ABSTRACT: Since the discovery of a natural catalytic antibody in 1989, many catalytic antibodies targeting peptides, nucleotides, virus and bacterial proteins, and many molecules have been prepared. Although catalytic antibodies should have features superior to non-catalytic monoclonal antibodies, the reports on catalytic antibodies are far fewer than those on normal (non-catalytic) antibodies. Nowadays, we can obtain any monoclonal antibody we want, which is not the case for catalytic antibodies. To overcome this drawback of catalytic antibodies, much basic research must be done. As one way to attain this purpose, we have been making a protein bank of human antibody light chains, in which the light chains were expressed, purified, and stored for use in screening against many kinds of antigen, to then get clues to introducing a catalytic function in normal antibodies. As the number of stored light chains in the above protein bank has reached the hundreds, in this study, we screened them against amyloid-beta (A?), which is well-known as one of the molecules causing Alzheimer's disease. We found two interesting light chains, #7TR and #7GY. The former could degrade both a fluorescence resonance energy transfer-A? substrate and A?1-40 full peptide. In contrast, #7GY, whose sequence is identical to that of #7TR except for the amino acids at the 29th and 30th positions, did not degrade the FRET-A? substrate at all. By using a synthetic substrate, Arg-pNA, the difference between the chemical features of the two light chains was investigated in detail. In addition, we found that the presence of Zn(II) ion hugely influenced the catalytic activity of the #7TR light chain but not #7GY. Through these facts and the discussion, we propose one of the clues to how to put a catalytic function in a normal (non-catalytic) antibody.