<<

Engineering Conferences International ECI Digital Archives

Vaccine Technology IV Proceedings

Spring 5-24-2012 Monolithic Columns for Purification and In- process Control of and -like Particles Lidija Urbas BIA Separations

Milos Barut BIA Separations

Follow this and additional works at: http://dc.engconfintl.org/vaccine_iv Part of the Biomedical Engineering and Bioengineering Commons

Recommended Citation Lidija Urbas and Milos Barut, "Monolithic Columns for Purification and In-process Control of Viruses and Virus-like Particles" in "Vaccine Technology IV", B. Buckland, University College London, UK; J. Aunins, Janis Biologics, LLC; P. Alves , ITQB/IBET; K. Jansen, Wyeth Vaccine Research Eds, ECI Symposium Series, (2013). http://dc.engconfintl.org/vaccine_iv/43

This Conference Proceeding is brought to you for free and open access by the Proceedings at ECI Digital Archives. It has been accepted for inclusion in Vaccine Technology IV by an authorized administrator of ECI Digital Archives. For more information, please contact [email protected]. Monolithic Columns for Purification and In-process Control of Viruses and Virus-like Particles

Lidija Urbas, Miloš Barut Faro, May 24th 2012 Outline

• Conventional particle based • Monolith based chromatography • Downstream processing of Ad3 VLPs – Case study • Convective interaction media (CIM) monoliths for in-process control – CIMac monolithic analytical columns as a tool for HPLC based analytical methods BIA Separations

• BIA Separations was founded in September 1998. • Headquartes in Austria, R&D and Production in Slovenia. • BIA Separations USA established in September 2007 - sales and tech support office. • BIA Separations China established in January 2011 - sales and tech support office. • Main focus: To develop and sell methacrylate monolithic columns & develop methods and processes for large separation and purification . Moved to a new 4.200 m 2 facility in October 2011 Downstream Processing of Viruses and Virus-like Particles • Conventional • Filtration • Ultracentrifugation • UF/TFF • Alternative • Ion exchange chromatography A method of choice • Size exclusion chromatography for • (Pseudo) . • Hydrophobic interaction chromatography What about for larger biomolecules? Chromatographic Separations

Principle of chromatography: molecule from the mobile phase binds to the stationary phase.

Mass transport: •Diffusion – Migration of the solutes from an area of high to an area of low concentration

•Convection – Movement induced by external force Conventional Liquid Chromatography Media

Mass transport in packed porous particle columns: convection + diffusion

Interparticle void volume Intraparticle void volume (preferential flow path) (contains majority of binding sites: > 90 %) DiffusionLimitations ConventionalChromatographyLiquid Media Binding capacity at at high flow rate: f3 f2> f1> f > Binding capacity

Binding capacity [mg/ml] f 1 Flowrate [ml/min] f 2 0 f 3 Porous Particles – Flow Dependent Dynamic Binding Capacity

1.0 Source 30 S 100 cm/h 200 cm/h IgG (c=0.1 mg/mL) 0.8 400 cm/h 600 cm/h

C/C 0.6 0

0.4

0.2

0.0 20 40 60 80 100 120 140 elution volume [ml]

Courtesy of Prof. A. Jungbauer, IAM, BOKU, Vienna, Austria Diffusion Limitations

Resolution in linear gradient elution at high flow rate: f2 > f1

f1

f2 Detector response response Detector Detector

Eluted volume [ml] Larger the molecule wider the peak – lower the resolution Difference Between Diffusive and Convective Mass Transport

2 molecule MW De (cm /s) small molecule 58 Da 1.4 x 10 -5 hemoglobin 64kDa 7 x 10 -7 BSA 66 kDa 6.1 x 10 -7 urease 482 kDa 3.5 x 10 -7 Tomatomosaicvirus (ToMV) 40 000kDa 5 x 10 -8 DNA 4.4 kbp 1.9 x 10 -8 DNA 3.7kbp 4,5x 10 -8

The bigger the molecule the lower the rate of diffusion Another Challenge – the Size of the Molecule of Interest

Pores become too small for large solutes! (Binding mostly on the outer surface) results in Very low binding capacity for large solutes (behave like nonporous particles) Working with big molecules

 Molecule size: surface accessibility

Molecule nm Proteins 1-5 IgM 25 150-250 130 Poxvirus 200 x 500 T4 phage 220 x 85

Courtesy P. Gagnon www.validated.com What Are the Alternatives?

