Quantitative Proteomics Wailap Victor Ng May 20, 2004

1
Quantitative Proteomics
Wailap Victor Ng Institute of Biotechnology in
Medicine Institute of Bioinformatics Department
of Biotechnology and Laboratory Science in
Medicine National Yang Ming University
2
Driving force of Technology Development

• Identification of differentially expressed
  (tissue specific) genes and proteins,
• Monitor expression difference upon perturbations

3
Differential Expression Profiling Tools

• Microarray measure the changes in
  transcrtipomes
• Proteomics measure the relative level of
  proteins and modifications of the final effectors
  in different cellular states
• Other methods
• Subtracted cDNA hybridizations
• Differential display
• cDNA sequencing
• RDA (representation difference assay e.g. PCR
  Select)

4
Why do we need Q-proteomics?

• The levels of mRNA changes can not be translated
  directly into equivalent levels of protein
  changes
• Serum markers, for instance, can only be directly
  identified by proteomics approaches
• Detection of post-translational modifications

5
Applications of Quantitative Proteomics

• Indentify differenial expressed protein in
  different states
• Detect alternation in protein phosphorylation
• Glycosylation of proteins in different states
• Protein complexe characterization
• Protein-protein interactions

6
Quantitative Proteomics Methods

• 2-D polyacrylamide gel electrophoresis
• - conventional 2-D gel
• - Cy3 and Cy5 labeled protein 2-D gel
• Stable isotope labeling of proteins
• - ICAT
• - ICAT derivatives
• Label free quantification
• - SELDI protein profiles
• - Peptide profiles

7
Stable Isotope Labeling Methods

• Metabolic labeling
• Chemical labeling
• Enzymatic labeling

8
Issues to considered

• Which isotope should be used?
• What is the purity of the labeling reagent?
• How many isotope labeled residues will be present
  in each peptide?
• Are there labeling biases?
• Will the labeling tag remain intact during
  peptide ion fragmentation?

9
Metabolic LabelingSILAC (Stable isotope
labeling with amino acids in cell culture)
10
Labeling Reagents

• 13C- and 12C-Lysine (heavy light 6 D)
• 13C- and 12C-Arginine (heavy light 6 D)

