Poster Note 64 8 0 2
Complete Characterization of a Cysteinelinked Antibody-Drug Conjugate Performed
on a Hybrid Quadrupole-Orbitrap Mass
Spectrometer with High Mass Range
Aaron O. Bailey , Eugen Damoc , Stephaneof
Houela
, Kai
Scheffler , Jonathan L. Josephs
Complete Characterization
Cysteine-linked
Antibody-Drug Con
1
2
1
3
1
Thermo Fisher Scientific, 1San Jose, CA, USA; 2Bremen, Germany; 3Dreieich, Germany
Aaron
O.
Bailey
Eugen Damoc
Damoc22,, Stephane
Stephane Houel
Houel11,, Kai
Kai Scheffler
Scheffler33,, Jonathan
Jonathan L.
L. Josephs
Josephs11
Aaron O. Bailey11,, Eugen
Thermo Fisher
Fisher Scientific,
Scientific, 11San
San Jose,
Jose, CA,
CA, USA;
USA; 22Bremen,
Bremen, Germany;
Germany; 33Dreieich,
Dreieich, Germany
Germany
Thermo
ABSTRACT
ABSTRACT
RESULTS
RESULTS
Wehave
havemodified
modifiedthe
theinstrument
instrumentcontrol
controlsoftware
softwareof
ofaabenchtop
benchtopquadrupole-Orbitrap
quadrupole-Orbitrapmass
mass
We
spectrometerto
toadd
addnative
nativeMS
MScapability.
capability.In
Inthis
thisstudy
studywe
wedemonstrate
demonstratecomplete
complete
spectrometer
characterizationof
ofBrentuximab
Brentuximabvedotin,
vedotin,aacysteine-linked
cysteine-linkedADC,
ADC,which
whichrequires
requiresnative
nativeMS
MS
characterization
conditionsfor
forintact
intactanalysis.
analysis.We
Wedemonstrate
demonstratepreservation
preservationof
ofnon-covalent
non-covalentbonding
bondingof
of
conditions
antibody
subunits
during
electrospray
ionization.
HMR
mode
can
be
turned
off
for
peptide
antibody subunits during electrospray ionization. HMR mode can be turned off for peptide
mapping.We
Weuse
usetrypsin
trypsinpeptide
peptidemapping
mappingapproach
approachwith
withHCD
HCDfragmentation
fragmentationto
toachieve
achieve99%
99%
mapping.
coverageof
ofthe
theBrentuximab
Brentuximabvedotin
vedotinsequence
sequenceusing
usingaasingle
singleLC-MS
LC-MSanalysis
analysisof
ofaa90
90min
min
coverage
reversephase
phasegradient.
gradient.Finally,
Finally,we
wedemonstrate
demonstratethat
thatsignature
signatureions
ionsspecific
specificfor
forHCD
HCD
reverse
fragmentationof
ofBrentuximab
Brentuximabvedotin
vedotincan
canbe
beutilized
utilizedto
toincrease
increaseMS/MS
MS/MSassignment
assignment
fragmentation
confidence.
confidence.
DENATURING LC-MS,
LC-MS, CYSTEINE-LINKED
CYSTEINE-LINKED ADC
ADC
DENATURING
INTRODUCTION
INTRODUCTION
Intactprotein
proteinLC/MS
LC/MSanalysis
analysisconventionally
conventionallyinvolves
involvesusing
usingmobile
mobilephases
phaseswhich
whichare
arecomprised
comprisedof
of
Intact
organicand
andacidic/basic
acidic/basicpH,
pH,often
oftensuited
suitedspecifically
specificallyfor
forreverse
reversephase
phasechromatography.
chromatography.This
This
organic
strategycan
canbe
beuseful
usefulfor
forachieving
achievinghigh
highresolution
resolutionprotein
proteinseparations.
separations.Conditions
Conditionssuch
suchas
asthese,
these,
strategy
however,are
arenot
notcompatible
compatiblefor
forperforming
performingintact
intactanalysis
analysison
oncertain
certainclasses
classesof
ofcompounds
compoundswhich
which
however,
require
preservation
of
non-covalent
bonds
to
maintain
structural
integrity,
such
as
cysteine-linked
require preservation of non-covalent bonds to maintain structural integrity, such as cysteine-linked
ADCs.We
Wedemonstrate
demonstratethis
thisphenomenon
phenomenonusing
usingthe
thecysteine-linked
cysteine-linked ADC
ADCBrentuximab
Brentuximabvendotin.
vendotin.
ADCs.
Denaturing(reverse
(reversephase)
phase)LC/MS
LC/MSanalysis
analysisof
ofBrentuximab
Brentuximabvedotin
vedotinresults
resultsinindetection
detectionof
ofroughly
roughly
Denaturing
1
collisionally-inducedm/z
m/z
sixunraveled
unraveledforms
forms(Figure
(Figure3A-C).
3A-C).We
Weobserved
observedaapreviously-reported
previously-reported1 collisionally-induced
six
718
fragment
of
the
fragile
vcMMAE
linker-drug
(Figure
3B).
