Hello,

What about your primary antibody? Are you sure it's still detecting properly GAPDH? Do you have any positive control to check it? Indeed, antidodies are proteins and they can be denatured and lose their recognition features if not stored properly or undergone repeated freezing/thawing cycles for instance. You should also check through control that your detection system (enzyme + substrate) is still OK.

Good luck! 

I agree with above.  If you've successfully used this GAPDH antibody before, add some samples from that previous experiment onto your gel.  If you've never successfully used this GAPDH before, try a different loading control (actin, tubulin, etc).  If you have to have GAPDH and your trials show that this specific antibody is no good, either buy a fresh one from the same vendor or try a different vendor.

Another thing to check is to increase the concentration of your primary antibody to see if you are able to detect any protein bands. Though their might be possibility that you get ghost bands but then you can make changes in your concentration usage. Also, why not try with the fresh samples and perform the experiment. Is your protein stable at the temperature you are heating before loading it to gel? Best way to find out is to perform a series of samples with different heating temperature along with different heating time. Try one at a time. Hope you find it useful. All the best.

Thank you all for your help. I used a positive control and got band there. So, now I am isolating proteins from new sample and will do western blotting form these fresh samples. Hopefully, this time I will be successful. 

yes heating of the sample before loading also matters a lot. try to look into that too. 

There can be a possibility that your protein of interest is not transferred on nitrocellulose membrane at higher concentration , you can try to increase your transfer time (~ 1 h 30 min) and voltage (~100 V). Make sure that the transfer buffer has optimum pH and your buffers are pre-chilled before use, check for precipitation, if present then prepare fresh buffer.Immerse you transferring unit in ice or plug it and keep at 4 degrees in an refrigerator as transfer liberates heat. Hope this helps you along with the above suggestions Cheers!!!

Thank you Kamalakshi di and Nehil Chaturvedi. I will surely follow your suggestions also.

n hey it can be problem with ur developing solution too.. check with that by either running same set of samples in replicate in the same blot.. if developing solution is causing problem then it will get clear. at times if u r using chemiluminescence chemistry then it is possible to get different result based on development time and amount of developing solution used. 

What if the first run was an artifact and the second run was the good one?

Bernard, I think he answered that question with the set-up of his second run.  By loading a control sample from a blot that worked along with the ones that didn't appear to work, he can test two things at once: 1. that the antibody works and 2. that it was his prep of the newer samples that caused the band not to show up.  If the control had a band, and the other samples didn't, then he knows that the antibody works but that the newer samples are no good.

Thank you all for your suggestions.  The problem was with my lysis buffer. May be protease Inhibitor have stopped working. I am now getting good results with the same buffers and solutions but with new lysis buffer.  Once again thank you all. I got to learn many things from your answers.