no cross-reactivity at 81% homology using polyclonal antibodies? - (May/24/2006 )

Hi guys, Im a student almost finishing my thesis and looking for some clues in my thesis-discussion

I have a problem with my Westerns. I'm trying to detect recombinant human folate-binding protein (hFBP) using polyclonal anti-bovine FBP. The manufacturer claims cross-reactivity with human FBP because of 81% homology. The positive control (commercial bovine FBP) is readily detected but not the hFBP. Is this possible given 81% homology?
We tried a few monoclonal anti-bovine FBP's and they showed some detection but lots of aspecific bands.

So= if two proteins have 81 sequence homology, and you raise polyclonal aB to one of them, what is the chance that it does NOT cross-react with the other?

thx in advance
cheers

unfortunately, the 80% rule that companies state is just a bit of hit and miss, they rarely test the cross reactivity but assume at plus 80% should give signal

however thsi is not usually the case with polyclonal antibodies there is more of a 50:50% chance of getting cross reactivity with samples that are 80-90% related while 90% plus is actually where you should claim some degree of cross reactivity

have you checked the homology yourself? (www.expasy.org) or are you taking the maunfacturers word for it? i would check

its simply a case of that at 80% homology it is difficult to predict cross-reactivity without testing for it, due to the differences in amino acids can affect protein folding even though certain areas are conserved they may not be presented in the same way, obviously the reason you are not observing cross-reactivity it hard to identify without furhter molecular characterisation

i hope this helps

thanx man,

I did found 81% myself by alignment. But I just found a really interesting article

Immunoreactivity and Cross-Reactivity of Human and Bovine Retinal S-Antigen

that explains a lot. I don't have time to explain it myself now due to thesis deadline, but check the abstract.

and thanx again
cheers

interesting article, and no problem