Testing the specificity of your antibody is the first thing you should do. I assume you have produced the antigen in E.coli, purified it and then generated antibodies. The first control must be to load an extract from untransformed E.coli as control (or E.coli transformed with the expression vector without insert) directly next to the same amount of extract from your overexpressor, and when you see bands, they should only appear in the overexpressor samples. If this is successful, then you need to test them in the tissue of interest (the human tissue) and you can use an excess of the purified antigen as competitor in westerns. Those bands that correspond to the same epitops should disappear in the presence of competitor (the competitor is in solution and titrates all the antibodies that bind to your desired antigen). If these two tests are both positive, then you have to ask questions about the expected size. Could it contain hydrophobic transmembrane domains, could it be glycosylated, could it be a precursor that is cleaved to smaller parts? If all this doesn't explain the unexpected size on gel, well then you can try sequencing it, and that means purifying it first via a combination of approaches (ion-exchange chromatography, gel-filtration, hydrophobic interaction, affinity). That's a small project, but if you have it purified, even a small amount is enough to get someone to do Mass-Spec or microsequencing...it's a human protein, so even a short sequence can lead to gene identification from the sequenced genome. Long answer because I don't know what you are working on. I hope this helps.

One can do protein immunoprecipitation (using cell lysate and specific antibody) and futher one could run the eluted protein on SDS gel and comapre it with control. On the gel SDS gel a particular protein band which is unmatched with control can be excised and further protein sequencing can be done to identify the protein.

Testing the specificity of your antibody is the first thing you should do. I assume you have produced the antigen in E.coli, purified it and then generated antibodies. The first control must be to load an extract from untransformed E.coli as control (or E.coli transformed with the expression vector without insert) directly next to the same amount of extract from your overexpressor, and when you see bands, they should only appear in the overexpressor samples. If this is successful, then you need to test them in the tissue of interest (the human tissue) and you can use an excess of the purified antigen as competitor in westerns. Those bands that correspond to the same epitops should disappear in the presence of competitor (the competitor is in solution and titrates all the antibodies that bind to your desired antigen). If these two tests are both positive, then you have to ask questions about the expected size. Could it contain hydrophobic transmembrane domains, could it be glycosylated, could it be a precursor that is cleaved to smaller parts? If all this doesn't explain the unexpected size on gel, well then you can try sequencing it, and that means purifying it first via a combination of approaches (ion-exchange chromatography, gel-filtration, hydrophobic interaction, affinity). That's a small project, but if you have it purified, even a small amount is enough to get someone to do Mass-Spec or microsequencing...it's a human protein, so even a short sequence can lead to gene identification from the sequenced genome. Long answer because I don't know what you are working on. I hope this helps.

To sequence this protein after all of the above to recapture, use LC-MS for sequencing. Loads of techniques for this, just do a google search to find a up to date technique.

First, antibodies quite often recognize conformational epitopes. Western blotting denatures proteins, so an antibody is more likely to recognize linear peptide sequences that could be part of many proteins. Was your antibody made to denatured or purified intact native protein? This could be part of the answer.

If your antibody was generated using native protein, you could do something like the following. Couple your antibody to beads so that you can do pull-downs of the protein to purify it. The "old-school" method is to use Protein A and/or Protein G Sepharose, but you can couple antibody to other kinds of beads (eg. Luminex) as well. Check to see if you have one or more than one protein. If there is only one (major protein) you could recover and concentrate. This could be trypsin-treated and the peptides identified by nano-LC MS-MS or MALDI-TOF MS. If there are multiple proteins in the pull-downs they can be run on 1D or 2D gel electrophoresis, bands or spots excised, and identification could be done by MS of one sort or another.

We found that nano-LC-MS is best, Sorry for leaving out trypsinization of the protein. Thanks Clifford.

