is this feasible? searching for an autoantibody - (Sep/26/2008 )
hello all,
ive posted about this before but i have a different idea that i want to run by everyone. basically, im trying to search for an autoantibody (to adipocytes) in patients with a particular form of lipodystrophy. i was initially going to use a cdna library to screen with patient serum but im looking for alternative methods.
what i was thinking is to run westerns using whole cell protein lysates from adipose tissue and liver tissue. my primary antibody (which is unknown) would be in patient serum and the secondary would be a anti-human IgG HRP. if i see extra bands present in the westerns of the adipose tissue lysates but not in liver tissue lysates, i can assume that these new bands represent an antigen not present in the liver samples.
what i am curious about is identifying the protein sequence of this antigen. can i cut these new bands out of my gel and send them for protein sequencing? is this possible/feasible? if so, i could compare the sequence to online protein databases and go from there i suppose. does anyone know a company which does protein sequencing from gels?
alternative, does anyone (perhaps immunologists) have any other ideas on how to approach this problem?
thanks!
ive posted about this before but i have a different idea that i want to run by everyone. basically, im trying to search for an autoantibody (to adipocytes) in patients with a particular form of lipodystrophy. i was initially going to use a cdna library to screen with patient serum but im looking for alternative methods.
what i was thinking is to run westerns using whole cell protein lysates from adipose tissue and liver tissue. my primary antibody (which is unknown) would be in patient serum and the secondary would be a anti-human IgG HRP. if i see extra bands present in the westerns of the adipose tissue lysates but not in liver tissue lysates, i can assume that these new bands represent an antigen not present in the liver samples.
what i am curious about is identifying the protein sequence of this antigen. can i cut these new bands out of my gel and send them for protein sequencing? is this possible/feasible? if so, i could compare the sequence to online protein databases and go from there i suppose. does anyone know a company which does protein sequencing from gels?
alternative, does anyone (perhaps immunologists) have any other ideas on how to approach this problem?
thanks!
I would think that screening a cDNA library would be more effective. For one, you don't know how sensitive your antibody is - it's possible that you could detect very small amounts of antigen and it would be hard to isolate that out of a gel for sequencing. You would also probably get a lot of bands migrating at the same position, which means that your protein sequence might not be all that clean - you would probably have to go to a 2D gel to get reliable data. The cDNA experiment seems much cleaner and easier - you might get a couple of different possible candidates, but then you could eliminate them if they don't correlate with western data - i.e. not the right size. Also, if you were to go straight for protein sequencing, then you might not see the antigen because it might be really high or really low molecular weight and could get buried amongst non-specific bands.
-I'm not an immunologist, but the cDNA library actually sounds like a pretty good idea.