm* Dyn. binding capacity Membrane chromatography Gel chromatography kda Size of molecule *E. Karlsson, L. Rydén and J. Brewer, Protein Purification, Principles, Jan-Christer Janson, Lars Rydén Eds., VCH Weinheim, pp 123, 1989 February 15, 2012 Page 6

7 Membranes Ion exchange Anion exchange Anion exchange HIC Metal chelate Strong: Q, S Weak: D Sartobind STIC PA (Primary Amine) Salt Tolerant Interaction Chromatography Phenyl Iminodiacetic acid (IDA) Coupling Affinity... Aldehyde Protein A February 15, 2012 Page 7

8 Sartobind STIC Sartobind STIC: Salt Tolerant Interaction Chromatography Combination of unique cellulose-based membrane structure and covalently attached salt tolerant primary amine (PA) ion exchange ligand Primary amine (Sartobind STIC PA) February 15, 2012 Page 8

9 Sartobind STIC - Next Generation of Membrane Adsorbers Shares same cellulose base membrane as Q: >3 μm pore size 0.5 μm Here is the binding capacity Sartobind Q Hydrogel grafted Quaternary ammonium Sartobind STIC Direct derivatisation Binding + ligand density capacity is distributed + pore accessibility more evenly Primary amine (Sartobind STIC PA) I. Tatárova, I., R. Fáber, R. Denoyel, M. Polakovic, J. Chromatography A 1216 (2009) 941 February 15, 2012 Page 9

10 Sartobind STIC PA and Q dynamic binding capacity at 10% 10% DBC BSA [mg/ml] Sartobind Q Sartobind STIC PA Salt conc. [mm] February 15, 2012 Page 10

11 Typical binding capacities Membrane Description Dynamic binding capacity 10 %* Quaternary ammonium (Q) Strong basic anion exchanger 29 mg/ml (0.8 mg/cm²) Sartobind STIC primary amine (PA) Weak basic anion exchanger 50 mg/ml (1.4 mg/cm²) in TRIS buffer+150 mm NaCl Sulfonic Acid (S) Strong acidic cation exchanger 26 mg/ml (0.7 mg/cm²) Phenyl Hydrophobic Interaction Chromatographic membrane *Standard proteins: BSA (Sartobind Q and STIC, ) 20 mm TRIS/HCl ph 7.5 lysozyme (S) 10 mm KPi ph 7, HIC: globulin, 0.9 M (NH 4 ) 2 SO 4 Membrane area: 36.4 cm² = 1 ml volume 14.6 mg/ml (0.4 mg/cm²) February 15, 2012 Page 11

12 Selection Guide Sartobind Polishing Q,S STIC HIC Low salt High salt Contaminant removal: Viruses, DNA, Host cell proteins, endotoxins, aggregates Membrane Adsorber step Capture Q Purification: large proteins (Factor VIII), viruses (vaccines), phages... February 15, 2012 Page 12

13 Agenda 1. Membranes 2. Formats 3. Applications February 15, 2012 Page 13

14 Flow rate: 0.5 bed volumes/min for the column 30 bed volumes/min for Sartobind Column 20 l 10 l / min 20 l Membrane Adsorber 0.5 l (~2 m²) 14 l/min 0.5 l 0.4 cm 15 cm results in a flow rate of 0.5 BV / min results in a flow rate of 28 BV /min February 15, 2012 Page 14

15 Need for Throughput Results in Oversizing Columns Bottleneck at production scale = flow rate Membrane based system Beads based system -0.5 L bedvolume l/h - BC 15 g sufficient for contaminant removal - 50 cm / 30L bed volume l/h - BC 1500 g oversized H.L. Knudsen et al., J. Chromatogr. A 907, , 2001 February 15, 2012 Page 15

16 Sartobind membrane adsorber portfolio process scale 4 mm bed height 8 mm bed height Q S STIC Q S HIC Size pico nano mini mega nano 150 ml Jumbo ml ml ml Contaminant removal: flowthrough mode to remove DNA, Host cell proteins, endotoxins, viruses Singe-use Purification: bind & elute of viruses and virus like particles, large proteins Single-use / intra-batch use / re-use February 15, 2012 Page 16

17 Sartobind 4 mm capsules for flowthrough operation with new size Sartobind pico 0.08 ml membrane volume Size ml pico 0.08 nano 1 mini Mega 1620 Q S STIC HIC * *coming soon nano 3 ml 150 ml 150 Jumbo 5 l 5000 February 15, 2012 Page 17