• Membranes • Monoliths

X X X

Stacks of thin polymeric layers – supplied in single piece but in fact they are discontinuous unit . CIM Monoliths

Monoliths are continuous stationary phases cast as a homogeneous column in a single piece.1

1 G. Iberer, R. Hahn, A. Jungbauer, LC-GC 17 (1999) 998. CIM Monoliths

Made of highly cross-linked porous rigid monolithic:

O O + O • poly(glycidyl methacrylate-co- O O O O GMA EDMA ethylene dimethacrylate) or Glycidyl methacrylate Ethylene glycol dimethacrylate

• poly(styrene-divinylbenzene) O O O O polymers O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O

Poly(glycidyl methacrylate-co -ethylene dimethacrylate) What distinguishes monoliths from conventional supports? Structure of the monoliths • High porosity (over 60 %): • Flow-through channels (“pores”) having large diameter (1.5 µm), • Uniform channel connectivity in 3D (homogeneous structure ). CONSEQUENCE • Low pressure drop • High surface accessibility – no dead end pores • High dynamic binding capacities for large molecules • Mass transfer predominantly based on CONVECTION What distinguishes monoliths from conventional supports?

Convective transport – flow independent properties

60 1 3 300 2 100 50 250 Flow rate

40 ml/min 80 35.9 ml/min 40 200 80 ml/min 76.2 ml/min 120 ml/min 152.7 ml/min 60 160 ml/min 224.6 ml/min 30 150 200 ml/min 681 ml/min

240 ml/min 40 BufferBuffer[%] [%] AA 20 100 Absorbance Absorbance Absorbance 280 280 atat(mAU) (mAU) nm nm 10 20 50 AbsorbanceAbsorbance[mAU] [mAU] atat280280 nm nm

0 0 0 0 50 100 150 200 250 300 0 100 200 300 400 500 600 700 800 900 1000 Volume [ml] Elution volume (ml)

Podgornik et al., Anal. Chem. 72 (2000) 5693 Avaliable CIM Monolithic Columns

0.1 0.34 1 8 80 800800 8000 ml

CIMmultus (multi-use disposable columns) CIM Monolith Column Capacity

Flu vaccine 24.000 doses/run pDNA .48 g/run 240.000 doses/run 4.8 g/run 2,400.000 doses/run 48 g/run Main CIM Monoliths Applications Area

Endotoxin DNA depletion

Large proteins CIM® Monoliths for Production of Complex Biomolecules • Drug Master Files (DMF) for CIM ® DEAE, QA and SO3 columns in place, HIC in preparation. • First drug purified using CIM ® monoliths passed CPIII trial (pDNA for ). • More than 25 projects in CPI – CPIII trials (various Influenza, various Adenovirus, , various IgMs, Inter -alpha - inhibitors). • More than 200 projects in pre-clinical trials (Influenza A and B virus (eggs, Vero and MDCK cells), , Rotavirus, AAV, various Adenovirus subtypes, , Mulv, MVM, Feline calicivirus, Japanese encephalitis, Crimean-Congo hemorrhagic fever, Hantaan virus, VLP (Hepatitis B, HPV, Influenza, Adenovirus), bacteriophages (Lambda, T4, VDX10, Pseudomonas phage), Tomato and Pepino Mosaic virus, pDNA, IgM, various proteins). Downstream Processing of Virus-like Particles (VLPs)

Case study

Purification of adenovirus type 3 dodecahedric VLPs for biomedical applications Work done with the Institute of Biochemistry and Biophysics, Polish Academy of Science (Ewa Szolajska and Jadwiga Chroboczek). Dodecahedric VLPs of Adenovirus Serotype 3 (Ad3 VLPs)

Ad3 capsid Ad3 VLPs

penton fiber

hexon penton base

MW (Ad3 VLPs): 3.6 MDa MW (base): 63 kDa Production of Ad3 VLPs

Expression in baculovirus/insect cells system

Cell lysis 5 days

Ultracentrifugation in a sucrose density gradient

Chromatography on an agarose based matrix

Ad3 VLPs A delivery vector for bleomycin (BLM) Goal: • Evaluation of CIM monoliths – Chromatographic step • Direct purification of crude lysate using CIM monoliths

