Facile synthesis of cyclohexanediones and dialkylresorcinols - Bioactive natural products from entomopathogenic bacteria

Article abstract of DOI:10.1002/ejoc.201403346

A synthesis for the recently identified, widespread bacterial natural product classes of dialkylresorcinols and cyclohexanediones was developed. The synthesis route is similar to the biosynthesis route in that the formation of the cyclohexanedione ring results from two parts, as exemplified by the synthesis of the multifunctional isopropylstilbenes identified in Photorhabdus luminescens. Testing of these compounds revealed good bioactivity against Trypanosoma brucei rhodesiense and T. cruzi, the causative agents of sleeping sickness and Chagas disease, respectively.

Full text of DOI:10.1002/ejoc.201403346

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201403346
Facile Synthesis of Cyclohexanediones and Dialkylresorcinols – Bioactive
Natural Products from Entomopathogenic Bacteria
Max Kronenwerth,[a] Christina Dauth,[a] Marcel Kaiser,[b] Iain Pemberton,[c] and
Helge B. Bode*[a,d]
Keywords: Natural products / Biosynthesis / Biological activity / Resorcinols / Cyclohexanedione
A synthesis for the recently identified, widespread bacterial
natural product classes of dialkylresorcinols and cyclohexa-
nediones was developed. The synthesis route is similar to the
biosynthesis route in that the formation of the cyclohexane-
dione ring results from two parts, as exemplified by the syn-
thesis of the multifunctional isopropylstilbenes identified in
Photorhabdus luminescens. Testing of these compounds
revealed good bioactivity against Trypanosoma brucei
rhodesiense and T. cruzi, the causative agents of sleeping
sickness and Chagas disease, respectively.
to be an interesting class of natural products that have been
overlooked in the past. Moreover, the actual identification
of these molecules is simple by using modern MS analysis[7]
owing to the characteristic fragmentation pattern of DARs,
but the actual isolation of such compounds might be diffi-
cult because of low production titers. Additionally, keto–
enol tautomerism of CHDs might complicate analysis by
NMR spectroscopy, and thus, derivatization is required.[7]
To overcome these challenges, facile synthetic access to
these two compound classes was developed and applied to
1 and reduced derivative 2, as well as CHD equivalents 3
and 4, all of which have been identified in different strains
of Photorhabdus.[7]
Introduction
Isopropylstilbene (1) was described as the first stilbene
compound not derived from plants but from bacteria.[1] It
is produced by all strains of Photorhabdus, which are
entomopathogenic bacteria that live in symbiosis with
nematodes of the genus Heterorhabditis. It is crucial for
nematode development,[2] shows insecticidal and antibiotic
activity,[3] and is also an inhibitor of the human soluble
epoxide hydrolase (sEH).[4] Owing to its multiple activities,
it has also been called the “Swiss army knife” of P. lumi-
nescens.[5] The biosynthesis of 1[2] and similar dialkylresor-
cinols[6] was originally postulated as a condensation of two
β-keto esters.[2,6] However, recently 2,5-dialkylcyclohexane-
1,3-diones (CHDs) were shown to be true biosynthesis in-
termediates resulting from condensation (Scheme 1) of a β-
keto acyl and an α,β-unsaturated acyl precursor (catalyzed
by the ketosynthase DarB) as the key biosynthesis step,[7]
as also proposed for the chiloglottone biosynthesis of or-
chids.[8] DarA was found to be the responsible enzyme to
aromatize the CHDs to 2,5-dialkylresorcinols (DARs) such
as 1. Given that 89 homologues for DarAB have been found
in other bacteria including several pathogens to humans or
animals such as Neisseria strains,[7] CHDs and DARs seem
Results and Discussion
[a] Merck Stiftungsprofessur für Molekulare Biotechnologie,
Fachbereich Biowissenschaften, Goethe Universität Frankfurt,
60438 Frankfurt am Main, Germany
The presented synthesis (Scheme 2) was “inspired” by the
biosynthesis, as it uses the same reaction types and provides
both CHDs and DARs; however, the precursors that are
condensed differ from the natural ones (Scheme 1). In the
first step, α,β-unsaturated ketones 5 and 6 were obtained
from Wittig olefination with an in situ formed ylide; meth-
yltriphenylphosphonium bromide was deprotonated with n-
butyllithium and treated with isovaleryl chloride. The ylide
underwent Wittig reaction with cinnamaldehyde (7) and
hydrocinnamaldehyde (8) to give 5 and 6.[9] The crucial step
E-mail: h.bode@bio.uni-frankfurt.de
http://www.bio.uni-frankfurt.de/48050101/ak-bode
[b] Swiss Tropical and Public Health Institute, Parasite
Chemotherapy,
Socinstrasse 57, P. O. Box, 4002 Basel, Switzerland
[c] Photeomix Protein Discovery, IP Research Consulting SAS,
34 Rue Carnot, 93160 Noisy le Grand, France
[d] Buchmann Institute for Molecular Life Sciences (BMLS), Goe-
the Universität Frankfurt,
60438 Frankfurt am Main, Germany
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201403346.
8026
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 8026–8028