18 Sartobind STIC PA, Q and S for process development Bed height 4 mm Frontal surface area 0.19 cm² Max pressure 6 bar Housing material Polypropylene Size 31 x 11 mm Void volume 0.4 ml *Phenyl coming soon February 15, 2012 Page 18

19 Sartobind STIC PA pico for application with limited sample material and virus spiking studies - Smallest scaleable membrane adsorber with 4 mm bed height - Decreases cost for virus spiking studies - Scales in binding capacity for protein, DNA and host cell proteins as well as endotoxins. Shows same log reduction for model virus as larger capsules and throughput / pressure flow relation 31 mm February 15, 2012 Page 19

20 The Sartobind pico construction realizes the smallest scaleable design with 4 mm bed height (15 layers) of membranes Upstream Downstream pico 5 February 15, 2012 Page 20

21 Connectivity of Sartobind pico is given by finger-tight Luer PEEK adapters (enclosed in package) February 15, 2012 Page 21

22 Sartobind Q, S and STIC 10 capsule cut-away Vent valve Sartobind membrane Sartobind membrane Fleece Fleece February 15, 2012 Page 22

23 Scale-up performance 1 0,8 0,6 nano lot A No.13 nano lot A No.35 5" lot B No.36 5" lot B No.30 c/c o 0,4 Flushing/equilibration: 10 mm phosphate buffer, ph 7 Loading: 2 g/l bovine serum albumin Washing: 10 mm phosphate buffer, ph 7 Elution 1 M NaCl in equlibration buffer MV = membrane volume (corrected by minus void volume, diff. membrane lots c/c o 0, ''_lot_504863_22 10 C No.22 MV 10''_lot_504863_07 10 C No.07 0,8 20''_lot_504663_03 20 D No.03 20''_lot_504663_10 20 D No.10 30''_lot_504563_07 30 E No.07 0,6 30''_lot_504563_04 30 E No.04 0,4 0, MV February 15, 2012 Page 23

24 Sartobind Q Pico scales to larger devices February 15, 2012 Page 24

25 Testing of integer and pinholed 10 capsule by: Diffusion test and HETP -> only diffusion testing works 2 mm Defect Capsule Result HETP Result Diffusion test None PASS 1.2 ml/min None PASS 0.4 ml/min 0,6 0,5 Comparison HETP with 5% acetone solution 10" Sartobind Q SingleSep with and without hole (diameter 2mm) 2 mm pinhole FAIL No value Extinction 280nm 0,4 0,3 0,2 0,1 2 mm pinhole FAIL No value , , ,5 V [l] capsule 1 with hole capsule 2 with hole capsule 1 without hole capsule 2 without hole February 15, 2012 Page 25

26 Screening device 96-well plate equipped with Sartobind STIC PA (coming soon) Build up from 12 x 8-strip well units Screening: 0.21 cm²/well per layer (3 layers: 0.7 cm² ml) Operation volumes up to 0.5 ml Fractionation using 96 well deep well plates Allows the implementation of all typical steps in chromatography February 15, 2012 Page 26

27 Comparison Sartobind Q and STIC PA Effect of citrate buffer and salt on protein binding 2.7 mg/cm² BSA were added in 2 steps of 0.5 ml, analyzing flow-through fraction Polyanionic buffers such as sodium citrate prevents binding than salt concentration less binding less binding February 15, 2012 Page 27

28 Comparison of Sartobind Q and STIC PA Effect of phosphate and salt on protein and DNA binding Evaluation ph 8 Contour plots: - Red: no binding - Green: high binding Sartobind Q: - Sensitive to [NaCl] Sartobind STIC: - Protein (GFP) is sensitive to [PO 4 ]phosphate - DNA is less impacted by phosphate buffers less binding less binding less binding February 15, 2012 Page 28

29 Sartobind 8 mm capsules for flowthrough and bind & elute applications Size ml pico 0.08 nano 1 mini Mega 1620 Q S STIC HIC * *coming soon nano 3 ml 150 ml 150 Jumbo 5 l 5000 February 15, 2012 Page 29

30 Jumbo void volume optimisation downscale nano 3 ml Sartobind nano 3 ml for bind & elute AND flow through Sartobind nano 1 ml for polishing flow through only Top view External channel Membrane (8 mm) Central core Internal channel Upstream flow channel Void volume optimized No optimization February 15, 2012 Page 30

