Application Note
USTR 2827(1)
Implementation of Mustang® Q Membrane
Chromatography as a Polishing Step (Residual
DNA Removal) in Monoclonal IgG1 Production
from CHO Cell Culture
In collaboration with ProBioGen AG, Berlin, Germany
1.
Summary
MustangQmembranechromatographywasevaluatedasapolishingstepfollowingproteinAaffinityand
cationexchangechromatographytoremoveresidualhostcellDNAduringaMonoclonalantibody(MAb)
purificationprocessat250Lcellcultureproduction-scale.DataindicatedefficientDNAclearancebythe
MustangQmembranechromatographystepduringtheprocess(<16pg/mL),whilekeepinganexcellent
MAbrecovery(>96%).Additionally,contributionoftheMustangQmembranechromatographypolishing
steptoHCPremovalwasshown.Basedonavirus-spikingscale-downstudyusingtwomodel-viruses
(MurineLeukemiaVirus(MuLV)andMinuteVirusofMice(MVM)),effectiveMuLVvirusremovalwas
demonstratedusingMustangQmembranechromatography,whileMVMvirusclearanceneededfurther
optimization.Overallviralclearancecapacitywassufficientformanufacturingofasafedrugsubstance
forclinicaltrials.
2.
Introduction
AGMP-compliantmanufacturingprocess(Figure1)wasestablishedat250Lculture-scaleforthesupply
ofamonoclonalgamma-globulinclass1(IgG1)materialfortoxicologicalstudiesandlater-onforphaseI
clinicaltrials.CurrentplatformsforMAbpurificationmostlyconsistofthreepackedbedchromatography
stepsincludingProteinAasacapturestepandtypicallycationexchangeorhydrophobicinteractionas
intermediatesteps.Membranechromatographyisincreasinglyusedasapolishingsteptoremovecontaminants(residualDNA,HostCellProteins)andprovidesadditionalprocess-specificvirusclearance.
Mustangmembranechromatographycapsulesaredisposable,easytouseandoffereffectivesolutions
forpolishing.
TheMustangQmembranechromatographystepimplementedhereforhostcellDNAremovalwas
conductedin“negativemode”.Lab-scaledevelopmentusingMustangQAcrodisc® unitaswell
asintermediate-scaleandproduction-scaleonMustangQcapsulesof10mLand60mLbedvolume
respectively,aredescribedhere.
Figure 1
ProBioGen’s Monoclonal IgG1 Production Process from CHO Fed-batch Cell Culture
CHO Fed Batch Cell Culture
Cell Removal
Affinity Chromatography: Protein A Capture
Virus Inactivation: 60 min at pH 3.5
Cation Exchange Chromatography
Mustang Q Membrane Chromatography (Flowthrough)
UF/DF (30 kDa cut-off)
Virus Removal by Nanofiltration
0.2 µm Filtration
FINAL BULK
2
3.
Materials and Methods
3.1
Cell culture
MonoclonalIgG1 wasproducedinserum-freeCHO(ChineseHamsterOvary)cellcultureat50L
and250Lfed-batchscalefortoxicologicaltrialsandGMPproductionrunsforphase1clinical
trialsbyProBioGenAG,Berlin,Germany.
3.2
Lab-scale optimization of process conditions on Mustang Q unit:
Unit: 0.18mLMustangQAcrodiscunit(Pall)
Running buffer: 20mMNaPhosphatepH7.0,0.2MNaCl
DNA elution buffer: 20mMNaPhosphatepH7.0,1MNaCl
Load for the DNA spiking experiment: 5.5µgCalfThymusDNAinrunningbuffer
Load for the process test experiment: 68mLofacationexchange(CEX)chromatography
columneluateat3.2mgIgG/mLand22-25mS/cm.
Flow rate: 10bedvolumes/min(1.8mL/min).
