Introducing the Octet as a reliable novel biosensor in characterizing

protein/protein interactionsYasmina Abdiche, PhD

Rinat Pfizer, South San Francisco, CA

Overview

1.The technology2.Octet v. Biacore3.“One-shot” kinetics4.Blocking as applied to epitope binning

a.Stepwiseb.Classical sandwichc.Premix

Array Biosensor Platforms

Rich & Myszka, Anal Biochem 361 (2007) 1-6

The TechnologyDip-and-read assay

using disposable fiber optic biosensors

Bio-Layer Interferometry

(BLI)

Parallel processing of

8 samples

free analyte

bound analyte

ligand

reflected beams

incident white light

tipWavelength (nm)

Rel

ativ

e in

tens

ity

Time (s)

Shi

ft (n

m)

bindunbindΔλ

Octet v. Biacore

BLI8 parallel sensors

Single-use tipsDip into open shaking plate

No microfluidicsNeed 200uL/well

Sample is reusable and recoveredHeat > room temp.

SPR4 serial flow cellsRe-usable chips

Inject from sealed vialsMicrofluidics

>25uL/injectionSample is consumed

4-40ºC

reference

“One-Shot” Kinetics

Res

pons

e (R

U)

Biacore

biotin-peptide (4kDa)

Fab

Octet

Uniformly load biotin-peptide (on tips)

Fab gradient

MethodOctet - 6 repeats over 5h with regen.

Biacore - 7 independent analyses

KD (nM) 25ºC1.81 ± 0.25 (14%)0.97 ± 0.27 (28%)

kon (1/Ms)1.14e52.43e5

koff (1/s)2.06e-42.16e-4

30mins10min Time (s) 10min1min

Shi

ft (n

m)

Benchmark Study

Shi

ft (n

m)

3h15minTime (s)

Res

pons

e (R

U)

30min3min

a/c GST-Ag

Fab

MethodMyszka et al. (n~150)

OctetBiacore (n=3)

Octet Biacore

DIPIA, Phoenix AZ, 2007

KD (pM) 25ºC470 (95% = 210-652pM)

329307 ± 3%

kon (1/Ms)1.2e5

1.51e51.39e5

koff (1/s)4.9e-5

4.97e-54.28e-5

Anti-Id Fab KineticsOctet

-0.20

0.20.40.60.8

11.21.41.6

1000 1500 2000 2500 3000 3500 4000 4500 5000-0.2

00.20.40.60.8

11.21.41.6

1400 1600 1800 2000 2200 2400 2600 2800 3000-0.2

00.20.40.60.8

11.21.4

1400 1600 1800 2000 2200 2400 2600 2800 3000

1min 5min

20min 40min

kon=3.83e4 (1/Ms)koff=7.64e-4 (1/s)

KD=20nM

1.96e41.81e-3

92

2.15e45.36e-3

249

2.15e45.35e-3

248

1.89e41.71e-3

91

3.52e49.89e-4

28

Shi

ft (n

m)

Time (s)

Res

pons

e (R

U)

Biacore 20min 5min

Dissociate into buffer

Dissociate into sinkExcellent fit to simple

model

Minimizing Rebinding Using a SinkOctet Biacore

Rebinding, sofit initial decay

Shi

ft (n

m)

Time (s)R

espo

nse

(RU

)

Time (s)KD (nM)

4226 ± 9 (n=3)

Biotin-peptide (4kDa)

Fab

kon (1/Ms)3.76e44.91e4

MethodOctet

Biacore

koff (1/s)1.57e-31.25e-3

15min 30min

1min

1h

Resolving a Tight InteractionS

hift

(nm

)

Time (s)

Octetbiotin-Ag on tip

kon=3.36e5 (1/Ms)koff<2e-6 (1/s)

Within noise (±0.1nm)KD<6pM

Res

pons

e (R

U)

Biacorea/c Ag on chip

kon=1.07e5 (1/Ms)koff<2e-6 (1/s)

Within noise (±1RU)5% rule

KD<19pM

Dissociate into sink 8h-dissociation phase

Direct Binding of Small Molecules

within noise

SS (-) and blank controls

Typically:a/c IgG = 4nm

Bind 25kDa Ag = 1nmNoise = 0.1nm

KD=0.64 ± 0.35nM (n=17)

a/c IgG on AR tips Bind 4kDa-peptide Detect with another mAb(sandwich)

Octet is 100fold less sensitive than Biacore

Blocking Strategies

1. “ABC” sandwiches where A does not bind C directly2. Choose assay orientation and surface chemistry3. Run proper controls, e.g., buffer, irrelevant partner, “self-sandwich”4. Where a mAb binds is often more important than its KD

Stepwise blocking

A

B

C

tip

Classical sandwich

A

BC

tip

Premix

A

BC

tip

Epitope Binning

Stepwise blocking

tip

Classical sandwich

Ag

tip

Premix

Ag

tipY YAg

1. Each strategy has its pro’s and con’s2. Consider MW and aggregation state of Ag3. Improve unattended throughput by preparing surfaces offline using

“batch immobilization”4. Can often transfer known regeneration cocktail from Biacore

Stepwise Blocking

Re-use tips x5 by regenerating them

a/c Ag (n=8) Primary IgG (array) Secondary IgG (fix)

Total run time = 3.5h

buffer

mAb #6mAb #6 (SB)

Time (s)

Shi

ft (n

m)

Primary mAb (array) - 5 repeat cycles

mAb#6 falls off

Shi

ft (n

m)

Time (s)

Secondary mAb - 2 3 4 5 6

Verifies binding to “free” Ag on tip

Stepwise Blocking

SBSB

SB

SB SB

-0.50

0.51

1.52

2.53

3.54

300 600 900 1200 1500 1800

a/c MAb #2

-0.2

0.2

0.6

1

1.4

1.8

2000 2200 2400 2600 2800 3000 3200

MAb #5 on tip

-0.2

0.2

0.6

1

1.4

1.8

2000 2200 2400 2600 2800 3000 3200

MAb #3 on tip

-0.2

0.2

0.6

1

1.4

1.8

2000 2200 2400 2600 2800 3000 3200

MAb #2 on tip

-0.50

0.51

1.52

2.53

3.5

200 400 600 800 1000 1200 1400 1600 1800

a/c MAb #3

-0.5

0

0.5

1

1.5

2

2.5

3

200 400 600 800 1000 1200 1400 1600 1800

a/c MAb #5

Premix Approacha/c IgG Premix Ag + mAb (array)

Ag alone

Confirmed 3 epitope bins using two blocking strategies

BIN 1: #1 and 3BIN 2: #5 and 6BIN 3: #2, 4, and 7

Binding (sandwich)UnclearNo binding (blocking)

“Traffic light” binding matrix

Time (s)

Shi

ft (n

m)

MAb#2

MAb#3

MAb#5

Classical SandwichRaw data

a/c IgGn=8

Bind Ag(monomer)

Bind IgGarray

Sandwich

Pooled 4 Abs

Self-sandwich (SS)

IgG

on ti

p

IgG in solution

Example shown

Summary

1. The Octet is a simple, yet versatile complement to Biacore2. “One-shot” kinetics provides an appealing assay format3. Returns accurate kinetic rate constants for a wide range of

interactions when compared head-to-head with Biacore4. The direct binding of small molecules is beyond the current

detection limit5. Well-suited to blocking, especially in the context of epitope

binning

Acknowledgements

Alanna Pinkerton Jaume PonsDirector

Protein Engineering

Dan Malashock