31 Computational Fluid Dynamics: Jumbo 5 l void volume optimization upstream downstream also published in: Pastor, A. and Barbe, S. (2010) Disposable Chromatography for Large-Scale Biomanufacturing, in Single-Use Technology in Biopharmaceutical Manufacture (eds R. Eibl and D. Eibl), John Wiley & Sons, Inc., Hoboken, NJ, USA. ch25. February 15, 2012 Page 31

32 Comparison of void volumes 4 mm, 70 ml and 8 mm, 150 ml capsules Upstream Membrane Downstream 70 ml Membrane volumes 150 ml 4.6 Void membrane volumes 1.3 February 15, 2012 Page 32

33 Breakthrough curve Sartobind nano to Jumbo Flow rate 4 bed volumes/min, sample BSA (2 g/l), elution 1 M NaCl in KPI 3 2,5 2 c [g/l] 1,5 1 0, BV 3 ml Sartobind Q nano 3 ml 150 ml Sartobind Q 150 ml 5000 ml Sartobind Q Jumbo 5 L February 15, 2012 Page 33

34 Pressure flow relation Sartobind Jumbo 5 l 40,00 35,00 30,00 25,00 flow [l/min] 20,00 15,00 10,00 5,00 0,00 0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,80 0,90 1,00 1,10 1,20 1,30 1,40 1,50 1,60 1,70 1,80 1,90 2,00 2,10 2,20 2,30 2,40 2,50 pressure difference [bar] February 15, 2012 Page 34

35 2 g/l BSA break through normalized to bed volume (BV) at 4 BV/min flow Breakthrough curve (BSA) and Elution peaks with Sartobind Q Jumbo and Sartobind Q 4 Three Sartobind Q Jumbo 5 liter lots BV/min vs. 3 of Sartobind Q nano 3 ml 3 2,5 c [g/l] 2 1, _Jumbo5L _Jumbo5L _Jumbo5L _1_nano _1_nano _1_nano 0, BV February 15, 2012 Page 35

36 The Jumbo and nano are suitable for bind elute applications Sartobind S nano 3 ml nano 3 ml Pre-conditioning 2 M NaCl in equilibration buffer 20 CV Equilibration 20 mm Tris ph 7.4, 1.8mS/cm 25 CV Loading of protein mixture ~ mg/ml for protein 1,2 and μl Wash 20 mm Tris ph 7.4, 1.8 ms/cm 4 CV Elution by linear gradient 2 M NaCl in equilibration buffer 16 CV February 15, 2012 Page 36

37 Agenda 1. Membranes 2. Formats 3. Applications - mab productioncontaminant removal - Virus purification February 15, 2012 Page 37

38 Why use Membrane Adsorber downstream in mab production major drivers No validation Saves 95 % buffer 10 times faster Effective removal of DNA, viruses, Host Cell Proteins February 15, 2012 Page 40

39 Gel versus Membrane when polishing of a monoclonal antibody Q Resin Q Membrane Flux / flow rate cm/h cm/h Capacity g/l > 3 kg/m² or > 10.9 kg/l Buffer used 100 % 5 % Operation time 8 9 h h Cleaning validation Yes Single use Zhou J X, Tressel T, AMGEN Inc.: Membrane chromatography as a robust purification system for large-scale antibody production. Bioprocess International, Sept. 2005, p February 15, 2012 Page 41

40 15 Reasons why use disposables: Product faster to market Reduction of complexity Reduce space No cleaning validation Decrease product contamination More flexible Simplify operations Reduce hazardous cleaning solutions More convenient Increase facility output Assure sterility Reduce water and buffer consumption Lower maintanance cost Reduce time to get facility running Faster optimization February 15, 2012 Page 42

41 Flowthrough: Contaminant removal in Downstream Processing Purification Cell-Harvest/Clarification Affinity Chrom. Capturing Step 2 nd column Sartobind Disposable UF Buffer Exchange 0.2μm Sterile Virus Filter Form and Fill Viruses (AEX) DNA (AEX) Host cell proteins (AEX, CEX) Endotoxins (AEX) Leached ligand (AEX) Aggregates (HIC) February 15, 2012 Page 43

42 Removal of contaminants during monoclonal antibody production Positvely (+) charged adsorber binds negatively (-) charged contaminants Endotoxins Isoelectric points Product Impurities Host cell proteins DNA Viruses Polishing: contaminants are bound - mabs ph buffer + Flow Through February 15, 2012 Page 44