3.3
Implementation of manufacturing process runs on Mustang Q unit at 50 L and
250 L cell culture-scale (Table 1, Figure 2)
Unit: 10mLMustangQCLM05capsuleor60mLMustangQCL3unit(Pall)
Initiallab-scaleconditionswereslightlymodifiedwithregardtotheconductivityoftherunning
bufferandtheprocesssolution:
Running buffer: 20mMNaPhosphatepH7.0
Load: CEXchromatographycolumneluatediluted2-foldinrunningbuffer(resultinginfinal
conductivityof~11-12.5mS/cm)
DNA elution buffer: 20mMNaPhosphatepH7.0,1MNaCl
Table 1
Process Conditions for Mustang Q membrane chromatography implemented at 50 L (Tox 2) and
250 L culture-scale (Tox 1 and GMP)
Process conditions
Culture-scale
Process volume
Mustang Q membrane volume
DNA binding capacity
Expected DNA load (process)
Flow rate
Production run
Tox 1
Tox 2
GMP
250 L
225 L
60 mL capsule (CL3)
1.2-1.5 g
Low mg-range
36 L/h
10 MV*/min
50 L
44 L
10 mL capsule (CLM05)
200-250 mg
µg-range
30 mL/min**
3 MV*/min**
250 L
238 L
60 mL capsule (CL3)
1.2-1.5 g
Low mg-range
36 L/h
10 MV*/min
* MV: membrane volume according to supplier specification
** Low flow rate due to pressure limitations at the chromatography system
Figure 2
Manufacturing run on a 60 mL bed volume Mustang Q capsule
Mustang Q 60 mL capsule
Courtesy of Dr Martin Suhr, ProBioGen AG.
AllchromatographyrunswereperformedusingÄKTAu systems(GEHealthcare).
www.pall.com/biopharm
3
3.4
Scale-down virus-spiking study on Mustang Q Acrodisc unit
Twomodelviruses,MurineLeukemiaVirus(MuLV)andMinuteVirusofMice(MVM)(Table2),
wereselectedforaspikingstudytodemonstratetheviralclearancecapacityofthefinalDSP
processinascale-downmodel.
Table 2
Model viruses used for the virus spiking study
Virus
Genome
Enveloped
Family
Size [nm]
Chemical
resistance
MuLV
MVM
RNA
DNA
yes
no
Retro
Parvo
80-110
~20
Low
Very high
Membrane chromatography unit: 0.18mLMustangQAcrodiscunit
Running/process buffer: 20mMNaPhosphatepH7.0
Load: ProcesssolutionspikedwithMuLVorMVMvirussolution(5%volumespike)
3.5
Analytics
IgG1 puritywasassessedusingaSizeExclusionChromatographycolumn(Superdexu 200
5/150GL,GEHealthcare)
IgG1 concentrationwasdeterminedbyUV280nm measurement.
ResidualDNAwasmeasuredbyuseofthePicogreenu Assay(Invitrogen)orbyquantitative
PCR(qPCR)performedbyanexternalserviceprovider.
QuantificationofHostCellProteins(HCPs)wasdonebyELISAusingakitfromCygnus
Technologies(HCPHostCellProteins,#F015).
VirusspikingstudywasperformedbyanexternalserviceproviderunderGLPusingaprocess
scale-downmodel.Thevirustiterwasdeterminedbyendpointtitrationand/orlargevolumeplating.
4.
Results and Discussion
ThefinalgoaloftheMustangQmembranechromatographystepevaluatedinthecurrentprocesswas
toachievelowlevelsofresidualDNAinthefinalMAbproductaccordingtothe<150pg/doserule.
4.1
Lab-scale optimization of Mustang Q membrane chromatography process conditions
for residual host cell DNA removal
Apreliminarylab-scaleDNAspikingexperimentusingcalfthymusDNAwascarriedoutto
investigateloadandelutionconductivityvaluesforDNAeliminationwithaMustangQAcrodisc
unit(typicalDNAbindingcapacityof3.6mg).ADNAloadof5.5µgDNAwaschosenas
aprocessrepresentativeimpurityloadofthisprocessstepbasedonPicogreendata.The
chromatographyprofileconfirmedadsorptionofDNAin20mMNaPhosphatepH7.0,0.2M
NaClfollowedbyadesorptioninthesamebuffercontaining1MNaCl(75mS/cm)(Figure3A,
zoomoftheÄKTAprofileshowingDNAelution).
Thesameconditionswereappliedinaprocesstestexperimentwhereacationexchange
(CEX)chromatographycolumneluatewasdirectlyloadedontotheMustangQmembrane
chromatographyunit(Figure3B,zoomoftheelutionstep).Itconfirmedtheretentionofresidual
DNAanditselutionat72mS/cm.
