w.m. keck foundation biotechnology resource laboratories erol … · 2019-09-29 · automated 2d...
TRANSCRIPT
W.M. Keck Foundation W.M. Keck Foundation Biotechnology Resource Laboratories Biotechnology Resource Laboratories
Erol GulcicekErol GulcicekChristopher ColangeloChristopher Colangelo
Terence WuTerence Wu
http://info.med.yale.edu/nida-proteomics/
- Site is currently up with basic information
- Overview of profiling technologies has recently been
added since the url was first distributed
- Ideas for future features are very welcome
Yale/NIDA Neuroproteomics Center Fact Sheet
Title of Center Grant: Yale/NIDA Neuroproteomics Research Center
Theme of Center: "Proteomics of Altered Signaling in Addiction"
Principal Investigator: Kenneth R. Williams, Ph.D.Director, W.M. Keck Foundation Biotechnology Resource LaboratoryProfessor (Adj,) Research Dept. of Molecular Biophysics & BiochemistryYale University School of Medicine
Co-Principal Investigator: Angus C. Nairn, Ph.D.Professor, Department of PsychiatryYale University School of Medicine
#Grants Awarded: 2 Nationally
Total Funding Awarded to Yale:
$7.7 million
Total Budget Period: 9/23/2004 – 5/31/2009
Center Cores: Administrative, Bioinformatics and Biostatistics, Protein Identification, Protein Microarray; andProtein and Lipid Separation and Profiling
Center Investigators: 14 from Yale University, The Veterans Administration Medical Center (West Haven) Rockefeller University
Overall Goal:
To substantially increase understanding of the biochemical mechanisms that underlie substance abuse and its treatment.
Specific Aims:
Bring together Yale faculty working at the forefront of such keyneuroscience areas as
Signal transduction, Plasticity, Neuronal morphogenesis, Lipid metabolism in neuronal signaling and Synaptic functionSynaptic response to psychotropic drugs
with experts in
Proteomics, Biostatistics, Bioinformatics.
Proteomics of Altered Signaling in Addiction
Translational Research
Receptor-Mediated Signaling
Transcriptional Regulation
Endocytosis and Exocytosis
Regulation of the Cytoskeleton
Neuronal Development
Neuronal Plasticity
Cell Death
DIGE ICAT 2D-LC Beckman Coulter System MudPIT
Phosphoproteome Profiling
Lipid Profiling
Identification and Characterization of Signaling Proteins Whose Expression is
Altered by Drug Addiction
SerumBiomarkers
Targeted Proteomics
Clinical Studies of Drug Addiction
Yale/NIDA Neuroproteomics ProgramYale/NIDA Neuroproteomics Program
Initial Stage:Initial Stage:- Meet individual PI’s to:
- Better understand the PI’s research area
- Jointly determine best to implement proteomics technologies available
from the Cores.
- Review of current technologies
- Development of new technologies
- Timelines
- Some already know Core personnel and have already initiated studies in which
case please continue working with these staff and additionally, I would be very glad
to talk with you also if you have questions about other technologies.
Erol E. [email protected]
Primary Biological SamplesOrgan, Tissue, cell extracts, biological fluids, etc.
Protein Sample PreparationSub cellular components, fractionation, affinity enrichment, Protein complexes, etc.
Protein Sample Simplification (or Purification):
Top Down MS or MS/MS
Protein Sequencing, and PTM
MudPit MS/MSHT Protein ID &
Quantitation
Digestion
RP LC MS/MSProtein sequence ID, PTM and Quantitation
Peptide Mass Fingerprint (PMF)
Protein ID
Protein Labeling ChemistriesICAT, PhIAT, SILAC,18O proteolysis, etc.
(Non LC) MS/MSProtein
sequence ID, PTM
Digestion
Predominantly ESI FTMS
Predominantly MALDI TOF MS
Predominantly ESI QTOF & IT (3D and 2D), MALDI TOF/TOF MS.
Also FTMS and Triple-Q (incl. Quad. Trap)
Predominantly ESI QTOF& IT (3D and 2D)
Enrichment step and/orPeptide Labeling Chemistries
(SCX, IMAC, iTrAQTM etc.)
1D-SDSPAGE
2D-SDSPAGE DIGE IEF FFEProtein LC
(Q, S, RP, SEX, etc.
Digestion
SELDI orMALDI
TOF MSBiomarker
patterns
Digestion
ProteinsPeptides
LC, M
S and Tandem-M
S
Commonly Used Protein Profiling Workflows in Mass Spectrometry
Primary Biological SamplesOrgan, Tissue, cell extracts, biological fluids, etc.
Protein Sample PreparationSub cellular components, fractionation, affinity enrichment, Protein complexes, etc.
