No abstract

Dedicated to Professor Louis A. Carpino on occasion of his 80th birthdayThe development of powerful methods for protein synthesis provides new ways to understand protein function, either by regioselective incorporation of modified amino acids as physical probes into natural proteins or de novo protein design to mimic some of the structural and functional properties of native proteins. Expressed protein ligation and protein trans splicing have emerged as efficient semisynthetic approaches for the preparation of engineered proteins with normal linear backbone structure.[1] However, the access to engineered proteins with an unnatural, multiply branched, and multicyclic backbone topology requires combinations of recombinant, enzymatic, and chemical synthesis (CRECS). Such proteins with nonlinear backbone topology which mimic extracellular binding domains (ECDs) of G protein-coupled receptors (GPCRs) may provide valuable insight into binding mechanisms of biologically important peptide ligands.GPCRs are targeted by most pharmaceuticals today, [2] but understanding how ligands bind to GPCRs at the molecular level is still hampered by a lack of high-resolution structural information. The extracellular domains of peptide receptors, namely the receptor N-terminus (ECD1) and the three loops (ECD2, 3, 4), often play a crucial role for ligand binding. Soluble nonglycosylated ECD1 of the corticotropin-releasing factor receptor type 1 (CRF 1 ), which belongs to the B1 subfamily of GPCRs, has been reported to be the major site of ligand interaction, as it shows high affinity to the natural peptide agonist urocortin 1 and the peptidic antagonist astressin.[3] Interestingly, we observed no binding of such a soluble ECD1 of CRF 1 to 125 I-labeled sauvagine, [4] another natural peptide agonist that binds like urocortin 1 with high affinity to wild-type CRF receptors.[5] This finding is indicative of a different contribution of the extracellular loops to high-affinity binding of the full-length receptor to the two ligands. To address this question, and to demonstrate the potential of a CRECS strategy with the first preparation of a protein mimic consisting of the four ectodomains of a GPCR, we synthesized such a mimic of CRF 1 , and determined the binding behavior of this receptor mimic for urocortin 1 and sauvagine.The template-assembled synthetic proteins (TASP) concept introduced by Mutter and Vuilleumier [6] proposes the use of topological templates as structure-supporting scaffolds, and was applied also to the assembly of a construct with three receptor loops. [7] As no experimental data are available on the topology of class B receptors, we initially placed no structural restrictions on the receptor model, but rather designed the template to allow a high degree of flexibility for the domains. Thus, it could be determined whether self-organization is possible in the construct by specific interactions between each other. Our primary efforts were aimed at developing a simple strategy for preparing single receptor domains (the cyc...

No abstract

The effect of tertiary amine (DIEA) on reaction rate and product purity of a carbodiimide/HOBt‐mediated peptide synthesis was studied. It was found that very rapid activation can be achieved using carbodiimide/HOBt in non‐polar solvents, such as DCM. Although the HOBt is poorly soluble in DCM, the activation proceeds within 2 min, probably forming the HOBt‐ester. By such a preactivation followed by a coupling in the presence of DIEA the rate of coupling is comparable with other rapid methods using BOP or TBTU, and no racemization was found in a model coupling (< 0.1%). For comparison, syntheses of neurotensin by means of different coupling reagents (BOP, TBTU, OPfp‐esters) and the DIEA‐catalyzed coupling after carbodiimide/HOBt‐activation under comparable conditions have shown that these procedures are of the same value in view of coupling efficiency and product purity.

No abstract