Understanding the factors responsible for the stability of common protein secondary structures has been a longstanding goal. Analysis of b sheets has lagged behind the study of a helices because the development of suitable model systems has been more challenging for the former than for the latter. Over the past decade, however, rules for the design of b sheets that fold autonomously in water have been elucidated, with a particular emphasis on two-stranded antiparallel b sheets (“b hairpins”), which represent a minimum increment of this secondary structure. These model systems have proved useful for evaluating the contributions of factors such as side-chain–side-chain interactions, interstrand linker composition, strand length, and strand number to the stability of b sheets. As an outgrowth of this work, b hairpins have emerged as excellent platforms for the evaluation of noncovalent interactions that do not necessarily occur naturally in b sheets. bHairpins and hairpin-like molecules stabilized by cyclization are also attractive systems for biomedical applications and fundamental studies. The use of designed peptides to probe the origins of bsheet stability, or to study noncovalent attractions between moieties that become spatially juxtaposed upon folding, requires the ability to determine the extent of b-sheet folding in solution. If only two conformational states are populated, unfolded and b sheet, then determining the population of these two states gives the folding equilibrium constant (Kfold), which provides insight into the stability of the folded state (DGfold= RT lnKfold). For proteins that adopt defined tertiary structures, conformational stability is often assessed by using heat or a chemical denaturant to disrupt the folded state while monitoring the extent of folding by a conformationally sensitive spectroscopic probe (for example, circular dichroism). This approach is convenient because globular proteins are typically completely folded near room temperature and in the absence of denaturant (“native conditions”), which establishes the spectroscopic signature of the folded state, and because it is often straightforward to identify the spectroscopic signature of a fully unfolded state generated at high temperature or high denaturant concentration. In contrast, most of the autonomously folding b-sheet model systems described to date cannot be driven by changing the conditions to the limiting states. Therefore, identifying the spectroscopic signatures for the fully unfolded and fully folded states of b-sheet model systems has frequently required the preparation and characterization of distinct reference peptides or the implementation of elaborate data analysis techniques. We have recently developed a new approach for studying the conformational stability of small proteins, and here we describe the extension of this approach to b hairpins. This method involves polypeptide analogues in which one backbone amide group has been replaced by a thioester (namely, thiodepsipeptides). The conformational stability is assessed by monitoring the equilibrium constant for a thiol–thioester exchange reaction that causes the full-length thiodepsipeptide to be reversibly cleaved, which precludes adoption of a native-like fold. The “backbone thioester exchange” (BTE) measurements can be conducted under native conditions, and it is not necessary for the full-length molecule to be completely folded under these conditions. Our previous BTE studies have focused on polypeptides that adopt a discrete tertiary structure. Here we show that the BTE method can be extended to a secondary structure model system (Figure 1). Our implementation of BTE began with a designed b hairpin designated HP5W4 (Figure 2) that was developed by Andersen and co-workers. 12] The sequence of HP5W4 incorporates key design features introduced originally by Cochran et al. in their development of the “tryptophan zipper” peptides. The turn segment of a b hairpin appears to be an optimal region for the amide-to-thioester modification that is required for implementation of BTE. Thioesters are poorer hydrogenbonding partners than are secondary amides, and our previous efforts have therefore involved replacement of the backbone amide groups that do not form intramolecular hydrogen bonds in the folded state so as to minimize the conformational destabilization that might result from the modification. The optimized b hairpin turn segment of Andersen and co-workers, Asn-Pro-Ala-Thr-Gly-Arg, seemed to offer a particularly attractive site for thioester insertion at the Thr Gly bond. Our first thiodepsipeptide design (NT-C’) contains a thioester linkage at this position as well as three conservative side-chain modifications compared to HP5W4: all three Lys residues were mutated to Arg. These mutations were made to avoid the possibility of acylation of [*] E. B. Hadley, A. M. Witek, Dr. F. Freire, A. J. Peoples, Prof. S. H. Gellman Department of Chemistry University of Wisconsin–Madison 1101 University Avenue, Madison, WI 53726 (USA) Fax: (+1)608-265-4534 E-mail: gellman@chem.wisc.edu