DOI: 10.1002/adsc.201200694 Enzymatic Fragment Condensation of Side Chain-Protected Peptides using Subtilisin A in Anhydrous Organic Solvents: A General Strategy for Industrial Peptide Synthesis Timo Nuijens, a Annette H. M. Schepers, a Claudia Cusan, a John A. W. Kruijtzer, b Dirk T. S. Rijkers, b Rob M. J. Liskamp, b and Peter J. L. M. Quaedflieg a, * a DSM Innovative Synthesis B.V., P.O. Box 18, NL-6160 MD Geleen, The Netherlands E-mail: peter.quaedflieg@dsm.com b MedicinalChemistry and Chemical Biology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, P.O. Box 80082, NL-3508 TB Utrecht, The Netherlands Received: August 3, 2012; Published online: January 4, 2013 Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201200694. Abstract: Herein, the enzymatic condensation of side chain-protected peptide fragments using subti- lisin A in anhydrous organic solvents is described. A screening with dipeptide Cbz-Val-Xxx carboxam- ACHTUNGTRENNUNGidomethyl esters with H-Xxx-Val-NH 2 nucleophiles was performed, wherein Xxx stands for every (side chain-protected) amino acid residue. Finally, it was demonstrated that it is feasible to enzymatically condense larger peptide fragments (up to the 10- mer level) bearing multiple side chain-protecting groups with very high conversion. Keywords: anhydrous conditions; fragment conden- sation; organic solvents; peptide synthesis; subtilisin A Although a few pharmaceutically relevant peptides can be produced relatively cost-efficiently on a large scale by fermentation, whereas most pharmaceutical peptides on the large scale for short peptide sequen- ces are synthesized by chemical means. [1] Solution phase stepwise chemical peptide synthesis is only fea- sible for medium-sized and long peptides (containing 10–50 amino acid residues), solid phase peptide syn- thesis (SPPS) is most commonly applied. [2] The main disadvantage of SPPS is that extremely high yields should be realized for each coupling and deprotection cycle. [3] For instance, when per step (coupling, Fmoc- deprotection and final cleavage) a yield of 95% is re- alized in the synthesis of a decamer, an overall yield of only 34% of the crude peptide is obtained, which should be purified by a laborious and cost-inefficient HPLC method. Therefore, on the industrial scale, peptides longer than 10–15 amino acids are not cost- efficiently synthesized in one run on the solid phase. Furthermore, during synthesis, long side chain-pro- tected peptides tend to form tertiary structures (a process which is called “hydrophobic collapse”) [4] making peptide elongation very troublesome so that a large excess of reagents and amino acid building blocks is needed. Additionally the purification of the final product is often very difficult due to the pres- ence of significant amounts of truncated peptides and peptides containing deletions. [5] Taking into account the drawbacks of both SPPS and solution phase peptide synthesis, it is generally impossible to synthesize, for instance, a 30-mer pep- tide solely by only one of these two strategies. To remedy these limitations, a hybrid approach can be used wherein protected peptide fragments are synthe- sized by means of SPPS and subsequently chemically coupled in solution. On paper, the “ideal” approach for the synthesis of a 30-mer peptide would be a fully symmetric (convergent) fragment condensation strat- egy wherein the length of the protected peptides syn- thesized by SPPS does not exceed 10 amino acids, e.g., a 10 + 10 + 10 strategy (Figure 1, A). However, if the desired peptide does not contain Gly or Pro resi- dues at the C-terminal coupling position of the frag- ments, racemization is inevitable. Although with care- fully selected coupling positions and reagents or con- densation techniques such as the azide method, [6] rac- emization can be minimized, it can never be fully eliminated. Especially in the case of pharmaceutical peptides, even a few tenths of percent of diastereoiso- mers in the final product is unacceptable. Therefore, a convergent approach is most often not feasible and one has to adapt the peptide fragment length to the positions of the Gly and Pro residues, if they are pres- ent at all. This leads to the SPPS synthesis of undesir- ably short (< 5 amino acids) or long (> 10 amino Adv. Synth. Catal. 2013, 355, 287 – 293 # 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 287 COMMUNICATIONS