GE Healthcare
Outstanding sensitivity
for confident label-free
interaction analysis
Outstanding sensitivity for confident
label-free interaction analysis
The precision of data from surface plasmon resonance (SPR)based analyses, and thus the confidence with which results
can be interpreted, is strongly coupled to the sensitivity of the
instrument.
Higher sensitivity means that interactions involving low
molecular weight analytes can be studied with more
precision and confidence. It means that you can precisely
measure very low concentrations of analytes. Unstable
molecules such as membrane proteins can be more readily
studied because preservation of function is necessary in
only a fraction of the immobilized total. Importantly, the
confidence with which measured on- and off-rates can be
interpreted is optimal when using the lowest possible level of
immobilized interacting partner.
The range of kinetic rate constants that can be determined
using Biacore T200 is wider than ever before (Fig 1). This
allows resolution and ranking of strongly binding antibodies,
as well as the most rapidly associating compounds, an
important parameter where bioavailability is limited in vivo; in
such cases association must be faster than clearance for any
biological effect to occur.
As on- and off-rates may be precisely measured even at the
extremities of kinetic behavior, researchers are not restricted
to steady state analysis. One major benefit of using kinetic
analysis rather than the steady state model is that it is not
necessary to use high concentrations of interacting partners
in solution. Further, the use of low concentrations enables the
analysis of low-solubility molecules as well as the avoidance
of phenomena such as promiscuous binding.
M
1f
Biacore™ T200 is an extremely sensitive, label-free
interaction analysis system offering the opportunity to work
confidently at the limits of kinetic, molecular weight and
concentration ranges, bringing improvements in data quality
to a wide range of new application areas.
• Measure active concentrations at levels previously
beyond quantitation limits
• Perform precise affinity determinations of interactions
involving the smallest compounds with confidence
• Analyze interactions characterized by on- and off-rates
at the extremities of the kinetic scale
• Characterize antibodies, without the complicating effects
of avidity
Expanding the measureable kinetic space
Interactions characterized by similar affinities can have very
different on- and off-rates. On-rates may be regarded as a
measurement of recognition between interacting partners,
or readiness with which interacting partners associate,
while off-rates indicate the stability of the complex. By
deconstructing affinity into on- and off-rates, a much deeper
picture is obtained on how the dynamics of molecular
interactions relate to protein function.
2
M
0f
fM
10
M
1p
10
M
0p
pM
10
M
1n
9
10
8
nM
M
0n
10
Association: log (ka)
Here, we showcase some examples of how researchers in
academia and biotechnology have discovered the benefits of
the sensitivity of Biacore T200.
10
7
M
1µ
6
1
10
µM
5
4
µM
0
10
2
M
1m
3
-5
-4
-3
-2
-1
0
Dissociation: log (kd)
Fig 1. Kinetic measurements over the broadest range, from the fastest
on-rates to the slowest off-rates. (1) Interactions with apparently similar
affinities can have very different kinetic profiles. Resolution into component
on- and off-rates can improve candidate selection. (2) Even interactions at
the extremes of kinetic behavior, for example, with very slow off-rates, can
be detected and differentiated with confidence.
Working with sensitive targets
10 000
8000
Response (RU)
3.5
3
2.5
2
1.5
1
0.5
0
-0.5
-1
16
12
8
4
0
50
150
250
Time (s)
10
8
6
4
2
0
0.001
6000
4000
2000
LOD
0
0.001
0.01
Concentration (µg/ml)
0
20
40
60
80
100
120
1000
140
Time (s)
3.5
3
2.5
2
1.5
1
0.5
0
-0.5
-1
1
Concentration (µg/ml)
Measurement 1
Measurement 2
0 µg/ml
0.002 µg/ml
0.004 µg/ml
0.008 µg/ml
-20
Fig 3. Concentration analysis of the therapeutic human antibody, Xolair.
The limit of detection for Xolair was determined from the mean of 10 blanks
+ 3 SD (0.8 RU).
