Synthesis of pyrrolnitrin and related halogenated phenylpyrroles

Article abstract of DOI:10.1021/ol8026957

A general approach to halogenated arylpyrroles, including the antifungal natural product pyrrolnitrin, is described using newly synthesized halogenated pyrroles and 2,6-disubstituted nitrobenzenes or 2,6-disubstituted anilines.

Full text of DOI:10.1021/ol8026957

ORGANIC
LETTERS
2009
Vol. 11, No. 5
1051-1054
Synthesis of Pyrrolnitrin and Related
Halogenated Phenylpyrroles
Matthew D. Morrison, Jason J. Hanthorn, and Derek A. Pratt*
Department of Chemistry, Queen’s UniVersity, Kingston, Ontario, Canada K7L 3N6
pratt@chem.queensu.ca
Received November 21, 2008
ABSTRACT
A general approach to halogenated arylpyrroles, including the antifungal natural product pyrrolnitrin, is described using newly synthesized
halogenated pyrroles and 2,6-disubstituted nitrobenzenes or 2,6-disubstituted anilines.
Halogenations are a particularly interesting class of enzymatic
reactions, especially in light of the high percentage of
medicinal natural products and pharmaceuticals that bear
halogen atoms.1 Enzymatic incorporation of halogens during
natural product biosynthesis allows for the fine-tuning of
electronic and steric properties which determine the affinity
and selectivity of a molecule’s interactions with its biological
target. Enzymes can install halogen atoms on aliphatic
carbons, olefinic centers, and a wide variety of aromatic and
heterocyclic rings.2 For some time, it was thought that
haloperoxidases were Nature’s only halogenation catalysts,3
but more recent work has uncovered two further classes of
oxidative enzymatic catalysts: R-ketoglutarate-dependent4
and flavin-dependent halogenases.5
from tryptophan (1) is believed to involve two regioselective
chlorinations of unactivated positions carried out by the
flavin-dependent halogenases PrnA and PrnC (Scheme 1).6
Scheme 1. Proposed Mechanism of Pyrrolinitrin Biosynthesis
Few of these enzymes have been structurally and/or
mechanistically characterized and many questions surround
their activities. Our interests lie in the mechanisms of the
halogenation steps and the strategies employed by the
enzymes to ensure regioselectivity; especially in cases
where the “unexpected” regioisomer is formed. For example,
the biosynthesis of the antifungal antibiotic pyrrolnitrin (5)
The lack of commercial availability of the putative
biosynthetic precursors to pyrrolnitrin or suitable sub-
strate analogs thereof has precluded detailed studies of these
enzymes beyond PrnA.7 Although two total syntheses of 5
have been reported,8,9 they are cumbersome, involving de
(1) Gribble, G. W. J. Chem. Educ. 2004, 81, 1441–1449.
(2) Vaillancourt, F. H.; Yeh, E.; Vosburg, D. A.; Garneau-Tsodikova,
S.; Walsh, C. T. Chem. ReV. 2006, 106, 3364–3378.
(3) Morrison, M.; Schonbaum, G. R. Annu. ReV. Biochem. 1976, 45,
861–888.
(6) Kirner, S.; Hammer, P. E.; Hill, D. S.; Altmann, A.; Fischer, I.;
Weislo, L. J.; Lanahan, M.; van Pee, K.-H.; Ligon, J. M. J. Bacteriol. 1998,
180, 1939–1943.
(4) Vaillancourt, F. H.; Yin, J.; Walsh, C. T. Proc. Natl. Acad. Sci. U.S.A.
2005, 102, 10111–10116.
(5) Keller, S.; Wage, T.; Hohaus, K.; Holzer, M.; Eichhorn, E.; van
Pee, K.-H. Angew. Chem., Int. Ed. 2000, 39, 2300–2302.
(7) Dong, C. J.; Flecks, S.; Unversucht, S.; Haupt, C.; van Pee, K.-H.;
Naismith, J. H. Science 2005, 309, 2216–2219.
