Hi Elena , what company antibody are you using? Can you cross check with the antibody datasheet if the supplier mentions anything about non-specific bands.

You can use a blocking peptide that bind specifically to the target antibody and block antibody binding. This is because the peptide resembles the epitope recognized by the antibody. Antibodies bound to the blocking peptide no longer bind to the epitope in the protein. Whatever bands that show up would then be due to non-specific binding. This mechanism is useful when non-specific binding is an issue. (Source: http://www.antibodies-online.com/resources/17/1235/Blocking+peptides/ )

Following can be helpful

Multiple bands

Cell lines that have been frequently passaged gradually accumulate differences in their protein expression profiles.

Go back to the original non-passaged cell line and run these samples in parallel.

The protein sample has multiple modified forms in vivo such as acetylation, methylation, myristylation, phosphorylation, glycosylation etc.

Examine the literature and use an agent to dephosphorylatem, etc. the protein so that it runs at the expected size.

The target in your protein sample has been digested (more likely if the bands are of lower molecular weight).

Make sure that you incorporate sufficient protease inhibitors in your sample buffer.

Unreported novel proteins or different splice variants that share similar epitopes and could possibly be from the same protein family are being detected.

Check the literature for other reports and also perform a BLAST search; use the cell line or tissue reported on the datasheet.

Primary antibody concentration is too high - at high concentration multiple bands are often seen.

Try decreasing the concentration of the primary antibody. Run a secondary antibody control (without the primary).

The antibody has not been purified.

Try to use affinity purified antibody. This will often remove non-specific bands.

The bands may be non-specific.

Where possible use blocking peptides to differentiate between specific and non-specific bands. Only specific bands should be blocked (and thus disappear).

The protein target may form multimers.

Try boiling in Laemmli buffer for 10 minutes rather than 5 minutes to disrupt multimers.

(Source: http://www.abcam.com/protocols/western-blot-troubleshooting-tips#multiple%20bands)

Good luck

Hi Elena , what company antibody are you using? Can you cross check with the antibody datasheet if the supplier mentions anything about non-specific bands.

You can use a blocking peptide that bind specifically to the target antibody and block antibody binding. This is because the peptide resembles the epitope recognized by the antibody. Antibodies bound to the blocking peptide no longer bind to the epitope in the protein. Whatever bands that show up would then be due to non-specific binding. This mechanism is useful when non-specific binding is an issue. (Source: http://www.antibodies-online.com/resources/17/1235/Blocking+peptides/ )

Following can be helpful

Multiple bands

Cell lines that have been frequently passaged gradually accumulate differences in their protein expression profiles.

Go back to the original non-passaged cell line and run these samples in parallel.

The protein sample has multiple modified forms in vivo such as acetylation, methylation, myristylation, phosphorylation, glycosylation etc.

Examine the literature and use an agent to dephosphorylatem, etc. the protein so that it runs at the expected size.

The target in your protein sample has been digested (more likely if the bands are of lower molecular weight).

Make sure that you incorporate sufficient protease inhibitors in your sample buffer.

Unreported novel proteins or different splice variants that share similar epitopes and could possibly be from the same protein family are being detected.

Check the literature for other reports and also perform a BLAST search; use the cell line or tissue reported on the datasheet.

Primary antibody concentration is too high - at high concentration multiple bands are often seen.

Try decreasing the concentration of the primary antibody. Run a secondary antibody control (without the primary).

The antibody has not been purified.

Try to use affinity purified antibody. This will often remove non-specific bands.

The bands may be non-specific.

Where possible use blocking peptides to differentiate between specific and non-specific bands. Only specific bands should be blocked (and thus disappear).

The protein target may form multimers.

Try boiling in Laemmli buffer for 10 minutes rather than 5 minutes to disrupt multimers.

(Source: http://www.abcam.com/protocols/western-blot-troubleshooting-tips#multiple%20bands)

Good luck

Hi Elena

Its strange that you see the bands in all the antibodies except tubulin.

As far as I know, BSA should not interfere with an extra bands if membrane blocking is done after transfer. Which is the blocking reagent you are using?? 

Another reason could be too much primary antibodies. Sometimes, in excess they bind non-specifically and have multiple bands. 

But don't we too worries until you are getting the band of interest at right size.

Hi there. I can't really add to what has been said in the first answer. Unexpected high molecular bands are not uncommon in Western blotting although I am surprised that you see the same band for different antibodies ( presumably raised to different synthetic epitopes) suggesting the product might be a specific variant of what you are looking at. For what it's worth I am pretty certain that 5%  milk versus BSA has nothing to do with what you are seeing. Thus: 

1. Examine the synthetic peptide used to raise antibody. If common to all of your antibodies regardless of source deliberately select an antibody raised to a different epitope ( synthetic peptide)

2. Try and access some synthetic peptide and use as a blocking peptide; that is preincubate antibody with peptide before hybridising to a western blot. Reduction in the signal from both bands will imply that the upper unexpected band is indeed a specific derivative of the protein of interest; that is a post translationally modified or splice form variant 

3. On the assumption that the extra band is a specific variant make lysate from the positive control line mentioned in ab data sheet and see if this variant goes away. That is a function of the cell line/ tissue under investigation

4.  make fresh lysate and repeat western blot

5. If not already doing so incubate with primary antibody over night at 4C at lowest recommended dilution instead of for 1-2 hours at room temp  in general irrespective of other problems this results in best signal to noise 

Thank you Vishwa, I am using Pierce antibodies, but they seem specific, and the strange is  that the first time it didn't ocurr with tubulin. The the supplier does not mentions anything about non-specific bands, but I will try to block the membrane more intensively and dilute a bit more the antibodies.