Zochowska et al, Plos One, 2009 Method Development

• HPLC system – detection at 280 nm • CIMac analytical column (5 x 5.2 mm ID, V = 0.1 mL) • Ad3 VLPs samples: – UC sample (sample that was prepurified with ultracentrifugation) – lysate sample • Methods: – Ad3 VLPs content - SDS PAGE – content - BCA, Bradford assay – DNA content - PicoGreen Analysis, agarose – Vector internalisation - Confocal Microscopy Screening of Stationary and Mobile Phases

• Initial screenings performed using the UC (purified) sample, • Various cation- and anion-exchange monolithic columns, • Various buffering systems - composition and pH. Cation Exchangers

1200 100

1000 Ad3 VLPs 80 800 60 600

A [mAU] [mAU] AA 40 400 Gradient B [%] [%] B B Gradient Gradient 200 20

0 0 0 0,5 1 1,5 2 2,5 3 t [min]

Conditions: buffer A: 20 mM Tris, pH 7.5 + 2 mM EDTA + 5% glycerol, buffer B: buffer A + 1 M NaCl, pH 7.5; V: 150 mL 2 ×diluted UC sample; detection at 280 nm. Anion Exchangers

200 100 180 Ad3 VLPs DNA 160 80 140 120 60 100 A[mAU] A[mAU] A[mAU] 80 40 GradientGradient B B [%] [%] 60 GradientGradientGradientGradient BBBB [%] [%] [%] [%] 40 20 20 0 0 0 2 4 6 8 t[min] DEAEDEAEDEAEQA QA EDA

Conditions: buffer A: 20 mM Tris, pH 7.5 + 2 mM EDTA + 5% glycerol, buffer B: buffer A + 1 M NaCl, pH 7.5; V: 60 mL 2 ×diluted UC sample; detection at 280 nm. Agarose based matrix and CIMac QA

90 FT E1 100 80 70 80 Ad3 VLPs 60 DNA 60 50 E2 E4 40 E3

A[mAU] 40 30 Gradient B [%] Gradient 20 20 TEM analysis of E2. 10 0 0 0 5 10 15 20 25 SDS PAGE; M: protein t [min] marker, L: loading sample, FT-E: collected fractions. 400 Ad3 VLPs DNA 100 FT 350 80 300 E3 250 60 E4 E5 200

A[mAU] 150 40

100 E1 GradientB [%] E2 20 50 TEM analysis of E3. 0 0 0 2 4 6 8 10 t[min] Dynamic Binding Capacity

SDS PAGE; M: protein marker, L: loading sample, 1-10: collected flow-through fractions. The dynamic binding capacity of Ad3 VLPs is 1.38 × 10 16 Ad3 VLPs/mL. Lysate Sample Treatment

800 Ad3 VLPs DNA 120 Conditions: buffer A: 20 mM Tris, FT pH 7.5 + 2 mM EDTA + 5% 700 E5 E8 100 glycerol, buffer B: buffer A+ 1M 600 E7 NaCl, pH 7.5; V: 60 µl 3 ×R cell 80 500 lysate ; detection at 280 nm. 400 60 E1 E3 E4

A [mAU] A 300 E2 40 200 E6 Gradient B[%] 20 100 SDS PAGE. M: protein marker, 0 0 FT1 -E8: collected fractions. 0 2 4 6 8 t [min]

Agarose FT E1 E2 E3 E4 E5 E6 E7 E8 L1 L2 M electrophoresis. M: protein marker, FT1-E8: collected fractions. Step Gradient

Ad3 VLPs DNA 2500 100 FT1 E3 2000 E2 80

1500 60

1000 40

A[mAU] FT2 E1 Gradient B[%] Gradient 500 20

0 0 0 2 4 6 8 10 12 SDS PAGE. M: protein marker, FT1-E3: collected fractions. t[min]

• less than 1% DNA, • purification protocol: 1 day.