10.1021/ol8026957 CCC: $40.75
Published on Web 02/05/2009
2009 American Chemical Society

noVo construction of the substituted pyrrole ring and are not
easily amenable to access related derivatives. We therefore
developed a new approach to 5, as well as its biosynthetic
precursors 3 and 4, in parallel with our efforts to heterolo-
gously overexpress the four individual members of the
pyrrolnitrin biosynthetic gene cluster in E. coli10 to facilitate
in vitro mechanistic and structural studies of these enzymes.
We report our synthetic efforts here.
Scheme 3. Regioselective Halogenation of Pyrroles
Since 3-arylpyrroles have been constructed by Pd-
catalyzed cross-coupling of appropriate 3-boronated pyrroles
with aryl halides,13 we settled on the retrosynthetic approach
to 5 shown in Scheme 2. This would allow us to more easily
Scheme 2. Retrosynthetic Approach to Pyrrolnitrin
rrole 11, following treatment with NBS and NIS, respec-
tively. The pinacolboronate ester 12 could then be easily
prepared from 10 or 11 using the procedure of Billingsley
and Buchwald (Scheme 4).16
Scheme 4. Preparation of 3-Chloropyrroles
prepare 5, its putative biosynthetic precursors, and substrate
analogs by simply changing the substitution on the pyrrole
or aryl halide undergoing cross-coupling.
Unfortunately, a regioselective preparation of 3-chloro-
pyrrole has hitherto not been described in the literature.14,15
While regioselective bromination and iodination of pyrrole
can be directed to the 3-position conveniently using the
procedure of Muchowski and co-workers (Scheme 3),14 this
yields a mixture of products when attemepted with common
electrophilic chlorinating agents. However, we found that
by first installing bromine regioselectively as in Scheme 3,
subsequent lithium-halogen exchange followed by quenching
of the organolithium with an electrophilic chlorine source
(e.g., hexachloroethane) allowed access to new halogenated
pyrroles in good yields, such as the 3-chloropyrrole 8 as well
as the 3-bromo-4-chloropyrrole 10 and 3-chloro-4-iodopy-
Preparation of the desired 2-bromo-6-chloroaniline 17 (and
2-bromo-6-chloronitrobenzene 18) was somewhat challeng-
ing. While it has been reported that 17 can be prepared in
71% yield by hydrolysis of the anilide produced from the
treatment of N-(2-bromophenyl) benzohydroxamic acid with
thionyl chloride,17 this reaction proved difficult to reproduce
in our hands.18 We next attempted the preparation of 17 using
directed ortho metalation of 2-chloroaniline. Both tert-
butoxycarbonyl (Boc)19 and pivaloyl (Piv)20 were used as
directed metalation groups on the aniline moiety. Unfortu-
(8) Nakano, H.; Umio, S.; Kariyone, K.; Tanaka, K.; Kishimot, T.;
Noguchi, H.; Ueda, I.; Nakamura, H.; Morimoto, Y. Tetrahedron Lett. 1966,
7, 737–740.
(9) Gosteli, J. HelV. Chim. Acta 1972, 55, 451–460.
(10) To date, overexpression of PrnA7 and PrnB11 in Pseudomonas
fluorescens and PrnD12 in E. coli has been reported, but enzyme activity
has only been reconstituted in Vitro for PrnA and PrnD.
(11) De Laurentis, W.; Khim, L.; Anderson, J. L.; Adam, A.; Johnson,
K. A.; Phillips, R. S.; Chapman, S. K.; van Pee, K. H.; Naismith, J. H.
Biochemistry 2007, 46, 12393–12404.
(12) Lee, J.; Simurdiak, M.; Zhao, H. J. Biol. Chem. 2005, 280, 36719–
36728.
(16) Billingsley, K.; Buchwald, S. L. J. Am. Chem. Soc. 2007, 129,
3358–3366.
