Structure-Based Optimization of Peptide Inhibitors of Mammalian Ribonucleotide Reductase †,‡ Maria Pellegrini, §,|,⊥ Sebastian Liehr, ⊥,# Alison L. Fisher, #,3 Paul B. Laub, 1,O Barry S. Cooperman,* ,# and Dale F. Mierke* ,§,b Department of Molecular Pharmacology, DiVision of Biology and Medicine, and Department of Chemistry, Brown UniVersity, ProVidence, Rhode Island 02912, Department of Chemistry, UniVersity of PennsylVania, Philadelphia, PennsylVania 19104, and Fox Chase Cancer Center, Philadelphia, PennsylVania 19111 ReceiVed June 9, 2000 ABSTRACT: Mammalian ribonucleotide reductase (mRR), a potential target for cancer intervention, is composed of two subunits, mR1 and mR2, whose association is critical for enzyme activity. In this article we describe the structural features of the mRR-inhibitor Ac-F-c[ELAK]-DF (Peptide 3) while bound to the mR1 subunit as determined by transferred NOEs. Peptide 3 is a cyclic analogue of the N-acetylated form of the heptapeptide C-terminus of the mR2 subunit (Ac-FTLDADF), which is the link between the two subunits and previously shown to be the minimal sequence inhibitor mRR by competing with mR2 for binding to mR1. Structural refinement employing an ensemble-based, full-relaxation matrix approach resulted in two structures varying in the conformations of F 1 and the cyclic lactam side chains of E 2 and K 5 . The remainder of the molecule, both backbone and side chains, is extremely well-defined, with an RMSD of 0.54 Å. The structural features of this conformationally constrained analogue provide unique insight into the requirements for binding to mR1, critical for further inhibitor development. Mammalian ribonucleotide reductase (mRR) 1 catalyzes the radical deoxygenation of ribonucleotides to 2′-deoxyribo- nucleotides, which is the rate-determining step in de novo DNA synthesis (1). As such, it is a potential target for cancer intervention (2). The enzyme is composed of two different subunits, mR1 and mR2, with masses of 90 and 45 kDa, respectively. The larger mR1 subunit carries the substrate binding site as well as two allosteric sites while the smaller mR2 subunit contains two µ-oxygen-bridged high-spin Fe- (III)s and a stable tyrosine radical. For turnover to occur, mR2 must bind to mR1, allowing an electron to be transferred between the tyrosine radical and the substrate site (3-5). This binding takes place via the C-terminal residues of mR2, and can be inhibited by peptides mimicking the C-terminal sequence of mR2 (6). The linear heptapeptide Ac-FTLDADF (Peptide 1), corresponding to the seven C-terminal amino acids of mR2, was found to have the minimum length necessary for full inhibitory activity (7). Based on the structural features of this linear analogue (Peptide 1), and the closely related heptapeptide Ac- YTLDADF (Peptide 2), while bound to mR1, as determined by transferred nuclear Overhauser effects (NOEs) (8, 9), a series of cyclic analogues, employing a lactam bridge between the side chains of residues 2 and 5, were synthesized and tested for binding affinity to mR1 (10). Variations of the length of these side chains, thereby altering the ring size, as well as the direction of the lactam amide bond, affect the inhibitory activity and are therefore significant variables in the design of mRR inhibitors (10). To obtain structural insight into these findings, we have undertaken the characterization of the conformational preferences of the most active cyclic analogue, Ac-F-c[ELDK]-DF (Peptide 3), while bound to mR1. This peptide, containing an 18-membered lactam ring, had superior (∼2.5-fold) binding and inhibitory activity toward mR1 and mRR, respectively, than Peptide 1. The structural features of this analogue, structurally constrained by the cyclization, provide important properties for the rational design of optimized inhibitors of mRR activity. EXPERIMENTAL PROCEDURES Sample Preparation. The preparation of the mR1 subunit of mRR was carried out following published procedures (11). Peptides 3-7 were synthesized and cyclized on solid-support, † This work was supported in part by National Institutes of Health Grants GM-54082 (D.F.M.) and CA-58567 (B.S.C.), and by the Research Corporation through a Cottrell Scholars Award (D.F.M.). ‡ Coordinates have been deposited in the Protein Data Bank (accession code 1foz). * To whom correspondence should be addressed at the Department of Molecular Pharmacology, Division of Biology and Medicine, Brown University, Providence, RI 02912. Voice: (401)863-2139; Fax: (401)- 863-1595; e-mail: dale_mierke@brown.edu. § Department of Molecular Pharmacology, Brown University. | Present address: BASF Bioresearch Corp., Worcester, MA 01605. ⊥ M.P. and S.L. contributed equally to this work. # Department of Chemistry, University of Pennsylvania. 3 Present address: Department of Drug Metabolism, Merck Research Laboratories, West Point, PA 19486. 1 Fox Chase Cancer Center. O Present address: Incyte Pharmaceuticals, Inc., Palo Alto, CA 94304. b Department of Chemistry, Brown University. 1 Abbreviations: DTT, dithiothreitol; DG, distance geometry; DQF- COSY, double quantum filtered correlation spectroscopy; EDTA, ethylendiaminetetraacetate; IRMA, iterative relaxation matrix approach; Mamb, m-aminobenzoic acid; NMR, nuclear magnetic resonance; NOE, nuclear Overhauser enhancement; NOESY, nuclear Overhauser en- hancement spectroscopy; mRR, mammalian ribonucleotide reductase; RMSD, root-mean-square deviation; ROESY, rotational-Overhauser enhancement spectroscopy; TOCSY, total-correlation spectroscopy. 12210 Biochemistry 2000, 39, 12210-12215 10.1021/bi001323a CCC: $19.00 © 2000 American Chemical Society Published on Web 09/13/2000