Sign up to receive email notification of new issues

ForteBio Interactions Newsletter Biosensor photo

SPRING 2012    VOLUME 5    ISSUE 1


Technical Tip:

Measuring High-Affinity Interactions on the Octet System

Darick Dayne, Ph.D., Product Manager, Marketing,

As more drug candidates are affinity-matured through the development process, accurately measuring their affinity becomes a crucial part of characterization. In this article, we discuss some tips and tricks that can help Octet and BLItz platform users measure high-affinity molecular interactions more successfully.

There are always two binding partners involved in any kinetic measurement, and the essence of the measurement is to characterize how strongly the two molecules interact. Using a surface-based technology platform such as the Octet system, one of the two binding partners is immobilized onto the biosensor surface (the target) while the other binding partner is in solution (the analyte). When the interaction between the target and the analyte is expected to be very strong (e.g., very high affinity), precautions must be taken to ensure the kinetic measurement is set up and performed correctly.

The two kinetic terms that describe affinity of an interaction are the association constant (ka) and dissociation constant (kd). The ratio of these two terms (kd/a) give rise to the affinity constant KD, which is used for systematic comparison of affinity between different molecular interactions. In the overall affinity spectrum, KD values generally in the 1x10-3 molar range or above indicate very weak interactions. KD values in the 1x10-12 molar range indicate very strong interactions. The challenge of measuring a very high affinity interaction arises from the slow dissociation of target-bound analyte back into solution. For interactions with KD values in the picomolar range, the dissociation is so slow that the slope of the dissociation curve is extremely small. To measure and fit this slope accurately for an off-rate determination, some necessary precautions should be taken.

  • Ensure that the slope only comes from the dissociation of the target-bound analyte and nothing else. Signal drift of any kind during measurement, whether a result of the surface-bound target detaching from the biosensor surface or other buffer/detergent effects, should be strictly minimized or controlled well enough to be corrected during analysis (i.e. reference subtraction).
  • The method of target immobilization should be firm and non-reversible, ideally done either through amine coupling chemistry with the AR2G Biosensor or through biotin-avidin binding with the Streptavidin Biosensor.
  • The measurement should include a reference biosensor immobilized with the same target molecule but taking measurement in running buffer during analyte association/dissociation steps to account for any drift. This reference measurement can be used for subtraction during data analysis to eliminate any drift due to background effects.
  • When measuring a titration series of the analyte for a rigorous KD determination, start with a concentration that is ~10–20X above the expected KD value and titrate down by 2–3X. For example, if the expected KD value of the interaction is 10 pM, then the concentration series of the analyte would be 200 pM, 100 pM, 50 pM, 25 pM, and so forth.
  • The dissociation step should be substantially lengthened depending on the expected KD value of the interaction being measured. For interactions with KD values in the sub-nanomolar range, dissociation steps longer than one hour are generally required.
  • The stronger the interaction, the longer it will take to measure the dissociation step. Figure 1 shows a reference-corrected kinetic measurement of a very high-affinity interaction, where the dissociation is so tight that a dissociation time of 900 seconds is insufficient to observe any appreciable drop in signal. In such cases, a much longer dissociation such as one hour (3600 seconds) or more is needed.
  • Binding capacity of the assay and thus the resulting signal response of association should be maximized so that the drop in signal during dissociation is still greater than the instrument background noise. For users of the Octet RED96/384 instruments the average background noise is approximately 0.005 nm. Higher signal response during the association step will always increase the reliability of the statistically calculated dissociation constant kd and therefore the accuracy of the affinity constant KD.


Biomolecules with very high binding affinities are becoming more common in biopharmaceutical R&D, and their accurate kinetic determination is very important in drug discovery. These technical tips will enable researchers to accurately and easily measure high-affinity molecular interactions while also taking advantage of the streamlined workflow and ease-of-use provided by the Octet and BLItz platforms.