11
SILAC (Stable isotope labeling with amino acids
in cell culture)
(98)
Prostate cancer cell line PC3
PC3M (low metastatic potential)
PC3M-LN4 (high metastatic potential)
Everley et al., MCP 2004
12
Labeling Efficiency of both isotopes
Ratio 0.98
Figure 3 C. Control of SILAC peptide (IISLDAK)
pair from glutamyl-propyl tRNA synthetase. PC3M
cells were cultured under both C12 and C13
conditions. Equal amount of proteins was mixed
for SILAC.
Everley et al., MCP 2004
13
Difference of Io and I6 3 mass units ? MH 2
Mascot ? MSQuant Ratio areas of monoisotopic
peaks
Figure 3. Identification and quantification of
SILAC peptides
Everley et al., MCP 2004 http//msquant.sourceforg
e.net
14
Figure 2. Distribution of protein expression
ratios (PC3M-LN4/PC3M) as determined by SILAC.
Total number of quantified protein, 444 60
upregulated and 22 downregulated in the highly
metastatic PC3M-LN4.
Everley et al., MCP 2004
15
Everley et al., MCP 2004
16
Everley et al., MCP 2004
17
Figure 4. Immunoblots of selected proteins from
microsomal lysates of PC3M and PC3M-LN4 cell lines
Everley et al., MCP 2004
18
Chemical LabelingIsotope Coded Affinity
Tag(ICAT)
19
Proteins
TCEP Tris(2-carboxyethyl)phosphate (reduction)
Iodoacetamide ICH2CONH2 (alkylation)
20
Isotope Coded Affinity Tag (ICAT)
Heavy reagent d8-ICAT ( X deuterium ) Light
reagent d0-ICAT ( X hydrogen )
442.2 or 450.2
O
N
N
O
O
X
X
X
X
O
I
cys
S H
N
O
O
N
S
X
X
X
X
Biotin tag
Linker (heavy or light)
Thiol specific reactive group
Gygi et al. Nature Biotech, 17994, 1999
21
ICAT DATA Analysis Software
Sequest Express (Finnigan) Sequest/Comet,
Express, ASAP (ISB) Pro ICAT (ABI/MDS Sciex)
22
The ICAT strategy for quantifying differential
protein expression. Two protein mixtures
representing two different cell states have been
treated with the isotopically light and heavy
ICAT reagents, respectively an ICAT reagent is
covalently attached to each cysteinyl residue in
every protein. The protein mixtures are combined
and proteolyzed to peptides, and ICAT-labeled
peptides are isolated utilizing the biotin tag.
These peptides are separated by microcapillary
high-performance liquid chromatography. A pair of
ICAT-labeled peptides are chemically identical
and are easily visualized because they
essentially coelute, and there is an 8 Da mass
difference measured in a scanning mass
spectrometer (four m/z units difference for a
doubly charged ion). The ratios of the original
amounts of proteins from the two cell states are
strictly maintained in the peptide fragments. The
relative quantification is determined by the
ratio of the peptide pairs. Every other scan is
devoted to fragmenting and then recording
sequence information about an eluting peptide
(tandem mass spectrum). The protein is identified
by computer-searching the recorded sequence
information against large protein databases.
Gygi et al. Nature Biotech, 17994, 1999
23
MS
MS/MS
Figure 3 Isotope-coded affinity tag quantitative
analysis of a protein from the mixture. (A)
Full-scan (5001,500 m/z) mass spectrum at time
19.76 min of the microcapillary LC-MS and
LC-MS/MS mixture analysis. Shown are at least
four different peptide doublets eluting from the
column. Each doublet corresponded to a pair of
ICAT-labeled peptides of identical sequence. The
mass-to-charge (m/z) ratio difference between
peptides is dependent on the charge state (number
of hydrogen ions) and is typically either 4.0 or
8.0 (mass difference of 8 Da and a charge state
of one or two). (B) Expanded view of full-scan
mass spectrum showing the ion abundances for each
species of an ICAT-labeled peptide eluting from
the column at 19.76 min. (C) Reconstructed ion
chromatograms for the peptide ions measured in
(B). The ratio of the calculated areas (0.54) was
used to determine the relative peptide
concentrations in the two mixtures.
Gygi et al. Nature Biotech, 17994, 1999
24
Gygi et al. Nature Biotech, 17994, 1999
25
Quantitative proteomics analysis of yeast grown
in ethanol versus galactose
26
Figure 5 Isotope-coded affinity tag analysis of
alcohol dehydrogenase isozymes in yeast.(A) ADH1
converts acetaldehyde to ethanol when cells are
growing on hexose sugars. When other carbon
sources are not available, ethanol can be
utilized as the sole carbon source in yeast. This
depends on ADH2 for conversion of ethanol back
into acetaldehyde as the first step in the
pathway where acetyl-CoA is used to fuel the
tricarboxylic acid (TCA) cycle and the glyoxylate
cycle for energy utilization in the cell24. (B)
The peptides identified from ADH1 and ADH2 are
95 identical and differ only by a single amino
acid substitution (valine for threonine). The
substitution shifted the retention time of the
ADH2 peptide pair by 2 min and shifted the mass
by 2 Da, which allowed for unambiguous
identification and analysis of gene expression
levels.
Gygi et al. Nature Biotech, 17994, 1999
27
Gygi et al. Nature Biotech, 17994, 1999
28
Characterization of yeast RNA polymerase II
transcription preinitiation complex
29
RNA polymerase II transcription complex isolation
Pst I
HIS4 promoter linked to magnetic beads

M
ACT
TATA
activator

rTBP

CRC
CRC
M
M
TATA
TATA
CRC
CRC


CRC
CRC
M
M
TATA
TATA
CRC
CRC
Combine
Proteolyze Fractionate mLC/MS/MS
Ranish et al. 2003. Nat Genet 33349
30
Protein composition after affinity purification
B
TOA2 peptide K.NCQVTVEDSHR.D
A
C
kD
TBP
-TBP
Light Heavy 1.0 36.5
TBP
TOA2 (TFIIA)
TFIIB peptide K.ITMLCDAAELPK.I
TFIIB
Light Heavy 1.0 5.2
SRB4
Kin28 peptide R.WTAVQCLESDYFK.E
KIN28 (TFIIH)
1
2
Light Heavy 1.0 4.1
silver stain
western
ICAT
Ranish et al. 2003. Nat Genet 33349
31
Systematic analysis of Microarray and ICAT data
32
Comparison of mRNA and protein expression
patterns.
PNAS 9914913, 2002
33

• The bR regulon and the isoprenoid, carotenoid,
  and bR biomodules
• The arginine synthesis and fermentation biomodules

Ratio bat-bat
PNAS 9914913, 2002
34
Problems of the Prototype ICAT Reagents

• Chromatographic property of heavy (deuterium) and
  light (hydrogen) ICAT labeled peptides are not
  identical
• The tag is bulky ( 442.2 or 450.2 mass unit)
• Complicate labeling procedure leads to low
  recovery of labeled peptides

