Help in western blotting GAPDH bands are not even - (May/23/2015 )

Hello, 

Recently I am having too much troubles in getting even bands in my western blot. I did the BSA protein assay with the modified Lawry commercial kit from Fisher Scientific, the standard curve was fine but the reading for the protein concentration was lower than normal. This is mice brain tissues homogenized in tris buffer, usually I have the protein concentration between 5-6ug per ul this time I had only between 1-2ug per ul. I did the dilution estimating that the reading in the protein essay was around 3 times less than the normal reading. The homogenized tissue and the BSA solutions were in the same 1x Laemelli buffer. When I did Western blot today I found out that the there was a big variation in the GAPDH bands, they were uneven. I am excluding a pipetting problem but could it be poor preparation of the samples or there is some mysterious mistake that I am doing is letting me get these results. This problem was repeated two times in totally different samples. Could also protein degradation be the issue and why does it happen. Attached is a picture of my GAPDH bands. Any help, I would really appreciate it. 

Thanks

Narjes

gapdh is used to normalize between samples. it is not unusual to see differences and then correct.