Role and substrate specificity of the Streptomyces coelicolor RedH enzyme in undecylprodiginine biosynthesis

Article abstract of DOI:10.1039/b801677a

The function of RedH from Streptomyces coelicolor as an enzyme that catalyses the condensation of 4-methoxy-2,2′-bipyrrole-5-carboxaldehyde (MBC) and 2-undecylpyrrole to form the natural product undecylprodiginine has been experimentally proven, and the s

Full text of DOI:10.1039/b801677a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Role and substrate specificity of the Streptomyces coelicolor RedH
enzyme in undecylprodiginine biosynthesisw
Stuart W. Haynes, Paulina K. Sydor, Anna E. Stanley, Lijiang Song and Gregory L. Challis*
Received (in Cambridge, UK) 30th January 2008, Accepted 3rd March 2008
First published as an Advance Article on the web 14th March 2008
DOI: 10.1039/b801677a
The function of RedH from Streptomyces coelicolor as an
enzyme that catalyses the condensation of 4-methoxy-2,20-bi-
pyrrole-5-carboxaldehyde (MBC) and 2-undecylpyrrole to form
the natural product undecylprodiginine has been experimentally
proven, and the substrate specificity of RedH has been probed in
vivo by examining its ability to condense chemically-synthesised
MBC analogues with 2-undecylpyrrole to afford undecylprodi-
ginine analogues.
esis of 1 and the elucidation of the substrate specificity of this
protein using a variety of chemically-synthesised analogues of
MBC 3.
Prodiginines are a group of red tripyrrole antibiotics, with
potent immunosuppressant,1 antimalarial2 and anti-cancer
activities,3 which are produced by several actinobacteria and
other eubacteria. The biological activities of prodiginines have
stimulated much recent interest in their chemical synthesis,
biosynthesis and mode of action.4 Recently it has been dis-
covered that some prodiginines bind to anti-apoptotic mem-
bers of the BCL-2 protein family,5 thus blocking the binding
of pro-apoptotic BH3-only proteins, resulting in induction of
apoptosis. These discoveries stimulated the development of
obatoclax (GX150-70), a synthetic prodiginine analogue,
which is currently in phase 1b and 2 clinical trials for treatment
of a variety of leukaemia, lymphoma and solid tumour
malignancies.6 A thorough understanding of prodiginine bio-
synthesis may facilitate manipulation of the pathway to
produce novel analogues with potent anti-cancer activity.
Streptomyces coelicolor A3(2) produces undecylprodiginine
17 and the cyclic derivative streptorubin B 2.8 A cluster of 23
genes in S. coelicolor (the red cluster) directs the biosynthesis
of 1 and 2 (Fig. 1).9 4-Methoxy-2,20-bipyrrole-5-carboxalde-
hyde (MBC) 3 and 2-undecylpyrrole 4 are intermediates in the
biosynthesis of 1 and 2.8,10 Most of the genes in the red cluster
required for the biosynthesis of 1 and 2 have been experimen-
tally identified8,10 and the enzymology of early steps in MBC
biosynthesis has been investigated.11,12 The product of the
redH gene has been proposed to catalyse the condensation of 3
and 4 to yield 1,13 but this hypothesis has not been investigated
experimentally. Here we report the first experimental investi-
gation of the role played by the RedH protein in the biosynth-
To examine the function of redH in prodiginine biosynth-
esis, we replaced it on the chromosome of S. coelicolor M511
with a ‘cassette’ containing oriT and the aac(3)IV gene, which
confers apramycin resistance, using PCR-targeting methodo-
logy to create S. coelicolor W33.14 When grown on agar media,
the W33 mutant produces no visible red pigment, indicating
that redH is required for prodiginine biosynthesis. LC-MS/MS
analysis of organic extracts of agar-grown mycelia of the W33
mutant revealed low levels of 1. Control experiments suggested
this was an artefact resulting from the extraction procedure.
