A R T I C L E S
Nicolaou et al.
Scheme 9. Use of Advanced Intermediate 38 from the First Total
Synthesis of Diazonamide A as a Platform to Generate Structural
Analogs 39-43 for Biological Testinga
providing materials separately possessing the exocyclic 2-hy-
droxyisovaleryl residue or an aminal function in addition to the
12-membered AG macrocycle. Surprisingly, however, when we
screened these and other analogues, as well as protected and
fully deprotected compounds from the established sequences
toward 1, we were unable to discern any additional structure-
activity features from what we had already established. The only
real finding was the discovery that in the absence of the
complete diazonamide architecture, no compound possessed
cytotoxicity that broke the low micromolar barrier. Indeed, it
was only 1 and its C-37 epimer that ever possessed levels of
activity in the low nanomolar range. While a slightly disap-
pointing outcome from the standpoint of improving diazonamide
A’s activity or finding a simplified analogue that could become
a clinical agent, this result is in concert with studies toward
other bis-oxazole containing natural products where only minor
structural changes are tolerated before activity drops off
precipitously.
Methodology Development
Gratifyingly, our efforts to create diazonamide A’s architec-
ture were rewarded with a wealth of new synthetic methodolo-
gies. Some that we have already described, such as a means to
create 3-arylbenzofurans21 and the deoxygenation of sulfoxides
using titanocene methylidenes,22 were the direct result of insights
into chemical reactivity afforded by the unique diazonamide
skeleton. Others, such as the heteropinacol macrocyclization
sequence used in the successful total synthesis described in this
article, were the result of challenges imposed by the difficulty
in accessing its most challenging domains.23 As a final entry
into this collection of methods, we wish to expand in this section
on the generality of our unique conditions to accomplish
Robinson-Gabriel cyclodehydration on hindered substrates, a
protocol that succeeded where all others failed on three different
occasions in our diazonamide campaign, including both total
syntheses.
As indicated in Table 2, several keto amides (entries 1-6)
were smoothly converted into their oxazole counterparts through
the action of POCl3/pyridine (1:5) at 25 °C over the course of
3 h. The success of the dehydration in entry 3 serves as a useful
diagnostic measure for the power of this reaction on relatively
simple substrates, since this same dehydration was accomplished
in similar yield with a variety of other protocols during our
efforts to prepare BCD fragments for both the original and
revised structures of diazonamide A. The most important
example, however, may be the one in entry 7 that comes from
the elegant studies of Vedejs and Zajac24 toward the synthesis
of the 12-membered heterocyclic core of 1. On a compound
bearing a highly unstable aminal, an acid-labile Boc group, and
a benzyl ether, POCl3/pyridine proved to be the only set of
a Reagents and conditions: (a) DIBAL-H (1.0 M in toluene, 4.0 equiv,
added portionwise), CH2Cl2, 25 °C, 12 h, 73%; (b) MnO2 (10 equiv),
benzene, 80 °C, 1 h, 48% for 40, 23% for 41; (c) H2 (2.0 atm), Pd(OH)2/C
(20 wt %, catalytic), EtOH, 25 °C, 1.5 h; (d) 2 (1.1 equiv), EDC (1.5 equiv),
HOBt (1.5 equiv), NaHCO3 (7.5 equiv), DMF, 25 °C, 12 h, 47% over two
steps for 42, 21% over two steps for 43.
for the drug-efflux pump P-glycoprotein. Similar findings were
observed with epi-C-37 diazonamide A, except that its activity
levels were approximately 3- to 5-fold less potent (10-15 nM)
in the five cell lines that 1 was active against. Perhaps the most
important finding, however, was the ability of both of these
compounds to combat Taxol-resistant 1A9/PTX10 cells, sug-
gesting that the â-tubulin mutation these cells harbor has no
effect on their activity. As such, this outcome lends further
support for diazonamide A possessing a tubulin binding site
distinct from that of other chemotherapeutic agents with a similar
mechanism of action.
Spurred by these findings, as well as the preliminary
structure-activity relationships that we had established during
our studies toward the originally reported structure of diazon-
amide A,2b we sought to expand the scope of our tests by
screening intermediates obtained from both of our successful
total syntheses of 1. Moreover, since our developed sequences
approached the framework of diazonamide A from two com-
pletely different directions, we also prepared a series of
analogues of each of the macrocyclic domains of 1. For example,
as shown in Scheme 9, we were able to convert intermediate
38 into analogues 40-43 through a simple series of steps,
(21) Nicolaou, K. C.; Snyder, S. A.; Bigot, A.; Pfefferkorn, J. A. Angew. Chem.,
Int. Ed. 2000, 39, 1036-1039.
(22) Nicolaou, K. C.; Koumbis, A. E.; Snyder, S. A.; Simonsen, K. B. Angew.
Chem., Int. Ed. 2000, 39, 2529-2533.
(23) Challenges encountered during A-ring oxazole formation also inspired a
series of new synthetic methodologies related to the chemistry of the
Burgess reagent. (a) Nicolaou, K. C.; Huang, X.; Snyder, S. A.; Bheema
Rao, P.; Bella, M.; Reddy, M. V. Angew. Chem., Int. Ed. 2002, 41, 834-
838. (b) Nicolaou, K. C.; Longbottom, D. A.; Snyder, S. A.: Nalbandian,
A. Z.; Huang, X. Angew. Chem., Int. Ed. 2002, 41, 3866-3870. (c)
Nicolaou, K. C.; Snyder, S. A.; Nalbandian, A. Z.; Longbottom, D. A. J.
Am. Chem. Soc. 2004, 126, 6234-6235. (d) Nicolaou, K. C.; Snyder, S.
A.; Longbottom, D. A.; Nalbandian, A. Z.; Huang, X. Chem. Eur. J. 2004,
13, in press.
(24) Vedejs, E.; Zajac, M. A. Org. Lett. 2004, 6, 237-240.
9
12904 J. AM. CHEM. SOC. VOL. 126, NO. 40, 2004