(13) Alvarez, A.; Guzman, A.; Ruiz, A.; Velarde, E.; Muchowski, J. M.
J. Org. Chem. 1992, 57, 1653–1656.
(17) Ayyangar, N. R.; Kalkote, U. R.; Nikrad, P. V. Ind. J. Chem. B
1983, 22, 872–877.
(14) Bray, B. L.; Mathies, P. H.; Naef, R.; Solas, D. R.; Tidwell, T. T.;
Artis, D. R.; Muchowski, J. M. J. Org. Chem. 1990, 55, 6317–6328.
(15) 3-Chloropyrrole has been isolated and characterized from mixtures
obtained upon acidification of N-chloropyrrole (De Rosa, M. J. Org. Chem.
1982, 47, 1008-1010) and photolysis of 3-chloropyridine N-oxide
(Bellamy, F.; Streith, J.; Fritz, H. NouV. J. Chim. 1979, 3, 115-122).
(18) Attempts carried out at low temperature in Et2O, THF, or benzene
gave no reaction, whereas reactions carried out at elevated temperatures or
with excess SOCl2 gave complex product mixtures.
(19) Muchowski, J. M.; Venuti, M. C. J. Org. Chem. 1980, 45, 4798–
4801.
(20) Fuhrer, W.; Gschwend, H. W. J. Org. Chem. 1979, 44, 1133–1136.
1052
Org. Lett., Vol. 11, No. 5, 2009

nately, while the 2-chloropivanilide could be metallated and
halogenated in excellent yield, the subsequent removal of
the Piv group proved difficult. In contrast, the Boc group
could be easily removed, but the metalation/halogenation
proceeded in poor yields.
ester (25, prepared from 9 in 80% yield via the analogous
conditions in Scheme 4) with commercially available 2-bro-
moaniline and 2-bromonitrobenzene as well as our newly
prepared 17 and 18. The results are shown in Table 2.
We then came across the work of Roe,21 which showed
that 1,3-difluorobenzene and 1,3-dichlorobenzene could be
converted to 2,6-difluorobenzoic acid and 2,6-dichloroben-
zoic acid in 81% and 75% yield, respectively, via an
aryllithium intermediate subjected to carboxylation and
acidification.21a A subsequent Schmidt rearrangement was
shown to yield 2,6-difluoroaniline in 86% yield.21b
Table 2. Suzuki-Miyaura Cross-Couplings of the Pyrrole
Pinacolboronote Ester 25 and its Chlorinated Derivative 12 with
Relevant 2-Bromoanilines and 2-Bromonitrobenzenes
While Roe’s route was encouraging, Hickey and co-
workers22 have shown that benzyne-type decomposition of
the 2-bromo-6-chlorophenyl lithium intermediate occurs
under these conditions. Therefore, following Hickey et al.,
we switched from n-butyllithium to lithium di-iso-propyl
amide as the base and from -50 to -78 °C as the
temperature for the lithiation step and were very pleased to
see quantitative formation of 2-bromo-6-chlorobenzoic acid
following quenching with CO2 and acidification.
Indeed, these conditions improve on Roe’s yields of 2,6-
difluorobenzoic acid and 2,6-dichlorobenzoic acid (not
shown) and also allowed the peparation of 2,6-dibromoben-
zoic acid from 1,3-dibromobenzene. We found that dropping
the reaction temperature further to -100 °C allowed the
lithiation/carboxylation sequence to be carried out on even
1-bromo-3-iodobenzene.23 Subjecting these dihalobenzoic
acids to Schmidt conditions afforded the 2,6-dihaloanilines
in good to excellent yields, including the desired 2-bromo-
6-chloroaniline 17. Furthermore, oxidation of the 2,6-
dihaloanilines with m-chloroperbenzoic acid in dichloroet-
hane afforded the 2,6-dihalonitrobenzenes in good yield,
including the desired 2-bromo-6-chloronitrobenzene 18. The
results of the lithiation-carboxylation-Schmidt-oxidation
sequence are summarized in Table 1.
a Carried out at 100 °C for 16 h. b Carried out at 0 °C for 16 h. c Carried
out at 100 °C for 5 h.
