Efficient synthesis of a peculiar vicinal diamine semiochemical from streptomyces natalensis

Article abstract of DOI:10.1021/ol1013763

(Equation Presented). The pimaricin-inducing (PI) factor, produced by Streptomyces natalensis is a proposed pheromone with a peculiar vicinal diamine structure. The first synthesis of this molecule is reported. It features oxidative dimerization of an aci-nitro anion derived from tris(hydroxymethyl) nitromethane and disproportionation catalyst-facilitated hydrogenation of the resulting vicinal tertiary dinitro compound. As the synthesis requires only four steps with no chromatographic separations, it provides a convenient route to prepare PI factor for biological studies and industrial applications.

Full text of DOI:10.1021/ol1013763

ORGANIC
LETTERS
2010
Vol. 12, No. 15
3522-3524
Efficient Synthesis of a Peculiar Vicinal
Diamine Semiochemical from
Streptomyces natalensis
Jesse B. Morin and Jason K. Sello*
Department of Chemistry, Brown UniVersity, 324 Brook Street,
ProVidence, Rhode Island 02912
jason_sello@brown.edu
Received June 14, 2010
ABSTRACT
The pimaricin-inducing (PI) factor, produced by Streptomyces natalensis is a proposed pheromone with a peculiar vicinal diamine structure.
The first synthesis of this molecule is reported. It features oxidative dimerization of an aci-nitro anion derived from tris(hydroxymethyl)nitromethane
and disproportionation catalyst-facilitated hydrogenation of the resulting vicinal tertiary dinitro compound. As the synthesis requires only four
steps with no chromatographic separations, it provides a convenient route to prepare PI factor for biological studies and industrial applications.
Streptomyces natalensis is a soil-dwelling bacterium that
produces a commercially important antifungal polyene mac-
rolide called pimaricin (Figure 1).1 This natural product is
widely used in clinical and veterinary medicine for treatment
of fungal infections and in the food and agricultural industries
as a preservative. Notably, pimaricin is the only FDA-
approved natural antifungal food preservative on the market.
Tons of pimaricin are produced annually by fermentation of
Streptomyces bacteria.
In 2004, a small molecule termed the “pimaricin-inducing
Figure 1. Pimaricin and the pimaricin-inducing factor (PI factor,
factor” (PI factor, 1) was isolated from wild type S. natalensis
(Figure 1).2 It is believed to be a critical regulator of
pimaricin biosynthesis in S. natalensis.3 The proposed
structure of PI factor is a tetrahydroxy vicinal diamine.
Interestingly, no other bacterial chemical messenger with a
similar structure has been reported.4 PI factor is reported to
increase pimaricin production in the wild type S. natalensis
by up to 40% at concentrations as low as 50 nM.5 The
1) from S. natalensis.
mechanism of action of PI factor is entirely unknown, and its
utility in industrial fermentations has yet to be established.
Studies of PI factor have been limited by access to pure
preparations of the semiochemical. The low quantities in which
S. natalensis produces PI factor (nanograms/liter of culture) and
the high polarity of the molecule make large-scale isolation
impractical.2 The only viable means for producing useful
quantities of PI factor is by chemical synthesis.
Although PI factor is a very low-molecular weight compound
lacking stereocenters, its density of polar functional groups
makes it a challenging synthetic target. In exploring synthetic
routes toward PI factor, we found a lack of general and efficient
methods for the preparation of vicinal diamines. The most
(1) Aparicio, J. F.; Fouces, R.; Mendes, M. V.; Olivera, N.; Martin,
J. F. Chem. Biol. 2001, 7, 895–905.
(2) Recio, E.; Colinas, A.; Rumbero, A.; Aparicio, J. F.; Martin, J. F.
J. Biol. Chem. 2004, 279, 41586–41593.
(3) Vicente, C. M.; Santos-Aberturas, J.; Guerra, S. M.; Payero, T. D.;
Martin, J. F.; Aparicio, J. F. Microb. Cell Fact. 2009, 8, 33–46.
(4) Bassler, B. L.; Losick, R. Cell 2006, 125, 237–246.
(5) Martin, J. F.; Aparicio, J. F. Methods Enzymol. 2009, 459, 215–
242.
10.1021/ol1013763  2010 American Chemical Society
Published on Web 07/15/2010