Upon
deconvolution
we
also
718 fragment of the fragile vcMMAE linker-drug (Figure 3B). Upon deconvolution we also
observed
a
mass
corresponding
to
light
chain
with
addition
of
one
vcMMAE
and
a
loss
of
observed a mass corresponding to light chain with addition of one vcMMAE and a loss of
approximately762
762Da.
Da.
approximately
Thecomplexity
complexity of
ofmodern
modern therapeutic
therapeuticproteins,
proteins,such
suchas
asantibody-drug
antibody-drugconjugates
conjugates(ADCs),
(ADCs),
The
Figure3.
3. Denaturing
DenaturingLC-MS
LC-MSanalysis
analysis
Figure
(A)Unmodified
Unmodifiedsample
sample(1
(1ug)
ug)was
wasanalyzed
analyzedby
byreverse
reversephase
phase
(A)
chromatographycoupled
coupledtotoaaQQExactive
ExactivePlus
PlusOrbitrap
OrbitrapMS
MS
chromatography
operatingininHMR
HMRmode
modeand
andproduced
producedseveral
severalpeaks.
peaks.(B)
(B)The
The
operating
resultingaveraged
averagedMS
MSspectrum
spectrumisisaacomplex
complexmixture
mixtureofof
resulting
chargestate
stateenvelopes
envelopesas
aswell
wellas
asaapreviously
previouslydescribed
described11
charge
vcMMAE-specificreporter
reporterfragment
fragmention
ionatatm/z
m/z718.
718.(C)
(C)Data
Data
vcMMAE-specific
analysiswith
withReSpect
ReSpectdeconvolution
deconvolutionand
andSliding
SlidingWindow
Window
analysis
integration
show
roughly
six
covalently-structured
forms
integration show roughly six covalently-structured forms ofof
unraveledcysteine-linked
cysteine-linkedADC.
ADC.We
Wedetect
detectaaprotein
proteinspecies
species
unraveled
whichcorresponds
correspondstotoaalight
lightchain
chainwith
withaddition
additionofofone
onelinker
linker
which
drugand
andaaloss
lossofof762
762Da,
Da,which
whichisisalso
alsopresent
presentininthe
theraw
raw
drug
spectrum.
spectrum.
AA
100
100
80
80
Relative
RelativeAbundance
Abundance
presentaagreat
greatanalytical
analyticalchallenge
challengewhich
whichrequires
requireshigh
highresolution
resolutionchromatography
chromatographycombined
combined
present
withhigh
highresolution
resolutionmass
massspectrometry.
spectrometry.Complementary
ComplementaryMS
MSapproaches
approachessuch
suchas
aspeptide
peptide
with
mapping
and
intact
mass
analysis
are
needed
for
complete
characterization
of
therapeutic
mapping and intact mass analysis are needed for complete characterization of therapeutic
proteins.Cysteine-linked
Cysteine-linkedADCs
ADCspresent
presentaaunique
uniquechallenge
challengefor
forcharacterization
characterizationas
asproper
proper
proteins.
intactanalysis
analysisrequires
requiresnative
nativeMS
MSconditions
conditionsto
topreserve
preservestructurally-critical
structurally-criticalnon-covalent
non-covalent
intact
bindingbetween
betweenantibody
antibodychains.
chains.We
Wehave
havemodified
modifiedcommercially-available
commercially-availableThermo
Thermo
binding
TM Q ExactiveTM
TM Plus and Q ExactiveTM
TM HF OrbitrapTM
TM mass spectrometers to
ScientificTM
Scientific
Q Exactive Plus
and Q Exactive HF
Orbitrap mass
spectrometers to
perform
native
LC-MS
experiments.
In
the
present
study,
we
demonstrate
this
capabilitywith
with
perform native LC-MS experiments. In the present study, we demonstrate this capability
intactanalysis
analysisof
ofBrentuximab
Brentuximabvedotin,
vedotin,aacysteine-linked
cysteine-linkedADC
ADC(Figure
(Figure1).
1).Additionally,
Additionally,we
wehave
have
intact
performeddenaturing
denaturingLC-MS
LC-MSand
andpeptide
peptidemapping
mappingon
onthese
thesesame
sameinstruments
instrumentsto
togenerate
generate
performed
complementarydatasets
datasetsfor
forcomplete
completecharacterization.
characterization.
complementary
Reversephase
phasechromatography
chromatography
Reverse
20
20
00 6
6
Figure1.
1. Schematic
Schematicfor
forConstructing
ConstructingCysteine-Linked
Cysteine-LinkedADC
ADC
Figure
mAb
mAb
drug
drug
linker
linker
linker-drug
linker-drug
Cys-linked
Cys-linked
ADC
ADC
DrugtotoAntibody
AntibodyRatio
Ratio
Drug
(DAR)distribution
distribution==0-8
0-8
(DAR)
vcMMAE
vcMMAE
“linker”
“linker”
mAb SS
mAb
Pairsofoflinker-drugs
linker-drugsare
are
Pairs
attachedatatsites
sitesofof
attached
interchain-disulfide
bonds
interchain-disulfide bonds
“drug”
“drug”
BB
80
80
70
70
60
60
50
50
00
500
500
valinespacer
valinespacer
citrulline
citrulline
Brentuximabvedotin
vedotinwas
wasprepared
preparedfor
forpeptide
peptidemapping
mapping(reduction,
(reduction,alkylation,
alkylation,and
andtrypsin
trypsindigestion)
digestion)or
orintact
intact
Brentuximab
analysis(no
(notreatment).
treatment).For
Fordenaturing
denaturingLC-MS
LC-MSintact
intactanalysis
analysis11μg
μgofofprotein
proteinsamples
sampleswere
wereseparated
separatedusing
usingaa10
10
analysis
and0.1%
0.1%formic
formicacid
acid(Thermo
(ThermoMAb-Pac
MAb-PacRP;
RP;flow
flowrate
rate250
250μL/min).