We found the conditions recently reported by Billingsley
and Buchwald,16 which employ the newer SPhos ligand in
a 2:1 ratio with Pd(OAc)2 to be superior to those reported
by Muchowski et al.13 for the preparation of simple arylpyr-
roles some time ago, which employed Pd(PPh3)4. The use
of a butanol:water cosolvent system was key to minimize
the amount of reduced aryl halide formed and maximize the
overall yield. While couplings of 2-bromoaniline required
100 °C to achieve respectable yields for an overnight
reaction, 2-bromonitrobenzene, 17 and 18 coupled with good
yield overnight when heated to only 35 °C. Fluoro-desily-
lations also proceeded smoothly to yield the aryl pyrroles
Table 1. 2,6-Dihaloanilines (and 2,6-Dihalonitrobenzenes) by
the Lithiation-Carboxylation-Schmidt (Oxidation) Sequence
aryl halide
yield (product) yield (product) yield (product)
X, Y ) Br, F
X, Y ) Br, Cl
X, Y ) Br, Br
X, Y ) Br, I
92 (13)
99 (16)
85 (19)
80 (22)
86 (14)
92 (17)
78 (20)
83 (23)
88 (15)
94 (18)
82 (21)
78 (24)
a (i) LDA, THF, -78 °C, 1 h; (ii) CO2, -78 °C to rt, 2 h (-100 °C to
rt for 22). b (i) H2SO4, 60 °C, 1.5 h; (ii) NaN3, rt, 42 h. c m-CPBA, DCE,
70 °C, 2 h.
(21) (a) Roe, A. M.; Burton, R. A.; Reavill, D. R. Chem. Commun. 1965,
52. (b) Roe, A. M.; Burton, R. A.; Willey, G. L.; Baines, M. W.; Rasmussen,
A. C. J. Med. Chem. 1968, 11, 814–819.
(22) Hickey, M. R.; Allwein, S. P.; Nelson, T. D.; Kress, M. H.; Sudah,
O. S.; Moment, A. J.; Rodgers, S. D.; Kaba, M.; Fernandez, P. Org. Proc.
Res. DeV. 2005, 9, 764–767.
With the necessary coupling partners for 3, 4, and 5 as
well as related halogenated pyrroles in hand, Suzuki-Miyaura
cross-couplings of the various pyrrole pinacolboronate esters
and aryl bromides were carried out. We began with a series
of reactions of the unchlorinated pyrrole pinacolboronate
(23) While some lithium-halogen exchange does occur at this temper-
ature, the desired 2-bromo-6-iodobenzoic acid is the major product, formed
in a 5:1 ratio with the 3-bromobenzoic acid arising from initial lithium-
halogen exchange.
Org. Lett., Vol. 11, No. 5, 2009
1053

27,27 29,27 and 32,25 as well as monodechloroaminopyr-
rolnitrin 324 in good yields.
Scheme 5. Preparation of Brominated and Deuterated Pyrroles
It is interesting to note that the cross-coupling of 18 with
25 had to be carried out at 0 °C, since at 35 °C, a significant
amount of 2,6-dipyrrole substituted nitrobenzene is formed.
Thus, cross-coupling of both the aryl bromide and the aryl
chloride are possible under these conditions. This was not
observed in cross-couplings of 25 with the less reactive
dihalogenated aniline 17.
Since the cross-couplings of 25 with the various aryl
bromides in Table 2 proceeded as hoped, we next turned
our attention to cross-couplings of the chlorinated pyrrole
pinacolboronote ester 12. The results are also shown in Table
2. Again, both the cross-couplings and the fluoro-desilyla-
tions proceeded smoothly and in good yield, allowing access
to the arylpyrroles 34 and 36,28 as well as aminopyrrolnitrin
429 and pyrrolinitrin 5.