Complementation of the redH:aac(3)IV mutation by integra-
tion of a plasmid (pPKS1) containing redH under the control
of the constitutive ermE* promoter into the chromosome of
S. coelicolor W33 restored wild type levels of prodiginine
production.w Accumulation of 4 but not 3 could be detected
in the W33 mutant by LC-MS/MS comparisons with chemi-
cally-synthesised8,15 authentic standards.w However, the tran-
sient accumulation of 3 in S. coelicolor W33 could be inferred
by the ability of this mutant to restore red pigment production
in previously reported10 MBC-deficient mutants of S. coelico-
lor by cross-feeding.w Unlike mutants of S. coelicolor blocked
in the biosynthesis of 3 or 4,8,10 where feeding of chemically-
synthesised MBC15 and 2-undecylpyrrole,8 respectively, re-
stored production of 1 and 2, feeding of synthetic 3 or 4 did
not restore prodiginine production in the W33 mutant. To
further investigate the role of RedH in the biosynthesis of 1,
we integrated pPKS1 (containing redH under the control of
the constitutive ermE* promoter) into the chromosome of
Streptomyces venezuelae ATCC10712. LC-MS/MS analysis of
Department of Chemistry, University of Warwick, Coventry, UK CV4
7AL. E-mail: g.l.challis@warwick.ac.uk; Fax: +44 024 7652 4112;
Tel: +44 024 7657 4024
w Electronic supplementary information (ESI) available: Experimental
procedures for the construction and genetic complementation of the S.
coelicolor W33 mutant, the construction of S. venezuelae/pPKS1, the
synthesis of 7a–d, 13 and 14a–d and the feeding experiments; LC-MS
data showing accumulation of 4 and 14a–d; cross-feeding data show-
ing accumulation of 3; spectroscopic data for 3, 5b, 6, 7a–d, 11–13 and
14a–d. See DOI: 10.1039/b801677a
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 1865–1867 | 1865

View Article Online
Fig. 1 Organisation of the red cluster that directs biosynthesis of 1 and 2 in S. coelicolor. Genes known to participate in assembly of MBC 3 are
shaded black; genes known to participate in the assembly of 2-undecylpyrrole 4 are shaded grey. The redH gene investigated in this study is
highlighted with stripes.
wild type S. venezuelae showed that it does not produce
prodiginines on agar media and analysis of the complete
genome sequence of this organism indicates that it does not
contain a similar gene cluster to the red cluster, nor does it
contain a gene with significant similarity to redH (M. J. Bibb,
personal communication). Feeding of MBC 3 and 2-undecyl-
pyrrole 4 to wild type S. venezuelae grown on agar media did
not result in visible red-pigment production. However LC-
MS/MS analysis of acidified organic extracts of the mycelia
from this experiment revealed trace quantities of 1, likely
resulting from the extraction procedure. On the other hand
feeding of MBC 3 and 2-undecylpyrrole 4 to S. venezuelae/
pPKS1 (redH+) grown on agar media resulted in visible red-
pigment production and LC-MS/MS analysis of acidified
organic extracts of the mycelia from this experiment revealed
substantial quantities of 1 (Fig. 2). Together these data
provide strong evidence that RedH catalyses condensation of
the known intermediates MBC 3 and 2-undecylpyrrole 4 to
form 1 in the S. coelicolor prodiginine biosynthetic pathway.
RedH is predicted to encode a protein with three functional
domains.9 These domains show sequence similarity to the
ATP-binding domains and the phosphotransfer domains,
respectively, of phosphoenol pyruvate synthase (PEPS) and
pyruvate-phosphate dikinase (PPDK).9 The third functional
domain of RedH does not show sequence similarity to any
proteins or protein domains of known function.9 By analogy
with the known catalytic mechanism of PEPS and PPDK, we
hypothesised that RedH catalyses the ATP-dependent phos-
phorylation of the aldehyde oxygen atom of MBC 3, thus
activating the carbonyl carbon atom towards nucleophilic
attack by C-5 of 2-undecylpyrrole 4 and assisting subsequent
loss of the aldehyde oxygen atom as phosphate. We thus
thought it would be interesting to probe the substrate speci-
ficity of RedH by examining the ability of our previously
constructed S. coelicolor W39 (redM::aac(3)IV) mutant
(which is blocked in the biosynthesis of 3) to catalyse con-
densation of chemically-synthesised MBC analogues with the
2-undecylpyrrole that accumulates in this mutant.
