The Mtb Proteome Library:
Development and application of assays for targeted MS analysis
of the complete proteome of Mycobacterium tuberculosis by
SRM and SWATH-MS
Olga Schubert
Group of Prof. Ruedi Aebersold
ETH Zurich
Skyline User Meeting 2013 Minneapolis
Mycobacterium tuberculosis (Mtb)
•
Mycobacterium tuberculosis is the causative agent of Tuberculosis (TB)
•
One third of the world’s population latently infected with Mtb
•
1.7 million deaths from TB each year
 More efficient treatments urgently needed
•
Limited availability of techniques to measure proteins with high
sensitivity, selectivity and reproducibility
Traditional approach: Antibodies
Only for few targets,
high cost, low
throughput
Aim
To generate a resource of validated assays for
the sensitive detection and accurate
quantification of every protein of
Mycobacterium tuberculosis, even in complex
backgrounds.
The Mtb Proteome Library contains SRM
assays for the entire proteome of Mtb
A
Phase I
Proteome Mapping
Bacterial
cultures
B
Phase II
Proteome Library Generation
Prediction
Scoring
C
Phase III
Proteome Library Validation
Bacterial
cultures
Peptide
selection
Protein
extraction
Protein
extraction
Peptide
synthesis
Proteolysis
+
Proteolysis
-
pI = pH
MS
Rv0001
Rv0002
Rv0003
Rv0004
Rv0005
Rv0006
Rv0007
MS accessible
proteome
Peptide
fractionation
Discovery MS
Pools of
96 peptides
MS
Peptide and
protein
identification
SRM-MS2
No
fractionation
MS
Fragment
ion selection
Synthetic
proteome library
SRM
Quantitative
SRM trace
Validated
proteome library
Schubert et al. Cell Host & Microbe 2013
The Mtb Proteome Library: A Resource of Assays to Quantify the Complete Proteome of Mycobacterium tuberculosis.
Saturation
Definition of the MS-accessible
Mtb proteome by discovery MS
A
B
Extensive fractionation and shotgun MS on Orbitrap XL
≥3 pep/prot
2 pep/prot
1 pep/prot
0 pep/prot
23%
4%
4%
69%
Isoelectric focusing (off-gel electrophoresis)
3074
proteins
(77%)
Proteome
mapping
C
100%
3,074 (77%)
Discovery MS
80%
80%
60%
40%
20%
0%
RNA sequencing
456
(13%)
60%
Saturation
100%
414
(12%)
2,618
(75%)
40%
Acquired data
Simulation all
Simulation true
20%
0%
0
0
0
0
0
0
0
00 ,00 ,00 ,00 ,00 ,00
,
50 100 150 200 250 300
RNA-seq
data: Arnvig et al.,matches
PLoS Pathogens 2011
Peptide-spectrum
Total: 3,488 (87%)
100% corresponds to the
4,012 annotated ORFs in Mtb
(TubercuList v2.3)
Generation of SRM assays
using crude synthetic peptides
100%
80%
60%
85%
≥3 pep/prot
2 pep/prot
1 pep/prot
0 pep/prot
C
20%
Not detected
28%
39%
16%
≥3 pep/prot
2 pep/prot
1 pep/prot
0 pep/prot
17%
20%
SRM validated
Mtb Proteome Library
0%
62,884
7 8 9
10 11 12 13 14 15 16 17 18 19 20 21
(72%)
Fraction
peptidesper
per bin
bin
Fraction
of of
peptides
Fraction of peptides per bin
Peptides with assay
100%
40%
60%
40%
60%
No MS Previously
evidence observed
Total
Mtb unscheduled
Mtb scheduled
C
Mtb+human unsche
20%
Peptides
with assay
Not detected Mtb+human schedu
Peptides
with assay
Not detected
100%
0%
100%
10
0
1
2
3
4
5
6
80%
80%
8
Transitions
40%
60%
60%
40%
40%
20%
20%
0%
0%
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75
Fraction of peptides per bin
B
C
60%
80%
80% 0%
B
80%
100%
100%20%
40%
Synthetic 0%
Mtb Proteome Library
No MS (97%)
Total
3894 proteins
3,894
(97%) Previously
evidence
observed
B
Fraction of peptides
with SRM assay
8%
Fraction of peptides
with SRM assay
A
A
3%
4%
Peptides
A
6 7Hydrophobicity
8 9 10 11 12(SSRCalc)
13 14 15 16 17 18 19 20 21
Length (amino acids)
Length (amino acids)
d-Score < 1.78
m-Score < 0.01
d-Score < 1.78
1
.0
m-Score < 0.01
17,463 crude synthetic
peptides (JPT) measured in pools ofE ~100 on a Qtrap 4000 in SRM-triggered
MS2 mode
D
6
4
2
Increasing SRM/SWATH assay specificity and
throughput by using iRTs and scheduled SRM
iRT peptides spiked into each sample allow to
determine a chromatography-independent
retention time (iRT) for each peptide.
