The Kinamycins

The structure elucidation of the kinamycins was a formidable challenge, and the information presented below draws from the work of several research groups over a period of more than 20 years. As will be shown, the originally proposed structure of the kinamycins contained a cyanamide rather than a diazo function. Subsequent synthetic and biosynthetic studies led to replacement of the cyanamide with a diazo function. The structural elucidation was challenging, in part, because of the high degree of unsaturation of the kinamycins, which limits the utility of H and 2D NMR analysis. In addition, because these structures were unprecedented, there were no clear benchmarks for comparison at the time. The pathway from isolation to determination of the correct structure is described below. 

Subsequent biosynthetic and synthetic studies, however, provided data that were inconsistent with the (V-cyanocarbazole function. Gould synthesized 15(V2-kinamycin 

Chart 3.2 13 Originally proposed structure for kinamycin C 14 structure an /V-cyanoindo-loquinone synthesized by Dmitrienko and coworkers 15 originally proposed structure for the metabolite prekinamycin 

---

Figure 7.23 shows the prototype diazobenzo[Z ]fluorene-based natural products kinamycin A and prekinamycin. The kinamycin A-D family were first isolated from Streptomyces murayamaensis, but the structures were incorrectly characterized as having a cyanobenzo[Z ]carbazole ring. Since the initial discovery of the kinamycins, many new analogues have been discovered from natural sources.88-92... 

Marco-Contelles, J. Molina, M. T. Naturally occurring diazo compounds the kinamycins. Curr. Org. Chem. 2003, 7, 1433-1442. 

Feldman and Eastman have suggested that the kinamycins may by reductively activated to form reactive vinyl radical (25) and orf/to-quinone methide (26) intermediates (Scheme 3.2c) [16]. The authors provided convincing evidence that the alkenyl radical 25 is generated when the model substrate dimethyl prekinamycin (24) is exposed to reducing conditions (tri-n-butyltin hydride, AIBN). Products that may arise from addition of this radical (25) to aromatic solvents (benzene, anisole, and benzonitrile) were isolated. The ort/io-quinone methide 26 was also formed,... 

Gould and coworkers have extensively studied the biosyntheses of the kinamycins, and this work was recently reviewed [5a]. Feeding studies established that the carbo-cyclic skeletons of the kinamycins are constructed from 10 equivalents of 5-acetyl coenzyme A, and the pathway shown in Scheme 3.4 was proposed. The pathway begins with formation of the natural product dehydrorabelomycin (29). A novel ring contraction then occurs to form the cyclopentadienone 30. Feeding studies with /V-15-ammonium sulfate established that the diazo functional group is then installed... 

Porco and Lei reported the first synthesis of (—)-kinamycin C (3) [24]. Their route constitutes the first completed pathway to any of the kinamycins and provides several powerful insights into the strategies that are viable for construction of... 

Porco s synthesis of ( )-kinamycin C (3) constituted the first reported route to any of the diazofluorene antitumor antibiotics. This synthesis invokes several powerful transformations, including a modified Baylis-Hillman reaction, a catalyst-controlled asymmetric nucleophilic epoxidation, and a regioselective epoxide opening to establish the D-ring of the kinamycins. The tetracyclic skeleton was constructed by an... 

Ishikawa s synthesis of ( )-0-methylkinamycin C (54) represents a distinct approach to the kinamycins that hinges on a key Diels-Alder reaction to establish the tetracyclic skeleton of the natural products. Additional key steps in the sequence include a substrate-directed dihydroxylation, substrate-directed reduction, and spontaneous epimerization of an a-hydroxyketone intermediate. 

Nicolaou and coworkers reported efficient enantioselective syntheses of ( )-kinamycin C (3), ( )-kinamycin F (6), and ( )-kinamycin J (10) [39], Nicolaou s retrosyntheses of these targets are shown in Scheme 3.13. The authors envisioned that all three metabolites could be accessed from the common precursor 82. The a-hydroxyketone function of 82 was envisioned to arise from an intramolecular benzoin reaction of the ketoaldehyde 83. This key bond disconnection would serve to forge the cyclopentyl ring of the kinamycin skeleton. The ketoaldehyde 83 was deconstructed by an Ullmann coupling of the aryl bromide 84 and the a-iodoenone 85. The latter were anticipated to arise from the bromojuglone derivative 86 and the enantiomerically enriched enone 87, respectively. 

The presence of 9-diazofluorene groups in kinamycin antitumor natural products would lead one to think of an active role for the diazo group. The hypothesis may be substantiated by the fact that one of the precursors in kinamycin biosynthesis, kinafluorenone 10 [44], which lacks the diazo moiety, shows no antibiotic activity against B. subtilis ATCC 6633, known to be very sensitive to the kinamycins. However, prekinamycin (9) [49], which is similar to kinafluorenone but retains the diazo group, shows activity to-... 

The isolation of diazobenzo[fr ]fluorenes as stable antitumor natural products raises several questions about their mode of action. The inability to cleave DNA by diazotization of 9-aminofluorene may imply that if the diazo functionality is involved in the mode of interaction of kinamycins with DNA, its conversion to diazonium and the ensuing reduction may seem to be of negligible importance. An additional possibility, which will be discussed later, is that 9-diazofluorene may not be the ideal model for these natural products. In exploring DNA cleavage as a possible route to the kinamycins role as a stable antitumor agent, which may supplement their speculative and as yet unconfirmed role as alkylating molecules [67], this early model seemed to suggest that the well-established activation of diazonium may not be relevant. 

---