I need to chemically biotinylate a protein with 10 lysine. How can I just conjugate one or two biotin on the molecule? Is there a way I can detect the level of biotinylation?
What is the size of the protein you're planning to biotinylate?
I guess it should be possible to 'minimally' label a protein, as there might be a preferential lysine you could target by tuning the concentration ratio (protein:label).
The level of biotinylation can be checked using mass spectrometric analysis of the intact protein; the site(s) of labeling could be found via a 'bottom-up' (peptide mapping) LC-MS/MS approach.
Some lysine residues will be more readily labeled than others, but there is no way to ensure in advance that labeling will occur on any particular residue and not on others. One exception to this is that if there is a lysine in the active site of an enzyme, it may be possible to block it by performing the labeling reaction in the presence of a saturating concentration of the substrate.
The average labeling stoichiometry can be adjusted by varying the concentrations of labeling reagent and protein.
In order to get a chemical label to attach at a specific site, you could perform site-directed mutagenesis on the protein to introduce a Cys residue at the desired site and remove any other free, surface-exposed Cys residues. Then, biotin-maleimide can be used to put the biotin on the single surface-exposed Cys residue.
Unless you use a site-directed approach, as discussed by Adam B Shapiro, it is impossible to label just one residue on every molecule. If you are unlikely to inactivate your protein, you should probably aim for three sites of modification (which is a population average of course), which ensures that the probability of molecules having no biotin is ~zero.
Another approach you can try is to deliberately under label the protein i.e., aim for a population average of around 0.3 biotin per protein. Obviously much of the protein cannot then bind to streptavidin, and is wasted, but of those molecules that can bind to streptavidin the majority will have one biotin label, with a small proportion having two. You will still have a heterogenous population of singly-labeled molecules, as the biotin is likely to be found in different positions, but probably with some bias towards certain lysines. The only way to find this out is mass spec as noted by Eef Dirksen.
An other possible alternarnative to perform selective single site biotinilation is modify the clone by add the Avi tag (GLNDIFEAQKIEWHE) at the C- or N-term of your protein and perform enzimatic biotinilation (exploiting the BirA activitty) which specifically biotinilate it.
i have done it several time inserting the AVi tag at the C_term between the GOI and the HIs tag
eg
original: protein-GS-HHHHHH
modified protein- GS- GLNDIFEAQKIEWHE-GS-HHHHHH
Since the Avitag sequence is quite short you can simply insert it on your clones by PIPE cloning perfvorming a vector PCR using primers with extended overlapping regions
Both Manuele Martinelli and Adam B Shapiro are correct. The typical NHS chemistry will result in a Poisson distribution of label incorporation that depends on the molar ratio of reactants.
The Avi-tag method works, c-terminal cys incorporation followed by maleimide chemistry works, along with more complex reactions (e.g. Sortagging).
There are kits available for biotin incorporation quant (HABA / avidin; check out Pierce biotin quant kit), but the result will be an average of the total biotin.
In general, for most assay development, I find that using fresh NHS-LC-biotin at a 3:1 molar ratio will provide a solid reagent with minimal effect on protein function.