μL/min).For
For
mingradient
gradientofof10-90%
10-90%ACN
ACNininHH2OOand
min
2
native
LC-MS
intact
analysis
10
μg
of
sample
was
desalted
online
using
size
exclusion
chromatography
native LC-MS intact analysis 10 μg of sample was desalted online using size exclusion chromatography
TM BEH SEC 4.6x150mm; 50 mM NH OAc isocratic elution, flow rate 300 μL/min) and directly presented
(WatersTM
BEH SEC 4.6x150mm; 50 mM NH44OAc isocratic elution, flow rate 300 μL/min) and directly presented
(Waters
themass
massspectrometer
spectrometervia
viaelectrospray
electrosprayionization.
ionization.Peptide
Peptidemapping
mappingwas
wasperformed
performedusing
using2.5
2.5μg
μgofofsample
sample
totothe
and0.1%
0.1%formic
formicacid
acid(Acclaim
(AcclaimRSLC
RSLC120
120C18;
C18;flow
flowrate
rate
separatedusing
usingaa90
90min
mingradient
gradientofof2-90%
2-90%ACN
ACNininHH2OOand
separated
2
250
μL/min).
Commercially-available
Orbitrap
mass
spectrometers
(Q
Exactive
HF
and
Q
Exactive
Plus)
which
250 μL/min). Commercially-available Orbitrap mass spectrometers (Q Exactive HF and Q Exactive Plus) which
weremodified
modifiedtotoinclude
includeHigh
HighMass
MassRange
Range(HMR)
(HMR)mode
modetotoallow
allowimproved
improvedhigh
highmass
masstransmission
transmissionand
andscanning
scanning
were
uptotom/z
m/z8000.
8000.Native
Nativeintact
intactand
anddenaturing
denaturingMS
MSdata
datawere
wereacquired
acquiredininHMR
HMRmode
modeatatsetting
settingofofR=15k
R=15kor
or17.5k
17.5k
up
TM algorithm and Sliding Window integration in Thermo ScientificTM
anddeconvolved
deconvolvedusing
usingthe
theReSpect
ReSpectTM
algorithm and Sliding Window integration in Thermo ScientificTM
and
TM 1.0 SP1 software. Deconvolution species were identified automatically using the publiclyBioPharma
Finder
TM
BioPharma Finder 1.0 SP1 software. Deconvolution species were identified automatically using the publiclyavailableFASTA
FASTAsequence
sequencefor
forBrentuximab
Brentuximabvedotin,
vedotin,aamass
masstolerance
toleranceofof50
50ppm,
ppm,and
andaastatic
staticmodification
modificationofof
available
Gln>Pyro-Glufor
forthe
theheavy
heavychain.
chain.Peptide
Peptidemapping
mapping data
datawere
wereacquired
acquiredby
bydata
datadependent
dependentselection
selectionwith
withR=60k
R=60k
Gln>Pyro-Glu
or70k
70kfor
forFull
FullMS
MSand
andR=15k
R=15kor
or17.5k
17.5kfor
forMS/MS.
MS/MS.Peptide
Peptidemapping
mappingdata
datawere
weresearched
searchedusing
usingthe
theMassAnalyzer
MassAnalyzer
or
algorithm
in
BioPharma
Finder
software
with
a
tolerance
of
5
ppm.
algorithm in BioPharma Finder software with a tolerance of 5 ppm.
Figure2.
2. LC-MS
LC-MSInstrumentation
Instrumentationfor
forComplete
CompleteADC
ADCCharacterization
Characterization
Figure
Allexperiments
experimentswere
wereperformed
performedusing
usingaaVanquish
VanquishUHPLC
UHPLCconnected
connectedtotoaaExactive
ExactiveHF
HFor
orQQExactive
ExactivePlus
Pluswith
with
All
HighMass
MassRange
Range(HMR)
(HMR)mode.
mode.
