Surface Plasmon Resonance

Section 1 Introduction to SPR

Class 1 Dr. Sandeep Kale Bioprocessing Group

DBT-ICT-Centre for Energy Biosciences, Institute of Chemical Technology, Mumbai-400 019, India

Labeling is Commonly Used

Traditional Biochemical Assays Label the reaction with color

Traditional Assays Labeling requires time and may impact the molecular interaction ELISA: Enzyme-Linked Immunosorbent Assay FIA: Fluorescent Immunoassay End-point measurement - not suitable for kinetics Not easy for small molecules

Power of Label Free Biosensor

Light Absorbed/Generated

Report tags on the reagent molecules generate light signal from the side of reaction. Is there another way of detection?
Polymer Surface

Metal Surface

Instead of detecting light signal from the side of reaction, detecting the refractive index change from the side of substrate. SPR is the most widely adopted label-free technology for detecting biomolecular interactions.
Refractive Index/Light Interference

What is SPR?

Metal Surface Refractive Index/Light Interference SPR Definition - SPR is a collective electron oscillation happening at metal/dielectric interface. It senses the refractive index change (mass change) within a thin layer on the surface.
SPR Features - Measures refractive index changes on the surface of a sensor chip - Delivers kinetics, equilibrium and concentration data SPR System - Consists of fluidics, optics, sensor chips, controllers and software - Implements experiment steps and analyzes sensorgrams

Basic SPR Terminology

Two molecules and a surface are required: Surface: usually a polymer with functional carboxyl groups. Ligand: an interaction reagent immobilized on the surface. Analyte: an interaction reagent flowed over the surface.
Analyte

Ligand

Y
Surface

Power of Label Free

How specific is the interaction? How fast is the interaction (ka)? How stable is the complex (kd)? How strong is the interaction (KD)?

Biomolecular Interactions Protein-Protein Protein-Peptide Antibody-Antigen Protein-Small Molecules

DNA-Protein DNA-DNA CarbohydrateProtein Lipids

How does SPR work?

Kretschmann Configuration
The surface is attached to a prism

Flow channel

Gold layer

Sensor surface

Prism

Polarized light

Optics

Detector

Surface Plasmon Resonance

100%

Plasmon

Gold Prism

Reflectivity

500 nm

Signal = SPR dip shift

0%

Incident angle of light

- A change in the refractive index of the medium shifts the angle of light in which it is totally absorbed the SPR dip is shifted. - The prism is made of high refractive index optical glass. - A variety of metals can be used for SPR device, but gold is the most common. - The sensitivity is determined by the accuracy of SPR dip determination.

SPR Signal
RU

[A]1 [A]2 [A]3 R=Response @ [A]

A Analyte B Ligand

R refractive index ( mass )

R molecular weight R mass on surface [ AB ]complex

Time

1 RU = 1106 RIU

1000 RU 1 ng mm 2 ( protein)

SPR Biosensor Workflow

1. 2. 3. 4. Chip coated with a polysaccharide layer Activate the COOH groups on the alginate layer Covalently bind Ligand to the COOH groups through amine coupling Deactivate the COOH groups

Ligand Surface

SPR Signal over Time

SPR Biosensor Workflow

5. 6. 7. 8. Flow over the Analyte and measure how fast it binds to the Ligand Wait for equilibrium of the binding Remove Analyte from flow Regenerate the Ligand layer

Analyte Ligand Surface

SPR Signal over Time

A +B

Kinetic and Equilibrium Constants ka

kd

A B

How fast is the interaction (ka)? Association Rate Constant: Rate of complex formation, per second, in 1 molar of A and B. units: 1/(M s)

d [ AB ] How stable is the complex (kd)? = k a [ A][ B ] k d [ AB ] Dissociation Rate Constant: dt The fraction of complex decays, per second. Units: 1/s
How strong is the interaction (KD)? Equilibrium/Steady State Constant: The strength of the interaction/complex. Units: M

d [ AB ] kd = 0 KD = dt ka

SPR for Kinetics

Models
ka & kd kd
Association Dissociation
Langmuir Mass Transport Limited Langmuir-Drift Heterogeneous ligand Heterogeneous analyte Bivalent analyte Two State Conformation Off-Rate Analysis

RU

Baseline

Time

Proteon has 8 different models for sensorgram fitting to obtain kinetics.

Why Kinetics?
100 KD = 1.0 nM 80 [A]=10 nM

60 Response, RU

5 Interactions Different ka and kd Same KD

40

Ka, M-1 s-1

20

Kd, s-1 1.00E-06 1.00E-05 1.00E-04 1.00E-03 1.00E-02

1.00E+03 1.00E+04 1.00E+05

Association Phase Dissociation Phase 400 600 time, seconds 800 1000 1200 1400

1.00E+06 1.00E+07

-20

200

SPR for Equilibrium

The relationship between equilibrium constant and temperature provides important thermodynamic parameters.
Gm = RT ln K D

RU

KD
Equilibrium

Baseline

Time

The change of KD at different temperatures can be used to determine Gibbs free energy change of the reaction.

Equilibrium Curve

100

Concentration range: 0.1KD-10KD

80

% [B] Bound

4-5 concentrations are sufficient

60 40

20

0 0.01 0.1 1

KD=10 nM
10 100 1000

[A], nM

Summary of SPR Technology

Advantages 1. Label Free 2. Real Time 3. Reuse of Ligand 4. Low Consumption 5. Automated 6. High Throughput

RU

KD ka & kd
Association Equilibrium

kd
Dissociation

Baseline

Regeneration

Time

Section 2 ProteOn XPR36

ProteOn XPR36 Technology

Flow System for SPR

Microfluidics is often employed in SPR biosensor.

