ARBRE: Scientific output consolidation

At the Dresden meeting we discussed the ARBRE Working Groups' outputs. Working Group 2 concerns scientific output consolidation.

You can view the presentation slides from the Working Groups session here.

You can view the proposals received for Working Group 2 here.

We are now asking for comments in particular with respect to the issue of a standard protein system for biophysical techniques benchmarking. Do you have a favourite test sample you think could be ideal for widespread use? Some of the points raised are listed in the WG2 proposals under point 4.

Please comment, adding to this thread!

Rob Gilbert (convenor, WG2)


Posted on 27-Jul-2015 14:00 CEST
Robert Gilbert

I think point 2 and point 4 from Working group 2 proposal can be combined. We could think about benchmarking set of samples that can be tested using most available techniques to study molecular interaction such as ITC, SPR, MST, FP AUC. Did I missed anything?

Posted on 27-Jul-2015 16:26 CEST
Alexander Fish

I think you make a good point Alex; we need to have in mind samples which will be robust in multiple techniques.



Posted on 28-Jul-2015 12:39 CEST
Robert Gilbert

FCS might also fit in here. What about thermal shift assays to study interaction in CD, DSC, DSF, Prometheus, etc. ? the more different techniques, the better ;)

What about a streptavidin variant interactin with a PEG linked biotin to whatever reporter/catcher you need?

Using a mutant which has reasonable binding affinites (~1 µM) ( in addition to the versatility of the PEG linker on the biotin can make this a robust system for a variety of methods. Conjugating a dye to the biotin-PEG will allow all fluorescent techniques to follow the binding (MST, FP, FCS, FL-AUC?, ...), conjugating the bioting-PEG to the surface will allow all surface techniques to use the same system (SPR, SAW, Octet, ...).

Just an idea ;)



Posted on 10-Aug-2015 11:28 CEST
Arthur Sedivy

Dear all,

As discussed in Dresden and in this forum, we are in need of an easy 1:1 binding standard, that can be measured in various techniques (see Forum above, feel free to add more techniques!) The standard Streptavid-Biotin system is much too complex, since streptavidin forms tetrameres with four binding pockets, a collegue of mine (Herman Gruber, JKU Linz) spent a lot of time in characterizing this system.

I think Biotin-PEG-with a chemical reactive site is a very versatile tool, that we should look forward to use, since it can be labelled (fluorescence, EPR, NMR, goldparticle, you name it) and covalently bound to surfaces (gold, ...) which would enable this molecule to be traceable with almost all techniques I can think of.

The binder of choice should be a 1:1 binder with low unspecific interaction and an affinity with a Kd in the high nM to low µM range, maybe even two or three variants with different Kds as refereces!

One molecule that might fit the criteria is a monomer mutant of Streptavidin (, that might need to be mutated in the binding pocket to create lower affinity binding variants ((

Another possible binder might be a specifically designed DARPIN or Nanobody.

Jonas Schaefer from Zurich has a facility that creates DARPINs (, I will ask him if he already has a DARPIN against biotin or if he can generate one.

Since I am not only in a protein characterization but also protein production facility, we can try variant one (monomeric Streptavidin, different affinity mutants) and express them and we can also express the DARPINs, once we we know the sequence.

Do you think this is a reasonable way to go? What might go wrong? What other interactions can you think of as alternatives? Do want to help to:

-create a construct

-express and purify it

-characterize it

-characterize the interaction with a technique that you own


Posted on 13-Oct-2015 10:23 CEST
Arthur Sedivy

I would like to launch a general survey among ARBRE members to know which of you have pursued contacts with the Dynamic Biosensors company (

If so, have you actually tried their switchSENSE instrument, and what is your opinion about it (novelty, applications of interest, user-friendliness, …).

We will be trying it at the end of october ay Institut Pasteur, and I will let those of you who are interested know what is our opinion.


Posted on 13-Oct-2015 10:46 CEST

Hi Arthur

I am very interested in this topic and think that, as you say, we need to define what the most important aspects of an ideal 'standard' are. Firstly what we use standards for varies between facilities. They can be used as calibration standards, ie for calibrating delay volumes in viscometers or the constants in your MALLLS system, which requires a small 'round' monodisperse protein to be most effective. Also these standards should be able to cross most of the technologies so that for once we are able to compare results between very different instrumentation. For example, using MS and NMR as well as more standard interaction technologies.

I don’t think that we will be able to get a single pair of molecules that covers all the bases, but we can probably come close. One problem with the avidins is that the interaction is far too tight, even the reduced affinity avidins are still very tight. They are also multimeric, although I have some rhizavidin constructs which is dimeric. Monomeric avidin could be ok, if the binding is not too tight. And perhaps with another mono-biotinylated protein rather than PEG. PEG is a bit of a problem for several technologies such as MS and CD.

Some of the important factors that I can think of are:

Cheap to produce, small (but greater than 10 kDa otherwise AUC is a pain), stable (can you use them for DSC/Optim etc), the constructs need to be free from IP or licences, be able to be labelled without affecting interaction and not too tight or too weak, which would preclude some analysis techniques.

I have been in talks with a group here who has a stable Vhh domain called HEL4, obviously raised against hen egg lysozyme. It is a 13 kDa molecule which is very stable and very easy to produce in e.coli. It binds both lysozyme and proteinA but it has not been extensively tested for its binding as it was not originally made for this purpose, but we are in the process of seeing if it is suitable. The construct will be made available if it is suitable.


Best wishes


Posted on 13-Oct-2015 12:45 CEST
Thomas Jowitt
Displaying 1-7 of 71