Side Reactions During Coupling

Murphy s law certainly prevails in peptide synthesis. Some important side reactions, such as the formation of urethanes in coupling via alkylcarbonic acid mixed anhydrides or the generation of A-acylureas and dehydration of asparagine side chains when DCC is used for peptide bond formation, have already been mentioned in this chapter. Yet, countless additional side reactions and by-products have been observed and reported, often only as a footnote. Thus, it would be difficult to give a historical account of their discovery. A review [50] of side reactions noted in peptide synthesis reveals that most of them are caused by strong acids and bases, by excessive protonation or deprotonation of the amino add and peptide derivatives brought into reaction. 

Activation of the carboxyl group in itself is conducive to racemization. The otherwise chirally stable acylamino acids and peptides can lose chiral homogeneity once their free carboxyl group is converted into a reactive derivative. The electron-withdrawing effect of the activating group (X) renders the proton on the a-carbon atom slightly acidic and hence abstractable by base and chirality is, of course, lost in the carbanion. This simple mechanism, however, is operative only 

In the most commonly detected pathway of racemization azlactones (5(4if)-oxazolones, 5-oxazolinones) are implicated. Proton abstraction from these cyclic intermediates results in a resonance stabilized carbanion 

One of the most important practical results of these studies is the conclusion, drawn from theory and supported by experimental evidence, that racemization is greatly diminished if the amino component is acylated as such and not as a mixture of its salt with a tertiary amine. 

Individual amino acids show considerable differences in their propensity for racemization. This is exceptionally pronounced in derivatives of S-benzyl-cysteine, 0-benzyl-serine and S-cyanoalanine. The role of the substituent on the jS-carbon atom is not obvious. Early assumptions of j8-elimination, that is the reversible elimination-addition of benzyl mercaptane, benzyl alcohol or HCN, were not supported by the extensive studies of J. Kovacs. This example shows, however, that the azlactone mechanism, while it appears to be the most important pathway of racemization, is not the only process which leads to diminished chiral purity. For instance it is reasonable to consider that 

---

All That Can Go Wrong, Will Side Reactions During Coupling 95... 

The cycle is repeated until the required sequence is prepared. In general, a capping step (for example, treatment with acetic anhydride as in the phosphoramidite method) is not required. Truncated sequences are not normally the result of the presence of unreacted 5 -hydroxyl groups from the previous coupling reaction, but rather arising from the side reactions during coupling, such as sulfonation. 

The solid phase method possess certain inherent sources of inaccuracy resulting for the most part from the slight incompleteness of coupling that may occur at each step of amino acid addition to the growing chain and from certain side reactions during the final deprotection steps. However, careful purification of synthetic products has been carried out (84, 85), and synthetic variations in sequence have been restricted to those that yield essentially all-or-none answers about questions of enzymic function and structure. 

Fig. 26 Formation of isourea ethers as a side reaction during conversion of dextran using DCC as coupling reagent...

Formation of a C-N bond with cleavage of a C-C bond occurs occasionally as undesired side reaction during nitration,166,1183 nitrosation, or azo coupling. Some examples of this have been mentioned in the respective Sections. Such reactions have preparative interest only when they form the main reaction, as in the Japp-Klingemann reaction (see page 436) and in certain oximations by nitrous acid (see page 429). 

There is a risk of undesirable side reactions during activation, coupling and deprotection. 

The synthesis described met some difficulties. D-Valyl-L-prolyl resin was found to undergo intramolecular aminoiysis during the coupling step with DCC. 70< o of the dipeptide was cleaved from the polymer, and the diketopiperazine of D-valyl-L-proline was excreted into solution. The reaction was catalyzed by small amounts of acetic acid and inhibited by a higher concentration (protonation of amine). This side-reaction can be suppressed by adding the DCC prior to the carboxyl component. In this way, the carboxyl component is "consumed immediately to form the DCC adduct and cannot catalyze the cyclization. 

As a result of various side reactions, the yields are relatively low. However, in no case was ring fission found during the oxidations. Specially noteworthy is the ease with which the two methine groups in the 5-position of the 2-hydrazino-selenazoles are coupled together. Reference to models indicates that the quinonoid dyes exist in the trans form. 

Cyclic structures can form as a result of side reactions. One of the most common examples is the formation of diketopiperazines during the coupling of the third amino acid onto the peptide chain (Fig. 7). Intramolecular amide bond formation gives rise to a cyclic dipeptide of a six-membered ring structure, causing losses to the sequence and regeneration of the hydroxyl sites on the resin. The nucleophilic group on the resin can lead to fiuther unwanted reactions [14]. 

The danger of epimerization during the coupling of segments exists for all cases, except when me activated residue is Pro and Gly, with a few exceptions (see Section 7.23). The obvious is to design a strategy mat involves activation at these residues only. Options to try to minimize me side reaction for activation at other residues are as follows ... 

FMF Chen, NL Benoiton. Identification of the side-reaction of Boc-decomposition during the coupling of Boc-amino acids with amino acid ester salts, in JA Smith, JR Rivier, eds. Chemistry and Biology. Proceedings of the 12th American Peptide Symposium, Escom, Leiden, 1992, pp. 542-543. 

---