Precision through sensitivity:
detect the smallest molecules
-20
0
20
40
60
80
100
120
140
Time (s)
Fig 2. Binding of a small molecule, xanthine amine congener (XAC), to
stabilized His-tagged GPCR (StaR®) A2. Data of higher quality is generated
by Biacore T200 (lower sensorgram) compared to Biacore T100 (upper
sensorgram) when using low levels of immobilized GPCR. StaR® was
transiently expressed and isolated from the solubilized membranes of HEK
cells and captured on Sensor Chip NTA. A concentration series of XAC was
injected at a flow rate of 30 µl/min. The analysis temperature was 10˚C.
Data courtesy of Dr. Andrei Zhukov, Heptares Therapeutics Ltd, Welwyn
Garden City, UK.
Precision through sensitivity: quantitate the
lowest concentrations of analyte with confidence
Confident measurement of very low protein concentrations
is important in areas such as quality control during
production of biotherapeutics or in the detection of
trace quantities of unwanted proteins. In addition, many
intracellular proteins are expressed at extremely low
concentrations and their quantitative analysis requires high
precision even at the limits of sensitivity. The detection limit,
as well as the measurable dynamic concentration range for
a human antibody (Xolair™, from Novartis) was investigated
in a concentration assay on Biacore T200 together with
Human Antibody Capture Kit from GE Healthcare.
The need for technologies that can be applied to reliably
detect and profile interactions involving very small
compounds is increasing in areas such as small
molecule- or fragment-based drug discovery. The sensitivity
of Biacore T200 allows the user to detect and precisely
characterize any organic compound, regardless of its
molecular weight. This means that an affinity analysis of
the simplest analytes can be confidently performed (Fig 4).
9
7
12
10
Response (RU)
Response (RU)
Xolair was precisely quantitated at the lowest
concentration (2 ng/ml) and over the entire concentration
range tested (Fig 3).
Response (RU)
Response (RU)
The possibility to derive high quality data from low levels
of immobilized interaction partners is advantageous in
the analysis of sensitive proteins such as G proteincoupled receptors (GPCRs). GPCRs are among the most
important classes of drug targets, and the high sensitivity
of Biacore T200 means that only a fraction of the
immobilized target need remain active following
immobilization (Fig 2). In addition, rare targets may
be used sparingly, allowing for reduced consumption
without compromising data quality. Sensitive targets
may thus be studied with greater confidence, reducing
time-to-results in the drug discovery process.
Anti-human antibody was immobilized on Sensor Chip CM5
using amine coupling chemistry. Xolair was then injected
for 3 min over the prepared surfaces at concentrations
between 2 ng/ml and 1 mg/ml. The surface was
regenerated according to the instructions provided with
the kit.
5
8
6
4
2
0
-2
-10
3
0
10
20
30
Time (s)
40
50
60
70
1
0
5 × 10-4
10-3
Concentration (µM)
1.5 × 10-3
2 × 10-3
Fig 4. Binding of methanesulfonamide (Mr 95) to carbonic anhydrase
immobilized at a level of 7000 RU. Affinity (KD) was calculated to 0.49 mM.
3
Flexibility in assay design: antibody
characterization without avidity
Full characterization of antigen-antibody interactions
is of great importance when assessing the suitability of
antibodies as therapeutic, analytical, or diagnostic tools.
When running Biacore assays involving antibodies, it is
usually recommended to immobilize the antibody on the
sensor surface in order to avoid avidity effects that might
complicate data interpretation.
In certain cases, however, it can be preferable to immobilize
the antigen, for example, in order to save precious target
or because the antigen may be easier to immobilize.
High sensitivity instrumentation allows the antigen to be
immobilized at a very low density on the sensor surface,
resulting in clean, avidity-free interaction studies (Fig 5).
Bivalent binding also causes the measured dissociation
rate to appear slower than in reality, creating difficulties
in ranking strong binders. For example, it was possible
to resolve seven slowly dissociating antibodies with very
similar dissociation profiles when the target antigen was
immobilized on a sensor surface at very low density.
Figure 7 shows that at higher response levels, little
or no difference was detectable among seven tested
antibodies. To clearly differentiate the antibodies in terms
of dissociation rates it was necessary to use low levels of
immobilized binding partner.
