There are many things that could have gone wrong.

One possibility is that the ECL has been activated by someone contaminating bottle B, with some of the solution from bottle A. That would lead to the activation of the reagent in the bottle and so it wouldn't matter when you mixed the two together prior to exposing. So that is where I would start with - change the ECL.

I don't know what molecular weight standards you use - but you can try using anti-penta his to see if some of the standards react to confirm that everything is working. 

Dear Carolin,

If you don't detect your regular proteins anymore, try to prepare almost all solution afresh. May be there were some contaminaton in your PBST (TBST) or milk or something else. We had this problem when used PBS stored for 2 weeks in the fridge. And check your ECL too!

good luck!

If you are using a new batch of antibodies, you may also contact the manufacturer of the antibody. Sometimes you get a bad batch (it happened to me) and manufacturer are normally willing to help you out with a different batch. 

Usually if you don't see bands which you saw before with same antibodies and conditions. The first thing, you've to check about ECL. We also had same problem recently, started changing everything including secondary antibodies yielded no clues. Once after we replaced ECL  we are started seeing bands like as before. Later we found that our ECL was too old(Check date of expiry).  our Western are fine now. 

All The Best.

Thank you very much for your answers!

The confusing thing is that I tried a different blot with the same solutions (including tbst and ecl) and antibodies, but on a different membrane afterwards  and it worked really well. I also detected the phosphorylated version of the protein on the same membrane, so I suppose that the membrane is fine and that the transfer worked.

So, you are confident that all your solutions were fresh and antibody worked properly...

Membrane contamination also can be a reason. Did you try to strip and re-block your membrane after previous detections?

I haven't tried stripping and reblocking yet. I didn't use to do it in previous blots either though, and never had any problems with the detection of other proteins on the same blot. Seems definitely worth a try in this case, though!

Hi Carolin, 

your current issue could be Ab's related , samples' related or blot related.

I would suggest you to run your samples with a positive as well a negative control when you can, to understand if the mainly issue is due to the antibody or your sample only. After the blot you could use a Red Ponce solution to double check your transfer (don't worry it's a revertible staining: you can wash it away). If you didn't change the protocol for the proteins' isolation it should work. Bare in mind that some peculiar proteins needs a specific kind of membrane : if you were working on PVDF and the pore size accidentally changed (wrong order) also that aspect could affect your western. Currently it's hard to tell you more if you didn't run controls with your samples. I'm sure you will sort it out: just change one thing for time and you'll nail it! 

I hope it helped somehow. 

Cheers,

Micol

Along with all the suggestions given, you must also take into account that the nature of blocking solution used during western blotting also plays a crucial role in detection of phosphorylation and unphosphorylated proteins. Generally, one should use skimmed milk while detecting unphosphorylated protein whereas BSA works fine with phosphorylated proteins. 

Here, you can try probing one antibody on the membrane, stripping the proteins and re-probing the same membrane again but using the different blocking solution. If all other reagents are working fine, the blocking solution might be interfering with the phosphorylation sites and thus preventing generation of any signal. 

http://www.abcam.com/protocols/western-blot-troubleshooting-tips#No%20signal

 Hi Caroline

Since you could observe bands previously, you can also try checking your protein samples' integrity by simply staining them with CBB or Ponceau S stain and proceed to probe with a housekeeping protein. This will eliminate the chances that either your protein samples, transfer technique or ECL is working or not.

Another thing that you can try is to use two different primary antibodies for your phosphorylated and un-phosphorylated proteins to be probed each having different host specificity. For example, say if the antibody for phosphorylated protein was raised in rabbit, you can try using the un-phosphorylated antibody raised in mouse. Or even if you are not able to get these kind of specificities, try to keep the IgG profiles discrete, say IgG-1 for one target and IgG-2a or IgG-A for another target. 

This can help because there will be minimized cross-linkage of non-specific antibodies and the signal will be solely coming from the target protein. 

Best of luck :)