Facile Synthesis of Cyclohexanediones and Dialkylresorcinols
Scheme 1. Comparison between biosynthesis in P. luminescens and the chemical synthesis of 1. ACP = acyl carrier protein.
in the synthesis (and also in the biosynthesis) is the forma- yields of the CHD forming steps varied with the substrate.
tion of the six-membered ring with the isopropyl group at Whereas good yields were achieved for nonconjugated
the 2-position of the final ring. In a one-pot reaction, this system 6, only low amounts were yielded if the ketone was
ring can be produced from α,β-unsaturated ketones 5 and conjugated to the phenyl ring in 5. The fact that the conju-
6 with diethyl malonate. Michael addition generated the gated system in 5 is less suitable for the initial Michael ad-
starting material (not shown) for the cyclizing Dieckmann dition than α,β-unsaturated ketone 6 explains the variations
condensation step. Saponification of resulting esters 9 and in the yields.
10 (not isolated) followed by decarboxylation of the β-keto
An alternative route to the DARs was described in a
acid[10] resulted in CHD derivatives 3 and 4. Aromatization patent:[12] After permethylation, a dihydroxylated benzoic
of diketones 3 and 4 was performed with mercury(II) acet- acid or a brominated derivative was treated with nBuLi to
ate in acetic acid to generate 1 and 2, respectively.[11] The enable nucleophilic attack onto isopropyl bromide to intro-
duce the isopropyl group in between the hydroxy groups.
Subsequent reduction of the carboxylic acid group to the
aldehyde gave the starting material for the reaction with a
benzyl phosphonate to form isopropylstilbene after a final
deprotection step. However, this route did not work well in
our hands, probably because of steric effects of the substi-
tuted benzoic acid. Alternative routes to CHD derivatives
have also been described: Schiestl et al. showed a similar
synthesis of a CHD in 2003 but reported lower yields for a
comparable substrate.[13] Furthermore, Poldy et al. de-
scribed two other routes by using either Birch reduction to
produce the CHD with linear side chains from a dihydroxy
benzoic acid[14] or an organocadmium-mediated variant
starting from glutaric anhydride to yield a CHD with an
isopropyl group.[15]
The synthesis made compounds 1–4 available for ad-
ditional bioactivity testing, and therefore, the compounds
were tested against targets other than those described in the
literature for these compounds. Given that Photorhabdus
spp. have to compete with other soil living organisms such
Scheme 2. Synthetic route to natural compounds 1–4. Position of
the double bond is marked with a dashed line. The first compound
number refers to the structure with the double bond structure, the
second one to the single bond compound.
Eur. J. Org. Chem. 2014, 8026–8028
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
8027