Thanks for your help

 Hi Madhur,

I'am blocking with PBS-Tween-5% non-fat milk. I think is good idea to dilute a bit more the antibody, but what worries me is that it seems like this 100 kDa band is "stealing" the signal of interest.

Thank you!!

Hi Laurence,

I think you are right, but I don't have more tissue to prepare new lysate, but I will try to look for information about the similarities between peptides used for antibodies generation. Yesterday, a friend of mine, has told me that she had the same problem using another kit of extraction from Macherey (mine is Qiagen), but se didn't solve it.

Thank you for your help.

Hi Elena, if your target protein is far from 100 kd, you can cut out the gel containing desired protein and transfer the gel on a membrane. By doing this the antibody will only bind to its target protein rather then having a cross-reaction with another antigen.

Make a small aliquot from stock and try boiling the lysate which will break protein aggregates in case the target protein is trapped due to protein aggregation. See if this helps.

I see that a lot is already written. I had similar problem in my lab and I solved it, but it was quite tough... 2 questions:

1. What kind of blotting system and gels do you use? (key question)

2. Have you ever try to put an antioxidant in your lysis buffer from the beginning on. Might be that somehow the proteins get oxidized and form 'clusters' with each other.

I hope it helps somehow

Hi there

Dimimanovs point is a salient one. You upper band might indeed be specific - blocking peptide would show that as explained - and could either be an unusual post translatinally modified form or as suggested (and considering the size differnce) a multimer or concatemer form associated with either the sample or the gel itself. Apart from what Dirimanov has suggested you could try use of DTT instead of B mercaptethanol when you boil the sample and make sure you boil for 5-10min; Some persons only boil to denature for 2 min followed by snap cooling and most of the time this is adequate. However in situations where multimers might be forming varying the reuction regime by boiling for longer or trying DTT instead of b ME can sometimes help

Laurence, thank you very much for the positive comment.

I think that the unexpected band is likely to represent the protein of interest (better control with a blocking peptide as Laurence mentioned, to be sure) - the lanes seem to be of similar size with those of expected one... I think the reason can be: 1. this band is a result of posttranslational modifications (lysate of cells expresing those proteins can also be tryied as a control); 2. there are any modifications after extraction (oxidation, building an aducts with non-polymerized acrylamide- http://www.ncbi.nlm.nih.gov/pubmed/15935323) 3. proteins are not separated properly on the molecular weight (trying different gels and I would say maybe also blotting system...)

Many thanks Dirimanov and Laurence!!

I use the Mini-PROTEAN Tetra cell for electroforesis and the transblot turbo transfer system for transfer. Both from Bio-Rad. I will try to use DDT instead of BME as you suggested, it's a good idea. Maybe I'm wrong, but I don't think it's a multimer o posttranslational modifications, since it's the same band for some antibodies. But I will change the percentage of gels to improve protein separation to see if the molecular weight is the same or not. I can use another blotting system, thanks!

Yesterday I used another antibody (anti-fibronectin from another supplier) and it works. Blocking with milk instead of BSA. It's curious. Today I will try with some of the antibodies I tried last week. 

Thank you all. I will told you the end of the story, I hope soon.

Hello, Elena! Thank you for your reply!

I used gels and blotting system from another company, so I cannot blame them. Anyway, if nothing else works, I think it worths that you try that. In my view, if you have the same problem with antibodies of one manufacturer, feel free to contact them. Normally, they help you to solve the problem - providing new AB, secondary AB, blocking peptides for testing the specificity or refunding...

Good luck with that labyrinthine story and keep us informed :) 

Thank you  again Dirimanov!!

Hello Elena,

It's possible BSA is an inproper blocking solution for this reaction and shows a cross-reactifity.  In this case some of the antibodies are binding unspecifically to a compound in the BSA, which results in uncomplete blocking and the unspecifically bound fraction is showing on your gels. One possible way to check this out is treating the correct working antibody originally blocked with 5% milk the same way as all the other antibodies in diluting it in BSA instead of milk. If it's showing the same band at 100kDa than all the others, BSA is causing an unspecific reaction and you shouldn't use it any more for this type of western-blots. The band showing at 100kDa might be really unspecific, just showing an antibody-antigen-reaction of albumin from different species. This is very possible, because the band is very strong and intensely.

The other possibility is, that just this one single antibody is really working properly and all the others show cross-reactions. This happens very frequently. You should contact the manufacturers in this case.

The density and concentration of your gel seems to be okay, the bands show sufficient separation, as far as I can see on your picture.

Blocking systems are essential and are not always applicable for all types of blotting and antibodies. You can find some information here:

https://www.lifetechnologies.com/de/de/home/life-science/protein-biology/protein-biology-learning-center/protein-biology-resource-library/pierce-protein-methods/blocking-buffers-western-blot-elisa.html

You'll surely mange to optimize your western-blots. Good luck!

Best regards,

Irmhild

Dear Irmhild,

Thank you for your advice. I have contacted the manufacturers and I'm waiting for their answer. Anyway I've repeated the same blot I showed you but incubating the antibody with milk and it seemed to work (Signal is OK but a bit weak, but there isn't the 100 kDa band). Now, I want to check the others and add some changes to improve the signal.

Many thanks!!