Cotranslational Incorporation of a Structurally Diverse Series of Proline Analogues in an Escherichia coli Expression System Wookhyun Kim, Anna George, Melissa Evans, and Vincent P. Conticello* [a] Introduction The imino acid proline occupies a unique niche in protein structural biology as a consequence of its distinctive structural and conformational properties vis-‡-vis the other canonical amino acids. [1] In contrast to the latter, proline has an endocy- clic amino group that narrowly limits its accessible conforma- tional space, as well as that of surrounding residues when placed in the context of a polypeptide chain, and prevents par- ticipation of the prolyl amide group in the hydrogen-bonding interactions that normally act to stabilize protein structure. [2, 3] Therefore, proline residues are not easily accommodated within periodic, hydrogen-bonded secondary structure and are often associated with aperiodic features such as turns and loops or with non-a/b periodic structures. Nonetheless, proline serves multiple roles of critical importance for protein structure and function; [4] these roles include delimitation of periodic sec- ondary structure elements, [5] alteration of the polypeptide- backbone trajectory through the formation of reverse turns or helical kinks, [6, 7] restriction of conformational entropy in the folded and unfolded states, [8] preferential destabilization of misfolded protein structures, [9] and conformational modulation through the thermodynamically accessible cis/trans isomeriza- tion of the Xaa±Pro peptide bond (Xaa = another amino acid). [10] In addition, proline residues are often observed within repetitive oligopeptide motifs [11] as central features in which the unique conformational characteristics of the imino acid largely determine the structural and functional properties of the repetitive domain. These proline-rich domains usually occur as either intrinsically unstructured elements [12] associated with protein±protein and protein±ligand recognition [13] or as the structural components of extracellular matrix proteins. [14, 15] Despite the structural importance of proline residues, few experimental methods are available to directly investigate local conformational effects that arise due to the presence of pro- line residues in polypeptide sequences on the structure and function of the corresponding native, folded proteins. Directed mutagenesis techniques have been applied routinely to re- place proline with other canonical amino acids, however these studies are limited in that the intrinsic structural differences between proline and the other canonical amino acids must be carefully considered during interpretation of the experimental results. As a point of contrast, scanning proline mutagenesis has been employed to identify functionally critical regions of protein secondary structure through the disruptive effect of the proline residues on hydrogen-bonding interactions and local peptide conformation. [16] Recently, chemosynthetic [17] and biosynthetic [18] methods have been described that permit sub- stitution of noncanonical amino acids into full-length, native protein sequences in place of canonical amino acids. In princi- ple, these methods enable the substitution of proline residues in polypeptide sequences with noncanonical imino acids that more closely mimic the structural and conformational proper- ties of the canonical amino acid. These investigations can pro- vide important information on the role of proline residues within specific structural contexts, particularly under conditions in which the differences in protein structure and function can be interpreted on the basis of the often subtle stereoelectronic distinctions that are observed between the canonical amino acid and its analogues. [19±24] Of the available methods, the bio- synthetic approach holds the most promise for large-scale syn- thesis of native proteins in which specific canonical amino acids, or sets of amino acids, have been replaced with structur- ally similar amino acid analogues. [18, 25] Several bacterial expres- sion systems have been described that enable global substitu- tion of proline residues in protein sequences with imino acid analogues; [26±30] however, these expression systems have nei- A set of Escherichia coli expression strains have been defined that are competent for the incorporation of a structurally diverse series of proline analogues under culture conditions that are compatible with high levels of analogue substitution within a proline-rich protein substrate. These bacterial strains have been employed to assay the efficacy of incorporation of noncanonical amino acids into a recombinant-protein test substrate and to create variant polypeptides in which native protein sequences have been globally substituted with imino acid analogues in re- sponse to proline codons. We envision that these methods may be used to interrogate the effect of imino acid substitution on protein structure and function and may be particularly informa- tive in the context of structural comparison of a series of modi- fied proteins with respect to the stereoelectronic differences between the incorporated proline analogues. [a] W.Kim,A.George,M.Evans,Prof.V.P.Conticello Department of Chemistry, Emory University 1515 Dickey Drive, Atlanta, GA 30322 (USA) Fax:(+ 1)404-727-6586 E-mail:vcontic@emory.edu 928 ¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim DOI: 10.1002/cbic.200400052 ChemBioChem 2004,5,928±936