High Throughput Protein Domain Elucidation by Limited ProteolysisMass Spectrometry Jeff Bonanno and

1
High Throughput Protein Domain Elucidation by
Limited Proteolysis-Mass SpectrometryJeff
Bonanno and Xia Gao Structural GenomiX, Inc
2
Outline

• Overview of SGX technology platform
• Overview of NY and SGX research consortium
  (NYSGXRC)
• Mass Spectrometry applications
• Integration into SGX technology platform
• High Throughput Limited Proteolysis Mass
  Spectrometry

3
SGX Technology Platform
4
Overview of NY and SGX Research Consortium
(NYSGXRC)

• Vision
• The NYSGXRC will establish a cost-effective,
  high-throughput X-ray crystallography platform
  that serves as a model for the structural biology
  laboratory of the future.
• Mission
• To develop and use the technology for
  high-throughput structural and functional studies
  of proteins
• www.nysgxrc.org

5
NYSGXRC Members

• Albert Einstein College of Medicine (AECOM)
• Brookhaven National Laboratory (BNL)
• Columbia University (CU)
• Structural GenomiX, Inc. (SGX)
• Sloan Kettering Institute (SKI)
• University of California at San Diego (UCSD)
• University of California at San Francisco (UCSF)

6
MS Analysis from Gene Expression to Protein
Purification

• Molecular biology
• Verify expression and protein integrity
• Provide domain boundary information
• Fermentation
• Determine heavy atom incorporation
• Monitor progression
• Purification
• QA/QC on final pools
• Characterize post-translational modifications
• Fraction analysis to provide guidance

7
Methods for Domain Elucidation

• Bioinformatic approach
• Sequence alignment, e.g. BLAST, Pfam
• Secondary structure prediction
• Homology modeling
• Limited Proteolysis MS
• Probing protein structure in solution
• Provide termini information of protein functional
  domain(s)
• Provide information on solvent accessible or
  disordered loop regions (Cohen et al., 1995,
  Cohen and Chait, 1996, Marcotrigiano et al.,
  1997, Lee et al., 1996, Xie et al., 1996, Cabral,
  et al., 1998, Zhang et al., 1997)
• Hydrogen/Deuterium Exchange MS
• (Pantazatos et al., 2004)

8
Principle of Limited Proteolysis-Mass Spectrometry
9
A Successful Example
Domain 3
Domain 4
Domain 1
Domain 2
Cterm
1
Full length protein Low yield, 2mg/L High
tendency to aggregate Cannot be concentrated
above 1mg/ml
Domain 3
Domain4
Domain 1 ?
Domain 2
4
Cterm - 76
LP construct High yield, 30mg/L Stable
overtime Concentration of 5mg/ml Protein
crystallized
10
Proteolysis Experiment Conditions

• Proteases
• Trypsin, Lys-C, Chymotrypsin and GluC
• Buffer condition
• PH8, salt and detergent if necessary--Protein
  native condition
• Protein concentration
• 2mg/ml
• Time points
• 5min, 10min, 30min, 1h, 2hrs and 4hrs.
• Capacity
• Eight proteins per experiment

11
Sample Preparation for Automated MALDI-MS Analysis

• Thin-Layer Sample Preparation
• (Cadene and Chait, 2000)
• High homogeneity offers high success rate for
  automated data acquisition. Better than 95
  attempts result in satisfactory MS spectrum.
• Thin-Layer method affords high detection
  sensitivity,
• Automated Data Acquisition by Sequence Control
  from ABI

12
Data Interpretation with PAWS
13
Data Assembly by Digests Reader in Batch Mode
14
Throughput

• Total proteolysis experiments 270
• Total number of data sets acquired 250
• Total number of data sets analyzed 210
• Duration of this endeavor six months
• Total FTEs 1.0 on average

15
Representative Results

• Summary of MS analysis of crystallized proteins
  which diffract poorlyThree examples
• Proteins showed stability toward proteolysis
• Removal of His6-tag is recommended
• Removal of structural micro-heterogeneity
  necessary

• Large scale cloning, expression/solubility
  testing and resubmission to crystallization
  underway

16
Plans Forward
Automated Proteolysis Experiment
Automated Data Acquisition
Automated Data Assembly
Automated Molecular Biology
Need to Automate Sample preparation
Need to Automate Data Interpretation
17
Acknowledgements

• SGX
• Julie A. Reynes, Michelle Buchanan and Chau Thai
• Curtis Marsh and Boris Laubert
• Ken Schwinn and Michael Sauder
• Stephen K. Burley
• Spencer Emtage
• Rockefeller University
• Brian T. Chait and Martine Cadene
• Tecan
• Brian Smith
• NIH NIGMS PSI