43 Anion exchange step in flowthrough polishing 2 nd step: Sartobind Q may be sufficient, mab dependent both are options Q or Sb.STIC Bioreactor Centrifuge /Dept filter Protein A AEX step Membrane CEX Virus filter Cross Flow Production Clarification Capturing FT Polishing Capturing Virus removal UF / DF Reduces: Viruses DNA HCP Leached ligand Endotoxins February 15, 2012 Page 45

44 AEX as 2 nd chromatography step: higher contaminant level Harvest 1st Protein A 2nd Anion exchange membrane adsorber step: higher impurity load (limits load density) Low conductivity 4-7 ms/cm allows Q usage 3 rd Cation exchange February 15, 2012 Page 46

45 Anion exchange step in flowthrough polishing 3 rd step: Position for Sartobind STIC Sb. STIC or Q? Bioreactor Centrifuge /Dept filter Protein A CEX AEX step Membrane Virus filter Cross Flow Production Clarification Capturing Capturing FT Polishing Virus removal UF / DF Reduces: Viruses February 15, 2012 Page 47 DNA HCP Leached ligand Endotoxins

46 Purification bottleneck Facility / Tank size limitation at high mab titers: 5 g/l mab Liters 200 L Protein A 30 g/l, 8 cycles, 2.5 CV Pool 4000 Liters 500 L CEX 100 g/l, 6 CV Pool, ms/cm 3000 Liters Q needs 4-7 ms/cm, dilution 1: Liters AEX 10 kg/l FT February 15, 2012 Page 48

47 Purification bottleneck Facility / Tank size limitation at high titers, e.g.: 5 g/l mab Liters 200 L Protein A 30 g/l, 8 cycles, 2.5 CV Pool 4000 Liters 500 L CEX 100 g/l, 6 CV Pool, ms/cm 3000 Liters Sartobind STIC AEX 10 kg/l FT no need for 6000 Liter tank February 15, 2012 Page 49

48 Anion exchange step in flowthrough polishing 3 rd step: Position for Sartobind STIC Sb. STIC Bioreactor Centrifuge /Dept filter Protein A CEX AEX step Membrane Virus filter Cross Flow Production Clarification Capturing Capturing FT Polishing Virus removal UF / DF Reduces: Viruses February 15, 2012 Page 50 DNA HCP Leached ligand Endotoxins

49 Impurity levels after 1 st column and g Mab product : 1.5% aggregate 180 g : 1000 ppm CHOP 12 g : 200 ppm DNA 2.4 g : 30 ppm leached ProA 0.36 g Typical load value: > 2 kg protein per liter Sartobind Q After 1 st column Gerardo Zapata et al., Abgenix/AMGEN, Membrane Chromatography for Purification of Human Antibodies in Commercial Processes, IBC Antibody Meeting 2005 February 15, 2012 Page 51

50 Polishing Step: hmab after capture Virus Size (nm) Enveloped LRV Run 1 LRV Run 2 MVM: Minute Virus of Mice No 6.77 ± ± 0.37 Reo-3: Reovirus Type III No 7.28 ± ± 0.29 MuLV: Murine Leukemia Virus Yes 5.57 ± ± 0.32 PRV: Pseudorabies virus Yes 5.67 ± ± 0.23 (log 10 ) removal for DNA : >2 log host cell proteins: 1.9 log endotoxin: >2.8 log Zhang R, Bouamama T, Tabur P, Zapata G, AMGEN, Gottschalk U, Mora J, Reif O. Viral Clearance Feasibility Study With Sartobind Q Membrane Adsorber for Human Antibody Purification. IBC s 3rd European Event BioProduction February 15, 2012 Page 52

51 Polishing Step after Intermediate Purification After 2 nd column Impurity Levels after 2nd column: : << 100 ppm CHO protein : << 20 ppm DNA Typical load value: 10.9 kg antibody/l membrane (180 times more than Q resin) Feng Li et al., Amgen, J. BioProcessing, 09/10, 23-30, 2005 February 15, 2012 Page 53