commonly reported routes to these compounds involve reductive
couplings of imines, diamination of alkenes, conjugate additions
of amines to R,ꢀ-unsaturated nitro compounds, and ring-opening
of aziridines by nitrogen nucleophiles.6 None of these methods
are ideal for the synthesis of PI factor.
In light of the limitations of methods for vicinal diamine
synthesis, we reasoned that the best way to construct the
skeleton of PI factor, 1, would be through oxidative dimer-
ization of an aci-nitro anion followed by hydrogenation of
the resulting dinitro product, 3 (Scheme 1). We envisioned
sodium persulfate are virtually identical. Given the limited
solubility of nitroalkanes in water, we suspected that our
conditions for dimerization would be compatible with a wider
range of substrates. Thus, we carried out a series of CAN-
mediated oxidative dimerizations with both primary and
secondary nitroalkanes. We were gratified to find that inter-
and intramolecular oxidative couplings of aliphatic nitro
substrates were achieved in moderate yields (Table 1).
Table 1. Substrate Scope of Oxidative Nitro Dimerization
Scheme 1. Retrosynthetic Analysis of the PI Factor
that the requisite aci-nitro anion could be generated from
tris(hydroxymethyl)nitromethane via a retro-Henry reaction.
In fact, oxidative dimerization of the aci-nitro anion derived
from ketal-protected tris(hydroxymethyl)nitromethane,7 4,
using hydrogen peroxide as an oxidant was reported in a
patent.8 Although this reaction provided the desired product,
its yield was only 11%. We set out to develop reaction
conditions for the oxidative dimerization of the requisite
dinitro compound that were higher yielding.
a 2.4 equiv of NaH in THF, 8 h at rt; 2.2 equiv of CAN in MeOH, 15 h.
b 1.2 equiv of NaH in THF, 1.5-4 h at rt; 2.2 equiv of CAN in MeOH, 15 h.
c Reaction performed as in b but at 0.015 M to promote intramolecular reaction.
After extensive screening of bases and solvents, it was found
that the retro Henry reaction yielding the aci-nitro anion from
4 proceeded best with NaH in THF. To effect the single electron
oxidation of the anion, our oxidant of choice was ceric
ammonium nitrate (CAN). This reagent has been shown to be
effective in the oxidation of aci-nitro anions.9 CAN-promoted
oxidative dimerization yielded the desired product 3 in a 3-fold
higher yield than same reaction using a peroxide oxidant. We
reasoned that the yield of the CAN-promoted oxidation of the
aci-nitro anion was negatively effected by the presence of the
sodium counterion. We found that inclusion of the sodium
chelator 15-crown-5 increased the yield of the CAN-promoted
oxidation from 38% to 68%.
Having prepared the desired dinitro compound 3, the next
transformation in our synthetic scheme was hydrogenation
to yield the vicinal diamine 2. Although this transformation
seems trivial, it is fraught with difficulties. Accumulation of
dangerous hydroxylamine intermediates that are difficult to
reduce is a major problem in nitro reductions.11 Additionally,
tertiary vicinal dinitro compounds can undergo sequential
elimination of both nitro groups after single electron transfer
to generate an undesired alkene side product.12,13 There is a
single precedent for the reduction of hindered vinical dinitro
compounds, but it requires harsh conditions (Sn metal with
12 N hydrochloric acid) that would remove the ketal
protecting groups and complicate recovery of the desired
product.14 The efficacy of hydrazinium monoformate in
reductions of both aromatic and aliphatic nitro compounds
In the course of our work, Chavez and co-workers reported
oxidative dimerization of 4 using aqueous sodium persulfate
as the oxidant.10 The yield of the reaction under our
conditions and that of the reaction effected by aqueous
(11) (a) Baumeister, P.; Blaser, H.-U.; Studer, M. Catal. Lett. 1997, 49,
219–222. (b) Studer, M.; Neto, S.; Blaser, H.-U. Top. Catal. 2000, 13, 205–
212.
(6) Lucet, D.; Le Gall, T.; Mioskowski, C. Angew. Chem., Int. Ed. 1998,
37, 2580–2627.
(7) Quirk, J. M.; Harsy, S. G.; Hakansson, C. L. U. S. Patent 4,978,793,
1990.
(12) Bowyer, W. J.; Evans, D. H. J. Org. Chem. 1988, 53, 5234–5239.
(13) Indeed using SmI2/MeOH, NaBH4/NiCl2, NaBH4/CoCl2, Zn, Fe,
Mg, LiAlH4, and Pd/C led to the formation of alkene 2,2,2′,2′-tetramethyl-
4,4′,6,6′-tetrahydro-5,5′-bi(1,3-dioxinylidene) as the major product.
(14) Asaro, M. F.; Nakayama, I.; Wilson, R. B., Jr. J. Org. Chem. 1992,
57, 778–782.
(8) Senkus, M. US Patent 2,543,472, 1951.
(9) Arai, N.; Naraska, K. Bull. Chem. Soc. Jpn. 1997, 70, 2525–2534.
(10) Chavez, D. E.; Hiskey, M. A.; Naud, D. L.; Parrish, D. Angew.
Chem., Int. Ed. 2008, 47, 8307–8309.
Org. Lett., Vol. 12, No. 15, 2010
3523