High
TM
ExactivePlus/
Plus/
QQExactive
TM
HFOrbitrap
OrbitrapTM
HF
MS
MS
with HMR mode
with HMR mode
11
11
12
12
762.5026
762.5026
1000
1000
1500
1500
2000
2000
2500
2500
m/z
m/z
3000
3000
3500
3500
4000
4000
4500
4500
CC
monomethylauristatin
auristatinEE
monomethyl
MATERIALS AND
AND METHODS
METHODS
MATERIALS
VanquishTM
Vanquish
UHPLC
UHPLC
10
99
10
Time(min)
(min)
Time
BioPharma
BioPharma
TM
FinderTM
Finder
software
software
25041.47
G0F
25041.47
22xxG0F
(32.5 ppm)
125678.81
G0F
125678.81
11xxG0F
100 (32.5 ppm)
(16.0ppm)
ppm)
76675.44
100
(16.0
76675.44
90
(1.0ppm)
ppm)
90
(1.0
80
80
2xxG0F
G0F
2
70
(-761.96Da)
Da)
70
103272.01
(-761.96
G0F
G0F
103272.01
60
22xxG0F
11xxG0F
60
(5.6ppm)
ppm)
148086.47
54269.74
(5.6
50
148086.47
54269.74
50
2xxG0F
G0F
(28.1ppm)
ppm)
(4.6ppm)
ppm)
40
2
(28.1
(4.6
40
105907.20
24279.51
30
105907.20
24279.51
30
(23.5ppm)
ppm)
20
(23.5
20
10
10
00
20000 30000
30000 40000
40000 50000
50000 60000
60000 70000
70000 80000
80000 90000
90000 100000
100000 110000
110000 120000
120000 130000
130000 140000
140000 150000
150000 160000
160000
20000
Mass
Mass
Relative
RelativeIntensity
Intensity
attachment
attachment
group
group
88
SheathGas:
Gas:10
10
Sheath
AuxGas:
Gas:55
Aux
AuxGas
GasTemp:
Temp:50ºC
50ºC
Aux
CapTemp:
Temp:320ºC
320ºC
Cap
S-Lens:200
200
S-Lens:
In-sourceCID:
CID:60
60
In-source
R=17500
17500
R=
718.5121
718.5121
40
40
30
30
20
20
10
10
77
Denaturing LC-MS
LC-MS
Denaturing
100
100
90
90
Relative
RelativeAbundance
Abundance
rentuximabvedotin
vedotinisisaacysteine-linked
cysteine-linkedADC
ADCwhich
whichisisconstructed
constructedby
bymodifying
modifyingan
anantibody
antibodywith
withvcMMAE,
vcMMAE,aa
BBrentuximab
preformedlinker-drug
linker-drugcomprised
comprisedofofaavaline-cirtuline-based
valine-cirtuline-basedlinker
linkerand
andaamonomethyl
monomethylauristatin
auristatinEEtoxic
toxicdrug.
drug.
preformed
Saturaturated(8
(8drugs)
drugs)cys-linked
cys-linkedADCs
ADCsare
areheld
heldintact
intactwith
withonly
onlynon-covalent
non-covalentbinding.
binding.
Saturaturated
Brentuximab
Brentuximab
vedotin
vedotin
60
60
40
40
NATIVE INTACT
INTACT LC-MS,
LC-MS, CYSTEINE-LINKED
CYSTEINE-LINKED ADC
ADC
NATIVE
NativeMS
MSintact
intactprotein
proteinanalysis
analysisallows
allowsdirect
directobservation
observationof
ofmolecules
moleculeswhich
whichrely
relyon
onnonnonNative
covalentinteractions
interactionsto
topreserve
preservecritical
criticalstructural
structuralfeatures,
features,such
suchas
asmaintaining
maintaininginterchain
interchain
covalent
associations
which
hold
together
cysteine-linked
ADCs.
The
use
of
100%
aqueous
physiological
associations which hold together cysteine-linked ADCs. The use of 100% aqueous physiological
pHbuffers
buffersininnative
nativeMS
MSanalysis
analysisproduces
producesdecreased
decreasedcharge
chargestates
states(increased
(increasedm/z)
m/z)and
and
pH
improvesmass
massseparation
separationof
ofheterogeneous
heterogeneousmixtures.
mixtures. We
Weperformed
performednative
nativesize
sizeexclusion
exclusionLCLCimproves
MSand
andobserved
observed55distinct
distinctspecies
speciescorresponding
correspondingto
tointact
intactBrentuximab
Brentuximabvedotin
vedotinwith
with0,
0,2,
2,4,
4,6,
6,
MS
or
8
vcMMAE
linker-drugs
(Figure
4).
We
measured
an
average
drug-to-antibody
ratio
of
4.07,
or 8 vcMMAE linker-drugs (Figure 4). We measured an average drug-to-antibody ratio of 4.07,
2.
which
is
consistent
with
a
previously
published
studies
reporting
3.9-4.2
drugs
per
antibody
2
which is consistent with a previously published studies reporting 3.9-4.2 drugs per antibody .
PEPTIDE MAPPING, CYSTEINE-LINKED ADC
A fundamental component of biotherapeutic protein characterization is peptide mapping.
onjugate Performed on a Hybrid Quadrupole-Orbitrap Mass Spectrom
A
NATIVE INTACT LC-MS, CYSTEINE-LINKED
ADC
Figure
5. Peptide
of CysteineWhereas
intactMapping
mass analysis
aims to detect the abundances and distributions of mass
Linked
ADC Brentuxmab
vedotin
deviation
combinations,
peptide mapping allows highly sensitive analysis of site-specific
Figure 4. Native intact LC-MS analysis
Size exclusion chromatography
(A) Unmodified sample (10 ug) was analyzed using size
Native
MS intact protein
allows
direct observation
of molecules which rely on nonexclusion
chromatography
coupled analysis
to a Q Exactive
Plus
100
Brentuximab
covalent
interactions
to preserve
critical structural features,
such as maintaining
interchain
50 mM
Orbitrap
MS operating
in HMR mode.