0.0007
0.0006
0.0005
Ab(Ag)2
Rmax = 1 RU
AbAg
kd (s-1)
0.0004
Rmax = 70 RU
0.0003
0.0002
0.0001
0
A
Fig 5. Superimposed, normalized sensorgrams describing interactions
between anti-IgE in solution and IgE immobilized on a sensor surface at
high density (Rmax 70 RU) or low density (Rmax 1 RU). All data are fitted to a 1:1
binding model.
IgE was immobilized on a sensor surface at various
densities. A concentration series of an anti-IgE antibody
was then injected over each prepared surface. The effects
of avidity due to bivalency progressively reduced as IgE was
immobilized on the sensor surface at lower densities and
disappeared entirely below Rmax of ~ 5 RU (Fig 6).
kd (s-1)
ka (M-1 s-1)
2.0 × 10-3
1.75 × 105
1.5 × 10-3
1.50 × 105
1.0 × 10-3
1.25 × 105
5.0 × 10-4
1.00 × 105
1
10
log Rmax (RU)
1
10
log Rmax (RU)
Fig 6. Changes in ka and kd for interactions between anti-IgE in solution and
decreasing surface densities of immobilized IgE (indicated by Rmax).
4
B
C
D
E
F
Antibody
Response level: 3 to 6 RU
Response level: 20 to 40 RU
Response level: 55 to 85 RU
Fig 7. Reliable ranking at low immobilization levels: resolving antibodies
with highly similar, slow dissociation profiles.
Flexibility in assay design: stable capture
of histidine tags to Sensor Chip NTA
Histidine is the most widely used molecular tag today.
The high sensitivity of Biacore T200 allows the capture of
very low levels of tagged molecules, and promotes highly
stable binding to Sensor Chip NTA.
The surface of Sensor Chip NTA is constructed of a
carboxymethylated dextran layer pre-immobilized with
nitrilotriacetic acid (NTA). Histidine-tagged molecules are
immobilized on the sensor surface via Ni2+/NTA chelation.
By choosing Sensor Chip NTA, the user controls steric
orientation of the immobilized interaction partner for
optimum site exposure. In addition, the surface is easily
regenerated by injecting EDTA to remove Ni2+.
G
Histidine-tagged tankyrase 1 (TNKS1), an enzyme involved in
the response of cells to DNA damage, was first injected over
the surface of Sensor Chip NTA for 30, 60, 90, 120, and 180 s
to find stable binding conditions. An injection time of 30 s
showed that a response level of 1000 RU or less resulted
in stable histidine tag capture of the protein. The optimal
experimental conditions were thus created by immobilizing
TNKS1 at an extremely low level that was nevertheless still
within the sensitivity threshold of Biacore T200 (Fig 8).
7000
180 s
6000
120 s
Response (RU)
5000
90 s
60 s
2000
30 s
1000
0
0
80
40
120
Time (s)
160
200
Fig 8. Test of capture level to obtain stable capture of histide-tagged TNKS1
to Sensor Chip NTA. A capture level of < 1000 RU following 30 s injection
was found to be stable and was chosen for further kinetic analyses.
The interaction of a molecule (Mr 310) with TNKS1 was
then profiled using single-cycle kinetics (Fig 9). Singlecycle kinetics is an alternative to the more conventional
multicycle approach, where the requirement to regenerate
the sensor surface between injection cycles is eliminated.
This can simplify problematic analyses and reduce assay
development time. Biacore T200 software includes the
necessary tools for both types of kinetic analyses.
Response (RU)
Reliable characterization of antibody:
fibrillar protein interactions
Human rheumatoid arthritis (RA) is a common
autoimmune disorder that can be artificially induced in
mice by immunization with type II collagen, leading to the
appearance of potentially pathogenic autoantibodies.
Certain specific epitopes on type II collagen have been
found to be strongly related to the pathogenicity of the
induced antibodies. Biacore T200 was used in an analysis
of serum IgG to test whether the autoimmune responses of
mice were directed to these epitopes.