Y Y Y Y Y

Advantages of using microfluidics and flow (1) Low sample consumption (2) Instrument automation (3) Conservation of analyte concentration for accurate kinetics determination

Serial Flow System

Serial flow system work principle
Flow Channel 1 - Sample Flow Channel 2 - Reference Flow Channel 1 - Sample Flow Channel 2 - Reference

Activation
Flow Channel 1 - Sample Flow Channel 2 - Reference

Ligand Immobilization

Analyte Injection The disadvantages: slow workflow, sensorgrams not aligned temporally, real time double reference unavailable.

Parallel Flow System

ProteOn XPR36 with criss-cross fluidics design realizes parallel flow system.
6L ig a nd s 6A n s te y l a

Interaction at Intersection

Advantages
1. 2. 3. 4. 5. 6 x 6 Injections in perpendicular directions Simultaneous analysis of 36 interactions Reduced experimental time High throughput for less cost Interspot and real time reference options

Criss-Cross Design

MCM (Multichannel Module)

Sensor Chip

Parallel Channel (Criss-Cross Design)

One-Shot Kinetics

Activate

One-Shot Kinetics

Ligand

One-Shot Kinetics

Deactivate

One-Shot Kinetics

Change Channels

One-Shot Kinetics

Analyte Rinse

One-Shot Kinetics

Analyte Rinse
3 2 1

5 4

6 5 4 3 2 1

Power of XPR36
Gradient of 1 Ligand 6 Ligands

an A 1f o s no it u li D 6

1-to-1 (Optimization of Kinetics)

an A 1f o s no it u li D 6

6-to-1 (High Throughput Kinetics)

Power of XPR36
6 Ligands 36 Ligands

se ty l an A 6

6-to-6 (Multiplex Screening)

e ty l an A 1 36-to-1 (High Throughput Screening)

Efficiency
A typical 1-to-1 experiment (Proteon Small Molecule kit)
L1 L1 L1 L1 L1 L1 L1 L2 L2 L2 L2 L2 L2 L2 L3 L3 L3 L3 L3 L3 L3 L4 L4 L4 L4 L4 L4 L4 L5 L5 L5 L5 L5 L5 L5 L6

Ligand: CA II
A1 A2 A3 A4 A5 A6 L6 L1: CA II L6 L6 L6 L6 L6 L2: CA II L3: CA II L4: CA II L5: CA II L6: CA II Low Density Gradient High

Analyte: CBS Dilutions

A1: CBS 6.67 M A2: CBS 2.22 M A3: CBS 0.74 M A4: CBS 0.25 M A5: CBS 0.08 M A6: CBS 0.00 M

Efficiency

By 1 injection, obtained (1) reliable kinetic results (2) optimized experiment condition. No need of regeneration. This is One-Shot Kinetics.

Artifact-Free
In serial flow systems

In the ProteOn, microfluidic design of parallel channels prevents artifacts at injection switch.
Raw data - free of artifacts

we see data presented is illustrated in Figure 21A. Note how these figures have been scaled so that the spikes (which often occur at the transitions between running buffer and sample) are visible. Referenced data - free of artifacts

Rich & Myszka, Survey of 2008, J. Mol. Recognit., 23, 1-64, 2010.

Experiment Conditions
Multiple Ligand Conditions
L1 A1 A2 A3 A4 A5 L1 L1 L1 L1 L1 L1 L2 L2 L2 L2 L2 L2 L2 L3 L3 L3 L3 L3 L3 L3 L4 L4 L4 L4 L4 L4 L4 L5 L5 L5 L5 L5 L5 L5 L6 L6 L6 L6 L6 L6 L6

Parameters typically optimized: 1. Surface Activation Conditions 2. Ligand Buffer Conditions 3. Ligand Density 4. Analyte Flow Parameters 5. Analyte Dilutions 6. Analyte Buffer Conditions 7. Reference Surface Preparation 8. Interaction Reproducibility 9. Mass Transport Effect 10.Ligand Regeneration
Experiment conditions are usually optimized within a couple of hours unattended time on ProteOn system.

ty l an A e lp i tl u M

A6

ProteOns parallel fluidics are the most efficient tool to optimize experiment conditions. Other SPR systems may not ever achieve the best conditions because of the time it takes for optimization.

Regeneration Optimization
Regeneration of the ligand surface to remove the analyte is performed to allow the ligand to capture further analytes. The regeneration step must remove all the bound analyte but not cause any harm to the ligand. ProteOn system has the power to optimize regeneration conditions in a single injection. Reproducibility is the validation of regeneration conditions:
NaOH H3PO4 Response decay over interaction cycles Urea MgCl2 SDS Other Response overlap over interaction cycles

Sensor Chips

Sensor Chips
Chip Selector

GLC Capacity <8 kRU 2D Surface monolayer Protein-Protein Antibody-Antigen

GLM Capacity <13 kRU 3D Surface Protein-Protein Antibody Capture Small Molecules

GLH

NLC

HTG / HTE

Capacity >20 kRU Small Molecules

Biotinylated Molecules

Capturing Surface Tris-NTA for His-tag proteins

Bare-gold chips are also available Liposomes capturing chips are under development (2012)

GLC Chip

The ProteOn GLC sensor chip has active carboxyl groups for covalent immobilization of biomolecules via amine groups. This versatile sensor chip is designed to bind a monolayer of analyte, and is an ideal choice for protein-protein interaction analysis.

GLM Chip

The ProteOn GLM sensor chip has active carboxyl groups for covalent immobilization of biomolecules via primary amine groups. This versatile sensor chip displays a rough carboxyl surface layer to provide higher ligand binding capacity for protein-protein and some small molecule interactions.

