Synthesis of the pyrazolo[4,3-e][1,2,4]triazine family of natural products: Nostocine A, fluviol A, and pseudoiodinine

Article abstract of DOI:10.1021/ja060937l

The first syntheses of any members of the naturally occurring pyrazolo[4,3-e][1,2,4]triazine family are reported, verifying the structures of nostocine A and fluviol A, and leading to the revision of the structure of pseudoiodinine. The revised structure of pseudoiodinine was also established by total synthesis. Copyright

Full text of DOI:10.1021/ja060937l

Published on Web 04/05/2006
Synthesis of the Pyrazolo[4,3-e][1,2,4]triazine Family of Natural Products:
Nostocine A, Fluviol A, and Pseudoiodinine
T. Ross Kelly,* Eric L. Elliott, Rimma Lebedev, and Jaione Pagalday
E.F. Merkert Chemistry Center, Boston College, Chestnut Hill, Massachusetts 02467
Received February 15, 2006; E-mail: ross.kelly@bc.edu
Scheme 1. Synthesis of Proposed Structure of Pseudoiodinine
The pyrazolo[4,3-e][1,2,4]triazine ring system has the distinction
of being one of only two naturally occurring ring skeletons
containing more nitrogen than carbon atoms.1 In the past 30+ years,
scientists from Germany, Japan, and Russia have reported the
isolation and structural characterization of seven naturally occurring
pyrazolo[4,3-e][1,2,4]triazines: pseudoiodinine (1),2 nostocine A
(2),3 and fluviols A (3)-E.4 In addition to their high nitrogen
content, family members are notable for the palette of bright colors
they display, including red, violet, purple, and yellow. To date,
there has been no independent verification of any of the assigned
structures nor reports on activity to achieve the total synthesis of
any members of the family. We now record the total synthesis and
structure confirmation of nostocine A and fluviol A and a structure
reassignment for pseudoiodinine. That structure reassignment is
supported by syntheses of both the originally proposed structure
and the revised structure.
Scheme 2. Synthesis of Fluviol A and Nostocine A
The synthesis of pseudoiodinine was undertaken first, in part
because we suspected that the structure was incorrectly assigned.
The original structure assignment was based heavily on the X-ray
crystallographic characterization of a degradation product of
pseudoiodinine named normethylpseudoiodinine and assigned
structure 3. Pseudoiodinine undergoes facile demethylation to give
normethylpseudoiodinine.2 Two or so decades later, a newly isolated
natural product, fluviol A, was also assigned structure 3. Our
suspicions regarding the correctness of structure 1 were largely a
consequence of our skepticism that 1 would suffer preferential N-
(rather than O-) demethylation to give 3.5
The synthesis of structure 1 is summarized in Scheme 1. Reaction
of diethyl aminomalonate with dimethylformamide dimethyl acetal
gave amidine 5;6 condensation7 of 5 with hydrazine afforded
dihydrotriazinone 6. The latter could be oxidized to triazinone 7
with iodobenzene diacetate.8 Treatment of 7 with POCl3 under the
conditions of Robins9 delivered chlorotriazine 8 which was reacted
with the Boc-protected methylhydrazine 910 to give 10. Boc
protection of methylhydrazine was necessary because otherwise the
methylated nitrogen in methylhydrazine is the more nucleophilic
of the two nitrogens11 (a property of methylhydrazine that is
exploited to prepare the Boc derivative 9 itself).
might seem dubious, but reaction of 11 with the oxophilic,
commercially available (trimethylsilyl)diazomethane afforded 12
in 8% yield (13C NMR clearly distinguishes O- versus N-
methylation). Structure 12 is the structure (1) previously assigned
to pseudoiodinine, but spectra of the yellow synthetic 12 are
decidedly different from those of the red natural product. The
inescapable conclusion is that 1 is not the structure of pseudo-
iodinine.
With the demonstration that pseudoiodinine does not possess
structure 1, the next question was whether the pivotal structure (3)
of normethylpseudoiodinine (supposedly ) fluviol A) is correct.12
To that end, synthesis of 3 was undertaken (Scheme 2). Chloro-
triazine 8 was reacted with Boc-hydrazine13 (13) to give 14.