Protein Sample Simplification (or Purification):
Top Down MS or MS/MS
Protein Sequencing, and PTM
MudPit MS/MSHT Protein ID &
Quantitation
Digestion
RP LC MS/MSProtein sequence ID, PTM and Quantitation
Peptide Mass Fingerprint (PMF)
Protein ID
Protein Labeling ChemistriesICAT, PhIAT, SILAC,18O proteolysis, etc.
(Non LC) MS/MSProtein
sequence ID, PTM
Digestion
Predominantly ESI FTMS
Predominantly MALDI TOF MS
Predominantly ESI QTOF & IT (3D and 2D), MALDI TOF/TOF MS.
Also FTMS and Triple-Q (incl. Quad. Trap)
Predominantly ESI QTOF& IT (3D and 2D)
Enrichment step and/orPeptide Labeling Chemistries
(SCX, IMAC, iTrAQTM etc.)
1D-SDSPAGE
2D-SDSPAGE DIGE IEF FFEProtein LC
(Q, S, RP, SEX, etc.
Digestion
SELDI orMALDI
TOF MSBiomarker
patterns
Digestion
ProteinsPeptides
LC, M
S and Tandem-M
S
Primary Biological SamplesOrgan, Tissue, cell extracts, biological fluids, etc.
Protein Sample PreparationSub cellular components, fractionation, affinity enrichment, Protein complexes, etc.
Protein Sample Simplification (or Purification):
Top Down MS or MS/MS
Protein Sequencing, and PTM
MudPit MS/MSHT Protein ID &
Quantitation
Digestion
RP LC MS/MSProtein sequence ID, PTM and Quantitation
Peptide Mass Fingerprint (PMF)
Protein ID
Protein Labeling ChemistriesICAT, PhIAT, SILAC,18O proteolysis, etc.
(Non LC) MS/MSProtein
sequence ID, PTM
Digestion
Predominantly ESI FTMS
Predominantly MALDI TOF MS
Predominantly ESI QTOF & IT (3D and 2D), MALDI TOF/TOF MS.
Also FTMS and Triple-Q (incl. Quad. Trap)
Predominantly ESI QTOF& IT (3D and 2D)
Enrichment step and/orPeptide Labeling Chemistries
(SCX, IMAC, iTrAQTM etc.)
1D-SDSPAGE
2D-SDSPAGE DIGE IEF FFEProtein LC
(Q, S, RP, SEX, etc.
Digestion
SELDI orMALDI
TOF MSBiomarker
patterns
Digestion
ProteinsPeptides
LC, M
S and Tandem-M
S
Commonly Used Protein Profiling Workflows in Mass Spectrometry
Protein Profiling Overview
Christopher M. Colangelo, PhDDirector of Protein Profiling
300 George StreetRoom G006
(203) 737-2636 (W)[email protected]
Brief Overview of the Keck LaboratoryFounded in 1980:
– 40 Full time staff– 100 Major instrument systems purchased at a cost of >$11
million dollars– 25,000 ft2 of space– 12 individual Resources that provide a wide range of proteomic
and genomic biotechnologiesCompletes 270,000 protein and nucleic acid syntheses and analyses annually for 260 Yale and 640 non-Yale investigators at 300 institutions & companies in 27 countriesNHLBI Proteomics Center (2002)NIH High End Instrumentation Grant (2002): FTICR-MSNational Biodefense Center of Excellence: Proteomics Core (2003)NIH High End Instrumentation Grant: “Center for High Performance Computation In Biomedicine” (2004)NIDA Neuroproteomics Center (2004)
(http://keck.med.yale.edu/)
Disease Based ProteomicsTranslational R
esearch
Receptor-Mediated Signaling
Transcriptional Regulation
Endocytosis and Exocytosis
Regulation of the Cytoskeleton
DIGE ICAT 2D-LC Beckman Coulter System MudPIT
Phosphoproteome Profiling
Lipid Profiling
Identification and Characterization of Signaling Proteins Whose Expression is
Up/Down Regulated
SerumBiomarkers
Targeted Proteomics
Clinical Studies of Biomarker Efficacy
Keck MALDI-MS Based Disease Biomarker Discovery Sera from >50 disease patientsSera from >50 normal patients
Automatically desalt each sample (submitted in 96 well plate format) using a Micromass MassPrep robot and C-18 Zip-Tips.
Automated MALDI-MS on Micromass MALDI-L/R instrument: 130,000 m/z vs intensity datapoints acquired from 700 to 28,000 Da.
Optimize marker selection by increasing the size of the training set until the ability to correctly classify a testing set is maximized
Use a Random Forest Algorithm customized by Keck Biostatistics Resource (Wu et al , 2003) to identify 20-40 peptide markers whose intensities can be used to maximally
discriminate all normal from disease samples.