Earlier studies have shown that SPR analysis of interactions
of antibodies with collagen presents some practical
difficulties (1). Firstly, fibrillar proteins such as collagen
have a tendency to aggregate and, if used in solution, the
concentration would be underestimated, leading to a lower
calculated rate of association. Secondly, immobilizing the
collagen on the surface leads to problems with avidity.
4000
3000
The applied methodology allowed gentle neutral buffers
to be used throughout the experiment, and low capture
levels allow precious target molecules to be used sparingly.
In addition, the protein is captured in a consistent and
predictable orientation.
In this study, the increased sensitivity of Biacore T200
allowed confident characterization of the collagen:antibody
interaction, avoiding aggregation and avidity effects.
Collagen was immobilized at a non-detectable level and
the very low Rmax was verified using a reference antibody.
Antibodies to several epitopes on type II collagen, isolated
from mouse sera, were subsequently characterized
(Table 1).
Table 1. Collagen antibody characterization
Type II collagen
epitope
Antibody
Affinity
designation
Pathologic
effect
J1
M2139
M287
M284
High
+++
+++
++
D3
CIIC2
Medium
Unknown
8
C1
CIIC1
CB20
Low
Low
++
+
6
U1
UL1
Low
++
4
F4
F4
Low
-1
2
1
Antibody F4 has been shown to protect against collagen-induced arthritis.
0
0
200
400
Time (s)
600
800
Fig 9. Immobilizing low levels of histidine-tagged TNKS1 on Sensor Chip NTA
enabled a rapid single cycle kinetics analysis of an interaction with a
molecule of Mr 310.
5
The appearance of the high affinity antibody, M2139,
directed against the J1 epitope on type II collagen is linked
to the most severe disease, whereas the lower affinity
antibody F4 has been shown to protect against collageninduced arthritis in mice (Fig 10).
The high sensitivity of the instrument and the very low
surface density used in these experiments enabled precise
characterization the collagen:antibody interactions. As the
target epitopes were not repetitive, binding was predicted
to be monovalent. The ability to use extremely low-density
surfaces also enables investigations into how avidity can
affect binding of antibodies to large proteins with repetitive
epitopes.
2
Response (RU)
1.5
1
Avidity-free characterization and selection
of strongly binding antibodies
Comprehensive characterization and selection of antibodies
for use in a diagnostic ELISA-based allergy assay were
performed at Phadia AB using Biacore T200. To obtain
a robust diagnostic assay, the choice of antibodies with
appropriate characteristics is pivotal.
As the study involved characterization of strong
antibody:antibody interactions, it was important to avoid
avidity effects. High instrument sensitivity was therefore
crucial for the generation of reliable results. Using a
sandwich assay setup, the antibodies intended for use in
the ELISA assay (anti-IgE antibodies and a capture antibody)
were used as the interacting partners in solution while
low levels of IgE were captured via human Fab binders
immobilized at low density on the sensor surface (Fig 11).
In this way, all antibodies were characterized in one
experiment.
If, in contrast, IgE had been injected over antibodies
immobilized on the sensor surface, the assay would have
consumed multiple chips and would have taken much
longer.
0.5
0
-0.5
-20
180
380
580
780
980
Time (s)
Antibody to
Fab-enzyme
conjugate
Response (RU)
3.5
2.5
1.5
IgE
0.5
-0.5
-10
0
5
10
20
30
Human Fab
capture kit
40
Time (s)
Response (RU)
5
3
Fig 11. Assay setup for the characterization of anti-IgE and capture
antibodies intended for use in a commercial ELISA.
1
-1
-5
5
15
25
35
45
Time (s)
Fig 10. Different interaction profiles of high, medium, and low affinity
antibodies. In many cases, the kinetic profiles are related to severity of
disease induced by the antibodies generated after injection of mice with
type II collagen.
In the Biacore T200 assay, human Fab binders (from Human
Fab Capture Kit from GE Healthcare) at a concentration of
10 μg/ml in sodium acetate at pH 5.0 were immobilized to a
level of 2000 RU on Sensor Chip CM5 (2 min injection) using
standard amine coupling. IgE at 1 μg/ml was injected for
60 s before each analyte injection. The analytes were either
potential capture antibodies or the parent antibodies (antiIgE) to potential Fab-enzyme conjugates. Of five antibodies
tested, two were judged to show desirable kinetic profiles
that suggested the potential to form a stable conjugate.
