Download A new strategy for quantitative proteomics using isotope

A new strategy for quantitative proteomics using isotope-coded protein labels
Alexander Schmidt, Cornelia Ciosto, Josef Kellermann and Friedrich Lottspeich
Max-Planck-Institute of Biochemistry, Martinsried, Germany
1
3
Overview
The aim of the following studies was the development of a method
suitable for comparative analysis of two highly complex protein
mixtures. This includes differential quantitative determination of protein
expression, post-translational modifications and isoforms. In order to
obtain this information, the whole sequence of every differential
expressed protein has to be analyzed.
4
Workflow
Mass spectrometry
1) Quantitation
Comparative analysis of two
different cells or states (e.g.
comparision of healthy and
tumor cells or tissues)
Cell state 1
The ratio of an isotope labelled peptide pair is calculated by comparing
their intensities. This provides relative quantitative information about
the differential protein expression of the 2 cell states.
Cell state 2
2) Identification
This challenge was achieved by isotope labelling of the high abundant
free amino groups in proteins, which allows the usage of any
separation method for proteins and peptides. Quantitation and
identification of differential expressed proteins is then performed using
high throughput mass spectrometry.
Denaturation and
alkylation
Only peptides from differential expressed proteins (ratio not 1:1) are
then selected for identification by sequencing (MS/MS).
∆m
Detect labelled
peptide pairs
Introduction
Select differentially
expressed peptides
6
TOF TOF Reflector Spec #1[BP = 1685.6, 74102]
90
80
%%Intensity
Intensity
70
60
50
40
30
20
10
1520
2140
2760
3380
4000
m/z
Mass (m/z)
TOF TOF Reflector Spec #1[BP = 1685.6, 74102]
Peptide: K*VGINYWLAHK*ALCSE
%
%Intensity
Intensity
70
70
50
d4
40
60
40
30
20
20
10
10
2098.4
2101.8
2105.2
2108.6
2D
2736
2742
2748
2754
FFE
1D
TOF TOF MS/MS Precursor 2098.59 Spec #1=>BC=>NF0.7=>BC=>BC[BP = 949.4, 894]
y7
894.3
I
K*
S
E
m/z
Enzymatic
cleavage of proteins
into peptides
Reduction of
complexity on
peptide level
5
Mass spectrometry
Results
! The sequence coverage of the proteins
ranged from 32.5 to 82.3% (labelled
peptides only)
MS-parameters
Sample preparation:
The modified proteins were digested, diluted (except in-gel digests) and
mixed 1:1 with a α-cyano-4-hydroxycinnamic acid (CHCA) solution. 0.4 µl
of the mixture containing 500 fmol of peptides were then applied to a
stainless steel MALDI target.
Instrument:
Proteomics analyzer 4700 (MALDI-TOF/TOF)
Database search:
GPS-Explorer / Mascot
Comparative analysis of healthy and
apoptotic human HEP-G2 cells
7
The aim of this study was the discovery of new
proteins that change during apoptosis and are
specific for apoptotic cells (marker proteins).
(For details refer to poster “MONACO –
Proteomic ...“ presented by Cornelia Ciosto).
Workflow
Healthy (125µg)
For this experiment, the heavy 4-2H-nicotinic
acid was replaced by the 6-13C version. This
ensures co-elution of derivatized peptides
during reversed phase liquid chromatography
(RP-LC) analysis which is required for accurate
protein quantification.
Apoptosis (125µg)
Label with
Label with
12C-Nic-NHS
13C-Nic-NHS
Pre-fractionation using SCX-LC
Fraction 11
Combine &
digest (Glu-C)
SCX-LC
20 SCX-Fractions
RP-LC of SCX-Fraction 11
! The ratio of the 6 proteins could be
determined very precisely, having a
standard deviation of less than 6%
90
A
Protein
identification by
database search
After derivatization, the combined protein mixtures
were enzymatically cleaved using endopeptidase
Glu-C. The obtained peptide mixture was then
applied to a MALDI target without further
separation and analyzed. The extended views are
each showing one peptide pair origin from
myoglobin and α-lactalbumin. Identification of the
peptide/protein is finally carried out by MSsequencing followed by database search.