Escher et al., Proteomics 2012, Using iRT, a normalized retention time for more targeted measurement of peptides
Theoretical assessment of SRM assay
specificity using the SRMCollider
C
100%
Peptides
80%
60%
Mtb unscheduled
Mtb scheduled
Mtb+human unscheduled
Mtb+human scheduled
40%
20%
0%
0
1
2
3
4
Transitions
5
6
Röst et al., MCP 2012, A computational tool to detect and avoid redundancy in selected reaction monitoring
80%
60%
Validation of the Mtb Proteome
Library in
C
unfractionated whole cell lysates by SRM
20%
0%
40%
mProphet analysis
SRM
validatedMtb
MtbProteome
Proteome
Library
SRM2884
validated
Library
proteins
(72%)
2884
at(72%)
1% FDR
2,884
(72% of the Mtb proteome)
2
3
4
Transitions
0
.8
4
0
0
0
1
3
0
0
0
0
5
6
0
.2
17%
d
e
c
o
y
t
a
r
g
e
t
decoy
target
q
0
.0
16%
≥3 pep/prot
2 pep/prot
1 pep/prot
0 pep/prot
Mtb unscheduled
Mtb scheduled
d-Score < 1.78
m-Score < 0.01
E unsched
Mtb+human
Mtb+human scheduls
2
0
0
0
39%
Transition groups
28%
0%
1
.0
D
20%
0%
20%
Length (amino acids)
60%
1
0
0
0
B
40%
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
0
Synthetic Mtb Proteome Library
3,894 (97%)
Peptides
80%
60%
0
.6
100%
Fraction
fr
a
c
tio
n
85%
40%
80%
0
.4
≥3 pep/prot
2 pep/prot
1 pep/prot
0 pep/prot
Fraction of peptides per
Fraction of peptides per
8%
−3
−2
.
9
4
−2
−2
.
0
1
−1
−1
.
0
7
0
1
3
−0
.
1
30
.
4
91
.
1
11
.
7
42
2
.
3
62
.
9
9
n
o
r
m
a
liz
e
d
d
is
c
r
im
in
a
n
ts
c
o
r
e
(
d
_
s
c
o
r
e
)
Normalised
discriminant
score
(d-Score)
To validate all these SRM assays, over 200 scheduled SRM runs were needed.
Reiter et al., Nature Methods 2011, mProphet: automated data processing and statistical validation for large-scale SRM experiments
−3
The Mtb Proteome Library contains SRM
assays for the entire proteome of Mtb
A
Phase I
Proteome Mapping
Bacterial
cultures
B
Phase II
Proteome Library Generation
Prediction
Scoring
C
Phase III
Proteome Library Validation
Bacterial
cultures
Peptide
selection
Protein
extraction
Protein
extraction
Peptide
synthesis
Proteolysis
+
Proteolysis
-
pI = pH
MS
Rv0001
Rv0002
Rv0003
Rv0004
Rv0005
Rv0006
Rv0007
Peptide
fractionation
Discovery MS
Pools of
96 peptides
MS
Peptide and
protein
identification
MS accessible
proteome
3074 (77%) proteins
SRM-MS2
No
fractionation
MS
Quantitative
SRM trace
Fragment
ion selection
Synthetic
proteome library
3894 (97%) proteins
SRM
Validated
proteome library
2884 (72%) proteins
Schubert et al. Cell Host & Microbe 2013
The Mtb Proteome Library: A Resource of Assays to Quantify the Complete Proteome of Mycobacterium tuberculosis.
