Protein Chemistry Laboratory

1
Protein Chemistry Laboratory

• Robert J. Fisher
• Cancer Biotechnology
• April 5, 2007

2
PCL Capabilities

• Protein Characterization
• Edman amino acid sequencing
• HPLC purifications of proteins and peptides
• Maldi-Tof mass spectrometry
• Molecular Interactions
• SPR Spectroscopy
• Fluorescence anisotropy
• Fluorescence life-time measurements

3
Organization of the PCL

• Protein Chemistry (301-846-6745)
• Oleg Chertov, Ph.D. and Young Kim
• Maldi-Tof Mass Spectrometry (301-846-6702) John
  Simpson, Ph. D.
• Characterization of Ligand Interaction
  (301-846-1634)
• Andrew Stephen, Ph.D. Karen Worthy, MS and
  Lakshman Bindu, MS

4
Protein Chemistry Techniques and Core Activities

• Complete characterization of a protein
• Non-routine protein identification
• Purification of peptides, proteins and labeled
  oligonucleotides
• Identification of cross-linked amino acids,
  post-translational modifications including
  phosphorylation
• High sensitivity Edman amino acid sequencing
• MALDI-TOF mass spectrometry

5
Why do Edman Amino Acid Sequencing?

• Best method to verify the amino terminus of a
  protein
• Quantitative technique - PTH amino acid analysis
  by HPLC is precise and quantitative
• Important in the Quality Control ie Total
  Characterization of a purified or recombinant
  protein

6
Edman Amino Acid Sequencing
coupling
cleavage
Conversion to PTH Amino Acid
7
Complete Protein Characterization
8
Topics

• How does BIAcore work?
• microfluidics
• Solid Phase ligand binding
• Surface Plasmon Resonance
• What can BIAcore do?
• Realtime Binding Kinetics (on and off rates)
• Concentration measurement
• Equilibrium constant
• Data analysis .

9
Basic Principle

• A binding molecule is bound to the sensor
  surface.(ligand peptide, protein, sugar,
  oligonucleotide))
• Another (the analyte) is passed over the surface
  and binds to it.

10
Experimental Design
11
Sensor Chip CM-5 Carboxymethylated dextran
coated surface.
Allows covalent coupling via -NH2, -SH, and -CHO
12
The Flow Cell
Surface is divided into 4 channels, which can be
used individually or in a number of combinations
13
Microfluidic System

• Low reagents consumption
• Efficient mass transport
• Low dispersion
• Highly reproducible injections CV typically less
  than 1
• Wide range of contact times, 1 s - 12 h
• Sample recovery and fractionation

14
Measurement of Binding

• Binding is measured as a change in the refractive
  index at the surface of the sensor
• This is due to Surface Plasmon Resonance (SPR)
• The change in refractive index is essentially the
  same for a given mass concentration change
  (allows mass/concentration deductions to be made)
• Binding events are measured in real time
  (allowing separate on and off rates to be
  measured.)

15
Theoretical Considerations

• Binding is measured as a change in the refractive
  index at the surface of the sensor

How?
16
Total Internal Reflection
At a certain angle of incidence, light entering a
prism is totally internally reflected. (TIR).
Although no photons exit the reflecting surface,
their electric field extends 1/4 wavelength
beyond the surface.
17
Resonance Surface Plasmon
If a thin gold film is placed on the reflecting
surface, the photons can interact with free
electrons in the gold surface.
Under the right conditions, this causes the
photons to be converted into plasmons and the
light is no longer reflected.
18
Surface Plasmon Resonance

• This occurs when the incident light vector is
  equal to the surface plasmon vector.

19

• Effect of binding on SPR

• Plasmons create an electric field (evanescant)
  that extends into the medium surrounding the film
• This is affected by changes in the medium (eg
  binding of analyte), and results in a change in
  the velocity of the plasmons.
• This change in velocity alters the incident light
  vector required for SPR and minimum reflection.

20
How does BIACore Measure this?

• Fixed wavelength light, in a fan-shaped form, is
  directed at the sensor surface and binding events
  are detected as changes in the particular angle
  where SPR creates extinction of light.

21
The Sensorgram
22
Surface Plasmon Resonance
response
time
23
Binding Analysis

• How Much?