4
Figure 3
DNA-spiking and process test experiments on 0.18 mL Mustang Q Acrodisc unit
Conductivity (mS/cm)
mAU
140
120
100
Absorbance 280nm (mAU)
mAU
mS/cm 140
80
HCP
120
70
100
60
80
Absorbance 260nm (mAU)
mS/cm
80
70
60
80
60
50
60
50
40
40
40
40
30
20
20
24 mL
0
20
DNA
0
16
18
20
22
DNA
104
(A) - DNA Spiking Run
106
108
110
30
20
112 mL
(B) - Process Test Run
Running buffer: 20 mM Na Phosphate pH 7.0, 0.2 M NaCl
DNA elution buffer: 20 mM Na Phosphate pH 7.0, 1 M NaCl
Loading conditions for:
- DNA spiking run (A): 5.5 µg Calf Thymus DNA in running buffer
- Process test run (B): cation exchange chromatography column eluate
Conclusion of 4.1
OptimizationoftheprocessconditionsonaMustangQAcrodiscunitevidencedthe
eliminationofresidualDNAinnegativemode(retentionofDNAontotheMustangQmembrane
andelutionat75mS/cm).PotentialeliminationofHCPpeakinnegativemodewasalsoseen
elutingearlierinthesaltgradient.
4.2
Implementation of Mustang Q membrane chromatography process conditions
to manufacturing-scale
Processconditionsinprevioussectionwerebasedonaloadconductivityof~22-25mS/cmof
theCEXeluate.Effectivevirusclearanceusinganionexchangerinnegativemodeisdependent
ontheconductivityofthefeedstock.Thereforetheprocessconditionswereslightlymodifiedto
reducetheloadconductivityto~11mS/cmbyimplementationofa1:2dilutionoftheCEXeluate.Furthermoretherunningbufferwaschangedto20mMNaPhosphate,pH7.0.
Themodifiedconditionswereappliedtorunthemanufacturingprocesseson10mLand60mL
MustangQcapsulesat50Land250Lculture-scales,respectively.The50Lculture-scalewas
producedfortoxicologystudies(Tox-50L)andthe250Lscalewasproducedfortoxicology
(Tox-250L)andGMP(GMP-250L)batches.
Thefinalconditionswerenotfurtheroptimizedatthisdevelopmentstage.Thereforethe60mL
MustangQcapsulewasselectedforthe250Lculture-scalebytakingintoaccountasafety
factor(potentialbiologicalvariationofthefeedstockcomposition).
www.pall.com/biopharm
5
4.2.1
DNA reduction during IgG1 purification process including the Mustang Q
membrane chromatography polishing step
PriortotheTox-andGMP-productionruns,twodevelopmentrunsat10Lculturescale(10L–Dev)werecarriedout.ResidualDNAwasmeasuredbyuseofthe
Picogreenassay.DataforbothrunsshowedthattheDNAcontentwasalreadyinthe
lowngrangeaftertheCEXchromatographystep(Table3).Thereforethisassaycould
notbeappliedforanalysisafterMustangQmembranechromatographyduetothe
limitofdetection.
Table 3
DNA levels after the first and second chromatography steps of the process, measured
using Picogreen assay, during 2 development runs at 10 L culture-scale
Process step
Development run
10 L – Dev 1
DNA ng/mL IgG1 mg/mL
Cell culture supernatant
5951
After first chromatography (Protein A) 19.0
After second chromatography step
4.2
(Cation exchanger)
n/a
6.6
3.4
10 L – Dev 2
DNA ng/mL
IgG1 mg/mL
7538
22.2
2.2
n/a
7.2
3.4
n/a = Not applicable
SamplesderivedfromtheMustangQmembranechromatographystep(flowthrough)
hadtobeanalyzedbyuseofaqualifiedqPCRmethod(assaysensitivitybelow1pg).
Analysisofafinalproductsamplewithdrawnafterpoolingtheproductofbothdevelopmentruns(10Lscale)resultedin12pgresidualDNA/mLthusconfirmingefficient
DNAremovalbyMustangQmembranechromatography.
Thisfindingwasconfirmedlateronasthedataobtainedfortheproductionruns
showedthatresidualDNAwasefficientlyremovedtoaconcentrationclosetothe
pg/mLrange(Table4).
