Dehydro amino acid residues are a common feature of many microbial peptides. In particular, dehydroalanine occurs in a number of peptide antibiotics, inter alia, in nisin, alternariolide, subtilin, siamycin, thiostrepton, nosiheptide, and thiopeptin. Dehydro amino acids have also frequently been implicated as biochemical reaction intermediates. It has been suggested, for example, that dehydrocysteine- and dehydrovaline-containing peptides, produced from dehydrogenation of the relevant precursors, could be intermediates of penicillin and cephalosporin biosynthesis. In addition, enzyme-bound dehydroalanine has been proposed as an intermediate in the desulfuration of cysteine catalyzed by S-alkyl-L-cysteine lyase, in the dehydration of serine by serine dehydrolase, and in the metabolism of O-acetylserine. The biochemical origins of the peptide antibiotic dehydro amino acid residues have received little attention. There is some experimental evidence suggesting that the lanthionine and O-methyllanthionine residues found in nisin are derived, in part, from serine and threonine, respectively, and it was proposed that dehydro amino acids play a role as reaction intermediates. In addition, it was suggested that the dehydroalanine and dehydrobutyrine residues present in nisin may also be produced by dehydration of the relevant precursors. Furthermore, Bycroft has postulated that the biosynthesis of dehydroalanine residues found in peptide antibiotics results from either the dehydration of serine or the dehydrogenation of alanine, but until now no experimental evidence has provided a sensible choice between the two possibilities. The present communication describes results which demonstrate that dehydroalanine residues arise by dehydration of serine, at least in berninamycin A (1), a polypeptide antibiotic produced by Streptomyces bernensis which has been shown to inhibit protein synthesis at the ribosome level. The structure assigned in this laboratory contains five dehydroalanine residues; in addition, other dehydro amino acids are involved in the oxazole A and B units and the berninamycinic acid residue contains a dehydrocysteine unit and yet another potential dehydroalanine unit. We introduce 1,7-octadien-3-one (4) as an easily available and very convenient bisannulation reagent. Annulation reaction to form fused six-membered cyclic ketones using alkyl vinyl ketones or their equivalents is well known as the Robinson annulation and has wide application, particularly for stepwise construction of polycyclic compounds such as steroids and certain terpenoids. One important offshoot of the Robinson procedure is the construction of two fused six-membered cyclic ketones from one reagent, instead of repeating the common Robinson cyclization twice. This new method of cyclization is called bisannulation, and few compounds have been recommended for this purpose. According to Danishefsky, the bisannulation reagent is a synthetic equivalent to 7-octene-2,6-dione. In other words, the bisannulation reagent must have a terminal enone or its equivalent in one side and a masked ketone to generate a 1,5-diketone system in the other end of the molecule. As the masked ketone, Stork's isoxazole and Danishefsky's 6-vinyl-2-picoline are well known. The usefulness of the bisannulation reagent is determined by easy accessibility of the reagent itself and the facile procedure of unmasking. For example, the isoxazole ring is unmasked by hydrogenation and base-catalyzed hydrolysis. Birch reduction, followed by acid-catalyzed hydrolysis is the method of converting the picoline into 1,5-dione. The new bisannulation reagent, 1,7-octadien-3-one, that we now introduce, is very easily prepared in high yield from butadiene. After initial Michael reaction at the enone moiety, the terminal double bond is converted into the desired methyl ketone in one step by palladium-catalyzed oxidation in high yield under mild conditions as shown below. Thus, this compound is the most cheaply and easily available, convenient bisannulation reagent.