Using highly sensitive SPR analysis it was shown
that strongly binding antibodies can be confidently
characterized without complicating avidity effects.
Antibodies with desirable characteristics for specific
assays can be rapidly selected, bringing significant time
savings to the development process.
6
Selection and process/quality control of antibodies
From antibody to Fab-enzyme conjugate:
Antibody production: 3 weeks Selection and process/quality control of antibodies
Ab
Early selection and
characterization by Biacore
-at antibody level
Fab
Ordering information
Product
Code no.
Biacore T200 Processing Unit
28-9750-01
Series S Sensor Chip NTA
BR-1005-32
Human Antibody Capture Kit
BR-1008-39
Fab-enzyme conjugate
Late selection by sandwich
ELISA
-at conjugate level
Fig 12. Selection and process/quality control of antibodies: from antibody
to Fab-enzyme conjugate.
Researchers at Phadia AB estimate that in their search
to continually improve assay performance, the possibility
to carry out early quality control in the production of
Fab:enzyme conjugates will save weeks of development
time and costs (Fig 12).
Discussion and conclusions
Biacore T200 is a sensitive, label-free interaction analysis
system that offers not only excellent opportunities to work
confidently at the upper and lower limits of the kinetic
range, but also opens up possibilities to analyze interactions
complicated by phenomena such as avidity, which may be
problematic in antibody characterization. The opportunity
to immobilize interacting partners at very low density is
advantageous in applications such as the possibility to
confidently rank high affinity antibodies.
When analyzing interactions involving very small molecules
in drug discovery, there is no lower limit in terms of
molecular weight and previously borderline kinetic data can
now be confidently accepted, reducing the risk of missing
potentially interesting candidates as well as confirming
early failures.
Reference
1. Karlsson R. et al. Binding of autoreactive mouse anti-type II collagen antibodies
derived from the primary and the secondary immune response investigated with
the biosensor technique. J Immunol Methods 188, 63-71 (1995).
Acknowledgements
- Collagen antibodies: Professor Rikard Holmdahl and
Dr. Christoph Kessel, Department of Medical
Biochemistry & Biophysics, Medical Inflammation
Research, Karolinska Institute, Stockholm, Sweden.
- Sandwich assay: Dr. Pierre Leijon and Dr. Anders
Andersson, Phadia AB, Uppsala, Sweden.
- Histidine capture to Sensor Chip NTA: Dr. Johan Weigelt
and Dr. Natalia Markova, The Structural Genomics
Consortium, Karolinska Institute, Stockholm, Sweden.
7
GE, imagination at work, and GE monogram are trademarks
of General Electric Company.
For local office contact information, visit
www.gelifesciences.com/contact
www.gelifesciences.com/biacore
GE Healthcare Bio-Sciences AB
Björkgatan 30
751 84 Uppsala
Sweden
Biacore is a trademark of GE Healthcare companies.
All third party trademarks are the property of their respective owners.
Series S Sensor Chip NTA: The NTA ligand is manufactured by
QIAGEN GmbH and is under license from F. Hoffmann-La Roche Ltd
and QIAGEN GmbH.
©2010 General Electric Company—All rights reserved.
First published Jun. 2010
All goods and services are sold subject to the terms and conditions
of sale of the company within GE Healthcare which supplies them.
A copy of these terms and conditions is available on request. Contact
your local GE Healthcare representative for the most current
information.
GE Healthcare UK Limited
Amersham Place
Little Chalfont
Buckinghamshire, HP7 9NA
UK
GE Healthcare Europe, GmbH
Munzinger Strasse 5
D-79111 Freiburg
Germany
GE Healthcare Bio-Sciences Corp.
800 Centennial Avenue, P.O. Box 1327
Piscataway, NJ 08855-1327
USA
GE Healthcare Japan Corporation
Sanken Bldg., 3-25-1, Hyakunincho
Shinjuku-ku, Tokyo 169-0073
Japan
28-9794-20 AA 06/2010