2760
m/z
Mass (m/z)
2. Identification
MS/MS-spectra of selected peptide
100
LC
no
separation
d0
0
2730
2112.0
m/z
Mass (m/z)
d4
50
30
0
2095.0
Peptide:
LK*PLAQSHATK*HK*IPIK*YLE
80
d0
60
2.5E
Myoglobin (1:3)
90
%
%Intensity
Intensity
80
TOF TOF Reflector Spec #1[BP = 1685.6, 74102]
100
3.3E
α-Lactalbumin (2:1)
90
Reduction of
complexity on
protein level
To illustrate the ICPL workflow, 2 mixtures
containing the same 6 standard proteins in
various amounts were treated as described under
3. Isotopic labelling of all free amino groups was
7.4E+4
performed using activated nicotinic acid. As
nicotinic acid contains 4 hydrogens in the light
version which are replaced by 4 deuterium atoms
in the heavy version, the mass difference per
modified amino group is 4.
100
0
900
Peptide sequence
identification
by MS/MS
Labelling of standard proteins
1. Quantification
MS-spectra of labeled peptides
m/z
Relativ abundance
2
Combine
Quantitative proteome analyses usually are accomplished by 2Delectrophoresis (2DE) followed by mass spectrometric protein
identification. Although this method is well established, quantitative
determination is not accurate and the reproducibility of the 2D-gels is
very poor. Recent developments, like the ICAT reagent [1] or GIST [2]
methodology have shown to be powerful alternatives to comparative
2D gel imaging analysis. Nevertheless, these methods also have their
limitations. Here we describe a new method termed Isotope Coded
Protein Label (ICPL) which is based on isotopic labelling of all free
amino groups in proteins.
Compared to the ICAT reagent, that modifies the low abundant amino
acid cysteine in proteins, higher sequence coverage and thus more
information about post-translational modifications and isoforms are
obtained with ICPL.
With the GIST approaches isotope labelling of peptides is performed
after enzymatic cleavage of the proteins. Although almost every
peptide is modified using this strategy, the highly demanded
quantitatively controlled separation dimension on the protein level is
lost.
The efficiency of the ICPL method is demonstrated by comparative
analysis of two E. coli – proteomes spiked with different amounts of
five standard proteins.
100
Relativ abundance
ICPL label
MALDI-Spot 36
RP-LC
3840 MALDI-SPOTS
80
V G
I
N
Y
W
L
A
L
C
S
E
! All analyzed peptides containing a free
amino group have been modified
b11
b10
b6
30
y9
b5
20
0
70
K*
a1
40
10
H
y8
60
50
A
b2
b3
b4
486
b7
y6
b8
b12
y10
y11
b9
902
1318
y12
b13
y14
y13
1734
b14
*
b15
! No side reactions, that would result in
additional isotopic peptide pairs including
no free amino groups, could be identified
2150
m/z
Mass (m/z)
MALDI-TOF Spectrum of Spot 36
Results
! In total 3840 MS-spectra (13.3h) and 13462
pairs for MS/MS-analysis (37.4h) were
acquired
TOF TOF Reflector Spec #1 MC=>NF0.7[BP = 1412.7, 6612]
100
100
90
90
80
60
40
30
20
30
10
20
0
1141.0
In summary, the new ICPL approach presented here provides accurate quantitative
determination and high sequence coverage of differential expressed proteins even when
analyzing highly complex protein mixtures. With the opportunity of using any separation
method and protein sample, this new strategy is suitable to challenge comprehensive
quantitative proteome analysis.
1149.8
2760
1154.2
1158.6
1163.0
3380
y6
TOF TOF MS/MS Precursor 1144.56 Spec #1=>BC=>NF0.7[BP = 110.1, 16886]
SHHK*AK*GK*
100
! 25 acetylation, 2 phosphorylation and 2
methylation sites could be identified
2140
4000
MALDI-MS/MS Spectrum of the indicated Peptide
Pair identified as Histone H2AA (Ac at K127)
! This corresponds to 416 unique proteins
(average = 2.3 peptides/protein)
! 31 proteins were found to be up- and 33
down-regulated during apoptosis
1145.4
Mass (m/z)
1520
m/z
Mass
(m/z)
90
% Intensity
Conclusion
7777
650
13C
50
50
40
0
900
Relative abundance
8
TOF TOF Reflector Spec #1 MC=>NF0.7[BP = 1412
70
60
10
! Only 952 of 13462 pairs (7.07%) could be
correctly identified
12C
80
70
% Intensity
K*
% Intensity
Relative abundance
% Intensity
% Intensity
70
O
80
70
60
50
b2
N
b3
N
b4 b5
40
30
0
200
y5
y3
20
10
380
4885.6
O
Ac
560
740
Mass (m/z)
m/z
b6
920
1100