GLH Chip

The ProteOn GLH sensor chip also has active carboxyl groups for covalent immobilization via primary amine groups. This chip consists of a highly extended polymer layer for maximum binding capacity, and is an ideal choice for protein-small molecule and more difficult interactions.

NLC Chip

The ProteOn NLC sensor chip features Neutravidin bound to the polymer layer, and is an ideal choice for immobilization of biotinylated proteins, peptides, and nucleic acids.

HTG / HTE Chips

HTG

GLC

ProteOn Tris-NTA complexes contain three NTA functional groups for improved binding stability and increased binding selectivity to His-tag molecules.

Comparison between HTG and GLC chip performance with a model interaction. Ligand: 6His-Protein A Analyte: Human IgG1

HTG / HTE Selection

Chip Selector His-Tagged Protein Applications

HTG

HTE

Accomplishes high binding stability and selectivity Lower density Tris-NTA surface Binding Capacity ~6 kRU Resistance to non-specific binding: Higher Applications: Protein-Protein Protein-Peptide/DNA

Accomplishes high binding stability and selectivity Higher density Tris-NTA surface Binding Capacity ~12 kRU Resistance to non-specific binding: Lower Applications: Protein-Peptide/DNA Protein-Small Molecule

HTE Chip data

Ligand immobilization: His-tagged ERK2 (a MAP kinase protein) was captured on HTE surface. System preparation before analytes injections: running buffer was supplemented with 5% DMSO, the MCM rotated to horizontal orientation, and the baseline drift was evaluated. Analytes injections: six small molecules inhibitors (MW 300 - 500 Da) were tested sequentially. Five concentrations series of each inhibitor and running buffer in the sixth channel (for real time double referencing) were injected. Running buffer: Tris 50 mM, NaCl 150 mM , MgCl2 10 mM , MnCl2 1 mM + 5% DMSO, 0.005% Tween

His-ERK2 density reached to ~12,700 RU in saturation. Capturing uniformity and stability were very good in spite of the high density. Therefore, fixation of the ligand to the surface (by amine coupling) was not necessary. This is the same model system used in Technical Note 5965.

His ERK2 density (RU)

A1 A2 A3 A4 A5 A6 Ave CV% 12889 12769 12719 12766 12745 12827 12786 0.5%

HTE Baseline Drift and Data Processing

Raw data Interspot and EVC referenced A6 double referenced

Baseline drift before the first analyte injection was 4.3 RU/min. Real time referencing by the blank buffer channel was very accurate and allowed reliable data analysis.

HTE Interaction Analysis

ATP (MW 507.2)

* Staurosporin (MW 466.5)

* JAK3 inhibitor VI (MW 383.4)

* With some inhibitors, the highest concentration signal was omitted from the analysis since it was higher than the theoretical Rmax. This could
result from aggregates formation at high analyte concentrations.

HTE Interaction Analysis

Purvalanol B (MW 432.9) Amino Purvalanol A (MW 403.9)

* 1 Naphthyl PP1 (MW 317.4)

* With some inhibitors, the highest concentration signal was omitted from the analysis since it was higher than the theoretical Rmax. This could
result from aggregates formation at high analyte concentrations.

HTE Interaction Analysis

Inhibitor ATP (positive control) Staurosporin JAK3 inhibitor VI Purvalanol B Amino Purvalanol A Naphthyl PP1

MW (Da) 507.2 466.5 383.4 432.9 403.9 317.4

KD (M) 16.8 12.7 10.2 1.5 9.5 18.7

Rmax (RU) 41.3 156.7 103.6 56.2 93.7 55.1

Testimonials for HTE

"The higher capacity HTG chip proved to be very resilient. Candi Warner and Jim Carney (US Army) will be using them regularly and will be interested in further access to the high density surface -- Gary Ross, NASD FAS Were excited to see that Bio-Rad has developed a better His capture chip. Weve found that these new His-tag sensor chips are capable of capturing high densities of many Histagged proteins making it possible to characterize the binding of small molecules. Given the commonality of His tags, these chips will open up additional applications of the ProteOn technology. -- Dr David Myszka, University of Utah, Founder of Biosensor Tools

Referencing

What is Referencing?
Reference subtraction is needed to remove any contributions to the response, which are not results of the interaction we want to measure:
1. Bulk effects refractive index differences between running buffer and sample buffer 2. Baseline drift 3. Non-specific binding events

Life Science Group, Bulletin 3172

Reference Selection
There are two types of referencing: blank surface and blank buffer reference.
Ligand Surface + Analyte Injection = Interaction
Buffer Flow

Blank Surface + Analyte Injection

Buffer Flow

Analyte Ligand

Analyte

To subtract: bulk effect, NSB

Ligand Surface + Blank Injection

Buffer Flow

Ligand

To subtract: baseline drift

Reference Selection
Blank Surface Reference
Raw Data (Analyte Injection) Reference Subtracted (Analyte Injection)

RU

Blank Buffer Reference

RU

Double Referenced Sensorgram

Raw Data (Blank Injection)

RU

RU

Time

Time

Time

Time

Blank Surface Reference

ProteOn System Option 1 Channel Reference
This option is a traditional reference type in SPR biosensors. It reserves blank interaction spots for reference.

