High Field FTICR Instrumentation Usage

1
Proteomics Research Resource Center for
Integrative Biology Biological Sciences
Division and Environmental Molecular Sciences
Laboratory Pacific Northwest National
Laboratory Richland, WA
Environmental Molecular Sciences Laboratory at
PNNL
2
The PNNL Proteomics Resource Center

• P41 Biomedical Technology Research Resource
  Center funded in 2003 renewed in 2008
• Developing and applying ultra-sensitive and high
  throughput mass spectrometry based proteomics
  technologies and supporting informatics
  capabilities (http//ncrr.pnl.gov/)
• Leverages a large base of DOE instrumentation,
  infra-structure, and EMSL User Facility
  investments
• Collaborative and service projects supporting a
  growing number of clinical/translational
  proteomics applications (member UW CTSA OHSU
  CTSA membership pending)

3
The Flow of Biological Information
DNA Genome Plan
Readout Execution

• ? RNA
• Transcriptome

? Proteins ? Proteome
Measurement technologies enable new biomedical
applications
3
4
LC-tandem mass spectrometry (MS/MS) based shotgun
proteomics
Protein from tissue, blood, etc.
Proteolysis (trypsin) to form peptide fragments
from proteins
LC-MS/MS analysis
MS/MS Spectrum
LC Separation
MS Spectrum
l
i
300
1000
1700
200
600
1,000
1,400
1,800
LC separation elution time
Peptide mass to charge ratio (m/z)
Peptide mass to charge ratio (m/z)
Compare MS/MS spectrum to theoretical peptide
spectra (e.g. from DNA sequences)
Identify peptide and infer parent protein
Peptide identification, its accurate mass, and
its LC separation elution time provides an
Accurate Mass and Time (AMT) tag for use in
subsequent studies (i.e. without MS/MS)
5
Location of one peptide AMT tag derived from
LC-MS/MS analysis enabling identification of a
protein
Peptide GITINTAHVEYQTETR Mass1830.1987 LC NET
0.274
Peptide mass
.
Normalized LC separation elution times (NET)
6
Use of peptide AMT tags greatly increases LC-MS
proteome measurement throughput
Single LC-MS run with very high mass accuracy
analysis
Peptide AMT tags derived from multiple shotgun
LC-MS/MS runs
Mass
Mass
LC separation elution time (NET)
LC separation time (NET)
Peptides identified
7
PRISM Proteomics Research Information
Storage and Management system
8
Informatics capabilities are central to
successful proteomics applications
Data Management
Work flow Automation
Algorithm Development
Application Development
Multi Disciplinary Team
Aid for Experimental Design
Dissemination Data Tools
Meta data Capture
Data Facilitation
Biological knowledge
Terabytes of data
Informatics encompasses the capture, management,
processing, analysis, and dissemination of data
and related study information to facilitate
understanding
9
http//ncrr.pnl.gov/software/
10
Proteomics and the P41 Center

• Proteomics challenging, rapidly developing, and
  benefits from a larger Center infrastructure,
  team, and greater scale of efforts
• Major focus of Centers technology development
  efforts increasing measurement quality,
  throughput and sensitivity
• Presently thousands of proteins can be measured
  in single LC-MS analysis extended coverage
  of lower abundance proteins achievable with
  fractionation (trade-off with throughput)
• Major types of applications understanding
  biological systems and the identification of
  diagnostic or prognostic biomarkers
• Enabling new clinical applications microscale
  proteomics (e.g., of microbiopsies and small
  blood cell populations)

