ProNet Yeast TwoHybrid Technology and Protein Interaction Database

1
ProNet Yeast Two-Hybrid Technologyand Protein
Interaction Database
2
Outline of Presentation

• Myriads Collaboration with NIEHS
• ProNet Yeast Two-Hybrid Technology
• Yeast Two-Hybrid Example
• ProNet Web Interface
• RC Recombinational Cloning Process

3
Myriad - NIEHS Collaboration

• Three-year institution-wide partnership
• Yeast two-hybrid analysis of proteins for
  individual researchers
• Custom libraries constructed for the program
• Access to Myriads ProNet Web Interface including
  curation of public literature and AfCS yeast
  two-hybrid data
• Optional use of Myriads high-throughput
  expression cloning platform to facilitate
  validation experiments at NIEHS

4
ProNet Yeast Two-Hybrid Technology
5
Protein Interactions

• Protein interactions underlie cell biology
• Proteins modify each other modulate each others
  activity
• Protein complexes are cellular machines and
  structural components
• Mapping protein interactions
• Expands understanding of cell biology
• Associates proteins with functions and diseases
• Identifies novel therapeutic targets

6
The Yeast Two-Hybrid System

• Interaction of the test proteins localizes the
  activation domain
• to the reporter gene, causing it to be expressed.
• Yeast colonies can grow.

Reporter mRNA
Protein X
Reporter mRNA
Reporter mRNA
Reporter mRNA
Binding Domain
Reporter mRNA
Reporter Gene
7
Two-Hybrid Advantages

• Genetic no knowledge of biochemistry is required
• Extremely sensitive (Kd 10 mM)
• Measures direct interactions
• Can isolate and characterize novel proteins
• The procedure can be automated

Protein Y
Activation Domain
Reporter mRNA
Protein X
Reporter mRNA
Reporter mRNA
Reporter mRNA
Binding Domain
Reporter mRNA
Reporter Gene
8
Myriads Proprietary ProNet Process

• Roboticized 70-step process
• 42 custom protein libraries from a variety of
  tissues/cells
• Efficient strategy to survey millions of
  combinations
• Automated for quality and throughput
• Positive sample tracking
• Robots for media preparation, liquid handling,
  colony picking
• Statistical quality control

9
ProNet Process Flowchart
1
3
1 Bait Construction
40
2 Bait Verification
0
41
42
1
3 Two-Hybrid Screen
2
2
4
44
43
4 Identification and Confirmation of
Interactions
6
5
3
27
45
22
23
9
48
46
47
7
8
25
24
28
26
10
29
49
30
11
50
12
34
32
33
31
13
35
36
56
51
Seq
16
15
53
52
14
57
37
17
Seq
38
54
39
55
19
18
TRACKING
20
ELECTRONIC
60
59
58
ROBOTIC
21
62
61
64
SEQUENCING
Seq
4
63
Seq
10
DNA-Binding Domain Constructs

• Multiple baits for each protein
• Display recognizable protein domains
• Obtain coverage for entire protein as not all
  baits work well
• Smaller protein fragments typically work better
• Example - phosphatidylinositol 3-kinase,
  regulatory subunit (p85 alpha)

11
PI-3 Kinase (p85 alpha)
SH3
SH2
SH2
1
724
Baits
12
Activation Domain Libraries

• Characteristics
• Complex Greater than 5 million independent
  clones
• Comprehensive Random primed, directionally cloned
• Informative Small inserts (avg. 600-900 bp)
• Efficient Contain few false positives
• Tissue sources
• Spleen Liver Brain Breast cancer cells
• Thymus Kidney Substantia nigra Prostate
  cancer cells
• Macrophage Heart Hippocampus Acute hypoxia
  cells
• Endothelial cells Mouse myocyte Hypothalamus
  Chronic hypoxia cells
• Mouse B cells Skeletal muscle Placenta Total
  mouse embryo
• Mouse macrophage Adipose Mouse germ cells
• Custom libraries for collaborators

13
Reduction of False Positives

• False positives are eliminated in three steps
• Multiple reporter genes with dissimilar promoters
  50
• Two-hybrid test with heterologous protein
  baits 10
• Electronic removal of remaining false positives
  lt1

14
ABCA1 Interaction Network

• Two-hybrid analysis of ABCA1
• Brain, adipose, spleen, liver, macrophage, small
  intestine
• 17 protein partners
• Second round searches
• 130 interactors
• Apolipoproteins
• Numerous PDZ domain-mediated interactions

ABCA1
15
Custom Two-Hybrid Searches

• Searches in the presence of ligands or other
  small molecules
• Substrate trap versions of enzymes
• Expression of a third protein (phosphatase,
  kinase, regulatory subunit, etc.)

