George W. Anderson

NEW APPROACHES TO PEPTIDE SYNTHESIS

During the past decade, peptide synthesis has been improved by the development of new protecting groups and new reagents for forming the peptide bond. However, no development comparable in importance to Bergmann’sla-b “carbobenzoxy method” has emerged. The work of Bergmann and his associates in the 1930s combiaed the use of a new amine-protecting group, the benzyloxycarbonyl or “carbobenzoxy” group, with the azide carboxyl activating group of T. Curtiusla-b to make a general procedure for peptide synthesis. Today we have many useful new protecting groups and many new methods of forming the peptide bond, but definite combinations have not emerged as outstanding. I believe that such combinations will develop: for example, RI. Bodanszky’s nitrophenyl-ester activation with the benzyloxycarbonyl-group protection looks promising. My group a t Lederle laboratories has as one of its objectives the development of improved general procedures, and this paper is concerned in large part with our efforts towards that goal. I t would be fine if the use of protecting groups could be avoided, but in practice this cannot be done. In general, the a-amino group of one amino acid or peptide and the a-carboxylic group of another must be made unreactive, so that a peptide bond can be formed between the remaining a-amino and a carboxylic groups. Also, many amino acids have other carboxy, amino, hydroxy, or thiol groups that usually require protecting groups. An approach that appeals to us a t present is to use groups that can be removed by treatment with acids for temporary protection of a-amino and a-carboxylic groups, and groups removable by catalytic hydrogenation or chemical reduction for protection of other reactive sites. In this way, the latter protecting groups can be carried through a synthesis and removed a t the end. This is not a novel idea-it has been used on occasion-but perhaps its reconsideration as a general approach has value. We particularly wish to avoid the use of alkaline conditions for reasons to be given. The increased use of the benzyl group (FIGURE I), which can be removed by reduction and is relatively resistant to acid and base, appeals to us for protection of the imidazole-NH of histidine, the -OH of serine, threonine and tyrosine, and the -SH of cysteine. All of these are known and are being used in peptide synthesis. I t should be emphasized that such protection of side groups is not always necessary, but it appears useful for a generalized approach. For temporary protection of a-amino groups in the course of a peptide synthesis we have found, as have others, that Bergmann’s benzyloxycarbonyl’ has stood the test of time as a superior group. The original method of removal by catalytic hydrogenation remains valuable, and the hydrogen bromide in acetic acid method that has been developed in recent years2*3.4 has increased the utility of the group. Thus it is possible to use benzyloxycarbonyl as one of the groups to be removed a t the end of a synthesis by reduction in our general plan, or to use it for temporary protection during a synthesis. The tert. butyloxycarbonyl group, which has been developed independently

Chapter 28. Synthetic Peptides

Publisher Summary The primary reasons for the synthesis of peptides are to confirm structures of biologically active natural peptides and to study the effect of modification of such structures on the biological effects. A practical goal is the synthesis of peptides that can be used for drug purposes. The synthesis of reactive esters of acylamino acids and their use in lengthening a peptide chain by reaction at the amino end is a justifiably popular method; racemization is avoided, yields are satisfactory, and purification of the products is relatively easy. Illustrative of the complexity in evaluation of active esters, it has been observed that the mixed anhydride procedure that gives N-hydroxysuccinimide (HOSu) ester without racemization yields a partially racemized 8-hydromjquinoline (HOQ) ester. There are other indications that HOSu has a fine balance of properties for active ester use. Research is continuing on the synthesis of hormones and analogs. A complete report on the syntheses of p-corticotropins is noted here; this is of general interest because of the discussion of problems in synthesis and purification of products. An analog of a 25 amino acid corticotrophin (ACTH) fragment with improved properties for drug use is reported; resistance to aminopeptidase activity by providing a D-serine residue at the amino end and to carboxypeptidase by a C-terminal valinamide, and resistance to air oxidation by substituting norleucine for methionine are obtained. Biological assays show increased potency over natural ACTH, and more prolonged action in human beings. This suggests that synthetic variations are valuable in improving the medical properties of other biologically active peptides.

Chap. 26 Synthetic Peptides

Publisher Summary A new approach has been made to the rapid synthesis of peptides in aqueous solution using a water soluble carbodiimide as the reagent and for lengthening the peptide chain from the carboxylic end. Another procedure using a water soluble carbodiimide builds the peptide chain from the amino end. This method in its current form suffers from variable yields and the number of manipulative procedures. It is evident that there is definite advantage of an automatic machine for making small amounts of peptides, but there may be no advantage if large quantities are desired. The development of a safe general procedure (other than the azide method that has disadvantages) of combining peptides without racemization would diminish the importance of the one-at-a-time process for moderate-sized or large peptides. Further improvements of rapid conventional methods that give good yields might also nullify some of the advantages of the machine approach. The methods of synthesis discussed in this chapter are useful in their applications to the synthesis of biologically important peptides. Procedures are now good enough for the rapid synthesis of peptides containing up to a dozen amino acids and, with more difficulty, complex peptides such as insulin. The development of solid phase synthesis and its automation is significant. The synthesis of dozens of analogs of important peptides, such as oxytocin and bradykinin, has already been accomplished and structure–activity relationships are becoming clearer.