35
Acid Cleavable ICAT Reagents (ABI)
http//docs.appliedbiosystems.com/pebiodocs/043452
22.pdf
36

• Enhancements to the Cleavable ICAT Reagents
• The Heavy ICAT reagent is now 13C-based instead
  of deuterium based.
• Heavy- and Light-labeled peptides coelute, which
  allows quantification by mass spectrometry (MS).
• The mass difference between the Heavy and Light
  reagents is now 9 Da instead of 8 Da, eliminating
  potential confusion between oxidized methionine
  and doubly labeled peptides.
• The biotin portion of the ICAT reagent tag is
  cleaved with acid after the ICAT reagent-labeled
  peptides are eluted from the avidin cartridge.
  Biotin cleavage reduces the size of the reagent
  label on the peptide from 442 Da to 227 Da, which
  allows analysis of larger peptides.
• MS/MS sequence coverage is improved due to
  reduced fragmentation, which improves database
  searching and confidence in protein
  identifications.
• Tris(2-carboxyethyl)phosphine (TCEP)/ICAT reagent
  by-products of labeling are substantially
  reduced, which improves MS data quality.

37
Enzymatic LabelingTrypsin catalyzed 16O-to-18O
exchange
38

• Trypsin catalyzed 16O-to-18O exchange for
  comparative proteomics tandem mass spectrometry
  comparison using MALDI-TOF, ESI-QTOF, and ESI-ion
  trap mass spectrometers
• Heller M, Mattou H, Menzel C, Yao X.
• Am Soc Mass Spectrom. 14(7)704-18,2003

39
equal amount
40
Figure 1. Tuning of ESI-IT for optimal isolation
efficiency of the entire isotope cluster of
16O/18O-labeled peptide ions. A horse
apomyoglobin digest labeled with 16O- or
18O-water, respectively, was mixed at a
theoretical ratio of 11 in 1 acetic acid/MeCN
11 at 1 pmol/ l concentration. This solution was
delivered to the ESI-IT using a syringe pump at a
flow rate of 2 l/min. The doubly charged tryptic
peptide ions LFTGHPETLEK (m/z 636.34),
HGTVVLTALGGILK (m/z 689.93), VEADIAGHGQEVLIR
(m/z 803.93), and GHHEAELKPLAQSHATK (m/z
927.49), respectively were isolated in the ion
trap with the isolation parameters set as
described in the Experimental Methods section.
Isolation was done either on the I0 (left panel)
or the I4 isotope (right panel) resulting in mean
18O/16O-ratios of 0.94 0.17 and 1.13 0.31.
The small black diamonds denote the isolated
target mass.
41
(a) 1 2
(b) 2 1
42
Expression Ratio Calculation Methods

• Precursor ion intensities of MS scan
• y-ion intensity
• Extracted ion chromatogram (EIC) peak areas

43
Ion trap MS/MS and various ratio calculation
methods
44
Sample Apomyoglobin
45
Figure 4. Comparison of ion trap LC-ESI-MS and
MALDI-TOF-MS peak resolution for 18O/16O-ratio
calculations. A human plasma subfraction
containing low molecular weight proteins was
digested with trypsin and one half each was
enzymatically labeled with either 16O or 18O,
respectively. The peptides of the combined
aliquots were fractionated by cation exchange
chromatography and one particular fraction was
subjected to RP chromatography with on-line
ESI-MS/MS on an ion trap and fraction collection
for MALDI-TOF. On the left, the combined MS
survey scans of the precursor ion pairs submitted
to MS/MS in the ion trap are shown together with
the corresponding MALDI-MS of the same peptide
precursor ion on the right. At the top of each
pair of MS spectra, the protein identified by
means of the ESI-IT MS/MS spectra is given
together with the corresponding 18O/16O-ratio as
calculated by I4/I0 and I5/I1. The charge state
of the detected ion on the ESI-IT MS as well as
the m/z values for each of the first monoisotopic
peak in the 16O/18O peptide pair are marked as
well.
No exchange
Incomplete exchange
Space charge effect (IT)
46
Possible Problems in Enzymatic Labeling

• The kinetics for oxygen exchange can differ
  considerably between different peptides
• Space charge effect second isotopic peak can be
  disturbed and thus not detectable in ion trap MS

47
Enzymatic versus ICAT

• Enzymatic labeling
• All peptides are labeled ? higher coverage
• In vivo labeling of proteins not applicable to
  human studies
• ICAT
• Only cycteine containing peptides are labeled ?
  reduce peptide complexity