Treatment of 14 with CF3COOH followed by basification proceeded
in analogy to Scheme 1, delivering bicyclic hydrazide 15, conve-
niently manipulated as its hydrochloride salt (15a). Exposure of
15a to (trimethylsilyl)diazomethane (3 cycles @ 0.6 equiv14) gave
primarily two products, an O-methylated and an N-methylated
Cleavage of the Boc group in 10 with trifluoroacetic acid and
basification of the crude product leads directly to the bicyclic
pyrazolone 11. O-Methylation with accompanying tautomerization
of 11 should produce 1. The prospects of achieving selective
methylation of a single oxygen in the presence of five nitrogens
9
5646
J. AM. CHEM. SOC. 2006, 128, 5646-5647
10.1021/ja060937l CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
above) by reaction of 3 with diazomethane. The structure of
pseudoiodinine is thus revised from 1 to 19 and confirmed by total
synthesis.16
In conclusion, these first-reported syntheses of any members of
the naturally occurring pyrazolo[4,3-e][1,2,4]triazine family verify
the structures of nostocine A and fluviol A as 2 and 3, respectively,
and lead to the revision of the structure of pseudoiodinine from 1
to 19. Structures 1 and 19 were also established by total synthesis.
Acknowledgment. J. P. acknowledges the Ministerio de Edu-
cacio´n y Ciencia (Madrid) for a postdoctoral fellowship. We are
extremely grateful to Dr. Svetlana A. Dovzhenko for helpful
exchange of information, and Drs. Elena A. Kiprianova and Elena
P. Ivanova for assistance. We thank Professor John Frost (Michigan
State University) for bringing ref 18 to our attention.
Figure 1. Possible methylation products of 3 (other than 1).
species. The O-methylated species is identical to normethyl-
pseudoiodinine/fluviol A (3), thereby providing independent con-
firmation of their structures. The N-methylated species was an
unplanned bonus since it proved to be identical to nostocine A.
The structure of nostocine A (2) was originally assigned by X-ray
crystallographic analysis, and the location of the methyl group in
natural 2 was confirmed by NMR (heteronuclear multiple bond
correlation, i.e., HMBC) spectroscopy. HMBC studies15 on synthetic
nostocine A certified the earlier assignments.
Supporting Information Available: Experimental and character-
ization data for compounds 2, 3, 5-12, 14, 15, and 19-21; reproduc-
tions of selected NMR spectra. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
While the synthesis of 12 reported above establishes that the
structure 1 () 12) of pseudoiodinine is wrong, the original workers
provided one further piece of information: reaction of normeth-
ylpseudoiodinine (3) with diazomethane gives pseudoiodinine in
approximately 12% yield. With the hope that spectroscopic
techniques had advanced sufficiently in the 30+ years since the
structure of pseudoiodinine was originally examined and that they
might help clarify the situation, we reacted 3 with diazomethane
and obtained a 7% yield of synthetic material with properties
essentially identical to those reported for pseudoiodinine. It
remained only to locate the position of the newly installed methyl
group.
(1) That assertion is based on a thorough search of the “Dictionary of Natural
Products”17 and an informal polling of numerous members of the organic
community. To our knowledge, one 1,2,4-triazole has been found in
nature.18 The two most likely exceptions to the generalization, 1,2,3-
triazoles19 and azapurines,20 have both been stated by others not to occur
naturally.
(2) Lindner, H. J.; Schaden, G. Chem. Ber. 1972, 105, 1949-1955.
(3) Hirata, K.; Nakagami, H.; Takashina, J.; Mahmud, T.; Kobayashi, M.;
In, Y.; Ishida, T.; Miyamoo, K. Heterocycles 1996, 43, 1513-1519.
(4) Smirnov, V. V.; Kiprianova, E. A.; Garagulya, A. D.; Esipov, S. E.;
Dovjenko, S. A. FEMS Microbiol. Lett. 1997, 153, 357-361.
(5) Beak, P.; Bonham, J.; Lee, J. T., Jr. J. Am. Chem. Soc. 1968, 90, 1569-
1582. Note, however, ref 5 therein.
(6) Smodicˇ, J.; Zupet, R.; Petricˇ, A.; Stanovnik, B.; Ticˇler, M. Heterocycles
1990, 30, 393-405.
(7) Procedure adapted from the synthesis of 5-hydrazinocarbonyl-4,5-dihydro-
1,2,4-triazin-6(1H)-one reported in ref 6.
(8) Prakash, O.; Saini, N.; Sharma, P. K. Synlett 1994, 221-227.