Since there is much interest26 in whether chlorinating
enzymes such as PrnA and PrnC can incorporate bromine
into their substrates in lieu of chlorine, we also sought means
to prepare brominated analogs of pyrrolnitrin in order to
study the substrate scope of PrnC. This first required the
synthesis of the brominated pyrrole pinacolboronate ester 40,
which we obtained in two steps from 7 as in Scheme 5. We
also used an analogous approach to prepare the deuterated
pyrrole 42 whose cross-coupling product with 18 could be
used in mechanistic (kinetic isotope effect) studies with PrnC.
Not surprisingly, cross-couplings of 42 to the halogenated
anilines and nitrobenzenes of Table 2 were good reactions,
providing the 3-deuteriopyrrole analogs of pyrrolnitrin in
essentially the same yields (e.g., 90% for 42 + 2-bromoa-
niline, see Supporting Information). However, the yields of
the 3-bromopyrrole analogs from cross-couplings employing
40 were much lower (e.g., 18% from 40 + 2-bromonitroben-
zene, see Supporting Information) due to its competing self-
reaction. Attempts to optimize the conditions to favor the
cross-coupling over the homocoupling are underway.
In summary, we have developed a straightforward syn-
thesis of the antifungal agent pyrrolnitrin, its biosynthetic
precursors and related halogenated phenylpyrroles, which we
are now using in kinetic and mechanistic studies of heter-
ologously overexpressed pyrrolnitrin biosynthetic enzymes.
Along the way, we have accessed 3-chloropyrroles by a
tandem lithium-halogen exchange/electrophilic chlorination
pathway, generalized a lithiation/carboxylation/Schmidt se-
quence for the preparation of 2,6-dihaloanilines (and 2,6-
dinitrobenzenes following oxidation) bearing bromides and
iodides, and determined optimal conditions for the cross-
coupling conditions of chlorinated aryl bromides and bor-
onated pyrroles.
(24) Compound 3 has been prepared biosynthetically from 7-chlorot-
ryptophan as in van Pee, K.-H.; Salcher, O.; Lingens, F. Angew. Chem.,
Intl. Ed. 1980, 19, 828–829.
(25) Compound 32 has been prepared by reductive dechlorination of 5
with iodotrimethylsilane. See: Sako, M.; Kihara, T.; Okada, K.; Ohtani,
Y.; Kawamoto, H. J. Org. Chem. 2001, 66, 3610–3612.
(26) van Pee, K.-H.; Dong, C. J.; Flecks, S.; Naismith, J.; Patallo, E. P.;
Wage, T. AdV. Appl. Micro. 2006, 59, 127–157.
Acknowledgment. This work was supported by grants
from NSERC, Queen’s University and the Canada Research
Chairs program. D.A.P. and M.D.M. thank the Government
of Ontario for an Early Researcher Award and Ontario
Graduate Scholarship, respectively.
(27) Compounds 27 and 29 have been synthesized with tritium at the
2-position of the pyrrole ring using methods similar to those described in
ref 8. See: Chang, C. J.; Floss, H. G.; Hook, D. J.; Mabe, J. A.; Manni,
P. E.; Martin, L. L.; Schroder, K.; Shieh, T. L. J. Antibiot. 1981, 34, 555–
566.
Supporting Information Available: Details of the prepa-
ration and characterization of all intermediate and final
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(28) Compound 36 has been prepared using methods similar to those
described in ref 8. See: UmioS. KariyoneK. TanakaK. Ueda I. MorimotoY.
Chem. Pharm. Bull. 196917588595.
(29) Compound 4 has been obtained by reduction of 5 as in: van Pee,
K.-H.; Salcher, O.; Fischer, P.; Bokel, M.; Lingens, F. J. Antibiot. 1983,
36, 1735–1742.
OL8026957
1054
Org. Lett., Vol. 11, No. 5, 2009