Blank Ligand Channel

Blank Surface Reference

ProteOn System Option 1 Channel Reference
Raw Data (Analyte Injection)

RU

Time

Reference Subtracted (Analyte Injection)

RU

Time

Blank Surface Reference

ProteOn System Option 2 Interspot Reference
This option is a unique reference type in the ProteOn system. It has two special advantages. Advantage 1: Close to interaction spots Advantage 2: Saving interaction spots consumed in conventional channel reference

Blank Surface Reference

ProteOn System Option 2 Interspot Reference
Raw Data (Analyte Injection)

RU

Time

Reference Subtracted (Analyte Injection)

RU

Time

Interspot Reference
L1 A1 1 A2 7 A3 A4 A5 25 A6 31 32 33 34 35 36 26 27 28 29 30
36 Reaction Spots

L2 2 8 14 20

L3 3 9 15 21

L4 4 10 16 22

L5 5 11 17 23

L6 6 12 18 24
42 Interspot References

13 19

Validation of Interspot
The interspot reference allows all 36 spots to be used, if it works in the same way as channel reference. IL2 / Anti-IL2 Ab Interaction Interspot reference

Channel reference

Interspot Reference Subtraction ka (M-1s-1) 8.38 x105 kd (s-1) 1.31 x10-4 KD (M) 1.56 x10-10

Reference Channel Subtraction ka (M-1s-1) 9.16 x 105 kd (s-1) 1.18 x10-4 KD (M) 1.29 x10-10

Blank Buffer Reference

ProteOn System Option 1 Injection Reference
This option is a traditional reference type in SPR biosensors. It requires injection of blank buffer prior to analyte injection.

Blank Buffer Reference

ProteOn System Option 1 Injection Reference
Blank Surface Reference Subtracted

RU

Time

Prior Blank Buffer Injection

Double Referenced Sensorgram

RU

Time

Blank Buffer Reference

ProteOn System Option 2 Real-Time Reference
This option is a unique reference type in the ProteOn system. It has two special advantages. Advantage: Real-time monitoring over the ligand surface

Blank Analyte Channel

Blank Buffer Reference

ProteOn System Option 2 Real-Time Reference
Blank Surface Reference Subtracted

RU

Time

Double Referenced Sensorgram

RU

Time

Row Reference
Row reference is a real time injection reference which is unique in Proteon. The baseline correction is perfect because it subtracts artifact in real time. It is especially useful in capturing surface.
Small Molecule Kit No Double Reference Baseline Drift -0.7 RU/min

Small Molecule Kit Injection Double Reference Baseline Drift -0.3 RU/min

Small Molecule Kit Real Time Row Double Reference Baseline Drift 0.0 RU/min

Row Reference
Analyte Ligand Capture Surface

Interaction

Reference

Ligand is washed off in dissociation causing baseline drift.

Corrected Interaction

Real-Time Double Reference

Drift is caused because the ligand is dissociating from the capture reagent.

sec

sec

Running Buffer

Ligand surface decay causes exponential baseline drift.

Real-time blank injection reference completely corrects the baseline drift.

Kits for Assay Development

One-Shot Kinetics Kit

IL-2 cytokine and IL-2 antibody interaction to demonstrate a detailed kinetic analysis in a single injection cycle without regeneration

Advantages
Protocol optimization
Ligand can be immobilized to different levels in six parallel channels
Different levels of activation solution can be used, or different pH binding buffers

In a single injection cycle, an analyte concentration series is run in the six perpendicular channels A set of sensorgrams is produced for kinetic analysis of each interaction

Optimal immobilization conditions are rapidly and efficiently determined through parallel processing

One-Shot Kinetics Protocol

Step 1. Ligand immobilization 30 uL/min
EDC/NHS (1:1) IL-2 Ab Ethanolamine EDC/NHS (1:5) IL-2 Ab Ethanolamine EDC/NHS (1:25) IL-2 Ab Ethanolamine EDC/NHS (1:125) IL-2 Ab Ethanolamine EDC/NHS (1:500) IL-2 Ab Ethanolamine EDC/NHS (1:25) Acetate pH 4.5 Ethanolamine

Step 2. Analyte injections 50 uL/min

IL-2 600 nM IL-2 300 nM IL-2 150 nM IL-2 75 nM IL-2 37.5 nM PBST 120 uL 120 uL 120 uL 120 uL 120 uL 120 uL

0.0 nM

Multiple Protein Kit

TEM1 -lactamase mutants (5) & -lactamase inhibitor protein

Demonstrates a detailed kinetic analysis of multiple simultaneous interactions Map protein interfaces

Multiple Protein Protocol

Step 1. Ligand immobilization 30 uL/min
TEM1-TE 0.5 uM 165 uL TEM1 R243/S130 1 uM 165 uL TEM1 S130/S235 0.5 uM 165 uL TEM1 K234 1 uM 165 uL TEM1 K243/S235 1 uM 165 uL 10mM acetate buffer pH 4.0 165 uL

Step 2. Analyte injections 50 uL/min

BLIP 600 nM BLIP 300 nM BLIP 150 nM BLIP 75 nM BLIP 37.5 nM PBST 250 uL 250 uL 250 uL 250 uL 250 uL 250 uL

0.0 nM

Small Molecule Kit

Carbonic Anhydrase II (CAII, MW 31,000) 4-carboxybenzenesulfonamide (CBS, MW 201)

The affinity and kinetics of the interaction is well characterized by SPR Assess the ability of the system to measure small molecule interactions Establishes a benchmark for system performance

Small Molecule Protocol

Step 1. Ligand immobilization 30 uL/min
EDC/NHS (1:1) CA II EDC/NHS (1:5) CA II EDC/NHS (1:25) CA II EDC/NHS (1:125) EDC/NHS (1:500) Ethanolamine Ethanolamine Ethanolamine CA II Ethanolamine CA II Ethanolamine

EDC/NHS (1:25) Acetate pH 5.0 Ethanolamine

Step 2. Analyte injections 100 uL/min

CBS 20 uM CBS 6.67 uM CBS 2.22 uM CBS 0.74 uM CBS 0.25 uM PBST 100uL 100uL 100uL 100uL 100uL 100uL