Buffer exchange
80
vedotin physiological
occursassociations
online as ADCwhich
forms elute
a single peak,
holdas
together
cysteine-linked ADCs. NH
The
use of 100% aqueous
4OAc
60
isocratic
followed
by
a
second
peak
corresponding
separated
pH buffers in native MS analysis produces decreased charge states (increased m/z) and
buffer
elution
buffer salts. (B) Averaging 2 min chromatographic time
40
improves mass separation of heterogeneous mixtures.
We
performed native size exclusion LCproduces a native intact MS spectrum which includes all
salts
20to intact Brentuximab vedotin with 0, 2, 4, 6,
MS and
observed
5 distinct
species corresponding
DAR forms
(DAR
0-8). (C) ReSpect
deconvolution
and
8 vcMMAE
linker-drugs
(Figurepeak
4). We
measured
0 an average drug-to-antibody ratio of 4.07,
SlidingorWindow
integration
can accommodate
tailing
0
2
4
6
8
102
to report
quantitatively
accurate
abundances
for the
which
is consistent
with
a previously
published studies
reporting
3.9-4.2
drugs
per antibody
.
Time (min)
mixture of DAR forms which have diverse elution profiles.
A pattern of lower abundance species were detected
corresponding to a low abundance loss of 762 Da from
each glycoform at each DAR value (green arrows). (D)
Figure 4. Native intact LC-MS analysis
A
Based on the individual deconvolved abundances of the
Size exclusion chromatography
(A) Unmodified sample (10 ug) was analyzed using size
Sheath Gas: 40
G0F/G0F glycoform, we calculated an average DAR value
exclusion chromatography coupled to a Q Exactive Plus
100
Aux Gas: 5
of 4.07, which is consistent with previous reports2.
Brentuximab
50Aux
mM Gas Temp: 125ºC
Orbitrap MS operating in HMR mode. Buffer exchange
80
vedotin
NH4OAc
occurs online as ADC forms elute as a single peak,
Cap Temp: 275ºC
B 100 followed by a second peak corresponding separated
60
isocratic
S-Lens: 200
90 buffer salts. (B) Averaging 2 min chromatographic time
buffer
elution
40
In-source CID: 100
80 produces a native intact MS spectrum which includes all
salts
20
R= 17500
70 DAR forms (DAR 0-8). (C) ReSpect deconvolution and
0
Sliding
Window
integration
can
accommodate
peak
tailing
60
0
2
4
6
8
10
to report quantitatively accurate abundances for the
Time (min)
50
mixture of DAR forms which have diverse elution profiles.
40 A pattern of lower abundance species were detected
30 corresponding to a low abundance loss of 762 Da from
20 each glycoform at each DAR value (green arrows). (D)
Based on the individual deconvolved abundances of the
10
Sheath Gas: 40
G0F/G0F glycoform, we calculated an average DAR value
0
2.
Aux7200
Gas: 5 7400
of 4.07, 5400
which is 5600
consistent
with previous
reports
5200
5800
6000
6200
6400
6600
6800
7000
7600
Relative Abundance
Reduced, alkylated, and trypsin-digested sample (2.5
sequence
Thegradient
vcMMAE
linker-drug on Brentuximab vedotin poses particular
ug) was
separatedfeatures.
using a 90 min
and eluted
attempting
identify drug conjugation sites. We prepared a sample for
into achallenges
Q Exactive HFwhen
(equipped
with HMR to
mode)
operating
in Standard
mode.using
Using reduction
a mass accuracy
peptide
mapping
andcut
alkylation to block non-drug-conjugated cysteines,
off of 5 ppm, we achieved 99% sequence coverage of
followed by trypsin digestion. In one 90 min LC-MS gradient we were able to achieve 99%
both light and heavy chains. We detected known
sequence
both
light and heavy chains and detect peptides spanning all four
glycopeptides
and coverage
were able to for
detect
MMAEconjugated
peptides at all four
cysteines
circles)
drug conjugation
sites.
HCD(red
fragmentation
allowed detection of a peptide in the hinge
which are normally involved in interchain disulfide pairs
region of the heavy chain that is differentially modified with 0-2 vcMMAE drugs. As a result
in naked antibodies. A trypsin peptide sequence at the
elution
sustained
hingeefficient
region of the
heavyrequires
chain (red asterisk)
was delivery of high organic mobile phase.
present in forms ranging from 0-2 vcMMAE
conjugations.
A
missed
cleavage
peptide
contained
up
Figure 5. Peptide Mapping of Cysteineto 3 conjugations.