52 Virus removal study for MAb Virus Enveloped LRV Run 1 LRV Run 2 Virus recovery (%) MVM: Minute Virus of Mice no 6.03 ± ± Reo-3: Reovirus Type III no 7.00 ± ± MuLV*: Murine Leukemia Virus yes 5.35 ± ± 0.27 > 70 PRV: Pseudorabies virus yes 5.58 ± ± ph: 7.2 / conductivity: 4 ms/cm / flow rate: 450 cm/h / 1 % spike protein concentration: 4.3 g/l Mab load: 10.9 kg/l membrane (3 kg/m²) / *LRV = 5.59 at 600 cm/hr Zhou J and Tressel T, Amgen: Basic Concepts in Q Membrane Chromatography for Large-Scale Antibody Production. Biotechnol Prog., 22 (2) , February 15, 2012 Page 54

53 HCP binding on Sartobind Q vs. STIC PA at different salt concentrations: Salt tolerant STIC less dependent on salt->large window of operation Binding of HCP Sartobind Q Sartobind Q Sartobind STIC PA Binding of HCP Sartobind STIC PA B relative to the max [%] ,0 6,5 7,0 7,5 8,0 8,5 9,0 ph 0mM NaCl Binding relative to the max [%] ,0 6,5 7,0 7,5 8,0 8,5 9,0 ph 100mM NaCl 200mM NaCl 300mM NaCl 400mM NaCl 500mM NaCl February 15, 2012 Page 55

54 HCP breakthrough at 10 membrane volumes per min (resins 1 MV/min) UV (280nm) [mau] Sartobind STIC PA binds HCP most effectively Membrane volume [ml] Sartobind Q Sartobind STIC PA AEX resin 1 AEX resin 2 Mixed mode resin Buffer: 20 mm Tris ph 7.5, 150 mm NaCl February 15, 2012 Page 56

55 DNA removal from a therapeutic antibody. Campath 1-H pg DNA pro m g mg Protein / m l Average of three 12,500 liter batches with DNA binding m odule capacity After protein Sartobind AQ After Q Sartobind -Sepharose FF Q J. K. Walter, Boehringer Ingelheim Pharma, Bioseparation and Bioprocessing, G. Subramanian (ed.) Wiley VCH, Vol. II p , clears the DNA below detection limit superior pressure/flow relation Galliher P, Fowler E, Millennium Pharmaceuticals Inc.: Validation of Impurity Removal by the CAMPATH-1H Biomanufacturing Process. IBC s Biopharmaceutical Production Week, Paradise Point Resort San Diego, CA, November 12-15, the process time is reduced 23-fold, excluding the benefit in handling for set-up....diminish a loss of product...installed as in-line filters and can be disposed after use*. FDA approval March 2001 February 15, 2012 Page 57

56 DNA binding capacity at 10 % breakthrough and 16.8 ms/cm February 15, 2012 Page 56

57 HIC phenyl membrane works at flow through conditions between mm ammonium sulfate mm (NH 4 ) 2 SO 4 concentration February 15, 2012 Page 59

58 Aggregate removal from a recombinant protein by Sartobind Phenyl Ebert,S., Rentschler Biotechnologie, Fischer-Frühholz, S., SSB, Efficient Aggregate Removal from Impure Pharmaceutical Active Antibodies, BioProcess International, Vol. 9, No. 2, Feb 2011, pp February 15, 2012 Page 60

59 Endotoxins can be removed 5 LRV with Sartobind Q Run 1 Run 2 Process volume (Start) 50 L 47 L Process volume (End) 60 L 59 L Process time ~10 20 min ~10 20 min Protein concentration (Bradford) 4.10 g/l 4.14 g/l Conductivity ms/cm ms/cm Pre use endotoxin level 26,900 EU/mg 10,100 EU/mg Post use endotoxin level 0.13 EU/mg 0.18 EU/mg Protein recovery over step 81.4 % 84.6 % LRV endotoxins 5 5 Endotoxins were removed from GST-protease from levels EU/mg to 0.13 EU/mg protein solved in TRIS / 150 mm NaCl buffer ph 8 with Sartobind Q SingleSep 20 at cgmp runs. *A. Clutterbuck et al., Avecia, Endotoxin reduction using disposable membrane adsorption technology in cgmp manufacturing, BioPharm International 20 (5), (2007) February 15, 2012 Page 61

60 Endotoxin removal on Sartobind STIC nano 1 ml LRV >4.3 - Sample: 20 mm Tris ph 7.5, 150 mm NaCl spiked with 500 EU/ml - Device: Sartobind STIC PA nano 1 ml - Flow rate 10 ml/min - Sample load volume: 8000 ml - Total load 4,000,000 EU - Detection limit of test: 0.05 EU/ml 4 Million Endotoxin Units (1000 EU/ml) Before EU/ml After EU/ml LRV Concentration 1000 <0.05 >4.3 Complete removal February 15, 2012 Page 62