The Mtb Proteome Library is a publicly available resource
of MS reference data, SRM assays and their validation
www.PeptideAtlas.org/PASSEL
www.PeptideAtlas.org
www.SRMAtlas.org
Application of the Mtb Proteome Library to study the
dynamics of the DosR regulon of Mtb under hypoxia
Mycobacterium bovis BCG
A
Culture density (OD600)
4
No aeration
Re-aeration
A
B
3
2
1
0
-5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11
Time (days)
C
Re-aeration
Fold change
512
256
128
64
32
16
8
4
2
1
0.5
0.25
0
6h
1
2
3
4 5
Time (days)
6
7
8
9
Application of the Mtb Proteome Library to study the
dynamics of the DosR regulon of Mtb under hypoxia
Log2
Log
2
fold
change
fold
change
≤-2.0
≤-2.0
-1.5
-1.5
-1.0
-1.0
-0.5
-0.50.0
0.00.5
0.51.0
1.01.5
1.5
≥2.0
≥2.0
Rv1998c|Rv1998c
Rv1998c|Rv1998c
Rv1735c|Rv1735c
Rv1735c|Rv1735c
Rv2830c|vapB22
Rv2830c|vapB22
Rv1812c|Rv1812c
Rv1812c|Rv1812c
Rv2027c|dosT
Rv2027c|dosT
Rv0573c|pncB2
Rv0573c|pncB2
Rv0570|nrdZ
Rv0570|nrdZ
Rv1736c|narX
Rv1736c|narX
Rv0082|Rv0082
Rv0082|Rv0082
Rv2006|otsB1
Rv2006|otsB1
Rv3841|bfrB
Rv0083|Rv0083
Rv3841|bfrB
Rv2630|Rv2630
Rv0083|Rv0083
Rv2631|Rv2631
Rv2630|Rv2630
Rv0571c|Rv0571c
Rv2631|Rv2631
Rv2004c|Rv2004c
Rv0571c|Rv0571c
Rv2005c|Rv2005c
Rv2004c|Rv2004c
Rv0081|Rv0081
Rv2005c|Rv2005c
Rv2003c|Rv2003c
Rv0081|Rv0081
Rv1997|ctpF
Rv2003c|Rv2003c
Rv2629|Rv2629
Rv1997|ctpF
Rv3132c|devS
Rv2629|Rv2629
Rv3133c|devR
Rv3132c|devS
Rv1813c|Rv1813c
Rv3133c|devR
Rv2028c|Rv2028c
Rv1813c|Rv1813c
Rv2624c|Rv2624c
Rv2028c|Rv2028c
Rv1996|Rv1996
Rv2624c|Rv2624c
Rv0572c|Rv0572c
Rv0079|Rv0079
Rv1996|Rv1996
Rv2625c|Rv2625c
Rv0572c|Rv0572c
Rv3134c|Rv3134c
Rv0079|Rv0079
Rv0080|Rv0080
Rv2625c|Rv2625c
Rv2627c|Rv2627c
Rv3134c|Rv3134c
Rv0569|Rv0569
Rv0080|Rv0080
Rv2029c|pfkB
Rv2627c|Rv2627c
Rv2030c|Rv2030c
Rv0569|Rv0569
Rv2623|TB31.7
Rv2029c|pfkB
Rv1738|Rv1738
Rv2030c|Rv2030c
Rv3131|Rv3131
Rv2623|TB31.7
Rv2032|acg
Rv1738|Rv1738
Rv2007c|fdxA
Rv3131|Rv3131
Rv3127|Rv3127
0 6h 1 2 4 6 7 8 Rv2032|acg
Rv3130c|tgs1
Rv2626c|hrp1
Time (days) Rv2007c|fdxA
Rv2031c|hspX
Rv3127|Rv3127
0 6h 1 2 4 6 7 8 Rv3130c|tgs1
Time (days) Rv2626c|hrp1
Rv2031c|hspX
Statistical analysis of SRM data by SRMstats
Chang et al., Mol Cell Proteomics 2012. Protein significance analysis in SRM measurements
Absolute label-free quantification by SRM exploits the
linear correlation of the sum of the top transitions of the
top peptides per protein and the protein concentration
C
A
B
Mean
100
1.2
6
Concentration (fmol/µg)
5
2.30
0.1
2.15
1
2
2.25
1
3
0.0
2.20
10
4
1.0
0
2.0
2.35
6
y = 0.97x – 5.63
R2 = 0.83
n = 34
Density
0.4
0.8
3.0
fold error
1000
Transitions
2 3 4 5
Log10(concentration)
[fmol/ug]
A
-1.0
0.01
5
6
7
8
log10(MS intensity)
9
1
C
2
Proteins
3 4
6
21955 proteins
Peptides
Linear correlation established using 34 anchor proteins quantified by AQUA peptides
MS intensity: sum of 2 most intense transitions of the 3 most intense peptides per protein
Ludwig et al., MCP 2011, Estimation of absolute protein quantities of unlabeled samples by SRM
Proteome-wide absolute
abundance estimates for Mtb
2-Component
Systems
Ribosome
CH Met
Nucleotide Metabolism
Lipid Metabolism
Vit Met
AA Met
ABC
Transporter
Xenobiotics
Degradation
TCA
A
1000
Concentration (fmol/µg)
Bacterial
Secretion
100
10
1
Colour code corresponds to subfigure A.