Analogues 7a–d of MBC, with other aromatic rings in place
of the monosubstituted pyrrole ring, were synthesised via
Suzuki coupling of the appropriate aryl boronic acid 5a–d
with vinyl bromide 6 (Scheme 1). Similar conditions to those
recently reported for the synthesis of MBC via Suzuki coupling
of N-BOC-pyrrole-2-boronic acid with 6 were used.15
Roseophilin 8, a natural product from Streptomyces griseo-
viridis,16 contains an identical conjugated structure to the
prodiginines, except that the B-ring is a furan instead of a
pyrrole (Scheme 2). Reasoning that roseophilin and the pro-
diginines could be biosynthesised via analogous pathways, we
also synthesised the analogue 13 of MBC with the trisubsti-
tuted pyrrole replaced by a furan, to examine whether this
compound could be utilised by the S. coelicolor W39
(redM::aac(3)IV) mutant to biosynthesise a roseophilin–
undecylprodiginine hybrid structure
9
(Scheme 2).
4-Methoxy-2(5H)-furanone 10 was converted to enamine 11
by reaction with excess dimethylformamide dimethyl acetal,
distilling methanol from the reaction as it was formed. Com-
pound 11 was reacted with phosphorus oxybromide to form
the corresponding bromo-enamine which was converted to the
bromo-aldehyde 12 by hydrolysis. This compound was un-
stable and decomposed to a black tar on storage. Thus it was
used immediately in a Suzuki reaction with N-BOC-pyrrole-2-
boronic acid which afforded the N-BOC derivative of MBC
analogue 13. The BOC group was removed from this
compound by base-promoted hydrolysis (Scheme 2).
Feeding of MBC analogues 7a–d to the S. coelicolor W39
(redM::aac(3)IV) mutant resulted in production of the corre-
sponding undecylprodiginine analogues 14a–d (Scheme 3).w
Rather than relying solely on UV–Vis spectroscopic and low
Fig. 2 LC-MS analysis of the production of
1 by wild type
S. venezuelae and S. venezuelae/pPKS1 fed with synthetic 3 and 4.
Extracted ion chromatograms (ElCs) at m/z = 222 (top trace) and m/z
= 394 (second from top trace) for extracts from S. venezuelae/pPKS1.
ElCs at m/z = 222 (second from bottom trace) and m/z = 394 (bottom
trace) for extracts from wild type S. venezuelae.
Scheme 1 Synthesis of MBC analogues with other aromatic rings in
place of the monosubstituted pyrrole.
ꢀc
This journal is The Royal Society of Chemistry 2008
1866 | Chem. Commun., 2008, 1865–1867

View Article Online
In conclusion, we have demonstrated that the S. coelicolor RedH
enzyme catalyses condensation of MBC 3 and 2-undecylpyrrole 4
in the biosynthesis of undecylprodigine 1. We have also shown that
RedH can catalyse condensation of a variety of synthetic MBC
analogues 7a–d with 2-undecylpyrrole 4 to form several undecyl-
prodiginine analogues 14a–d, albeit with varying efficiences.18 The
results of these experiments indicate that the nitrogen atom in the
A-ring of MBC is not required for catalysis, but that a heteroatom
capable of donating a lone-pair to the aromatic ring is important.
They also indicate that RedH does not impose a strong steric
constraint on the A-ring of MBC. While we suspect that the
nitrogen atom in the B-ring of MBC may play an important role
in the catalytic mechanism of RedH, the data obtained here do not
allow us to conclusively confirm or deny this hypothesis. Finally
these experiments demonstrate the feasibility of preparing prodigi-
nine analogues via a mutasynthesis approach. It will be particularly
interesting to see whether this mutasynthesis approach can be
extended to the preparation of analogues of streptorubin B 2,
which unlike analogues of undecylprodiginine 1 and prodigiosin
are not readily accessible by total synthesis strategies.
This research was supported by grants from the UK BBSRC
(grant ref. BBSSK200310147), the US National Institutes of Health
(GM077147) and the European Union (Contract No. 005224), as
well as a Warwick Postgraduate Research Fellowship (to PKS).
Scheme 2 Structures of roseophilin 8 and the undecylprodiginine–
roseophilin hybrid 9, and synthesis of MBC analogue 13.