61 Bakteriophage Phi X 174 is considered as surrogate for viruses Microviridae ssdna nm diameter pi = K. Brorson, H. Shen, S. Lute, J.S. Pérez, D.D. Frey, Journal of Chromatography A 1207 (2008) M. Phillips, J. Cornier, J. Ferrence, C. Dowd, R. Kiss, H. Lutz, J. Carter, J. Chromatogr. A 1078 (2005) 74. February 15, 2012 Page 63

62 Applications flowthrough polishing on anion exchanger, typical spiked viruses to prove virus safety of Mabs Positvely (+) charged anion exchange adsorber binds negatively (-) charged viruses Isoelectric points Product Impurities Viruses Viruses are bound ph buffer Phage pi ΦX Source for pi of viruses: Daniel M. Strauss et al., Biotechnol. Bioeng. 2009, 104, February 15, 2012 Page 64

63 LRV of ΦX174 (pi ) of Sartobind Q and Sartobind STIC low and high conductivity 25 mm Tris ph 7.5, 0, 50, 150 mm NaCl February 15, 2012 Page 65

64 Phosphate and other multivalent anions inhibit the binding of ΦX174 Load 4*10 7 pfu/ml ΦX174, LP15, flow rate 20ml/min LRV 5 4,5 4 3,5 3 2,5 2 1,5 Sartobind Q Sartobind STIC 1 0,5 0 Buffer 25mM TRIS/HCl 25mM TRIS/HCl 25mM TRIS/HCl 25mM TRIS/HCl 20mM KPi 20mM KPi 20mM KPi 20mM KPi ph NaCl [mm] 7,5 7, ,5 7, February 15, 2012 Page 66

65 Phosphate inhibits phage binding but DNA contaminant is bound Sartobind STIC to be used at binding conditions of viruses Ortho phosphate mm Phage in flowthrough % of start material 0 < <1 2 < < < <1 DNA in flowthrough % of start material 0.41 ml (15 cm², 3 layers) Sartobind STIC PA were loaded with 150 ml ΦX 174, 1 x 10 7 PFU/ml, salmon sperm DNA 200 ng/ml at 10 ml/min, buffer 20 mm Tris/HCl ph 7.5, 150 mm NaCl plus phosphate. February 15, 2012 Page 67

66 Salt tolerant anion exchanger Sartobind STIC Primary Amine Protein Binding [g/l] Sartobind Q BSA in 200 mm NaCl (20 ms/cm) DNA Binding [g/l] DNA in 50 mm NaCl (7 ms/cm) Sartobind STIC LRV with Mouse Minute Virus (MMV) in 150 mm NaCl 1.81 > 4.96 Provides higher binding capacity compared to Q anion exchanger at higher conductivity Enables the polishing at broader operating conditions (high salt) February 15, 2012 Page 68

67 February 15, 2012 Page 69

68 Sartobind / AEX column Polishing February 15, 2012 Page 70

69 Sartobind in Virus Purification Influenza virus Adenovirus Lentivirus Adenoassociated virus Densonucleosis virus Baculovirus Foot and mouth desease virus Pseudrabies virus Bovine herpesvirus Rotavirus like particles Yellow fever virus Bacteriophages February 15, 2012 Page 71

70 Membrane Adsorbers vs. Density Gradient Adenovirus purification Membrane Adsorbers Density gradient ultra centrifugation 2 hours 36 hours Up to VP/ml No carryover, disposable, no validation, simple No contaminants 10 6 VP/ml Carryover validation Toxic CsCl 2, sucrose removal from finished product February 15, 2012 Page 72

71 Membrane Adsorbers vs. Columns Adenovirus purification Capacity membranes > 10 fold higher than beads 20 L module instead of 200 L column 25-fold higher volumetric flow rate February 15, 2012 Page 73

72 Purification of PPV virus spike PDA Technical Report No. 47: Preparation of virus Spikes used for Virus Clearance Studies 2010, p19-20 February 15, 2012 Page 74

73 Sartobind in virus purification targets Capture Influenza vaccine Rabies vaccine Adenovirus-vectored vaccine Polio vaccine Conjugated vaccine VLP Polishing DNA removal HCP removal February 15, 2012 Page 75

74 Thank you! February 15, 2012 Page 76