0.0 uM

Section 6 Application

ProteOn XPR36 Applications

Key Applications
Rapid Protocol Development (Guides, B3172, B5367, B5797) Antibody Characterization (B3172, B5360, B5449) Antibody Screening (B5540, B5820) High Throughput Screening (2 Publications of 36 Ligands) Epitope Mapping (B5540) Mutant Analysis (B5358, B5368) Ligand-Receptor Binding/Inhibition (B5449, B5367) DNA-Protein Interaction (B5449) Drug Compound Screening (B5679, B5960, B5965)

Antibody Screening
Blocking assays for epitope binning of antibody-antigen reactions Y. N. Abdiche, et al., Pfizer, USA, Analytical Biochemistry, 386, 172, 2009.
In tandem Premix Classical sandwich

Doesnt block Blocks

Each window showing 6 mAbs binding to Ag captured by a single mAb.

Faster!
Y, blocks N, doesnt block

Power of XPR36
A1 A2 A3 A4 A5 A6

L1

L7

L13

L19

L25

L31

How to exploit the power of 6 by 6 interaction spots design?

L2

L8

L14

L20

L26

L32

36 Ligands 6 x 6 Injections

L3

L9

L15

L21

L27

L33

High throughput screening!

L4

L10

L16

L22

L28

L34

L5

L11

L17

L23

L29

L35

L6

L12

L18

L24

L30

L36

6 analytes react with 6 ligands to realize 36 throughput screening. This feature is very useful in antibody screening, small molecule screening, etc.

36 Ligand Screening

Activate

Activate 1 channel instead of all 6 together.

36 Ligand Screening

Ligand Deactivate

Inject 6 ligands to put them in a column instead of the entire surface.

36 Ligand Screening

Activate

Activate 1 channel instead of all 6 together.

36 Ligand Screening

Ligand Deactivate

Inject 6 ligands to put them in a column instead of the entire surface.

36 Ligand Screening

After repeating all six channel

36 Ligand Screening

Analyte Rinse

36 Sensorgrams

36 Ligand Screening
Analytical Biochemistry Journal Front Cover

36 Ligand Screening
Spot-to-spot reproducibility Interaction of the same antibody ligand, with Analyte 1, on different locations on the 36 ligand array (L1A2 and L1A5).
NOTE: residuals are randomly clustered close to 0

36 Ligand Screening
Mass transport limited interaction generates good quality kinetic data Interaction of the Analyte 1 with antibody immobilized on position L1A3.
NOTE: residuals are randomly clustered close to 0

36 Ligand Screening
Shows excellent kinetic output for a low density ligand Interaction of the antibody ligand at location L5A6, with Analyte 1
NOTE: residuals are randomly clustered close to 0

36 Ligand Screening
Expanding the ProteOn XPR36 biosensor into a 36-ligand array expedites protein interaction analysis Y. N. Abdiche, et al., Pfizer, USA, Analytical Biochemistry, 411, 139, 2011.

18 rounds of binning on a single spot

Constructing the 36 ligand array for high throughput screening.

Protein Domain Interaction

Membrane protein domain interaction screening L. Jiang and A. N. Barclay, Immunology, 129, 55-61, 2010.

Search for interactions between leucocytes membrane protein extracellular domains with 6 x 6 sample injection configuration.

Protein Domain Interaction

Ligand Preparation: Recombinant proteins are immobilized to the sensor chip surface by capturing and fixing.
Extracellular domain of leukocyte surface proteins

Fixing step: EDC/NHS was applied to covalently fix the recombinant proteins to the OX68 monoclonal antibody to prevent dissociation.

rCD4 d3+4

Monoclonal antibody OX68 against rCD4d3 + 4

Protein Domain Interaction

Analyte Preparation Recombinant proteins Proteins coated on nanoparticles

Extracellular domain of surface proteins

rCD4 d3+4

Biotin

Avidin

36 Ligand Screening
Analytes
CEACAM8 (CD67) NCR3 (CD337) TACTILE (CD96) PVR (CD155, NECL5) CD226 (DNAM) NECL1 NECL2 PSG (CD66F) mGPNMB CD85 CD85C CD85E CD85H CD85K (LILRB4) CD83 CMRF35 (CD300C) CRTAM CD5 CD6 CD72 TOSO LAIR1 (CD305) ESAM CD99 mMAIR CD73 TREM1 TREM2 CD200 CEACAM1 (CD66) CEACAM5 CEACAM6

L1

L7

L13

L19

L25

L31

L2

L8

L14

L20

L26

L32

36 Ligands 6 x 6 Injections

L3

L9

L15

L21

L27

L33

L4

L10

L16

L22

L28

L34

L5

L11

L17

L23

L29

L35

L6

L12

L18

L24

L30

L36

36 Ligand Screening

Assay validation and sensitivity verification by analyzing known weak interaction.

36 Ligand Screening

new interactions

known interactions

Small Molecule Applications

High-throughput small molecule screening
L1 L1 L1 L1 L1 L1 L1 L2 L2 L2 L2 L2 L2 L2 L3 L3 L3 L3 L3 L3 L3 L4 L4 L4 L4 L4 L4 L4 L5 L5 L5 L5 L5 L5 L5 L6

Ligand: CA II
A1 A2 A3 A4 A5 A6 L6 L1: CA II L6 L6 L6 L6 L6 L2: CA II L3: CA II L4: CA II L5: CA II L6: CA II Low Density Gradient High

Analyte: Sample Dilutions

A1: Inhibitors A2: Inhibitors A3: Inhibitors A4: Inhibitors A5: Inhibitors A6: Inhibitors Low Concentration Gradient High

Small Molecule Applications

Kinetics of 9 sulfonamide inhibitors for carbonic anhydrase isozyme T. Bravman, et al., Bioanalytical Chemistry, 358, 281, 2006.
MW: 341 MW: 172 MW: 331

MW: 201

MW: 250

MW: 236

MW: 157

MW: 217

MW: 222

Molecule of 157 Dalton gives significant signal at 0.2 M.