Table 4
DNA levels after the Mustang Q membrane chromatography polishing step,
measured using qPCR assay, at 50 and 250 L culture-scale
Process step
Production run
Tox - 250 L
Tox - 50 L
GMP - 250 L
Final bulk
< 40 pg/mL*
< 17 pg/mL*
< 16 pg/mL*
* result below lower limit of quantification of the assay (LLOQ)
Conclusion of 4.2.1
TheMustangQmembranechromatographypolishingstepwasefficienttoremove
hostcellresidualDNAintheprocess.ThisstepwasthereforeimplementedintheGMP
purificationprocessat250LscalefortheproductionofIgG1 forearlyclinicaltrials.
4.2.2
6
IgG1 recovery after polishing step on Mustang Q capsules at manufacturing-scale
DeterminationoftheproductconcentrationbyUV280nmmeasurementshowed
highIgG1 recovery(>96%)fortheMustangQmembranechromatographystep,
independentlyoftheprocess-scale(Table5).
Table 5
Process conditions and IgG1 recovery for Mustang Q membrane chromatography
implemented at 10 L, 50 L and 250 L culture-scales
Run
Load
IgG1
Protein load
volume Mustang Q
Conductivity concentration (g product/mL
(mL)
capsule (mL) (mS/cm)
(g/L)
bed volume)
250 L - Tox 37600
50 L - Tox
5609
250 L - GMP 40150
60
10
60
12.9
11.8
12.5
1.7
2.4
1.7
1.07
1.34
1.14
IgG1
recovery
(%)
96.1%
97.9%
104.2%
Highconsistencywasachievedforthethreeproductionruns(Tox,GMP)regarding
loadconductivityandproteinload(gproduct/mLbedvolume).Thevariationofthe
productrecoverybetween96and104%maybeexplainedbyaccuracyofthe
volumedeterminationwhichwasdonebyweightmeasurementoftheprocess/
productsolutions.
Conclusion of 4.2.2
HighIgG1 recovery(>96%)wasachievedfortheMustangQmembranechromatography
polishingstepperformedinnegativemode.
4.2.3
HCP removal during IgG1 production process including the Mustang Q
membrane chromatography polishing step
Althoughthemembranechromatographystepwasnotspecificallydesignedfor
HCPremovalinthisprocess,theHCPdatashowedthattheMustangQmembrane
chromatographypolishingstepcontributedtoHCPdepletiontofinallymeetthe
establishedruleof<100ppmattheendoftheprocess(Table6).Additionally,HCP
removalwasquiteconsistentoverthedifferentcellculturebatches(50Land250L).
Table 6
Host Cell Protein (HCP) removal during the IgG1 production process at 50 and 250 L
culture-scales, including after the Mustang Q chromatography step
Process step
Clarified Cell Culture supernatant
After Protein A chromatography
After CEX chromatography
After Mustang Q membrane
chromatography
After nanofiltration
HCP level (ppm) of production run
250 L – Tox
50 L – Tox
250 L – GMP
181,592
1074
70
45
110,117
ND*
86
41
113,665
295
39
25
25
33
27
*ND: not determined
Conclusion of 4.2.3
TheMustangQmembranechromatographypolishingstepcontributed
toHCPdepletion.
www.pall.com/biopharm
7
4.2.4
Viral clearance during IgG1 production process including the Mustang Q
membrane chromatography polishing step
PotentialcontributionoftheMustangQmembranechromatographypolishingstep
tovirusdepletionwasstudiedusingtwovirus-spikingtestsonMustangQAcrodisc
units(Table7).
Theviralclearancecapacityofthemembranechromatographystepwasdemonstrated
forMuLV(log10reductionfactor>4).LowerLRFvalueswereobtainedforMVM
undertheprocessconditionstested(LRF~1).AstheMVMclearancemaybe
affectedbycompositionofthefeedstock,loadconductivitycouldbefurtheroptimized.
Table 7
Results of viral clearance during the virus-spiking scale-down study
Membrane
adsorber
Log10 reduction
factor (LRF) (MVM)
Log10 reduction
factor (LFR) (MuLV)
Test 1
Test 2
1.30 ± 0.41
1.12 ± 0.33
4.60 ± 0.29
> 5.16 ± 0.27
However,theoverallviralclearancecapacityoftheproductionprocesswasquitehigh
withaLRF≥9forMVMandLRF≥13forMuLV(Table8).Amoderatereductionwas
shownforMVMduringaffinitychromatography(LRF>3),effectivereductionforMuLV
duringlowpHtreatmentandduringtheMustangQmembranechromatography
polishingstep(LRF>4)andeffectivereduction,forbothvirusesapplyingthe20nm
nanofiltration(LRF>4).