Active Concentration
Kinetics

• How Fast?

Affinity

• How Strong?

Specificity

• How Specific?

24
Concentration

• Signal proportional to mass

• Same specific response for different proteins

25
Topics

• HIV nucleocapsid binding to short
  oligonucleotides
• HIV gag precursor protein binding to short
  oligonucleotides
• Antibody screening
• Affibody Characterization
• Small molecule binding to SH2 domains

26
HIV NC and Gag Precursor to Short
Oligonucleotides

• Alan Rein
• HIV Drug Resistance Program, CCR

27
Why study HIV NC interactions with
Oligonucleotides?

• NC is highly conserved
• NC alone or as part of Gag is involved in many
  stages of the HIV life cycle
• NC is a nucleic acid chaperone
• A static binding model is not sufficient to
  explain all of NCs activities
• Minimal step in HIV particle assembly
• Insights into binding of gag precursor

28
Nucleocapsid is a domain of the Gag polyprotein
Gag
Proteolysis
p6
MA
CA
NC
29
HIV NC binding to short oligonucleotides
d(TG)4 ligand
d(A)8 ligand
d(TG)4 ligand
d(TG)4 ligand
250mM NaCl
30
Working Model for NC Binding to Short
Oligonucleotides
31
Complex binding model Matt Fivash, DMS
N NC O oligonucleotide NO 11
complex NON 2N bound to oligonucleotide ONO
2 oligonucleotides bound to NC
32
Intrinsic tryptophan quenching by oligonucleotide
NON
NO
ONO
Oligonucleotide titrated into HIV NC (excess NC)
33
Evidence to Support Model

• SPR spectroscopyDirect binding experiments
  (Evidence for NO and ternary complexes NON, ONO)
• Tryptophan quenching (Evidence for NO and
  ternary complexes NON, ONO)
• Fluorescence Anisotropy (Evidence for NO and
  higher order complexes)
• The N-terminus of NC and intact zinc fingers are
  required for high affinity binding
• FTIR ESI Mass spectrometry identified NO and NON
  complexes

34
HIV-1 Gag - Basics

• Expression of a single viral protein (Gag) in
  mammalian cells, is sufficient for assembly of
  virus-like particles
• Gag specifically packages genomic RNA during
  virus assembly
• Gag can use genomic RNA or cellular mRNA as
  scaffolding during virus assembly
• Gag is responsible for annealing of nucleic acids
  at several stages of the viral replication cycle

35
Assembly of HIV-1 virus-like particles in vitro
RNA
Assembled particle
HIV Gag precursor protein

• HIV-1 Gag (missing the p6 domain) when incubated
  with nucleic acid can assemble into virus-like
  particles
• (Campbell and Rein, J. Virol. 1999 73 2270)

36
Eleven replicate injections of NC (200nM) used to
calibrate the density of a TGx10 Biacore chip
37
Kinetic binding of Gag (0.19 400nM) to 0.36
RUs of TGx10
38
Binding Site Analysis
39
Conclusions

• ESI-FTMS data suggest the binding site for NC is
  5 bases.
• NC injections can be used to calibrate the
  surface density of ultra-low oligo surfaces.
• Gag binds with high affinity to TGx10 oligos .
• Steady-state binding can be fit with a 2-site
  binding model (Kd1 1.5 nM, Kd2 160 nM).
• The binding of Gag to TGx10 does not saturate
  even at stoichiometries of gt6 Gag
  molecules/TGx10.
• These data suggest that once Gag has filled the
  TGx10, additional Gag can bind, forming
  TGx10-Gag-Gag complexes.

40
Kinetic Screening of Monoclonal Antibodies

• Ira Pastan
• Laboratory of Molecular Biology, CCR

41
How does kinetic screening work?

• Rabbit antimouse C domain polyclonal antibodies
  are co-valently attached to a CM-5 sensor chip.
• A 11 dilution of hybridoma supernatant is passed
  over the ramC, capturing 200-500 RUs of MAb.
  This process repeated on two other surfaces and
  the fourth is a control.
• An appropriate concentration of antigen is
  serially injected over all 4 flowcells, followed
  by surface regeneration.

