BIOORGANIC CHEMISTRY 14,417-428 (1986) Sesquiterpene Biosynthesis: The Biosynthesis of Quadrone and Terrecyclic Acid DAVID E. CANE, YVONNE G. WHITTLE, AND TZYY-CHYAU LIANG Department of Chemistry, Brown University, Providence, Rhode Island 02912 Received February 25, 1986 Feeding of [I-WI- and [l,2J3Cz]acetates to cultures of Aspergillus terreus gave labeled quadrone (1) and terrecyclic acid (2) which were analyzed by high-field “C NMR. The patterns of enhancements and couplings were used not only in the analysis of the biosyn- thetic origins of the two metabolites, but also in the assignment of the “C spectra them- selves. The latter assignments were confirmed and further extended by extensive analysis by a combination of ‘H COSY, rH13C heteroCOSY, and difference NOE spectra of quadrone. The biosynthetic pathway was further probed by incorporation of [3,4J3Cz]mevalonate, revealing that formation of 1 and 2 involves cleavage of the 3,4-bond of one of the three mevalonte precursors. The results are consistent with the formation of quadrone and terre- cyclic acid by cyclization of famesyl pyrophosphate (8). 0 1986 Academic Press. Inc. Quadrone (1) is a metabolite of Aspergillus terreus first isolated in 1978 and shown to have modest antitumor activity (1, 2) (Fig. 1). The challenge presented by the unusual tetracyclic lactone skeleton has prompted numerous synthetic approaches, resulting in more than a half-dozen successful syntheses by 1985 (3- 12). The first of these syntheses, reported by Danishefsky et al. (3), involved as the last step the pyrolysis of the enone acid 2 to afford quadrone. It was therefore of some interest when the same substance 2, dubbed terrecyclic acid, was subse- quently isolated from culture filtrates of A. terreus, raising the possibility that this cometabolite is in fact a biosynthetic precursor of quadrone (13, 14). Further investigations by the groups of Hirota and Isogai have led to the isolation of additional cometabolites, including terrecyclol (3), (2.5), isoquadrone (4) (also prepared previously by Danishefsky), %hydroxyquadrone (5), and 6-hydroxyiso- quadrone (6) (16). In the meantime, Smith and Konopelski (II) have carried out a total synthesis of (+)-quadrone, which proved to be opposite in sign of rotation to the naturally occurring levorotatory isomer, thereby establishing the absolute configuration of (-)-(l), a result independently confirmed by Kon et al. (17) who converted (+)-fenchone to (-)-terrecyclic acid, the enantiomer of the naturally occurring material. More recently, the latter assignment has been confirmed by CD exciton chirality methods (18). Our own interest in the biosynthesis of dimethylcyclopentane sesquiterpenes (19-23) has led us to consider the problem of quadrone and terrecyclic acid biosynthesis, The novel tricyclic carbon skeleton of this family of fungal metabo- lites suggested that these compounds might represent yet another variation in the 417 0045-2068/86 $3 .OO Copyright 0 1986 by Academic Press, Inc. All rights of reproduction in any form reserved.