M. Kronenwerth, C. Dauth, M. Kaiser, I. Pemberton, H. B. Bode
SHORT COMMUNICATION
as amoeba, clinical relevant protozoa might be a potential
target. Thus, bioactivity tests against protozoa causing
tropical diseases[16] were performed, and they showed, in
general, that DARs were more active than CHDs. The high-
est activity was observed for 2 against Trypanosoma brucei
Acknowledgments
This work was funded by the Deutsche Forschungsgemeinschaft
(DFG) and in part by the European Community’s Seventh Frame-
work Program (FP7/2007-2013) under grant agreement no. 602773.
rhodesiense (the causative agent of sleeping sickness) and The authors thank Dr. Nina Socorro Cortina and Prof. Dr. Martin
Grininger for HRMS (ESI) measurements, Friederike I. Nollmann
and Dr. Hélène Adihou for fruitful discussions, and Amale El Ar-
oussi for the purification of Li SIR2.
T. cruzi (the causative agent of Chagas disease) with IC50
values of 0.35 and 9.40 μm, respectively. Against malaria-
causing Plasmodium falciparum and Leishmaniasis-causing
Leishmania donovani, 1 was the most effective compound
with IC50 values of 7.67 and 3.77 μm, respectively. All bio-
activities are listed in Table S1 (Supporting Information).
Compound 1 was also tested against different NAD-de-
pendent protein acetylases, as the plant stilbene resveratrol
may serve as either a sirtuin activator or inhibitor in a sub-
strate-dependent manner.[17] Resveratrol inhibits deacetyl-
ation of the p53-acetylated peptide substrate TSPQPKK-
Ac by human SIRT2 and several kinetoplastid parasite
[1]
[2]
K. Hu, J. Li, B. Li, J. M. Webster, G. Chen, Bioorg. Med. Chem.
2006, 14, 4677–4681.
S. A. Joyce, A. O. Brachmann, I. Glazer, L. Lango, G. Schwär,
D. J. Clarke, H. B. Bode, Angew. Chem. Int. Ed. 2008, 47, 1942–
1945; Angew. Chem. 2008, 120, 1968–1971 .
[3]
I. Eleftherianos, S. Boundy, S. A. Joyce, S. Aslam, J. W. Mar-
shall, R. J. Cox, T. J. Simpson, D. J. Clarke, R. H. ffrench-Con-
stant, S. E. Reynolds, Proc. Natl. Acad. Sci. USA 2007, 104,
2419–2424.
SIR2 proteins (Pemberton, unpublished data). The concen- [4]
tration of 1 required to inhibit SIR2rp1 derived from Leish-
mania infantum (1, IC50 = 250 μm, 95%, CI = 198–315 μm)
E. Buscató, D. Büttner, A. Brüggerhoff, F. M. Klingler, J. We-
ber, B. Scholz, A. Zivkovic, R. Marschalek, H. Stark, D. Ste-
inhilber, H. B. Bode, E. Proschak, ChemMedChem 2013, 8,
919–923.
H. B. Bode, Curr. Opin. Chem. Biol. 2009, 13, 224–230.
B. Nowak-Thompson, P. E. Hammer, D. S. Hill, J. Stafford, N.
Torkewitz, T. D. Gaffney, S. T. Lam, I. Molnar, J. M. Ligon, J.
Bacteriol. 2003, 185, 860–869.
S. W. Fuchs, K. A. J. Bozhüyük, D. Kresovic, F. Grundmann,
V. Dill, A. O. Brachmann, N. R. Waterfield, H. B. Bode, An-
gew. Chem. Int. Ed. 2013, 52, 4108–4112; Angew. Chem. 2013,
125, 4202–4206.
S. Franke, F. Ibarra, C. M. Schulz, R. Twele, R. A. Barrow, R.
Peakall, F. P. Schiestl, W. Francke, Proc. Natl. Acad. Sci. USA
2009, 106, 8877–8882.
was similar to that observed for resveratrol (IC50 = 241 μm,
[5]
95%, CI = 217–268 μm). Whereas in both cases the inhibi-
tion was relatively weak, the small size of the mo-
lecules yielded a high binding efficiency index [for 1,
–log(IC50)/MW(kDa) = 15][18] that potentially could be op-
timized by fragment-based drug design. However, whereas
resveratrol is soluble to approximately 0.5 mm in aqueous
solution, compound 1 was found to be somewhat less solu-
ble under similar conditions (0.2 mm) and elicited a steeper
inhibition slope than that observed for resveratrol (Fig-
ure S2). Inhibition of LiSIR2rp1 by 1 may be attributed to
the formation of nonspecific compound aggregates coincid-
ing with its solubility limit, a common problem in small-
compound libraries.[19,20] Interestingly, comparatively poor
inhibition of the T. brucei SIR2RP1 (IC50 ≈ 1 mm) and
human sirtuin-2 (SIRT 2, IC50 Ͼ 1 mm; data not shown)
suggests that inhibition of kinetoplastid SIR2rp1 proteins
by 1 is not by a single promiscuous mechanism related to
its solubility limit.[21] Attempts to use 1 and other resvera-
trol-like stilbene derivatives as starting points in drug design
will inevitably require considerable attention towards the
inherent hydrophobicity of the compounds and the means
to improve solubility.
[6]
[7]
[8]
[9]
[10]
A. Zörb, R. Brückner, Eur. J. Org. Chem. 2010, 4785–4801.
D. Joseph, M. J. Queiroz, G. Kirsch, Org. Prep. Proced. Int.
1995, 27, 499–500.
J. E. Oliver, K. R. Wilzer, R. M. Waters, Synthesis 1990, 1117–
1119.
[11]
[12] J. M. Webster, J. Li, K. Hu, J. Zhu, patent international publi-
cation number WO 01–42231 A2, 2001.
[13] F. P. Schiestl, R. Peakall, J. G. Mant, F. Ibarra, C. Schulz, S.
Franke, W. Francke, Science 2003, 302, 437–438.
[14] J. Poldy, R. Peakall, R. A. Barrow, Tetrahedron Lett. 2008, 49,
2446–2449.
[15] J. Poldy, R. Peakall, R. A. Barrow, Org. Biomol. Chem. 2009,
7, 4296–4300.
[16] N. Feasey, M. Wansbrough-Jones, D. C. W. Mabey, A. W. Solo-
mon, