Journal of Biomolecular NMR, 11: 227–228, 1998. KLUWER/ESCOM © 1998 Kluwer Academic Publishers. Printed in Belgium. 227 Sequence-specific resonance assignments for a designed four-α-helix bundle protein Jack J. Skalicky a , Ramona J. Bieber a , Brian R. Gibney b , Francesc Rabanal b , P. Leslie Dutton b and A. Joshua Wand a,∗ a Departments of Chemistry, Biological Sciences, Biophysical Sciences and Center for Structural Biology, State University of New York at Buffalo, 815 Natural Sciences Complex, Buffalo, NY 14260-3000, U.S.A.; b The Johnson Research Foundation, Department of Biochemistry and Biophysics, University of Pennsylvania, Philadelphia, PA 19104-6089, U.S.A. Received 9 October 1997; Accepted 13 November 1997 Key words: maquette, NMR assignment, protein design Biological context We have undertaken to explore the basic structural fea- tures necessary to support specific biological function of electron transfer proteins by designing minimalist protein models of the photosynthetic reaction center, the bc1 complex, iron–sulfur proteins and others with small peptides that self-assemble to a four-α-helix- bundle architecture and bind the necessary prosthetic groups (Robertson et al., 1994). As designed, the pro- totype maquette (H10H24) folds to a four-α-helical structure, binds hemes with the expected stoichiome- try and affinity, and mimics the redox properties of the hemes in the cytochrome bc1 complex (Robertson et al., 1994). Successful as this prototype design was, the molecule still lacked a unique solution conformation as defined by NMR spectroscopy and probably exists as a mixture of different conformers that are intercon- verting slower than the microsecond timescale. This is a characteristic problem of most de novo protein designs (Betz et al., 1993) and hampers meaningful characterization of the protein functional properties. In order to improve upon the conformational speci- ficity, two of the heptad d-positions at the designed helix packing interface were targeted for substitu- tion. Single substitutions, H10H24-L6I and H10H24- L13F, resulted in molecules that exist primarily as single conformers in solution while the double vari- ant H10H24-L6I,L13F shows a single set of NMR resonances with narrow linewidths and chemical shift dispersion characteristic of a native protein. The lat- ∗ To whom correspondence should be addressed. ter protein will now serve as the parent molecule for a series of structural and functional investigations of itself and related proteins. This note presents the resonance assignment of the twofold symmetric mole- cule, provides a strong reference point for beginning to understand the origin of chemical shift perturba- tions brought by the attainment or loss of native-like structure, and describes our progress toward the solu- tion structure determination of H10H24-L6I,L13F by NMR spectroscopy. Methods and results H10H24-L6I,L13F [gsCGGGEIWKLHEEFLKKFEE LLKLHEERLKKL] 2 was cloned and expressed as a thioredoxin fusion protein. The peptide is released from the fusion protein with thrombin and the N- terminal cysteines are air oxidized to form a pair of linked helices. Protein solutions were prepared in a 20 mM sodium phosphate pH 6.60, 50 mM potassium chloride, 0.05 mM sodium azide, 92% H 2 O/8% D 2 O. All experiments were performed on a Varian Inova 500, Inova 600, or Inova 750 spectrometer. Main-chain C α , N, H N and side chain C β res- onances were assigned using HNCACB (Wittekind and Mueller, 1993) and CBCA(CO)NNH (Grzesiek and Bax, 1992) to establish segments of sequential connectivity. Main-chain H α and C ′ resonance assign- ments were then completed using CBCACO(CA)HA (Kay, 1993), HNCO (Kay et al., 1990) and 15 N-edited TOCSY (Zhang et al., 1994). Ambiguous connectiv- ity was resolved using the 15 N-edited TOCSY and