Table 8
Assessment of the virus clearance capacity of the manufacturing process.
Data expressed as Log10 values.
Process steps
Cell removal
Affinity chromatography
Low pH treatment
CEX chromatography
Mustang Q membrane
chromatography
TFF
Nanofiltration
0.22 µm filtration
Final assessment
Virus clearance capacity
Log10 reduction
factor (LRF) for MuLV
Log10 reduction
factor (LRF) for MVM
Not tested
Not tested
4.23 ± 0.27
Not tested
4.60 ± 0.29
Not tested
3.16 ± 0.36
Not tested
Not tested
<2
Not tested
≥ 4.49 ± 0.28
Not tested
Not tested
≥ 6.17 ± 0.26
Not tested
LRF ≥ 13.3
LRF ≥ 9.3
Conclusion of 4.2.4
Overallviralclearancecapacityoftheentireprocesswassufficientformanufacturing
ofasafedrugsubstanceforclinicaltrialsphase1.
8
5.
General Conclusions
•AMustangQmembranechromatographypolishingstepwassuccessfullyimplementedina
flowthroughmodeforremovalofresidualDNAinanIgG1 purificationprocessatproduction-scale,
followingtwoconventionalcolumnchromatographysteps(proteinAandcationexchange).
• ItwasefficienttoremovehostcellresidualDNAbelowtheLLOQoftheqPCRassayappliedforanalysis.
•Additionally,itshowedahighIgG1 recovery(>96%),independentofthescaleapplied.
•ContributionoftheMustangQmembranechromatographypolishingsteptoHCPremovalwasshown.
•EffectiveMuLVvirusremovalusingMustangQmembranechromatographywasdemonstratedbased
onavirus-spikingscale-downstudywhileMVMvirusclearanceneededfurtheroptimization.
•Overallviralclearancecapacitywassufficientformanufacturingofasafedrugsubstancefor
clinicaltrials.
•Mustangmembranechromatographycapsulesaredisposableandofferspeedandease-of-use
forpolishingapplications
Acknowledgements
Stefan Franke, Stefan Hartmann, Martin Suhr (ProBioGen AG, Berlin)
Virginie Brenac Brochier, Christophe Egrot, Russell Jones, Sylvio Bengio, Martin Glenz, Knut Thiele
(Pall Life Sciences)
Visit us on the Web at www.pall.com/biopharm
E-mail us at biopharm@pall.com
Corporate Headquarters
Port Washington, NY, USA
+1.800.717.7255tollfree(USA)
+1.516.484.5400phone
biopharm@pall.come-mail
European Headquarters
Fribourg, Switzerland
+41(0)263505300phone
LifeSciences.EU@pall.come-mail
Asia-Pacific Headquarters
Singapore
+6563896500phone
sgcustomerservice@pall.come-mail
International Offices
PallCorporationhasofficesandplantsthroughouttheworldinlocationssuchas:Argentina,Australia,
Austria,Belgium,Brazil,Canada,China,France,Germany,India,Indonesia,Ireland,Italy,Japan,Korea,
Malaysia,Mexico,theNetherlands,NewZealand,Norway,Poland,PuertoRico,Russia,Singapore,South
Africa,Spain,Sweden,Switzerland,Taiwan,Thailand,theUnitedKingdom,theUnitedStates,andVenezuela.
Distributorsinallmajorindustrialareasoftheworld.TolocatethePallofficeordistributornearestyou,visit
www.pall.com/contact.
Theinformationprovidedinthisliteraturewasreviewedforaccuracyatthetimeofpublication.Productdata
maybesubjecttochangewithoutnotice.ForcurrentinformationconsultyourlocalPalldistributororcontact
Palldirectly.
© 2012, Pall Corporation. Pall,
, Acrodisc and Mustang are trademarks of Pall Corporation.
® indicates a trademark registered in the USA and TM indicates a common law trademark.
Filtration.Separation.Solution. is a service mark of Pall Corporation. uSuperdex and ÄKTA are
trademarks of GE Healthcare. Picogreen is a trademark of Invitrogen.
2/12,PDF,GN11.7526
USTR2827(1)