42
Arrangement of surfaces in a CM5 sensor chip
utilizing Serial Flow Cell
Sample in
RAMC Control FC1
RAMC MAb1 FC2
RAMC MAb2 FC3
Sample out
RAMC MAb2 FC4
43
Pastan MAbs 1-60 blank referenced
44
Three distinct patterns observed

• Typical binding curve-although some individual
  MAbs have distinctly slower off rates (MAb47)
• Unusually low response
• Complex Binding

45
MAb47
Ka1.02e5/Ms kd1.1e-7/s KD1.08e-12M
46
MAb35
Ka5.0e4/Ms kd2.2e-3/s KD4.3e-8M
47
MAb12
Simple Model does not fit data
ABgtAB
48
MAb12 Complex Model
ka11.8e5/Ms kd11.2e-4/s KD16.4e-10M
ka28.5e5/Ms kd23.7e-3/s KD24.3e-9M
ABgtAB ACgtAC
49
Rapid Screening of Crude Hybridoma Supernatants

• 60 MAbs screened using two Biacore 2000
  instruments in an overnight experiment
• Possible to obtain MAb concentration, kon and
  koff rates
• Rapid qualitative assessment of MAb quality
• Biacore A100 would allow a fivefold increase
  in throughput
• Biorad ProteoN XPR 36

50
Biacore A100 Chip
51
Rapid screen of crude hybridomas - Assay setup
Biacore A100
Antigen
mAb (mIgG) from hybridoma
Capture Ab - Rabbit anti-mIgG(fc)
CM5
(1 of 4 flow cells shown)
300 samples tested against antigen for binding
and to provide dissociation rate ranking mAb
concentrations in hybridoma samples unknown
Immobilizing different levels of capture Ab on
each spot gives a range of mAb-binding
capacities. Maximum information extracted with
this strategy
52
Ultra-low Analyte Concentrations Affibody against
Her/Erb2

• Jacek Capala
• Radiation Oncology Branch, NIH/NCI

53
Targeting Molecules Affibody
54
Targeting Molecules Affibody
55
Affibody Molecule Diversity
Randomization of 13 selected positions
Protein A-domain scaffold
3x109 Affibodylibrary members
56
Experimental setup

• Rabbit Antimouse C (RamC) is covalently attached
  to each flow cell
• FC fusion protein her2/erb2 extracellular domain
  is captured onto flowcells 2 and 4
• An ultra low concentration affibody is injected
  serially over all four flowcells for 2000
  seconds, this is followed by two injections of
  buffer, but NO regeneration step is used

57
Titration of 7.8 to 500 picomoles of her2 z342
affibody on to captured her2/erb2 ECD
Sensorgram
70
40
60
30
50
40
20
Response Units (RU's)
Response Units (RU's)
30
10
20
10
0
kt5.5e8, kon3.3e6, koff9e-5 KD 27.2 pM
0
-10
-10
0
500
1000
1500
2000
2500
0
1000
2000
3000
4000
5000
s
Time
Time
Concatenated T100 Data
Unprocessed T100 Data
58
Biacore T100

• Latest version of high resolution SPR

59
Biacore T100 Features

• Online buffer degassing (allows up to four
  buffers to be used in an experiment)
• Out gassing in a varying temperature experiment
  is not a problem
• Two pairs of parallel flow cells which do not
  require hydrodynamic addressing
• Optimized for thermodynamic experiments

60
T100 experiments done at 8 different Temperatures
61
Beta2microglubulin binding to mAB 5 C
62
Beta2microgolbulin binding to mABat 40 degrees C
63
(No Transcript)
64
Parameter Name Parameter Value SE
?H kJ/mol -40 1.5
?S J/(Kmol) 11 5
T?S kJ/mol 3.2 1.5
?G kJ/mol -44 0.016
?Cp kJ/(Kmol) 0.65 0.26

?H ass. kJ/mol 36 4.4
?S ass. J/(Kmol) -29 15
T?S ass. kJ/mol -8.6 4.5
?G ass. kJ/mol 45 0.048
Ea ass. kJ/mol 38 4.4