Native LC-MS
Figure 6. Hinge Region Peptide of Cysteine-linked ADC is Site of Multiple Conjugations
60
(A) Our data analysis in BioPharma Finder software resulted in detection of peptides which covered the hinge region
of the heavy chain (red asterisk). A faithfully-trypsin-cleaved THT-KPK peptide was detected with 0-2 vcMMAE
conjugations at cysteines (red circles) normally involved in interchain disulfide pairs. (B) Addition of vcMMAE to
peptides dramatically increases hyphobicity which results in poor elution and increased retention time. MS/MS
analysis of the (C) 1 linker-drug (both positional isomers) and (D) two linker-drug forms in BioPharma Finder allowed
clear sequencing of y-ions in the hinge peptides, and thus facilitated automatic detection.
50
2 x G0F
153355.40
,
Relative Intensity
0
5200
60
40
20
2 x G0F
50720.06
10
C
5400
5600
2 x G0F
155987.81
5800
147000
148000
6200
6400
m/z
(-762.25 Da)
2 x G0F
148086.53
0
6000
149000
DAR2 152000
150000
151000
6600
6800
DAR8
7000
7200
DAR4
153000
154000
7400
7600
B
A
2 x G0F
158632.20
155000
156000
DAR6
157000
158000
159000
50
160000
*
Mass Accuracy
(ppm)
DAR
Relative Intensity
100
80
DAR0
2 x G0F
Relative
153355.40
Abundance
,
DAR0 2 x G0F
11.7
DAR2
23.1
69.23
22.4
100.00
50720.06
6.77
60
DAR4
40
DAR6
40.5 (-762.25 Da)
69.75
DAR8
17.6
20
0
147000
2 x G0F
148086.53
148000
149000
150000
151000
Average
2 x G0F
Drug-to-Antibody
155987.81
Ratio (DAR)
4.07
10.61
152000
153000
154000
DAR8
0
0
C
2 x G0F
158632.20
156000
157000
158000
159000
160000
Mass
D
G0F/G0F
DAR
Mass Accuracy
(ppm)
B
A
100
80
THT-KPK
60
155000
40
20
0
100
Relative
Abundance
Average
DAR0 CYSTEINE-LINKED
11.7
6.77
PEPTIDE MAPPING,
ADC
Drug-to-Antibody
23.1protein characterization
69.23
A fundamental component of DAR2
biotherapeutic
is peptide
mapping.
Ratio
(DAR)
Whereas intact mass analysisDAR4
aims to detect
and distributions of mass
22.4 the abundances
100.00
deviation combinations, peptide
allows highly69.75
sensitive analysis of site-specific
DAR6mapping 40.5
sequence features. The vcMMAE
linker-drug
vedotin poses particular
DAR8
17.6 on Brentuximab
10.61
challenges when attempting to identify drug conjugation sites. We prepared a sample for
peptide mapping using reduction and alkylation to block non-drug-conjugated cysteines,
followed by trypsin digestion. In one 90 min LC-MS gradient we were able to achieve 99%
sequence coverage for both light and heavy chains and detect peptides spanning all four
drug conjugation sites. HCD fragmentation allowed detection of a peptide in the hinge
region of the heavy chain that is differentially modified with 0-2 vcMMAE drugs. As a result
efficient
elution requires
sustained delivery
of high organic mobile phase.
PEPTIDE
MAPPING,
CYSTEINE-LINKED
ADC
A fundamental component of biotherapeutic protein characterization is peptide mapping.
Whereas intact mass analysis aims to detect the abundances and distributions of mass
deviation combinations, peptide mapping allows highly sensitive analysis of site-specific
sequence features. The vcMMAE linker-drug on Brentuximab vedotin poses particular
challenges when attempting to identify drug conjugation sites. We prepared a sample for
peptide mapping using reduction and alkylation to block non-drug-conjugated cysteines,
followed by trypsin digestion. In one 90 min LC-MS gradient we were able to achieve 99%
sequence coverage for both light and heavy chains and detect peptides spanning all four
drug conjugation sites. HCD fragmentation allowed detection of a peptide in the hinge
region of the heavy chain that is differentially modified with 0-2 vcMMAE drugs. As a result
efficient elution requires sustained delivery of high organic mobile phase.
4.07
0
0invcMMAE
50
(A) Our data analysis in BioPharma Finder software resulted
detection of peptides which covered the hinge region
0
of the heavy chain (red asterisk). A faithfully-trypsin-cleaved
THT-KPK peptide was detected with 0-2 vcMMAE
100
THTCPPCPAPELLGGPSVFLFPPKPK
conjugations at cysteines (red circles) normally involved in interchain
1 vcMMAEdisulfide pairs. (B) Addition of vcMMAE to
50
peptides dramatically increases hyphobicity which results
in poor elution and increased retention time. MS/MS
0
100 and (D) two linker-drug forms in BioPharma Finder allowed
analysis of the (C) 1 linker-drug (both positional isomers)
2 vcMMAE
50 facilitated
clear sequencing of y-ions in the hinge peptides, and thus
automatic detection.