with a variety of catalysts15 prompted us to attempt the
desired transformation using this hydrogen source and Raney
nickel as the catalyst. Under these reaction conditions only
a minor amount of the desired vicinal diamine was formed,
but the major products were partially hydrogenated (i.e.,
dihydroxylamine and aminohydroxylamine). We hypoth-
esized that this problem could be remedied by using
ammonium metavanadate as a disproportionation catalyst.11
This reagent catalyzes the conversion of two molecules of
hydroxylamine to the corresponding amine and the more
easily reduced nitroso compound.11 Indeed, inclusion of 5
mol % of ammonium metavanadate in the hydrogenation
afforded the diamine 2 in 71% yield with no detectable traces
of partially hydrogenated products.
doublet of doublets, especially due to the potential for intramo-
lecular hydrogen-bonding. We suspected that the chemical shifts
and coupling of these protons would be affected by pH. Using
an internally referenced D2O buffer at various pHs, the pH
1
dependence of the H NMR signals was investigated (Figure
3).16 We found that the signal from these methylene protons
Scheme 2. Total Synthesis of PI Factor
1
Figure 3
.
pH dependence of PI factor: H NMR spectra.
To complete the PI factor synthesis, the ketal protecting
groups were removed in quantitative yield using 1:1 6 M
HCl in methanol (Scheme 2). The recrystallized product was
analyzed by NMR, mass spectrometry, and X-ray crystal-
lography (Figure 2) confirming that our synthetic material
was identical to the structure proposed by Martin and co-
workers.2 Thus, PI factor was synthesized from tris(hy-
droxymethyl)nitromethane in 48% yield over four steps.
appears as a singlet in a pH range near 9 and as a doublet of
doublets with coupling constants of 12 Hz (typical of geminal
protons) at other pHs. The structural basis of this NMR behavior
is not entirely clear, but these observations suggest that the
ionization state influences molecular conformation.
We report the first chemical synthesis of the proposed
structure of the pimaricin-inducing factor from S. natalensis.
Given the polarity of the synthetic intermediates and the final
product, a strength of our route is that it requires no chromato-
graphic steps. Most importantly, it provides a means to prepare
sufficient quantities of the compound for biological testing and
for an assessment of its utility as an additive to pimaricin
fermentations.5
Acknowledgment. Financial support for this work was
generously provided by Brown University and a “Frontiers in
Chemistry Award” from the Brown Chemistry Department to
J.K.S. Prof. P. G. Williard is gratefully acknowledged for X-ray
analysis of diamine 2 and PI factor, 1. Dr. Russell Hopson and
Dr. Tun-Li Shen from the Brown Chemistry Department are
gratefully acknowledged for technical support in NMR and MS
analysis.
Figure 2
.
Crystal structure of the mono-HCl salt of PI factor, 1.
1
A curious feature of the reported H NMR spectrum of PI
factor was that the methylene protons appeared as a singlet.2
One would anticipate that these diastereotopic, geminal protons
would have nonequivalent chemical shifts and appear as a
Supporting Information Available: Procedures and
spectroscopic data for all compounds. Crystallographic data
for diamine 2 and PI factor, 1. This material is available
free of charge via the Internet at http://pubs.acs.org.
(15) Gowda, S.; Gowda, D. C. Tetrahedron 2002, 11, 2211–2213.
(16) Baryshnikova, O. K.; Williams, T. C.; Sykes, B. D. J. Biomol. NMR
2008, 41, 5–7.
OL1013763
3524
Org. Lett., Vol. 12, No. 15, 2010