The folded structure of a protein composed of D-amino acids is the mirror image of the all-L-amino acid protein molecule. Hence, for developing an all-D-protein ligand for a natural protein, the L-protein ligand has to be mirrored a second time (fig. 1).

The all-D-amino acid enantiomer of the target protein is prepared by total chemical synthesis and used as the target for phage display of a library of L-protein ligands. An L-protein, which binds to the all-D target with high affinity, is selected. Then, the corresponding all-D-protein is synthesized, which will bind to the native all-L-amino acid target protein.

Professor Stephen Kent from the University of Chicago, a pioneer in the chemical synthesis of proteins, evaluated this concept for its potential in drug development in collaboration with Reflexion Pharmaceuticals, a company he co-founded. Peptides and proteins consisting of D-amino acids resist degradation by proteases. Used as therapeutics, they combine an increased half-life with non-immunogenicity.

As an example for engineering a D-protein ligand to recognize an L-protein target vascular endothelial growth factor Type A (VEGF-A), an angiogenesis-inducing protein overexpressed in tumors, was chosen as the natural protein target. All-D VEGF-A was produced by total chemical synthesis using native chemical ligation.

Then, the laboratory of Professor Sachdev Sidhu at the University of Toronto used the all-D VEGF-A as the target for phage display of a library of small protein ligands, and the L-protein binding to all-D VEGF-A most efficiently was selected. The corresponding D-protein of the same amino acid sequence was prepared by total chemical synthesis and shown to bind to native VEGF-A and act as an antagonist preventing receptor binding.

Professor John Robinson’s group from the University of Zürich works on synthetic protein epitope mimetics. The group has developed a type of macrocyclic peptides acting as β‐hairpin mimetics. An L-Pro-D-Pro scaffold stabilizes the conformation of the loops. A β‐hairpin-mimetic tetradecapeptide derived from the antimicrobial peptide protegrin I showed a vast increase in activity. It acted specifically against Pseudomonas species. Pseudomonas bacteria such as P. aeruginosa pose a grave problem, as they are often detected in hospitals and have acquired multi-antibiotic resistance. The peptide antibiotic POL7080 developed by Polyphor and recently out-licensed to Roche is the result of further optimization of Robinson’s protegrin mimetic.

Well-Defined Glycoproteins and Glycopeptides

Chemical glycosylation is an innovative approach for the modification of peptides and proteins to improve drug properties such as pharmacokinetics, efficacy and solubility. The glycosylation of proteins is a most important post-translational process. Precise position and type of glycosyl residues are essential for the activity of the resulting glycoprotein. Glycoproteins are usually obtained by recombinant technologies employing mammalian cells, and the chemical synthesis even of simple glycopeptides is elaborate and cost-intensive.

GlyTech, a company in Kyoto, Japan, is a pioneer in the manufacture of glycoproteins and glycopeptides by chemical means. Professor Yasuhiro Kajihara from Osaka University and Dr Michael Haller from GlyTech described their approach to site-specific glycosylation. The synthetic approach developed by GlyTech allows choosing from a broad array of stable preformed glycan building blocks. It is perfectly suited for optimizing peptide leads and improving established peptide pharmaceuticals and for large-scale production. Their asparagine derivatives carrying linear or branched glycans are compatible with standard protocols of solid-phase peptide synthesis. Bromoacetylated glycans allow glycosylation of cysteine-containing proteins in solution.

Glycosylated proteins and peptides prepared by the method are well-defined compounds, more homogeneous compared with those obtained recombinantly in mammalian cells. Unambiguous structure is highly desirable for glycoproteins used in drugs and equal or improved activity could be demonstrated for glycoproteins obtained by GlyTech’s approach. Multiple site-specific glycosylation of proteins can be a more biocompatible alternative to PEGylation. Not unexpectedly, glycosylation improves the water solubility of peptides and proteins. With peptides, dramatic effects can be induced: glycosylation of the somatostatin analog octreotide not only increased the half-life of the drug, but it also markedly changed the receptor-binding affinity, which came close to the behavior of native somatostatin.

Peptide-Based Anti-Alzheimer Vaccine

Misfolding and aberrant processing of proteins play an important role in neurodegenerative diseases such as Alzheimer’s (amyloid β and Tau) or Parkinson’s (α-synuclein). Such proteins are potential therapeutic targets. AC Immune is a company specializing in the development of therapeutics for neurodegenerative diseases including Alzheimer’s disease (AD).

Professor Andrea Pfeifer, founder and CEO of AC Immune, considers the microtubule-stabilizing Tau proteins a promising target for a therapeutic vaccine for this devastating disease. Hyperphosphorylated Tau proteins form twisted fibers inside neuronal cells and build tangles, which are associated with the pathological conditions of AD and other neurodegenerative diseases. Pfeifer presented the company’s platform SupraAntigen it used to obtain a peptide-based anti-pTau vaccine. During pre-clinical development, this vaccine showed reduction of phospho-Tau aggregates and total pathological Tau and improvement of clinical parameters. The vaccine is also characterized by very specific and T-cell independent immune response, which is an important feature of the SupraAntigen technology platform. The vaccine is being tested in a Phase 1b clinical trial.

The heavily modified short peptide T3, an essential component of the anti-pTau vaccine, was produced at Bachem. The development and optimization of the synthesis and purification of the tetrapalmitoylated phosphopeptide was presented by Ralph Schönleber, vice president research & development at Bachem. Various strategies for preparing a peptide containing two phosphoserine residues and five lysines, four of them modified by palmitoylation, could be conceived. Solid-phase peptide synthesis was the method of choice, but modification steps can be performed on-resin or in solution, after cleavage from the carrier.

Purification of tetrapalmitoylated T3 was the major challenge due to the low solubility of the peptide, which could be facilitated by improving the quality of the crude product. Eventually, the palmitoylation of the purified partially protected peptide was conducted in solution followed by deprotection of the fifth lysine and the N-terminus. As the non-palmitoylated peptide can be purified very effectively, the final purity of T3 could be increased to more than 95%.