0.1
0.01
Proteins
SWATH-MS:
Data-independent
Targeted
mass spectrometric
techniques: SRM acquisition
and SWATH
with targeted data extraction
SRM
SWATH-MS
TARGETED data
acquisition
data-independent
acquisition
Q1
Q2
0.7 Da
Q3
Q1
0.7 Da
precursor
selection
fragmentation
Q2
TOF
10 ppm
25 Da
fragment
selection
precursor
selection
fragmentation
scanning
intensity
intensity
TARGETED data extraction
time
time
time
time
Gillet et al., MCP 2012, Targeted data extraction of the MS/MS spectra generated by data-independent acquisition: a new concept for consistent
and accurate proteome analysis.
Figure by Christina Ludwig
Generation of the Mtb SWATH library
Synthetic
peptides
Fractionated
lysates
Whole-cell lysates
of different stress
conditions
SWATH-MS
TripleTOF shotgun mode
Rv0569 GATIDQPDHR 2+
•
Search each dataset with Mascot and Sequest
• iProphet combination of all datasets
• Alignment of runs into a common RT space (iRT)
• Consensus spectral library generation with SpectraST
• Extraction of the top transitions per precursor
Rv0569 FGAVQSAILHAR 3+
Mtb SWATH library
3931 proteins (98%)
39,479 peptides
Rv1738 ELVGVGLAR 2+
SRM
SWATH-MS allows reproducible, high proteome
coverage measurements of Mtb in a single run
SWATH-MS
33%
44%
14%
9%
≥3 pep/prot
2 pep/prot
1 pep/prot
0 pep/prot
28%
39%
16%
Rv0569 GATIDQPDHR 2+
17%
Proteome coverage by
SWATH-MS
Proteome coverage by
SRM
single injection
2683 proteins (67%)
>200 injections
2884 proteins (72%)
(openSWATH software)
(mProphet software)
Rv0569 FGAVQSAILHAR 3+
Rv1738 ELVGVGLAR 2+
SRM
Summary
•
The Mtb Proteome Library is a public resource of SRM assays for the entire
proteome of Mtb
–
SRM assays generated from crude synthetic peptides and validated in whole cell lysates
–
Data analysis with Skyline and supporting tools: iRT peptides, SRMCollider, mProphet, SRMstats
–
Data can be browsed on and downloaded from www.SRMAtlas.org, www.PeptideAtlas.org/passel
•
Application of the Mtb Proteome Library to study the dynamics of the DosR
regulon of Mtb under hypoxia
•
Absolute label-free quantification by SRM exploits the linear correlation of the sum
of the top transitions of the top peptides per protein and its absolute concentration.
•
Expansion of the Mtb Proteome Library for use with SWATH-MS
•
SWATH-MS allows reproducible, high proteome coverage measurements of Mtb in
a single run
ETH Zurich
o Prof. Ruedi Aebersold
o Christina Ludwig
o Jeppe Mouritsen
o George Rosenberger
o Hannes Röst (Mon 5:45 pm and WP 595, Wed)
o Ludovic Gillet (TOD pm, Tue 5:30 pm)
o Ben Collins (WP 688, Wed)
o Alessio Maiolica, Mariette Matondo
Acknowledgements
University of Ghana
o Dr. Patrick K. Arthur
Max Planck Institute Berlin
o Prof. Stefan Kaufmann
o Dr. Martin Gengenbacher
ISB Seattle
o Prof. Rob Moritz
o Dave Campbell
o Zhi Sun
o Terry Farrah
o Samuel Bader (ABSciex, Mon 7 am)
orig. tríptico SysteMTb exterior ok traz.pdf
20/7/10
21:03:17
University of Washington
o Brendan MacLean (TP 499, Tue)
C
M
UBS
Promedica
Foundation
Institute of
Molecular
Systems
Biology