48
Enzymatic labeling de novo sequencing (5050
16O/18O labeling of the C-termini)
ESI-QTOF
ESI_Ion Trap
49
Solid phase ICAT
50
Schematic representation of the Solid-phase
isotope tagging
Zhou et al. Nature Biotechnology 20, 512 - 515
(2002)
51
Summary of the number of proteins identified and
quantified by the solid-phase and ICAT methods.
(A) Number of proteins identified from
large-scale experiment (L), in which 100 µg total
protein sample was labeled and 20 µg was analyzed
by LC-MS/MS. (B) Number of proteins identified
from small-scale experiment (S), in which 10 µg
of total sample was labeled and 5 µg analyzed.
(C) Number of proteins identified by the
solid-phase method in large-scale (L) and
small-scale (S) experiment.
Zhou et al. Nature Biotechnology 20, 512 - 515
(2002)
52
Chemical Labeling Glycosylated protein
profiling
53
Identification and quantification of N-linked
glycoproteins using hydrazide chemistry, stable
isotope labeling and mass spectrometry Hui
Zhang, Xiao-jun Li, Daniel B Martin Ruedi
Aebersold Nature Biotechnology 21660 - 666, 2003
54
Protein Glycosylation

• Most common form of post translational
  modification
• O-linked glycosylation
• - serine and threonine
• N-linked glycosylation
• - asparagine (- N - X S/T X proline)
• - PNGaseF (remove the sugar moiety)

55
Figure 1 Schematic diagram of quantitative
analysis of N-linked glycopeptides.(a) Proteins
from two biological samples are oxidized and
coupled to hydrazide resin. Nonglycosylated
peptides are removed by proteolysis and extensive
washes. The nonglycosylated peptides are
optionally collected and analyzed. The N-terminus
of glycopeptides are isotope labeled by succinic
anhydride carrying either d0 or d4. The beads are
then combined and the isotopically tagged
peptides are released by PNGase F and analyzed by
MS. (b) Oxidation of a carbohydrate to an
aldehyde followed by covalent coupling to
hydrazide resin.
Na periodate
hydrazide resin
56
(No Transcript)
57
Figure 2 Isolation of glycoproteins from
serum.(a) Specific capture of glycoproteins by
hydrazide resin. Total protein staining or
glycoprotein staining of serum before (-) and
after () capture of glycoproteins by hydrazide
resin. Proteins were separated by SDS-PAGE and
stained with silver (left) or Gel Code Blue
glycoprotein staining reagent (right). (b)
Comparison of total number of proteins or
peptides identified from serum samples isolated
by either cysteine-reactive tag or glycopeptide
capture method.
58
Figure 3 Quantitative analysis of formerly
N-linked glycosylated peptides by MALDI QTOF
MS.(a) The peptides were analyzed by MALDI-QqTOF
MS with an initial MS scan to quantify the
protein. The ratio of each peptide was analyzed
by comparison of the total area of the first
three monoisotopic peaks for the d0 and d4 forms
of the peptide. (b) An example of a peptide with
specific m/z value (1579.7431) was selected for
CID spectrum to identify the peptide and N-linked
glycosylation site.
59
Figure 4 Subcellular location of glycoproteins
identified from a crude microsomal fraction of
LNCaP prostate epithelial cells.
60

• Lectin affinity capture, isotope-coded tagging
  and mass spectrometry to identify N-linked
  glycoproteins
• Hiroyuki Kaji et al.
• Nature Biotechnology 21667 - 672, 2003

IGOT Isotope-coded glycosylation-site-specific
tagging
61
Figure 1 Schematic representation of the IGOT
strategy. (a) A subset of glycoproteins is
collected by lectin-affinity chromatography from
a biological mixture. After tryptic digestion of
the glycoproteins, the glycopeptides are
recovered by the same lectin and then are treated
with peptide-N-glycosidase in H218O to remove the
sugar moieties and to incorporate the
'glycosylation-site-specific' tag. The peptides
are finally analyzed by the multi-dimensional
LC-MS-based protein identification technology.
(b) PNGase-mediated conversion of the
glycosylated asparagine to aspartic acid and
concomitant incorporation of 18O from the solvent
H218O.
62
Label Free Quantification
63
SELDI by Ciphergen

• Based on patented Surface Enhanced Laser
  Desorption/Ionization (SELDI) technology,
  Ciphergens ProteinChip Systems offer a single,
  unified platform for a multitude of proteomics
  research applications, including characterize,
  and validate predictive protein biomarkers and
  biomarker patterns, right in their lab.

http//www.ciphergen.com/doclib/docFiles/262.pdf
64

• Biomarker discovery
• Expression Difference Mapping
• Interaction Difference Mapping
• Antibody-antigen interaction
• DNA-protein interaction
• Protein ID/peptide mapping
• Protein purification
• Epitope mapping
• Glycosylation analysis
• Phosphorylation/signal transduction
• Protein-protein interaction
• Receptor-ligand interaction
• Toxicity markers
• Clinical trials stratification

http//www.ciphergen.com/doclib/docFiles/262.pdf
65
http//www.ciphergen.com/doclib/docFiles/262.pdf
66
http//www.ciphergen.com/doclib/docFiles/262.pdf
67
http//www.ciphergen.com/doclib/docFiles/262.pdf
68
http//www.ciphergen.com/doclib/docFiles/262.pdf