Small Molecule Applications

Table of Results

Thermodynamic Analysis
transition state

RU

ka & kd
Association

Equilibrium

KD kd
Dissociation

1
Energy
unbound

2
bound

Baseline

Regeneration

Binding Coordinate

Time

van t Hoff plot: intercept = -/

ln(kah/kBT) lnKD

Eyring plot: intercept = -/ slope = /

slope = /

Thermodynamics are obtained through kinetics while temperature is varied.

1/T

1/T

Small Molecule Applications + Thermodynamic Analysis

Kinetics of 9 sulfonamide inhibitors for carbonic anhydrase isozyme T. Bravman, et al., Bioanalytical Chemistry, 358, 281, 2006.

Thermodynamic Analysis
Kinetics-thermodynamics of binding site-modified proteins. J. Kulman, Seattle, USA, Bio-Rad Webinar Series, 2010. Collagen
(Blood vessel subendothelium)

GPIb

(Platelet)

mAb NMC4

SPR Purpose: Find the epitope of the mAb Ligand: VWF A1 Domain Analyte: mAb Fab

Thermodynamic Analysis
Kinetics-thermodynamics of binding site-modified proteins.

Complications:
Low reagent quality Ligand heterogeneity Ligand degeneration Time Expense

Soutions:
Novel expression / capture method Throughput enhancement VWF A1 Domain

Thermodynamic Analysis
Purification from Lysates: Ca2+-dependent immunocapture of PC-tagged peptides

EDQVDPRLIDGK
g Ta
PC
Ca

PC Ca

Ta g

HPC4

Regeneration with EDTA

Interaction

Recycle is facilitated by capturing surface.

Thermodynamic Analysis
1st Experiment protocol optimization on accuracy and reproducibility.
Rmax Serial Replicates KD (10-10M) 2

170.1 1.0

6.38 0.34

12.1 2.3

150.3 1.2

6.23 0.40

9.1 2.1

107.0 0.3

6.23 0.12

5.7 1.1

71.1 1.4

7.19 0.77

3.8 0.3

40.9 2.1

8.26 1.95

2.9 0.6

Thermodynamic Analysis
2nd Experiment validate the reproducibility while temperature varies.
Temp (C): 15.0 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0

Parallel Replicates

Thermodynamic Analysis
3rd Experiment Obtain kinetic constants of all proteins while temperature varies.
transition state

Energy
unbound

bound

Binding Coordinate
KD kd
Dissociation

RU

ka & kd
Association

Equilibrium

Baseline

Regeneration

Time

Thermodynamic Analysis
Experiment time: 32 hours The customer mentioned that the throughput of XPR36 significantly reduces the experiment time from a month in conventional biosensors to days.

Ligand WT Q628A N633A K660A L664A Ligand WT Q628A N633A K660A L664A Ligand WT Q628A N633A K660A L664A

G 53.8 0.1 50.2 0.1 50.3 0.0 49.3 0.1 50.2 0.1 H 72.8 4.1 67.6 2.4 63.5 2.1 67.8 2.6 58.2 2.7 TS -19.0 4.1 -17.4 2.4 -13.2 2.1 -18.5 2.6 -8.0 2.7

G a 35.3 0.0 35.6 0.1 35.9 0.0 36.7 0.0 36.1 9.1 H a 15.7 1.5 17.5 2.7 15.9 1.1 19.8 2.2 18.9 2.4 TS a -19.6 1.5 -18.0 2.7 -20.0 1.1 -16.9 2.2 -17.1 2.4

G d 89.1 0.1 85.8 0.1 86.1 0.1 86.0 0.0 86.3 0.0 H d 88.6 3.2 85.3 2.3 79.6 2.4 87.0 1.9 76.5 2.1 TS d -0.5 3.2 -0.5 2.3 -6.5 2.4 1.0 1.9 -9.8 2.1

HTG Chip Applications

(A) Mono-NTA is the current standard method for capturing His-tagged proteins. Mono-NTA achieves weaker binding and results in: baseline drift, distorted kinetic results and inaccurate sensorgram fitting. (B) ProteOn Tris-NTA complexes contain three NTA molecules for improved binding stability and increased binding selectivity to His-tag molecules.

Enhanced Binding Stability

Visual difference: Much higher performance of the ProteOn HTG sensor chip. Quantitative difference: HTG chip achieves 3 to 50% improvement over the mono NTA chip

Tris-NTA

Mono-NTA

Compare Tris-NTA to Mono-NTA 1. Enhanced binding stability for more reliable experiment results 2. Improved binding selectivity for better data quality and online purification 3. Good regeneration capability for reduced cost

Chip Regeneration
Examples of regeneration 5 ug/mL Protein A Sequential repeats with longer injections

Overlap Consistency Overlap

5 ug/mL Ubiquitin Sequential repeats with longer injections

HTG Advantages
Ligand: 6His-Protein A Flow Rate: 50 L/min Analyte: Human IgG1 (100, 50, 25, 12.5, 6.25, 0 nM)

HTG

GLC

Chip Type HTG chip GLC amine coupled Protein A

Ligand Density RU 58 101

ka 1/Ms 2.11E+05 2.26E+05

kd 1/s 1.45E-04 1.01E-04

KD M 6.86E-10 4.46E-10

Rmax RU 358 355

The baseline drift of the captured protein A is insignificant ~0.7 RU/min. The kinetic constants of the his-tagged captured and the amine-coupled protein A are very similar. The activity of the his-tagged captured protein A is 1.8 fold higher than the amine-coupled protein A activity.