11
Application projects scattered across the country
Locations of Present Center Collaborators (21)
12
Application projects scattered across the country
Locations of Present Center Collaborators (21)
2
2
2
Locations of Major Service Project PIs
13
Service project Providing Proteomics Core for
WNPRC
14
Heat map showing protein abundance changes in a
time course study of lung fluid from Monkeypox
infected Macaques
Macaque 1
Macaque 2
Time- Course
Post-infection
Post-infection
Pre-infection
Pre-infection
Glyco-proteins, surfactant-associated proteins,
actin associated proteins
Host Viral Proteins
Viral and Inflammatory proteins
Classical Plasma Proteins and Immunoglobulins
Lower
Higher
Abundance
In conjunction with Prof. Scott Wong, and Oregon
National Primate Research Center (OHSU)
15
Greater throughput and sensitivity enable new
applications
Sample size and MS dynamic range

• As throughput increases
• Cost/analysis decreases
• More replicates practical
• Complexity more addressable

Mass accuracy, resolution, and separation power

• As sensitivity increases
• More replicates feasible
• Complexity more addressable
• New applications enabled

Lower limit of detection
16
3-D mapping of proteins in mouse brain enabled by
voxelation and quantitative HTP
proteomics Analysis of one voxelated mouse brain
at 1 mm resolution requires proteome analysis of
700 tissue samples
Quantitation and spatial distributions obtained
for gt1000 distinct proteins
diazepam binding inhibitor
Collaboration with Prof. Desmond Smith UCLA
17
Proteomics of ten mouse striata following
neurotoxin treatments (Comparison with 10
controls)
215 of 1971 proteins show significant changes
for treatment vs. control
Methamphetamine (MA)-treated
MPTP-treated
F
G
A
I
B
E
C
D
J
H
Proteins
Down
Up
Proteins with differing responses
Log2Ratio (Stimulated/Control)
18
Addressing the challenge of biological variation
in human populations

• Key challenge distinguishing between proteins
  that are effective candidate biomarkers for a
  disease state and those that vary due to other
  factors, including normal biological variation
• Requires sufficiently large sample cohorts for
  statistical confidence
• Biomarker development requires high quality
  proteomics data from large numbers of carefully
  selected samples

19
Quantitative comparisons of blood plasma proteomes

• Thousands of peptides covering 300 to gt1000
  proteins can be measured in each blood plasma
  proteome analysis
• Differences can arise from issues related to
  the measurements, normal biological variation, as
  well as possible disease states
• Pearson correlation coefficient for peptides
  between two different analyses No difference 1.0

Background color code
20
Variation between Macaque serum and plasma
measurements Comparison of plasma and serum
levels for 325 proteins in three technical
replicates (including sample preparation) for
three animals (controls for longitudinal study of
influenza infection)
Plasma Serum
Plasma Serum
With UWNPC, Prof. Michael Katze (UW), and Prof.
Carol Baskin (ASU)
21
Biological variation in blood plasma proteome
measurements
Nine healthy humans
Nine mice (same strain, sex, and age)
0.85 0.06
0.92 0.05
Pearson Correlation Coefficient
22
Diversity of plasma proteomes for 38 burn
patients
In progress proteomics measurements for plasma
and three blood cell types (T cells,
Neutrophils, and Macrophages) for 350 patients
Service Project Host Response to Injury Glue
Grant Prof. Ron Tompkins, Harvard Medical
School, PI
23
Examples of plasma protein concentration
variations
10 healthy humans
10 severe burn patients
Gene
0
0.4
1.2
Relative concentration, deviation from mean (log2)
24
Targeted Proteomics Measurements

• Biomarker verification generally uses a different
  targeted measurement platform providing higher
  throughput and for follow-up to broad discovery
  proteomics
• Multiple-reaction monitoring using LC-MS/MS with
  a triple quadrupole mass spectrometer is
  presently platform of choice
• Provides higher selectivity and sensitivity, but
  only for limited numbers of pre-selected proteins
• Circumventing this limitation is a major Center
  focus for technology development efforts
• Goal measurement platform combining attributes
  of both broad discovery and targeted platforms

25
A new proteomics platform combining fast LC with
much faster ion mobility separations (IMS) for
much higher throughput proteomics
Prototype LC-IMS-MS
IMS dimension
20 30 40 msec
MS
IMS
LC
MS dimension