?H diss. kJ/mol 78 3
?S diss. J/(Kmol) -33 10
T?S diss. kJ/mol -9.8 3
?G diss. kJ/mol 88 0.032
Ea diss. kJ/mol 81 3
1KJ 0.239005736 Kcal
65
Sample and Time consumption

• Experiment took about 48 hours run and about one
  hour of data processing
• 3 micrograms of mAb against beta2microglubulin
  was amine coupled to spot 2
• 1ml of a 85nM solution of beta2microglobulin was
  consumed (concentration series of 0.1 - 85nM) for
  entire experiment
• Need about 100 fold more reagents for ITC

66
Peptide Binding to SHC-SH2 Domain

• Terrence Burke
• Laboratory of Medicinal Chemistry, CCR

67
SHC protein

• Non-catalytic SH2 domain-containing docking
  modules
• Associated with proliferation, survival and
  apoptosis
• Downstream signaling of receptor tyrosine kinases
  (RTKs)
• Link activation of the cytoplasmic kinase domains
  with Ras effectors
• Disruption of Shc-dependent signaling through
  blockade of its SH2 domain interactions may
  indicate a new therapeutic approach

68
Phosphopeptide Binds Weakly to SHC Protein
WJC-01 8.40 x10-5M Ac-LpYQGLS-amide WJC-02 1.00
x10-2M Ac-LYQGLS-amide
Blue Phosphopeptide Red Nonphosphorylated
69
Fluorescence Anisotropy

• If a fluorescent molecule is excited with
  polarized light, the emission may also be
  polarized
• Molecule must contain a transition moment for
  absorption and emission within a defined
  orientation within the structure
• Homogeneous assay
• Fast plate reader 384 well format
• Method of choice for HTS

70
Protein binding using fluorescence anisotropy
Emitted Light is Depolarized
Polarized Excitation Light
Fluorescein
Fluorescein
Peptide
Peptide
Small peptide
Rapid Rotation
Fluorescein
Fluorescein
Emitted Light remains Polarized
Polarized Excitation Light
Peptide
Peptide
protein
protein
Slower Rotation
Large complex
71
Fluorescence Anisotropy
250
200
150
Milli-Polarization Units
WJC-03
100
50
0
0
2e-5
4e-5
6e-5
8e-5
1e-4
Column A
SHC Protein (Molar)
72
Eleven compounds tested in an afternoon
Won-Jun Choi KD Peptide
WJC-01 8.40E-05 Ac-LpYQGLS-amide
WJC-02 1.00E-02 Ac-LYQGLS-amide
WJC-03 3.15E-05 FITC-pYQGLS-amide
WJC-05 3.86E-06 FITC-NH-(CH2)2-CO-pYQGLS-amide
WJC-06 1.87E-05 FITC-NH-(CH2)3-CO-pYQGLS-amide
WJC-08 1.16E-05 FITC-NH-(CH2)5-CO-pYQGLS-amide
WJC-07 1.47E-06 FITC-NH-(CH2)4-CO-pYQGLS-amide
WJC-11 1.25E-05 FITC-NH-(CH2)2-CO-LpYQGLS-amide
WJC-12 1.13E-05 FITC-NH-(CH2)3-CO-LpYQGLS-amide
WJC-13 9.12E-06 FITC-NH-(CH2)5-CO-LpYQGLS-amide
WJC-16 No Binding FITC-NH-(CH2)4-CO-YQGLS-amide
73
Phospho-peptide shows high affinity binding in
Biacore experiment
74
Screening of unlabeled compounds using
fluorescence anisotropy
75
Conclusions FA experiments

• Position and linker length of fluorescein label
  are important for binding affinity for Shc
• No binding seen with non-phospho-peptide
• No binding to other SH2 domain protein
• SPR studies are consistent with fluorescence
  anisotropy results
• FA can be used to screen unlabeled peptides

76
Conclusions

• The PCL is positioned to measure molecular
  interactions of protein/nucleic acid,
  protein/protein and small molecule to protein
  ligand.
• The PCL has the highest resolution SPR
  instruments (noise is 0.2 RUs at 10 Hz)
• The PCL has an evolving expertise in fluorescence
  spectroscopy
• The PCL has the highest sensitivity Edman
  sequencing (low picomolar), can purify anything
  from anything by HPLC and can develop Maldi-tof
  methods