Notes and references
resolution MS data to characterise these undecylprodiginine
analogues, which do not unambiguously confirm their struc-
tures, authentic standards of 14a–d were prepared by acid-
catalysed condensation of 7a–d with 2-undecylpyrrole17 and
shown to be identical by LC-MS/MS analyses to the undecyl-
prodiginine analogues produced in the feeding experiments. In
critically important control experiments designed to discrimi-
nate between enzymatic and non-enzymatic condensation
reactions, 7a–d were fed to the S. coelicolor W33 (redH::
aac(3)IV) mutant. LC-MS/MS analyses indicated that
14a–d were produced in all of these experiments, probably as
a result of the isolation procedure.w While the quantity of 14a
produced relative to the accumulated 2-undecylpyrrole 4 was
modestly greater when 7a was fed to the W33
(redH::aac(3)IV) mutant than when 7a was fed to the W39
(redM::aac(3)IV) mutant, the quantities of 14b–d produced
by the W39 mutant were very much higher than the quantities
produced by the W33 mutant (relative to accumulated 4),
indicating unequivocally that 7b–d are substrates of RedH.w In
contrast, feeding of MBC analogue 13 to the S. coelicolor W39
mutant did not result in production of roseophilin–undecyl-
prodiginine hybrid 9. This may be because 13 is not a substrate
of RedH or because it cannot enter S. coelicolor cells.
1 A. Mortellaro, S. Songia, P. Gnocchi, M. Ferrari, C. Fornasiero,
R. D’Alessio, A. Isetta, F. Colotta and J. Golay, J. Immunol., 1999,
162, 7102–7109.
2 N. N. Gerber, J. Antibiot., 1975, 28, 194–199; M. Isaka, A.
Jaturapat, J. Kramyu, M. Tanticharoen and Y. Thebtaranonth,
Antimicrob. Agents Chemother., 2002, 46, 1112–1113.
3 B. Montaner and R. Perez-Tomas, Curr. Cancer Drug Targets,
2003, 3, 57–65.
4 A. Furtsner, Angew. Chem., Int. Ed., 2003, 42, 3582–3603.
¨
5 G. C. Shore, J. Bajorath, F. L. Stahura and M. S. R. Murthy,
International patent application WO 2003015788, 2003.
6 M. Nguyen, R. C. Marcellus, A. Roulston, M. Watson, L. Serfass,
S. R. M. Madiraju, D. Goulet, J. Viallet, L. Belec, X. Billot, S.
Acoca, E. Purisima, A. Wiegmans, L. Cluse, R. W. Johnstone, P.
Beauparlant and G. C. Shore, Proc. Natl. Acad. Sci. U. S. A., 2007,
104, 19512–19517.
7 S. W. Tsao, B. A. Rudd, X. G. He, C. J. Chang and H. G. Floss, J.
Antibiot., 1985, 38, 128–131.
8 S. J. Mo, P. K. Sydor, C. Corre, M. M. Alhamadsheh, A. E.
Stanley, S. W. Haynes, L. Song, K. A. Reynolds and G. L. Challis,
Chem. Biol., 2008, 15, 137–148.
9 A. M. Cerdeno, M. J. Bibb and G. L. Challis, Chem. Biol., 2001, 8,
211–224.
10 A. E. Stanley, L. J. Walton, M. Kourdi-Zerikly, C. Corre and G.
L. Challis, Chem. Commun., 2006, 3981–3983.
11 M. G. Thomas, M. D. Burkart and C. T. Walsh, Chem. Biol., 2002,
9, 171–184.
12 S. Garneau-Tsodikova, P. C. Dorrestein, N. L. Kelleher and C. T.
Walsh, J. Am. Chem. Soc., 2006, 128, 12600–12601.
13 N. R. Williamson, H. T. Simonsen, R. A. A. Ahmed, G. Goldet, H.
Slater, L. Woodley, F. J. Leeper and G. P. C. Salmond, Mol.
Microbiol., 2005, 56, 971–989.
14 B. Gust, G. L. Challis, K. Fowler, T. Kieser and K. F. Chater,
Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 1541–1546.
´
15 K. Dairi, S. Tripathy, G. Attardo and J.-F. Lavallee, Tetrahedron
Lett., 2006, 47, 2605–2606.
16 Y. Kayakawa, K. Kawakami, H. Seto and K. Furihata, Tetrahe-
dron Lett., 1992, 33, 2701–2704.
17 H. H. Wasserman, G. C. Rodgers and D. D. Keith, Tetrahedron,
1976, 32, 1851–1854.
Scheme 3 Structures of undecylprodiginine analogues produced
by feeding the corresponding synthetic MBC analogues to the
S. coelicolor W39 mutant.
18 These results were first presented on 20 June 2007 at the Chemistry
and Biology of Natural Products Biosynthesis III, University of
Bristol, UK.
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 1865–1867 | 1867