• Time scale for ion mobility separation lt 100
  msec
• Combination with fast LC separation provides
  high 2-D separation power

26
15 min LC-IMS-MS analysis of a mouse plasma
sample Combined 2-D separations with MS allows
detection of more proteins
27
Technology advances combine to enable higher
throughput

• Multiplexing IMS separations allows analysis of
  gt50-fold more ions than for conventional IMS-MS
• All benefits and data qualities of conventional
  IMS-MS retained
• Full benefits only gained in conjunction with
  improved nanoelectrospray ion source and ion
  funnel interface (providing improved sensitivity
  and quantitation) developed in Center

Enhancement of sensitivity with new ion
source-mass spectrometer interface
Nanoelectrospray emitter array
28
Coefficients of variation (CV) for triplicate
measurements of 20 peptides added to mouse plasma
Comparison of 15 min LC-IMS-MS vs. 120 min
LC-LTQ-FT MS analyses

• Spiking Level Peptide added

ND Not detected
29
Developing a broad spectrum of health biomarkers

• Technology advances open the opportunity for
  proteomics applied to large populations
• Blood proteomics may provide biomarkers for
  essentially every disease state
• As the studied population increases, a more
  detailed predictive capability should evolve

30
Our own populations within Human Microbiomes

• Largely uncharacterized, complex, and
  continually changing microbial communities exist
  in all of us, having mostly unknown health
  consequences
• NIH Roadmap Human Microbiome Projects are
  initially sequencing the DNA of selected
  microbiomes
• Proteomics studies of microbiome - host
  interactions are being enabled by this genomics
  data, as well as application of new informatics
  tools (e.g. developed by the new P41 Resource
  Center at UCSD Prof. Pavel Pevzner, PI)
• Initial Center microbiome studies aided by
  approaches and tools developed to support DOEs
  studies of environmental (e.g. ocean) microbial
  communities

31
Summary

• Proteomics capabilities will continue to evolve
  rapidly e.g. as measurement dynamic range and
  capabilities for identification of protein
  interaction partners and protein modifications
  are further extended
• The Center benefits from leveraging other
  investments, a diverse interdisciplinary team,
  the scale of operations, and an infrastructure
  facilitating remote interactions
• Advances in sensitivity and higher throughput
  are enabling new large scale proteomics
  applications
• Benefits particularly important for clinical
  applications and dealing with biological variation

32
The Proteomics Team
Co-PI Steven Wiley Associate Center Director
David Camp
Proteomics Applications Joshua
Adkins Ljiljana Pasa-Tolic Joseph Brown Juhi
Jain Joshua Turse Weijun Qian Jon Jacobs Tao
Liu Vlad Petyuk Charles Ansong David
Kaleta Nathen Manes Steve Callister Angela
Norbeck David Kaleta Feng Yang Sailful
Chowdhury Xu Zhang Mary Lipton Chris
Sorensen Jianying Zhao HTP Proteomics
Measurements Ron Moore Brianna Ogata David
Anderson Rui Zhao Danial Orten Anil Shukla Eric
Livesay Kim Hixson Heather Mottaz Carrie
Nicora Marina Gritsenko Therese Clauss Rui
Zhang David Prior Athena Schepmores Karl Weitz
Technology Development / Implementation Mike
Belov Keqi Tang Erin Baker Aleksey
Tolmachev Ryan Kelly Satendra Prasad Jason
Page Yehia Ibrahim Daniel Lopez-Ferrer Ridha
Mabrouki Ioan Marginean Errol Robinson Yufeng
Shen Kostas Petritis Alex
Shvartsburg Informatics Gordon Anderson
Matt Monroe Dave Clark Gary Kiebel John
Sandavol Deep Jaitly Bill Danielson Sam
Purvine Aleksey Tolmachev Nikola Tolic Ashoka
Polpitya Xiuxia Du Anoop Mayampurath Ken
Auberry Brian LaMarche Jaya Rao Andrei Liyu