(9) Robins, M. J.; Uznanski, B. Can. J. Chem. 1981, 59, 2601-2607.
(10) Metz, H. J.; Neunhoeffer, H. Chem. Ber. 1982, 115, 2807-2818.
(11) (a) Goldwhite, H. In Rodd’s Chemistry of Carbon Compounds, 2nd ed.;
Coffey, S., Ed.; Elsevier: New York, 1965; Vol. 1, p 151. (b) Atkinson,
R. S. In ComprehensiVe Organic Chemistry; Sutherland, I. O., Ed.;
Pergamon: Oxford, 1979; Vol. 2, p 219.
(12) The structure of normethylpseudoiodinine was determined by X-ray
crystallography, which is regarded as an absolute method, but the literature
is replete with erroneous X-ray structures: See, for example: (a) Jones,
P. G. Chem. Soc. ReV. 1984, 13, 157-172. (b) Parkin, G. Acc. Chem.
Res. 1992, 25, 455-460. (c) Marsh, R. E.; Kapon, M.; Hu, S.; Herbstein,
F. H. Acta Crystallogr., Sect. B: Struct. Sci. 2002, B58, 62-77. (d)
Nicolaou, K. C.; Snyder, S. A. Angew. Chem., Int. Ed. 2005, 44, 1012-
1044.
(13) Hydrazine itself is symmetrical. In contrast to the situation with methyl-
hydrazine, the reason for the Boc protection in the present case is to
suppress reaction of hydrazine with two molecules of 8, which was a
significant side reaction.
(14) Less than 1 equiv per cycle was used to suppress dimethylation.
(15) See Supporting Information.
(16) That 19 is the structure of pseudoiodinine means that our previously
mentioned skepticism about selective N- (rather than O-) demethylation
is misguided in the case of 19. The explanation may be related to the
finding that, unlike 2-pyridones,5 3-hydroxypyrazoles exist primarily as
the hydroxy rather than oxo tautomer in nonpolar media: Katritzky, A.
R.; Maine, F. W. Tetrahedron 1964, 20, 315-322.
(17) Dictionary of Natural Products, Chemical Databases Online. http://
www.chemnetbase.com/scripts/dnpweb.exe?welcome-main Web version
2004, Chapman & Hall/CRC Press.
Apart from 1, there are five possible methylation products of 3,
1
16-20 (Figure 1). The H NMR spectrum of synthetic pseudo-
iodinine contains a resonance for an aromatic hydrogen, eliminating
16 as a possible structure. HMBC spectroscopy reveals that the
hydrogens on the two methyl groups in pseudoiodinine each
correlate with only onesbut differentsring carbon, neither of which
bears a hydrogen. That finding excludes 17 and 18 as possible
structures for pseudoiodinine.
To determine whether 19 or 20 is the correct structure for
pseudoiodinine, we again turned to synthesis. The synthesis of 20
(eq 1) was begun by reaction of chloro ester 8 with methylhydrazine
to give the bicyclic 21 directly (shown to be different from its
previously synthesized regioisomer 11). Reaction of 21 with
(trimethylsilyl)diazomethane is capable of delivering only one
O-methylated product (20), which it does in 60% yield. The spectra
of yellow synthetic 20 show that it is not pseudoiodinine.
(18) Schmitzer, R. R.; Graupner, P. R.; Chapin, E. L.; Fields, S. C.; Gilbert,
J. R.; Gray, J. A.; Peacock, C. L.; Gerwick, C. J. Nat. Prod. 2000, 63,
777-781.
(19) Eicher, T.; Hauptmann, S. The Chemistry of Heterocycles; Wiley-VCH:
Weinheim, Germany, 2003; p 205.
By process of elimination, 19 must be the structure of pseudo-
iodinine. Nonetheless, it was desirable to positively affirm the
identity. That was done: O-methylation of nostocine A (2), which
we had already synthesized, with ethereal diazomethane gives
pseudoiodinine, identical with the material produced (as described
(20) Knowles, P. In ComprehensiVe Heterocyclic Chemistry II; Katritzky, A.
R., Rees, C. W., Scriven, E. F. V., Ramsden, C. A., Eds.; Pergamon:
Oxford, 1996; Vol. 7, p 511.
JA060937L
9
J. AM. CHEM. SOC. VOL. 128, NO. 17, 2006 5647