16
Nuclear Hormone Receptors
DNA-binding hinge
ligand-binding
1
2
3
4
5
Two-hybrid search with 4 mM ligand in selective
plate
ligand
Bait
- ligand
17
Ligand-Dependent Interactions

• Prey Bait No.
• Nuclear receptor coactivator 1 2 1
• 4 9
• Nuclear receptor coactivator 2 2 5
• 4 7
• Nuclear receptor coactivator 3 3 2
• 4 5
• Transcription factor 1 4 5
• 5 1
• Transcription factor 2 4 4
• 5 1
• Transcription factor 3 4 1

18
ProNet Program Summary

• In over six years of performing the ProNet yeast
  two-hybrid
• platform, Myriad has accomplished the following
• 4100 proteins have been analyzed from a number of
  organisms
• 23,000 protein constructs have been created
• 42 custom libraries have been constructed
• 55,000 searches have been conducted
• 280,000 positive clones have been analyzed
• 9200 interactions have been identified
• 98 of ProNet interactions are proprietary
• Extensive networks in the areas of heart disease,
  oncology, viral budding, dementia, diabetes,
  obesity, inflammation and general cell signaling

19
Yeast Two-Hybrid Interaction Validation

• Biochemical validation of selected interactions
• Reproducibility of interactions in additional
  yeast two-hybrid experiments
• Reversibility of interactions (bait is prey prey
  is bait)
• Interactions found using multiple libraries or
  different baits
• Functional validation of proteins in cell-based
  assays
• Compound development centered on proteins in
  interaction networks

20
Yeast Two-Hybrid Example
21
HIV-1 Viral Budding Pathway

• Discovery of viral pathogen-host interactions
• Validation of interactions using cell-based and
  biochemical assays (collaboration with Sundquist
  lab at University of Utah)
• Expansion of viral budding pathway by
  identification of interactions between key host
  proteins
• Identification of novel therapeutic strategies
  such as disruption of protein interactions and
  chemical modulation of newly identified protein
  regulators

22
Tsg101 and the VPS Pathway are Essential for
HIV-1 Budding
The p6 fragment of the HIV-1 Gag polypeptide was
used in two-hybrid screens of human
libraries. Six protein interaction partners were
identified, including the Tsg101 protein that was
later shown to be required for viral budding.
HIV-1 Gag p6
Tsg101
J.E. Garrus, U.K. von Schwedler, O.W. Pornillos,
S.G. Morham, K.H. Zavitz, H.E. Wang, D.A.
Wettstein, K.M. Stray, M. Côté, R.L. Rich, D.G.
Myszka, and W.I. Sundquist. Cell 107 55 (2001)
23
Elaboration of Viral Budding Pathway using
Two-Hybrid
U.K. von Schwedler, M. Stuchell, B. Müller, D.M.
Ward, H.-Y. Chung, E. Morita, H.E. Wang, T.
Davis, G.-P. He, D.M. Cimbora, A. Scott, H.-G.
Kräusslich, J. Kaplan, S.G. Morham and W.I.
Sundquist. Cell 114 701 (2003)
24
ProNet Web Interface
25
ProNet Informatics

• Secure, browser-based interface
• Weekly updates
• 36,000 Protein Home Pages (one-half are human)
• Graphical tools
• Project management tables
• BLAST function to identify similar proteins or
  genes in ProNet
• Curation of public two-hybrid literature

26
ProNet Curation Summary

• 4000 papers have been curated from the public
  literature.
• Links to abstracts are provided for easy
  reference.
• 10,000 protein-protein interactions have been
  curated and incorporated into the ProNet
  database.
• Majority of interactions found using yeast
  two-hybrid.
• A limited number of interactions found using
  alternative techniques such as mammalian
  two-hybrid, pulldown, immunoprecipitation and
  surface plasmon resonance are also included.
• Areas of emphasis heart disease, cancer,
  diabetes, obesity, CNS disorders, HIV pathology,
  inflammation, general cell signaling mechanisms

27
Logging on to the ProNet Interface

• Logging on
• The ProNet interface runs
• on Internet Explorer for all
• computer types.
• Each investigator has
• his or her own user name
• and password.