Linearly normalize baseline-corrected intensities in each spectrum to the overall median for all normal + disease spectra in the training set.
Filter out noise and all datapoints which fail to pass peak and unique peptde ion criteria.
Median MALDI-MS Spectra for Sera from 47 Ovarian Cancer (top) vs 42 Control (middle) Patients and the Resulting
Difference Spectrum (bottom)
m/z Versus Intensity Distribution for 129 Samples Around the 2084 Ovarian Cancer Marker
(relative importance = 2.9)
Median Cancer
Median Control
MarkerPosition
Impact of Training Set Size & Number of Biomarkers on Classification of 34 "Unknown" Ovarian Cancer Sera
Training Set Size: 102 Sera#Biomarkers = 25
50%
55%
60%
65%
70%
75%
80%
50 75 100 125 150
#Samples in Training Set
Cor
rect
ly C
lass
ified
Tes
t Ser
a
60%
65%
70%
75%
80%
0 20 40 60 80 100
Number of Biomarkers
%C
orre
ctly
Cla
ssifi
ed T
est S
era
Genomics – The blueprint of lifePCRDNA Central Dogma
Crick (1970) Nature 227, 561-563Crick (1958) Symp. Soc. Exp. Biol. 138
Mullis (1985)
Human Genome SequenceWatson and Crick (1953) Nature 171, 737-734
The International Human Genome Mapping Consortium (2001) Nature 409, 934 - 941 Venter et. al. (2001) Science Feb 16 2001: 1304-1351
What is proteomics?
Genomics – provides us with the words, but doesn’t provide us with any meaning
Proteomics - provides us with the definitions for the genome
The identification, characterization and quantification of all proteins involved in a particular pathway, organelle, cell, tissue, organ or organism that can be studied in concert to provide accurate and comprehensive data about that system.
Protein Profiling – provide us with the grammar and syntax to form a meaningful language
Understanding how proteins interact with each other, their environment, and function with outside molecules
Challenges - proteomics and protein profiling are dynamic languages
How do we bridge the gap from genomics to proteomics?
Genomics
Proteomics
Gene MicroarraysFirst step towards bridging the gap between human genome and function
Keck Human 4.6K cDNA Array Cluster Analysis
mRNA vs. Protein correlation
(R=0.66)
Comparing protein abundance and mRNA expression levels on a genomic scaleGreenbaum D, Colangelo C, Williams K, Gerstein MGenome Biol. 2003; 4(9): 117. Epub 2003 Aug 29.
Reality check for proteomicsAvogadro’s number - 6 x 1023 molecules/moleFor MS proteomics we want 100 fmoles (10-14 moles) = 6.02 x 1010 molecules
cell copy number cells needed for 100 fmoles10 6 billion (6.02 x 109)
100 600 million (6.02 x 108)1000 60 million (6.02 x 107)
10,000 6 million (6.02 x 106)100,000 0.6 million (6.02 x 105)
1g of tissue = 108 cells1000 copy number limit
10cm culture dish = 105 cells100,000 copy number limit
Levels of proteins in plasma
Molecular & Cellular Proteomics 2003 , Anderson and Anderson 2 (1): 50
Protein Profiling
Disease Based ProteomicsTranslational R
esearch
Receptor-Mediated Signaling
Transcriptional Regulation
Endocytosis and Exocytosis
Regulation of the Cytoskeleton
DIGE ICAT 2D-LC Beckman Coulter System MudPIT
Phosphoproteome Profiling
Lipid Profiling
Identification and Characterization of Signaling Proteins Whose Expression is
Up/Down Regulated
SerumBiomarkers
Targeted Proteomics
Clinical Studies of Biomarker Efficacy
Cleavable ICAT™ Reagents
Provide 9 Dawhich provides
ratio of heavy/light
peptide conc.
