Rapid Identification of Highly Active and Selective Substrates for Stromelysin and Matrilysin Using Bacteriophage Peptide Display Libraries (*)
- § To whom correspondence should be addressed: 3410 Central Expressway, Santa Clara, CA 95051. Tel.: 408-522-5718; Fax: 408-481-0522; Marc_Navre{at}qmgates.affymax.com
Abstract
The discovery of useful peptide substrates for proteases that recognize many amino acids in their active sites is often a
slow process due to the lack of initial substrate data and the expense of analyzing large numbers of peptide substrates. To
overcome these obstacles, we have made use of bacteriophage peptide display libraries. We prepared a random hexamer library
in the fd-derived vector fAFF-1 and included a “tether” sequence that could be recognized by monoclonal antibodies. We chose
the matrix metalloproteinases stromelysin and matrilysin as the targets for our studies, as they are known to require at least
6 amino acids in a peptide substrate for cleavage. The phage library was treated in solution with protease and cleaved phage
separated from uncleaved phage using a mixture of tether-binding monoclonal antibodies and Protein A-bearing cells followed
by precipitation. Clones were screened by the use of a rapid screening assay that identified phage encoding peptide sequences
susceptible to cleavage by the enzymes. The nucleotide sequence of the random hexamer region of 43 such clones was determined
for stromelysin and 23 for matrilysin. Synthetic peptides were prepared whose sequences were based on some of the positive
clones, as well as consensus sequences built from the positive clones. Many of the peptides have k
/K
values as good or better than those of previously reported substrates, and in fact, we were able to produce stromelysin and
matrilysin substrates that are both the most active and smallest reported to date. In addition, the phage data predicted selectivity in the P2 and P′1 positions of the two enzymes that were supported by the kinetic analysis of the peptides. This work demonstrates that the
phage selection techniques enable the rapid identification of highly active and selective protease substrates without making
any a priori assumptions about the specificity or the “physiological substrate” of the protease under study.
Footnotes
-
↵* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
-
↵1 The abbreviations used are:
- MMP
-
matrix metalloprotease
- HIV
-
human immunodeficiency virus
- mAb
-
monoclonal antibody
- ON
-
oligonucleotides
- KanR
-
kanamycin resistance
- tu
-
transducing units
- fTC
-
fAFF1-tetherC
- fTC-Good
-
fTC carrying a good substrate for stromelysin
- fTC-Bad
-
fTC carrying a bad substrate for stromelysin
- fTC-LIB
-
fTC derivative used for preparing the library
- DPA
-
3-(2,4-dinitrophenyl)-L-2,3-diaminopropionyl
- Ac
-
acetyl
- sfSTR
-
COOH terminally truncated form of stromelysin
- pro-sfSTR
-
the proenzyme form of sfSTR.
-
↵2R. Barrett, manuscript in preparation.
-
↵3R. J. Armstrong, L. Shi, and M. Navre, unpublished results.
-
↵4Niedzwiecki et al.(17) described the octapeptide Arg-Pro-Lys-Pro-Leu-Ala*Phe-TrpNH2 (* = site of cleavage), with a k
/K
of 18,000 M
s
, which was the best substrate described using only natural amino acids (interestingly, that substrate is very similar in
sequence to clone C4-33). The best stromelysin substrate currently reported is N-(2,4-dinitrophenyl)-Arg-Pro-Lys-Pro-Leu-Ala*Nva-Trp-NH2 (Nva = norvaline), also from Niedzwiecki et al.(17), which has a k
/K
of 45,000 M
s
.
-
↵5We have noted that the reactivity of the phage with the detecting antibody varies from clone to clone. Thus, no attempt was made to determine absolute phage concentrations.
-
↵6L. Ding, M. Smith, and M. Navre, unpublished results.
-
- Received December 1, 1994.
- Revision received January 9, 1995.
- © 1995 by The American Society for Biochemistry and Molecular Biology, Inc.
