I am working with self-assembled monolayers of gold nanoparticles on an aminosilane-functionalized glass surface. We are coating the gold with various carboxyl-terminated linkers, either polyethyleneglycol (PEGs) or alkanethiols. We've found, as expected, that the PEG molecules greatly reduce non-specific binding on the surface of our sensor; however, we're having trouble with the activation step through EDC/NHS.
We're following the activation protocol carefully: activating in MES buffer for optimal activation using a mM level solution, followed by pumping of PBS into the system (the surface never touches air), and then adding our protein, usually BSA (50-250uM). We observe no increase in binding from the activation. As I said, with PEG's, the non-specific binding is very low, so for the PEG coating, why is the covalent interaction not working? All measurements are done in real-time, so I'm very confident we're not getting any signal. Do you think our relatively charged surface could be introducing additional variable that compromise the activation process? Most publications seem to just activate on the bench, so it's not clear that the reaction is working well on surfaces, as least as far as I understand.
PS, we've tried this with pure carboxy-terminated coating, and carboxy/OH-terminated mixed coatings.
Thanks.
PS, when making NHS/EDC, I generally add MES to a microtube of pre-weighted EDC, and then pipette this directly into a microtube of NHS, then add this directly to the system. Is there a better way to make the EDC/NHS solution?
At the risk of overlooking other factors I will put my opinion here.
The NHS ester formed from the reaction of EDC/NHS with carboxylic groups is prone to hydrolysis. Such hydrolysis is favored at acidic and basic pH. In the past I simply activate the carboxyl groups by dissolving EDC/NHS in water. Also, the solutions are prepared right before use (which I think you are doing). After this the surface can be quickly rinsed with the buffer you will use and then introduce the solution of the biomolecule in buffer. Sometimes I have made a solution of EDC/NHS plus biomolecule and simply added it to the surface (as long as formation of NHS esters in your biomolecule is not a concern).
The other factor to consider is the isoelectric point of your protein. BSA will have a negative charge and your surface may have a slightly negative charge as well. Given that PEG will prevent other types of interactions with the proteins, the slight electrostatic repulsion could be preventing the interaction needed for the reaction to occur. You could confirm this by trying to immobilize a protein with high isoelectric point (e.g. avidin) or play with the pH (and you may need a different buffer).
This is a great answer thank you. Isn't MES actually recommended for this reaction though? I bought this buffer specifically for use with MES/PBS are recommended by pierce. Yes, we are indeed preparing solutions immediately.
What is the long term effct of forming NHS esters on a protein? Do you think over time, they will decay, or are they likely permanent? I will give this a try for sure.
Thanks for the suggestion, we will try avidin.
Over time the NHS ester will hydrolyze back to carboxylic groups. I'm just being cautious about unintended consequences of eliminating COOH groups, even if temporarily. I don't know how long will will it take for them to hydrolyze completely. You may want to look into the kinetics of the hydrolysis reaction (I know it has been published but can't remember the exact reference, look for David Grainger as an author).
MES is a good buffer to do this reaction. My recollection from the technology was that MES works well if the isolectric point of the protein is above the pH of your buffer when using carboxylic groups for immobilization. Good luck.
PS, how long do you usually wait between adding EDC/NHS to your solution, and rinsing out with the buffer? IE how long do you wait for the activation step?
For NHS-EDC catalyzed procedure to work the amine molecule should be in deprotonated form, it is possible that the amine groups of BSA are in protonated form, lowring the pH of protein introduction step might help. The book 'Bioconjugate Techniques' by Greg T. Hermanson has some useful tips for this procedure, it also deals with half.life of the activated carboxyls as a function of pH. Hope this helps
Cheers
Thanks, I will check out the book.
Yes, it did turn out that I was not considering the isoelectric point of the protein. By changing buffer, we were able to show a big increase in binding of BSA in a buffer that is more suitable for the electrostatic interaction between the sensor surface and the protein. Most of it did seem to rinse off, so we are going to try to up our EDC/NHS concentrations and hopefully will be able to see that most of the protein stays on the surface.
Dear Adam, to be frank, in past I have had spent immense time in trying to use EDC-NHS coupling without any success in cases of similar situations like your. However, you could try HBTU coupling and from my experience, this should work very well for your system and the process is also not very complicated.
Good luck.
Take a look at this link. It has specific protocol for the coupling reaction. I am not sure how much you can extrapolate to your system.
http://www.cytodiagnostics.com/pdf/Covalent-conjugation-carboxyl-gold-nanoparticles-TECH_NOTE_105.pdf
Hey guys,
Just to followup, we did finally fix the problem. For a negatively charged surface like our carboxy-coated nanoparticles, we needed a positively charged protein (hi PI) to accumulate enough proteins around the surface for the covalent-contact to happen. Frankly, I was shocked that even at mg/mL concentraitons, without the appropriate charge gradient, this reaction would not happen.
Please consider this note: The pH plays two important roles in the immobilization proce-dure. For the reaction with the activated ester to occur the ligand has to be uncharged and brought in close proximity of the surface by a process called preconcentration. This surface matrix-linked process is achieved by electrostatic attraction between the oppositely charged remaining surface carboxyls and the amino group of the ligand.The second role the pH is involved in is optimal reaction pH. The optimal reaction rate of the EDC/NHS chemistry proceeds around pH 8.5. Below pH 3.5 no reaction will take place. Due to this controversy between optimal reac-tion pH and required preconcentration, some molecules may be less effective to immobilize.
A couple of points that may be helpful. Phosphate is often a recommended buffer but it is very easy to show that phosphate interferes in EDC reactions. MES is a better option but take care with the pH. I've seen MES pH 4.7 recommended in many places but MES isn't an effective buffer at that pH, so you will not have good control of pH. For most reactions MES pH 6 is a good choice, allowing carboxyl activation (normally best at around pH 3.5) and reaction with amines (generally best with pH >8, as the unprotonated form of the amine reacts).
for EDC/NHS coupling pH has to be above 4 for reaction to occur. Above 4 pH carboxyl groups on the surface will be negatively charged. And your protein has to be at a pH that is below the protein's pI (isoelectric point), at a pH where the protein will be positively charged. The electrostatic interaction between negative carboxyl and positive protein will bring them close proximity and the coupling reaction will occur. For such use, recommended pH is 1 point below the pI of the protein. However, your protein is BSA. Its pI is 4.7 . Which is very close to the minimum possible pI for EDC/NHS to occur (pH=4). I would reccommend you to use another covalent coupling method instead of EDC/NHS or change the conjugation protein with a higher pI protein (if possible).
Best luck.