At least some say so…

While this is particularly the case in everyday life – not going into details here – it is definitely true in label-free sensing.

A quick introduction to the people new to label-free sensing.

Unlike labelled methods, like fluorescence-based assays, ELISAs, or radioimmuno-assays, the size of the analyte determines the signal strength, which is going to be detected using a label free technology. As an example, a fluorescence labelled steroid will have (approximately) the same signal as a fluorescence labelled antibody, which is roughly 500 times larger (if both molecules carry the same number of fluorescent molecules). In label-free sensing however, you can – if you are unlucky – expect the steroid (“small molecule”) to cause 500 times less signal than an antibody.

So, why is small molecule sensing such a big thing for label-free technologies?

Reason 1 – quality

Label-free sensing should – by theory – deliver the best and undisturbed physical properties of the interaction of two molecules. Ergo, less artefacts and closer to reality. In addition, flow-through systems also provide access to kinetic rate-constants, which are much more valuable than “only” the affinity (See figure 1). And remember, these constants are not influenced by labels.

Fast association, but also fast dissociation.

Slow dissociation, but also slow association.

Figure 1: Comparison of the same affinity, but different rate constants. 

Imagine attaching a fluorescence dye to a small molecule of approximately the same size (see figure 2): the properties and the behaviour of the small molecule will drastically change, rendering investigations on this artificial structure practically worthless.

Green mesh: Antibody. Light green sticks: Typical fluorescence label (cyanine). Dark blue sticks: Small molecule (steroid hormone).

Figure 2: Comparison of the size of an antibody to the size of a typical fluorescence label and a small molecule.


Reason 2 – competition

There is an ongoing competition between the different label-free technologies and technology providers. And the one question, which has been asked for decades is: “how low (in terms of molecular weight) can you get?”. Smaller would be better in this case.

Questioning “big things” (or small things – if you like)

Reason 1 – quality (of your data) – is without question a valid point. Reason 2 – competition – is kind of… “meh”

I recall a discussion I had with someone from marketing, let’s call him “Dave”:

Dave: ”Dude, we need to be able to measure molecules below 50 Dalton!”

Me: “Why?”

Dave: “Because company xyz just announced that the new xyz series can measure down to 50 Da!”

Me: “Ok, but how many physiologically relevant substances exist being below 50 Da and can serve as potential new drugs to cure… everything?”

Dave: “Doesn’t matter, we need to write something like 50 Dalton OR LESS now on our spec sheet.”

At this time, my inner mathematician was already going berserk. As a rule a of thumb: any limit where lower is better should not be stated as “smaller than…”. -52 Dalton is obviously less than 50 Dalton, but such molecules would probably rip the space/time continuum apart.

Let’s put this anecdote into scientific/biophysical context:

The tricky part

The question arises: what are you really detecting when you monitor the interaction of a small molecule with a protein? Do you measure “only” the binding or rather conformational changes induced by this molecule upon its binding? To me, the latter is much more likely. And then this whole “how low (in terms of molecular weight) can you detect” thing would become rather futile… Think of a polymer (e.g. PDMS), which swells under the influence of a small molecule (e.g. toluene, 92 Dalton) [1]. Or let’s take oxygen or nitrogen. Everyone working with flow-injection or microfluidics will tell you that it is amazingly easy to detect it when it reaches your sensor – especially when it’s not supposed to happen and screws up your entire experiment.

Ok, the last one has been a bit silly, but what about going real small? Take some protons (e.g. from hydrochloric acid – HCl) and bring it in contact with any protein-coated transducer. Boom! Amazing signal caused by a molecule with meagre 1 Dalton. But it’s all “conformational change”.


  • detecting small molecule binding is challenging for label-free technologies
  • you cannot be sure if you detect binding or rather secondary effects like conformational changes
  • you can cheat

What are your experiences with small molecules and label-free technology?

Join the discussion on Facebook, Twitter or LinkedIn.



[1] Leidner, L. and Gauglitz, G. (2011) Development of a modified grating coupler in application to geosciences. Analytical and Bioanalytical Chemistry. 400: 2783-2791.

This article was written by
When Peter is not trying to save the world, he is managing the R&D activities at Biametrics. Ancient legends tell that he studied biochemistry before receiving his PhD in physical chemistry and becoming an expert in kinetics.