Relative Abundance
G0F/G0F
Heavy chain
100
Figure 6. hinge
Hinge
Region Peptide of Cysteine-linked
ADC
Site(min)
of60Multiple
0
10 20
30 is40
70 80 Conjugations
90 100 110 120
region peptide
Time50
Mass
D
100
y5
Heavy chain
hinge region peptide
*
20
100
D
Experimental
THTCPPCPAPELLGGPSVFLFPPKPK
y13
spectra
80
60
40
0
y8
y6
b3
20
500
C
y9
y7
y14
1000
m/z
y17
0
100
50
0
100
50
0
100
40
20
0
100
40
20
0
0
20
500
D
100
Predicted
spectra
Experimental
spectra
y13
60
y8
y6
b3
y7
500
1000
m/z
y9
y14
1500
40
Time50
(min)60
80
90
100 110 120
y13
y
y7
40
y8
y11
60
1000
m/z (min)
Time
y14
y17
80
1500
100
80
120
THT-KPK
60
40
20
0
100
y5
80
y13
60
40
y17
70
801 vcMMAE
60
y5
80
120
THT-KPK
40
2 vcMMAE
20
50
0
THT-KPK
60
100
40
0 20 10 20 30
00 vcMMAE
100
y5
100
80
80
60
0
1500
60
Time (min)
80
50
Predicted
spectra
40
100
Relative Abundance
20
DAR0
80
%B
30
100
Relative Abundance
40
Relative Abundance
DAR2
70
Relative Abundance
Relative Abundance
80
%B
DAR4
90
*
Aux Gas Temp: 125ºC
Cap Temp: 275ºC
S-Lens: 200
In-source CID: 100
DAR6 R= 17500
m/z
100
*
Relative Abundance
Relative Abundance
Relative Abundance
Reduced, alkylated, and trypsin-digested sample (2.5
ug) was separated using a 90 min gradient and eluted
into a Q Exactive HF (equipped with HMR mode)
operating in Standard mode. Using a mass accuracy cut
off of 5 ppm, we achieved 99% sequence coverage of
both light and heavy chains. We detected known
glycopeptides and were able to detect MMAEconjugated peptides at all four cysteines (red circles)
which are normally involved in interchain disulfide pairs
in naked antibodies. A trypsin peptide sequence at the
hinge region of the heavy chain (red asterisk) was
present in forms ranging from 0-2 vcMMAE
conjugations. A missed cleavage peptide contained up
to 3 conjugations.
Native LC-MS
B
C
Linked ADC Brentuxmab vedotin
y
20
0
y7
500
y8
1000
m/z
y11
y14
1500
2 Complete Characterization of a Cysteine-linked Antibody-Drug Conjugate Performed on a Hybrid Quadrupole-Orbitrap Mass Spectrometer with High Mass Range
y17
%B
(A) Our data analysis in BioPharma Finder software resulted in detection of peptides which covered the hinge region
of the heavy chain (red asterisk). A faithfully-trypsin-cleaved THT-KPK peptide was detected with 0-2 vcMMAE
conjugations at cysteines (red circles) normally involved in interchain disulfide pairs. (B) Addition of vcMMAE to
peptides dramatically increases hyphobicity which results in poor elution and increased retention time. MS/MS
analysis of the (C) 1 linker-drug (both positional isomers) and (D) two linker-drug forms in BioPharma Finder allowed
Figure 7. HCD Signature Fragment Ions for vcMMAE Linker-Drug
clear sequencing of y-ions in the hinge peptides, and thus facilitated automatic detection.
The light chain C-terminal peptide SFN-GEC is a conjugation site for vcMMAE. This modified peptide was
automatically identified by BioPharma Finder (left side top panel). Further manual inspection produced additional
fragment assignments for vcMMAE signature ions (right side top panel). Theoretical masses (top panel) were
Bmasses
100
A
calculated manually and matched to experimental
(bottom panel) within 5ppm (green boxes). A cleavage site
for the loss of 762 Da is shown (black box; theoretical monoisotopic mass = 762.5017). We observed a high
abundance ion at m/z 1366.6189 (orange box, asterisk)
50 which corresponds to the peptide-retaining fragment pair of a
762 Da loss with an additional loss of 2 protons, presumably due to formation of a seven-membered aromatic ring.
Heavy chain
hinge region peptide
100
Relative Abundance
*
0
THTCPPCPAPELLGGPSVFLFPPKPK
S
50
0
100
0
10 20 30
0 vcMMAE
40
Time50
(min)60
762.5017 +1
70
506.3593 +1
80
90
100 110 120
321.2178 +1
1 vcMMAE
50
152.1075 +1
0
100
0
D
80
THT-KPK
40
20
0
100
80
60
40
20
0
100
Relative Abundance
Relative Abundance
60
80
60
40
20
b3
0
500
Predicted
spectra
a3 -H2O 2+
152.1070
y5
321.2170
a2
y8
y6
y7
506.3584
b4
y9
1000 500
m/z
y13
b6
y14
Experimental
718.5105
spectra
100
40
718.5119 +1
40
100
20
0
100
1366.6189
•Native LC/MS intact analysis of Brentuximab vedotin resulted in detection of intact ADC forms,
DAR0-8. ReSpect deconvolution and Sliding Window integration showed an average DAR of 4.07,
consistent with previous studies.
•Acquisition of MS/MS spectra with HCD fragmentation on Q Exactive Plus and Q Exactive HF
Orbitrap mass spectrometers followed by data analysis with BioPharma Finder resulted in 99%
sequence coverage from a single 90 min gradient using 5 ppm mass tolerance.
