Returned Posting

Jeffrey Weidner jeffrey_weidner at Merck.Com
Fri Feb 17 09:50:45 EST 1995


Mail*Link(r) SMTP               Returned Posting

In article <D3zun1.Jw2 at freenet.carleton.ca>, af423 at FreeNet.Carleton.CA
(Marc Savard) wrote:

> In a previous posting, Dirk Mielenz (fn162 at fim.uni-erlangen.de) writes:
> > Hi all,
> > has anyone ever heard of a monoclonal antibody recognizing
> > two completely different and unrelated proteins?
> > I don't think that the posibility of one certain epitope 
> > consisting of 4-6 amino acids (perhaps additionally N-
> > or O-glycosylated) occuring in two different proteins
> > is that low. If one looks at possible permutations for
> > a row of 4 to 6 amino acids out of 20, the number
> > should be relatively low compared to the number of
> > existing proteins. I've been thinking about that problem
> > because it might be that we have such a mAb 
> > in our lab.I'd be very glad if anyone has literature hints. 
> > Thanks in advance,
> > Dirk
> 
The number of hexamers that can be formed from just the 20 native amino
acids is 20 to the 6th power or 64,000,000 permutations.  When you also
consider post translational modification and 3-D structure, the
probability becomes very small that you would see such cross-reactivity
among 2 unrelated proteins.  If they share some common features such as
binding of similar cofactors, you may improve the odds since they may have
a high degree of local homology at a particular region, though they have
very poor global homology.

Jeff Weidner
Merck Research Laboratories
jeffrey_weidner at merck.com






More information about the Immuno mailing list