Antizyme / ODC Kinetics

Kinetics of Antizyme (Az) interaction with Ornithine Decarboxylase (ODC) Prof Gideon Schreibers lab at Weizmann Institute, J Mol Biol, 390, 503, 2009.

Az X ODC X

Az I Polyamine Synthesis

Where are the binding sites between ODC / AzI and Az? Binding sites were proposed by structural data. The mutagenesis approach is used to give an accurate answer, as an alternative to the challenging crystallography approach. (1) High throughput Ala scan of Az (41 Az mutants nonbiased panel). (2) Double-mutant cycles between 5 ODC mutants and Az.

Antizyme / ODC Kinetics

Chip: Custom made Tris-NTA chip (equivalent of HTG chip) Ligand: Az mutants with a 10-His tag attached to the N-terminus Analyte: ODC or AzI

Az WT / ODC WT

ProteOn XPR36

Az F213A / ODC WT

Workflow

Example

Antizyme / ODC Kinetics

Results of 9 Az mutants binding to ODC / AzI which are different from Az WT

The binding sites between Az and ODC / AzI are mapped. Identification of these residues drives computational modeling and double-mutant cycles (DMC) analysis.

Antizyme / ODC Kinetics

Results of DMC analysis of Az / ODC and comparison to the computational model
Green DMC Analysis Blue Computational

The ProteOn XPR36 is perfectly suited for such DMC measurements as six ligands (different Az mutants) can be measured in parallel against up to six analytes (or six concentrations of one ODC analyte as done here) in a single injection.

Antizyme / ODC Kinetics

Purified Az S191A 230 RU

Az crude extract + 15 mM imidazole 935 RU

Az crude extract no imidazole 1460 RU

Example of optimizing conditions on HTG chip (The Az crude extract samples was not purified.)
Purified Az S191A concentration 19 ug/mL. All crude extract samples were diluted 250-fold in running buffer. Ligand uniformity is excellent in all channels.

Antizyme / ODC Kinetics

Flow = 70 uL/min; Ch 1-6: 150 uL of ODC, 0 ,1.23, 3.7 ,11.1, 33.3 and 100 nM

Purified Az

Az crude extract + 15 mM imidazole

Ligand Density (RU) 230 935 1460 ka (1/Ms) 2.59E+05 1.87E+05 1.63E+05 kd (1/s) 2.54E-04 1.42E-04 1.41E-04 KD (M)

Az crude extract no imidazole

Rmax (RU) 149 389 428 Relative Activity 1.0 0.64 0.45

Sample Purified S191A Az Az WT crude extract + imidazole Az WT crude extract no imidazole

9.80E-10 7.60E-10 8.64E-10

The Az captured from crude lysate shows good binding activity. Non-specific binding of ODC to empty channel with Ni2+ or to interspots (without Ni2+) are not observed. These results show that the ratio between specific and non-specific binding of crude lysate samples could be improved by altering the binding conditions.

Cell Biology
Staphylococcus aureus is a single cell suspension with cells having a diameter of approximately 1 um. These cells are fixed and contain a high level of protein A on their surface. The protein A on the cell surface has a high affinity for the Fc domain of IgG. The data were collected by Japan sales team for demo purposes.

Ligand: Rabbit Polyclonal IgG Chip: GLC Buffer: Acetate pH 5.5 Concentration: 20 ug/mL Flow Rate: 30 uL/min

Analyte: Staphylococcus aureus Dilution Ratio: A1: x10 A2: x30 A3: x90 A4: x270 A5: x810 Flow Rate: 100 uL/min Contact Time : 180 sec Dissociation Time: 360 sec

Vaccine Production
Test of recombinant H1N1 virus hemagglutinin (HA) globular head for vaccines S. Khurana, et al., FDA, USA, PLoS ONE, 5, e11548, 2010.
Recombinant proteins are used to rapidly produce vaccines to control disease pandemic. In this case H1N1 A/California/07/2009 virus is the target. H1N1-HA (1-330) and H1N1-HA (1-480) are expressed in E. coli under controlled redox refolding conditions. H1N1-HA0 is a mammalian cell derived recombinant H1N1-HA with full length. Ligand: H1N1-HA (1-330), H1N1-HA (1-480), H1N1-HA0 (as labeled in the graph titles) Analyte: Infected sera with antibody from ferrets (B, C, D) Vaccined sera with antibody from human (E, F, G) SPR is used to quality control the expression of conformational native antigenic epitopes.

Vaccine Production

The CDC microneutralization method gives consistent results with ProteOn method on the antibody titer in infected sera from ferrets (no rise until day 5 and peak at day 7).

Conclusions of the paper

(1) The combination of bacterial expression system with established quality control methods could provide a mechanism for rapid large scale production of influenza vaccines in the face of influenza pandemic threat. (2) HA1 proteins derived from the newly spreading virus can be rapidly expressed in bacterial systems several months before the traditional approach using vaccine strains. With appropriate testing methods in place to monitor proper folding and biological activity, this simple and efficient approach may provide an early vaccine for large scale production to fulfill global vaccine needs in a much shorter time frame.

Section 7 Theory

Response Estimate
Use the equation below to estimate Rmax value. For protein-protein interaction

For all types of interaction Rmax Maximum theoretical response of analyte for a given ligand level n Stoichiometric number of analyte-ligand interaction MA Analyte molecular weight ML Ligand molecular weight NA Analyte mass refractive index increment over water NL Ligand mass refractive index increment over water

SPR for Kinetics

Models
ka & kd kd
Association Dissociation
Langmuir Mass Transport Limited Langmuir-Drift Heterogeneous ligand Heterogeneous analyte Bivalent analyte Two State Conformation

RU

Baseline

Time

There are different models for curve fitting in order to obtain ka and kd in various situations, Bio-Rad Proteon XPR36 has 7 fitting models in the software.