60
m/z
y13
y3 -H2O
y
20
y7
500
1500
y8
y11
1000
m/z
•Addition of vcMMAE linker drug dramatically increases peptide hydrophobicity and retention time.
•Signature HCD fragment ions of linker-drug may allow additional means for identifying drugconjugated peptides.
REFERENCES
1. Janin-Bussat MC, Dillenbourg M, Corvaia N, Beck A, Klinguer-Hamour C. Characterization of
antibody drug conjugate positional isomers at cysteine residues by peptide mapping LC-MS analysis.
J Chromatogr B Analyt Technol Biomed Life Sci. 2015 Feb 15;981-982:9-13.
y5
80
0
120
THT-KPK
40
1000
80
60
y17
1500
60
Time (min)
80
Relative Abundance
100
•We have modified the control software in Q Exactive Plus and Q Exactive HF mass spectrometers
to add native MS capability.
2 vcMMAE
50
*1366.6213
+1
0
20
C
CONCLUSIONS
y14
1500
y17
2000
2. Debaene F, Boeuf A, Wagner-Rousset E, Colas O, Ayoub D, Corvaïa N, Van Dorsselaer A, Beck A,
Cianférani S. Innovative native MS methodologies for antibody drug conjugate characterization: High
resolution native MS and IM-MS for average DAR and DAR distribution assessment. Anal Chem.
2014 Nov 4;86(21):10674-83.
© 2016 Thermo Fisher Scientific Inc. Waters™ is a trademark of Waters Corporation. ReSpect™ is a trademark of
Positive Probability Ltd. All rights reserved. All trademarks are the property of Thermo Fisher Scientific and its
subsidiaries. This information is not intended to encourage use of these products in any manner that might infringe the
intellectual property rights of others.
CONCLUSIONS
•We have modified the control software in Q Exactive Plus and Q Exactive HF mass spectrometers
to add native MS capability.
•Native LC/MS intact analysis of Brentuximab vedotin resulted in detection of intact ADC forms,
DAR0-8. ReSpect deconvolution and Sliding Window integration showed an average DAR of 4.07,
consistent with previous studies.
•Acquisition of MS/MS spectra with HCD fragmentation on Q Exactive Plus and Q Exactive HF
Orbitrap mass spectrometers followed by data analysis with BioPharma Finder resulted in 99%
sequence coverage from a single 90 min gradient using 5 ppm mass tolerance.
•Addition of vcMMAE linker drug dramatically increases peptide hydrophobicity and retention time.
•Signature HCD fragment ions of linker-drug may allow additional means for identifying drugconjugated peptides.
REFERENCES
1. Janin-Bussat MC, Dillenbourg M, Corvaia N, Beck A, Klinguer-Hamour C. Characterization of
antibody drug conjugate positional isomers at cysteine residues by peptide mapping LC-MS analysis.
J Chromatogr B Analyt Technol Biomed Life Sci. 2015 Feb 15;981-982:9-13.
2. Debaene F, Boeuf A, Wagner-Rousset E, Colas O, Ayoub D, Corvaïa N, Van Dorsselaer A, Beck A,
Cianférani S. Innovative native MS methodologies for antibody drug conjugate characterization: High
resolution native MS and IM-MS for average DAR and DAR distribution assessment. Anal Chem.
2014 Nov 4;86(21):10674-83.
© 2016 Thermo Fisher Scientific Inc. Waters™ is a trademark of Waters Corporation. ReSpect™ is a trademark of
Positive Probability Ltd. All rights reserved. All trademarks are the property of Thermo Fisher Scientific and its
subsidiaries. This information is not intended to encourage use of these products in any manner that might infringe the
intellectual property rights of others.
www.thermofisher.com
©2016 Thermo Fisher Scientific Inc. All rights reserved. Waters is a trademark of Waters Corporation. ReSpect is a trademark of Positive
Probability Ltd. ISO is a trademark of the International Standards Organization. All other trademarks are the property of Thermo Fisher
Scientific and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific products. It is not
intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications,
terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representative
for details.
Africa +43 1 333 50 34 0
Australia +61 3 9757 4300
Austria +43 810 282 206
Belgium +32 53 73 42 41
Brazil +55 11 2730 3006
Canada +1 800 530 8447
China 800 810 5118 (free call domestic)
400 650 5118
Denmark +45 70 23 62 60
Europe-Other +43 1 333 50 34 0
Finland +358 10 3292 200
France +33 1 60 92 48 00
Germany +49 6103 408 1014
India +91 22 6742 9494
Italy +39 02 950 591
Japan +81 6 6885 1213
Korea +82 2 3420 8600
Latin America +1 561 688 8700
Middle East +43 1 333 50 34 0
Netherlands +31 76 579 55 55
New Zealand +64 9 980 6700
Norway +46 8 556 468 00
Russia/CIS +43 1 333 50 34 0
Singapore +65 6289 1190
Sweden +46 8 556 468 00
Switzerland +41 61 716 77 00
Taiwan +886 2 8751 6655
UK/Ireland +44 1442 233555
USA +1 800 532 4752
PN64802-EN 0816S