28
ProNet Whats New Page

• Whats New Page
• Quick links to disease and project lists
• Description of important information
• List of recent interactions
• Tabs for site navigation
• Quick Search box
• Help link

29
Individualized Protein Lists

• Functions
• Sortable list of proteins by
• ProNet ID or short name
• Overview of project progress
• Links to Protein Home Pages
• Links to Search Information
• Pages

30
Protein Home Pages

• Protein Home Pages
• Protein names, aliases, taxon, cellular function,
  ProNet ID, alternative forms, orthologs
• List of protein interactors with links to data or
  literature reference
• Links to Interaction Home Pages
• Link to interaction graphic

31
Protein Home Pages

• Protein Home Pages
• 1) Graphical representation of protein
• 2) Link to Search Information Page

32
Protein Home Pages

• Protein Home Pages
• Links to public databases
• Nucleotide and amino acid sequences
• BLAST function

33
Links to Public Databases

• Databases included
• GenBank and GenPept
• LocusLink
• OMIM
• SwissProt
• UniGene
• PROSITE
• Others

34
Searching ProNet for Similar Genes and Proteins

• Blast function
• Search for similar genes
• Search for similar proteins
• Search for related domains

35
Blast Results

• Display output
• Graphical display of like proteins
• Alignment of sequences with scores shown
• Links to Protein Home Pages

36
Protein Interaction Graphical Display Tool

• Graphical display of protein
• interaction networks
• User-directed tool to elaborate networks around a
  protein of interest
• Includes proprietary and curated interaction data
• Capability to overlay other information (such as
  expression data) is in development

37
Interaction Home Pages

• Interaction Home Pages
• Summary of information about each of the two
  proteins
• Links to individual Protein Home Pages
• Table summarizing interaction information

38
Links to Abstracts of Curated Interactions

• Curation
• Directionality of protein interaction
• Domains/regions of contact between the protein
• Annotation of method of detection (such as yeast
  two-hybrid)

39
Interaction Home Pages

• Interaction Home Pages
• Graphical representation of protein interaction,
  fragments that interact and directionality of
  interaction
• Graphical display of protein features and domains
  (if annotated by InterPro)

40
Search Information Pages

• Search Information Pages
• Graphic display of baits ordered with bait status
  color-coded
• Table indicating individual interacting proteins
  found with each bait

41
Recombinational Cloning Platform
42
Recombinational Cloning Platform

• Expression construct generation
• Full-length or partial sequences
• Uses yeast recombination strategy for cloning
• Bacterial plasmid DNA is delivered within 6-8
  weeks
• Process automation
• 96-well format
• Minimal order of 20 constructs
• Process integration
• LIMS database
• Barcoded samples

43
Clone Bank Human cDNA Collection

• Collection of full-length human cDNA clones
  obtained from the Mammalian Gene Collection (MGC)
• DNA sequence verified
• 6836 distinct genes received by Myriad
• Additional genes to be received in the future as
  they are released by the MGC

44
Mammalian Expression Vectors

• There are currently 9 expression vectors
  available
• E. coli / yeast shuttle vectors
• Contain ampR (E. coli) and TRP1 (yeast)
  selectable markers
• Differ with respect to epitope tags and
  positioning
• New vectors include ones that are Gateway
  compatible and others that facilitate retroviral
  infection

45
Recombinational Cloning Process
46
Summary

• Collaboration Opportunity for NHLBI akin to NIEHS
• ProNet Yeast Two-Hybrid Technology
• Yeast Two-Hybrid Example HIV Viral Budding
• ProNet Web Interface
• RC Expression Cloning Platform for Validation
  Experiments

47
(No Transcript)