Synthesis of multi-module low density lipoprotein receptor class A domains with acid labile cyanopyridiniumylides (CyPY) as aspartic acid masking groups.

Low-density lipoprotein receptor class A domains (LA modules) are common ligand-binding domains of transmembrane receptors of approximately 40 amino acids that are involved in several cellular processes including endocytosis of extracellular targets. Due to their wide-ranging function and distribution among different transmembrane receptors, LA modules are of high interest for therapeutic applications. However, the efficient chemical synthesis of LA modules and derivatives is hindered by complications, many arising from the high abundance of aspartic acid and consequent aspartimide formation. Here, we report a robust, efficient and general applicable chemical synthesis route for accessing such LA modules, demonstrated by the synthesis and folding of the LA3 and LA4 modules of the low-density lipoprotein receptor, as well as a heterodimeric LA3-LA4 constructed by chemical ligation. The synthesis of the aspartic acid-rich LA domain peptides is made possible by the use of cyanopyridiniumylides (CyPY) - reported here for the first time - as a masking group for carboxylic acids. We show that cyanopyridiniumylide masked aspartic acid monomers are readily available building blocks for solid phase peptide synthesis and successfully suppress aspartimide formation. Unlike previously reported ylide-based carboxylic acid protecting groups, CyPY protected aspartic acids are converted to the free carboxylic acid by acidic hydrolysis and are compatible with all common residues and protecting groups. The chemical synthesis of Cys- and Asp-rich LA modules enables new access to a class of difficult to provide, but promising protein domains.

Prevention of aspartimide formation during peptide synthesis using cyanosulfurylides as carboxylic acid-protecting groups.

Although peptide chemistry has made great progress, the frequent occurrence of aspartimide formation during peptide synthesis remains a formidable challenge. Aspartimide formation leads to low yields in addition to costly purification or even inaccessible peptide sequences. Here, we report an alternative approach to address this longstanding challenge of peptide synthesis by utilizing cyanosulfurylides to mask carboxylic acids by a stable C-C bond. These functional groups-formally zwitterionic species-are exceptionally stable to all common manipulations and impart improved solubility during synthesis. Deprotection is readily and rapidly achieved under aqueous conditions with electrophilic halogenating agents via a highly selective C-C bond cleavage reaction. This protecting group is employed for the synthesis of a range of peptides and proteins including teduglutide, ubiquitin, and the low-density lipoprotein class A. This protecting group strategy has the potential to overcome one of the most difficult aspects of modern peptide chemistry.

Aspartic Acid Forming α-Ketoacid-Hydroxylamine (KAHA) Ligations with (S)-4,4-Difluoro-5-oxaproline.

The α-ketoacid-hydroxylamine (KAHA) ligation allows the coupling of unprotected peptide segments. Currently, the most applied hydroxylamine is the 5-membered cyclic hydroxylamine (S)-5-oxaproline, which forms a homoserine ester as the primary ligation product. In order to access native aspartic acid residues at the ligation site, we synthesized a 4,4-difluoro version of this monomer. Upon KAHA ligation, the resulting difluoro alcohol hydrolyzes to an aspartic acid residue with little or no formation of aspartamide. We applied this monomer for the synthesis of the hormone peptides glucagon and an insulin variant, and as well for segment ligation of the peptides UbcH5a and SUMO3.

A Threonine-Forming Oxazetidine Amino Acid for the Chemical Synthesis of Proteins through KAHA Ligation.

α-Ketoacid-hydroxylamine (KAHA) ligation allows the coupling of unprotected peptide segments through the chemoselective formation of an amide bond. Currently, the most widely used variant employs a 5-membered cyclic hydroxylamine that forms a homoserine ester as the primary ligation product. In order to directly form amide-linked threonine residues at the ligation site, we prepared a new 4-membered cyclic hydroxylamine building block. This monomer was applied to the synthesis of wild-type ubiquitin-conjugating enzyme UbcH5a (146 residues) and Titin protein domain TI I27 (89 residues). Both the resulting UbcH5a and the variant with two homoserine residues showed identical activity to a recombinant variant in a ubiquitination assay.

A short access to the skeleton of Elisabethin A and formal syntheses of Elisapterosin B and Colombiasin A.

A short stereoselective synthesis of the Elisabethin A skeleton 4 is described, which opens a formal access to the diterpenes Elisapterosin B and Colombiasin A as well. Key reactions were an intermolecular endo-selective Diels-Alder reaction to generate the decalin part of the molecule, a chemo- and diastereoselective allylation of an aldehyde with allylzinc, a palladium ene annulation of the cyclopentane ring, and a novel sulfonium ylide induced fragmentation of a polycyclic ketone. Additional insights have been gained for the crucial epimerization at C-2.