Conditions: buffer A: 20 mM Tris, pH 7.5 + 2 mM EDTA + 5% glycerol, buffer B: buffer A+ 1M NaCl, pH 7.5; V: 500 µl 5 ×R cell lysate ; detection at 280 nm. TEM analysis of E2. Vector Internalisation Incubation of HeLa cells with Ad3 VLPs: – control: cell nuclei (blue), – Ad3 VLP Q: cell lysate purified with ultracentrifugation and the agarose matrix. – Ad3 VLP QA : cell lysate purified on the CIMac QA column (E2 fraction),

Monitoring of Ad3 VLPs with confocal microscopy. Ad3 VLPs stained with green and cell nuclei counterstained in blue with DAPI. Confocal microscopy images show Ad3 VLP in the cytoplasm of HeLa cells after 90 minutes of incubation with both samples. Ad3 VLPs retain their cell penetration capacity. L. Urbas et al.,J.of Chromatography A, 2011 CIMac Analytical Columns for In-process Control Monolithic Analytical Columns for In-process Control (PAT)

10 ml/min = 2800 cm/h = 100 CV/min In-process control of Ad3 VLPs

Ad3 VLPs DNA Expression 800 100 700 80 600 Cell lysis 500 60 400 300 40 200 20 [%] B Gradient Ultracentrifugation Absorbance [mAU] 100 0 0 0 2 4 6 8 t [min] Chromatography

100 100 100 Ad3 VLPs DNA 100 80 Ad3 VLPs DNA 80 80 80

60 60 60 60

40 40 40 40 Gradient[%] B 20 20 BGradinet [%] 20 20 Absorbance [mAU] Absorbance [mAU] 0 0 0 0 0 2 4 6 8 0 2 4 6 8 t [min] t [min] In Process Control of the Purification

Ad3 VLPs DNA 2500 100 FT1 E3 2000 E2 80

1500 60

1000 40

A[mAU] FT2 E1 GradientGradient[%] [%] B B 500 20

0 0 0 2 4 6 8 10 12 t[min] In-process control of Ad3 VLPs

120 Ad3 VLPs DNA 100

80

60 A[mAU] 40

20

0 0 2 4 6 8 t [min] FT1 FT2 E1 E2 E3 Conditions: buffer A: 20 mM Tris, pH 7.5 + 2 mM EDTA + 5% glycerol, buffer B: buffer A+ 1M NaCl, pH 7.5; V: 200 µl of 5×disolved sample, method: gradient from 0-100% buffer in 8 min; detection at 280 nm. L. Urbas et al.,J.of Chromatography A, 2011 PAT of Ad5 Production Using Monolithic HPLC

Eden Biodesign Platform Ad5 Process

•Enzymatic degradation •Remove small proteins •Critical to prevent column fouling •Decrease process volume •Concentrate/ buffer exchange virus •Decrease [Benzonase] •Bind and elute •Virus concentration •Group separation •HCP, DNA and endotoxin removal •Removal of HCP •More cost and time efficient •Polish virus •Scalable

C. Sims et al., MSS 2010, Portoroz, Slovenia, 2010 Monolithic HPLC used for Ad5 Production Process Development – Basis for the PAT

Harvest 4.2 4.4 Option 1 Lysis

4.1 Clarification

DNA reduction

2 4 Filtration

Chromatography 1 Option 2 Chromatography 2

Final Formulation 4.4

2 4

P. Ball et al., Eden Biodesign, MSS 2008, Portoroz, Slovenia, 2008 Monolithic HPLC used for the Ad5 Production Process PAT

AEX load SEC eluate

AEX eluate Final Formulation

Whitfield RJ et al., J Chromatography A. 2009 Conclusions • Novel chromatography supports - monoliths - because of their advantages offer a good alternative to ultracentrifugation and conventional chromatographic support in the field of large molecules. • Ad3 VLPs were efficiently purified using monoliths: – Ad3 VLPs were separated from DNA, – Less than 1% of DNA in the purified fraction, – Ultracentrifugation could be omitted, – The morphology of the particles was unaltered, – Process time was shortened from 5 days to 1 day, – Ad3 VLPs retained their cell penetration capacity. • Monolithic analytical columns enable fast and reliable purification process development, PAT, and product characterization. Acknowledgements

BIA Separations d.o.o. Polish Academy of Sciences Aleš Štrancar Jadwiga Chroboczek Miloš Barut Ewa Szolajska Barbara Lah-Jarc Monika Zochowska Thank you for your attention! Slovenija Slovenija