What is in Kinetics?

How fast is the interaction (ka)? Association Rate Constant Rate of complex formation, per second, in 1 molar of A and B. units: 1/(M s) How stable is the complex (kd)? Dissociation Rate Constant The fraction of complex decays, per second. Units: 1/s How strong is the interaction (KD)? Equilibrium/Steady State Constant The strength of the interaction. Units: M

Constants

d [ AB ] = k a [ A][ B ] k d [ AB ] dt

First Order Reaction

d [ AB ] =0 dt

kd KD = ka

Steady State (Equilibrium)

dR = ka [ A]( Rmax R ) kd R, R( 0) = 0 dt

Kinetic Equation in RU

Association Phase

Req =

[ A] Rmax
[ A] + K D as t , R Req

1 R = Req 1 ( ka [ A ] +kd ) t e 1 Req = Rmax 2

at [ A] = K D ,

Reality Check!

Dissociation Phase

R = R0

R = R0 e kd ( t t 0 )

1 R = R0 2
R =0

as t ,
t0

R 0

t1 2

ln ( 2 ) = Reality Check! kd

Kinetic Models
Langmuir: Simple 1:1 bimolecular interaction Langmuir + Drift: Calculates linear drift /loss of mass from the surface of the chip Mass Transport: The diffusion of the analyte to the surface is slower than the interaction itself. Heterogeneous Ligand: One analyte is binding two separate ligand species. Heterogeneous Analyte: Two analytes compete for binding to one ligand site. Bivalent Analyte: The injected analyte has two binding sites to the ligand. Two State: Accounts for a complex that changes shape after interacting with the analyte

Equilibrium
According to Langmuir kinetics, equilibrium is reached when the forward and reverse reactions are equal to each other. At the case of high affinity binding, K is large and [L] at equilibrium is very low. Thus the equilibrium reaction rate is very low and it takes a long time to approach equilibrium. In contrast, low affinity binding approaches equilibrium very fast.

Sensorgrams of protein-small molecule interaction

Sensorgrams of antibody-antigen interaction

SPR for Equilibrium

The relationship between equilibrium constant and temperature gives useful thermodynamic parameters.
Gm = RT ln K D

RU

KD
Equilibrium

Baseline

Time

Besides the meaning of KD itself, the change of KD over temperature gives thermodynamic values, such as Gibbs free energy change of the reaction.

SPR for Thermodynamics

RU Equilibrium

KD

Gm = RT ln K D Gm = H m T S m

Baseline

Time

The change of KD over temperature gives thermodynamic values, such as Gibbs free energy change of the reaction.
vant Hoff plot: intercept = -/
lnKD

slope = /

H m 1 S m ln K D = + R T R

The usual method is vant Hoff analysis.

1/T

SPR for Thermodynamics

transition state

RU

ka & kd
Association

Equilibrium

KD kd
Dissociation

1
Energy
unbound

2
bound

Baseline

Regeneration

Binding Coordinate

Time

1 Gm k BT k1 = exp( ) hc RT

G k BT k2 = exp( 2 m ) h RT

Thermodynamics is measured by the change of kinetics over temperature.

Eyring plot: intercept = -/
ln(kah/kBT)

slope = /

H m kh 1 Sm ln = + k BT R T R

The usual method is Eyring analysis.

1/T

Synergy of Ligands
Antibody Characterization Using the ProteOn System Dr Hassan Issafras at Xoma, Bio-Rad Webinar Series, 2011.

IL-1 RI (receptor) binds to IL-1B (ligand) with the presence of RAcP (accessory protein). They need to figure out the signaling path because they have an Ab drug targeting the ligand.

Synergy of Ligands
Ligand: IL-1 RI Analyte: IL-1B

Problem The binding is biphasic. The fitting is not consistent with the sensorgrams.

Hypothesis The receptor sRI (soluble IL1-RI) has two states for binding.

Synergy of Ligands

Solution Immobilize the complex of the receptor, the accessory protein and the ligand. The accessory protein will quickly lock up the ligand structure when binding happens.

Enhancement The accessory protein is in excess to guarantee fast lock up.

Synergy of Ligands

Implementation The parallel injection of the ProteOn system was utilized to include the protein single forms and mixtures in a uniform environment for kinetic analysis.

Small Molecule Applications

Methodology of small molecule screening, Bio-Rad technical note 5679.
L1 L1 L1 L1 L1 L1 L1 L2 L2 L2 L2 L2 L2 L2 L3 L3 L3 L3 L3 L3 L3 L4 L4 L4 L4 L4 L4 L4 L5 L5 L5 L5 L5 L5 L5 L6

Ligand: CA II
A1 A2 A3 A4 A5 A6 L6 L1: CA II L6 L6 L6 L6 L6 L2: Anti-DNP Ab L3: HSA L4: Rabbit IgG L5: Blank L6: Blank

Analyte: Sample Dilutions

A1: Inhibitors, DNP, Digitoxin A2: Inhibitors, DNP, Digitoxin A3: Inhibitors, DNP, Digitoxin A4: Inhibitors, DNP, Digitoxin A5: Inhibitors, DNP, Digitoxin A6: Inhibitors, DNP, Digitoxin

Small Molecule Applications

Methodology of small molecule screening, Bio-Rad technical note 5679.
Digitoxin HSA MW 201 MW 157

DNP Anti-DNP Ab

High throughput allows automatic experiment with long walk away time.