peptide antibodies

Wed Mar 16 20:31:49 EST 1994

Peggy Kolm asked a question on parameters important to generate 
antipeptide antibodies.  Our experience is that it's best not to
use any less than 7 residues; we typically have used 12 to 20
residues/peptide.  How to choose a good region can be tricky.
Sometimes you must simply 'live' with a key region of interest.
However, if you have some latitude, try running an 'antigenicity'
calculation on the sequence; we have software to do this and
would be glad to help you out.  High scoring regions will often 
make good antipeptide antibodies (of course, only when conjugated
appropriately).  Generally, hydrophilic regions will score well,
especially those containing His, Lys, Arg, and Asp, but hydropho-
bics such as Ala, Leu, Tyr, and Val also seem to be correlated
with antigenicity.  I would recommend the papers below if you
wish to read more into this area:

     "Retrospective: 12 Years of Antigenic Determinant Predic-
     tions, and More" Tom Hopp (1993) Peptide Res. 6, 183-190.
     (Tom has written a number of other reviews on the same
     subject.  See, Methods Enzymol. 178, 571-585 (1990) and
     J. Immunol. Methods 88, 1-18 (1986)).

     "Prediction of Sequential Antigenic Regions of Proteins",
     Welling et al. (1985) FEBS Lett. 188, 215-218.

     "Mapping Antigenic Sites on Proteins: Implications for the
     Design of Synthetic Vaccines" Jemmerson and Paterson (1986)
     BioTechniques 4, 18-31.

Best wishes for success!  Cheers,  Shaun
 = Shaun D. Black, PhD   | Internet:  shaun%jason.decnet at = 
 = Dept. of Biochemistry | University of Texas Health Center, at Tyler = 

More information about the Methods mailing list