I think that the best approach is the proteomics one. Run an electrophoresis of your sample, excise the band in which your protein is probably located, and perform a proteomics mass spectra analysis. When you say sequencing, in proteomics is covering. Only edman technique allows for a N-terminal 20 aa sequencing.

sincere thanks to everybody for all the very useful information and suggestions. Just to provide a bit more information with regards my experiment as jugen requested. I am using a commercial antibody from abcam and the exact epitope/specific peptide that was used for antibody generation in rabbits is not known (commerical reasons) however they do provide a 50 amino acid sequence from which they say the epitope lies within.

I believe that some of the bands (not all of them) which i am seeing on my blots are glycosylated forms of the protein and so i was wondering would the glycosylation affect identifying the currently unknown bands if i was to do as has been suggested -

1 - do an IP,

2 - run the ip on a gel and cut out the bands

3 - then preform a mass spec analysis

one hypothesis is that my protein may be in alternative states of glycosylation which may affect its function and so i would like to keep the glycosylated state of the protein if possible during the mass spec analysis.

For the subsequent identification of the protein with the results obtained from mass spec analysis what online tools or software do people usually use.

Before you do sequencing first check for glycosylation by treating your extract with N-glycanase....if you get just one band then it might have the expected molecular weight....

hi,

As u told that not much information of proteins contains in your sample are available. therefor i would suggest that if u have SDS PAGE of those proteins, then u can cut those protein bands followed by proteolytic digetion. Then the digested sample u can analysis in MALDI-MS/MS and LC-ESI/MS/MS and try to generate protein sequences of as much as u can. Then at least u can achieve to put some information about few protein. but it may turn to laborious job for u and at the same time it has huge impact on work. all the best.

rajdeep

I think this paper may help you - Ohtsu et a.. Direct matrix-assisted

laser desorption/ionization time-of-flight mass spectrometric identification of

proteins on membrane detected by Western blotting and lectin blotting. J Proteome

Res. 4(4):1391-6, 2005. DOI: 10.1021/pr050073n.

Hi,

I would do as Prof. Deneke suggests: treatment with deglycosylation enzimes (Endoglycosydase H, PNGase F, A, etc) and check for size shifts in the western blot.

In addition I would also release the glycans from the protein prior to the Mass Spec analysis, since the glycosylated peptides are usually problematic for these techniques.

Hi everybody

I don't have a lot of experience with protein purification, but wouldn't pulling down the antigen using immobilized antibodies (as suggested by Clifford Calrk ) be superior to cutting the band out of a 1D PAGE?

I know that the apparantvsingle band in a 1D PAGE can potentially contain more than one peptide - with all the mess you have to deal with in downstream sequencing.

if possible, identify the band of your interest protein within an SDS-PAGE, cut it and analized by nano-LC-MS/MS. Since, it is a human protein a short peptide could be enough to identify the protein. However, you need to take some things into consideration:

1. is it easy to identify the protein in the SDS-PAGE? Otherwise you would need to do at least a preliminar purification

2. if possible stain your gel with Coomassie Blue, but if you need silver staining use a Glutaraldehyde-free method.

3. did you check the specificity of your antibody? If possible, include positive controls.

good luck with your experiments!!!

I'd cut it out of a gel, or PonceauS stain the protein on the membrane and use MS to sequence it, or just to get trypsinised fragment sizes. Talk to your local/favourite MS facility before starting and they can help you out in terms of how much they'll need and how to prep the sample (buffers, etc). Good Luck.

If you want to know the identity of the protein that the Ab recognizes, you could do an epitope mapping assay.

1) Tryptic digestion of the targeted band

2) Tryptic peptide capture by Ab fishing // wash // specific elution

3) MSMS identification of the peptide

So, this could be a way to know which of the proteins is identifying by the Ab.

If you are interesting in this assay I could give you the protocol

Hi ... if you do not want to perform a deep analysis and the only information that you need is the identification of your proteins. The easy way may be to prepare a 2DE, and cut the spots that present positive signal in the western analysis. Send this spots to MALDI TOF TOF MS identification ....