Removes biotin which
improves MS/MS
fragmentation
Reacts with free cysteines
Allows selection and concentration of cysteine-containing peptides
Applied Biosystems
Differential expression analysis on human leukemicprecursor cells vs. differentiated monocytes/neutrophils
16 SCX fractions collected
Cys-containing peptidesaffinity purified/ cleaved with TFA
100ug of protein is labeled with acid cleavable ICATTM
reagent/trypsinized
Online LC-MS/MSanalysis on QStar®
XL System
ICAT-Determined Expression Ratio of Zinc Finger Protein 9 in Two Human Cell Lines
Overall Average Expression Ratio of Protein 9 in Two Human Cell Lines:
3.4 fold decrease(average H/L = 0.297 +0.061)
Web-based query results from Yale Protein Expression Database (YPED)
http://genome2.biology.yale.edu/yp_results/logon.jsp
Disease Based ProteomicsTranslational R
esearch
Receptor-Mediated Signaling
Transcriptional Regulation
Endocytosis and Exocytosis
Regulation of the Cytoskeleton
DIGE ICAT 2D-LC Beckman Coulter System MudPIT
Phosphoproteome Profiling
Lipid Profiling
Identification and Characterization of Signaling Proteins Whose Expression is
Up/Down Regulated
SerumBiomarkers
Targeted Proteomics
Clinical Studies of Biomarker Efficacy
Automated 2D Comparative LC Protein Profiling(Beckman-Coulter PF2D Platform)
Partial pI/UV map of a colon cancer cell line before and after drug treatment. The RP-HPLC profiles illustrate differences in the pI 6.0-6.2 fraction which are shown also by the color coded band depictions of these RP-HPLC profiles (from Beckman-Coulter).
Two-dimensional Comparison Plot of
NB4 (Green) and NB4 + RA (Red)
(Neutrophil) from Beckman PF2D
System
| | | | | | | | | | | | | | | | | | | | |1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Fraction
pH8.6 7.9 6.4 5.2 4.3
Ret
entio
n Ti
me
22-
20-
16-
18-
12-
14-
10-
5 mg of total protein
The 2D overlay plot shows the relative differential expression for each of the 21 pH fractions based on overlay comparisons of 214 nm reversed phase chromatograms.
Overlay of NB4 and NB4+RA Reversed Phase Chromatograms from Fraction 7
The differential peaks in the UV chromatogram represent intact proteins that are expressed or modified between control and treated samples.
Retention Time
Abs
orba
nce
(211
nm
)
Life Science Tools is our Business™
APEX FTMSAPEX-Q Industrial Design
Typical Project Milestones
0 1 2 3 4 5 6 7 8 9 10 11 12Weeks
AAA results
Differential Gel
Preparative gel
Mass Spectrometry
Data AnalysisDIGE
AAA results
ICAT Labeling
Cation-exchange
Mass Spectrometry
Data AnalysisICAT
AAA results
2D Chromatography
ICAT Labeling
Mass Spectrometry
Data Analysis
PF2D
Disease Based ProteomicsTranslational R
esearch
Receptor-Mediated Signaling
Transcriptional Regulation
Endocytosis and Exocytosis
Regulation of the Cytoskeleton
DIGE ICAT 2D-LC Beckman Coulter System MudPIT
Phosphoproteome Profiling
Lipid Profiling
Identification and Characterization of Signaling Proteins Whose Expression is
Up/Down Regulated
SerumBiomarkers
Targeted Proteomics
Clinical Studies of Biomarker Efficacy
2D Electrophoresis
2D gel electrophoresis a “Gateway” proteomic tool for the separation of complex protein mixtures from many different biological sample types
Pros: -High resolution separation by pI and MW -Possible to resolve several thousand proteins on a single gel-IPG strips feature high reproducibility and resolution (single pH unit)-PTMs, such as phosphorylation and glycosylation are clearly visualized
Cons:- Membrane/ hydrophobic proteins have solubility problems
a) can be addressed by detergents, e.g., ASB-14, Triton X-100b) some proteins are prone to precipitation at their IE pointc) try an alternative technology, such as ICAT
-Gel-to-gel reproducibility and therefore Quantitation is a problem
DIFFERENTIAL FLUORESCENCE GEL ELECTROPHORESISPROTEOME PROFILING (DIGE)
DIGE Detection Limits
# of cells protein copies/cell total # of proteins moles of protein ng of 50kd
1 x 108 1000000 1014 160pmole 8,304(~1 g of tissue) 100000 1013 16pmole 830
10000 1012 1.6pmole 831000 1011 160fmole 8.2100 1010 16fmole 0.8
10 109 1.6fmol 0.08DIGEMS
DIGE method is highly reproducible…but only with good sample preparation!
-use of label compatible buffers7M Urea, 2M Thiourea, 4% CHAPS, in 30 mM TRIS adjusted to pH 8.5
-ensure that the sample preparation is homogenous
-”clean”-up of sample via precipitation, dialysis, nuclease
-low abundance issues are best addressed at the sample preparation stepa)subcellular fractionationb)directed biochemical enrichments/pre-fractionationc)microdissection/punches of specific tissue regionsd)removal of high abundance species
-quantitate samples prior to analysis to normalize, AAA is best
-ensure there is enough sample for downstream studies!
Albumin and IgG depletion from serum
cy 3 image
cy 5 image
spots with a greater or = to 3X increase (Blue) or Decrease (Red)In cy5/cy3 spot volume ratio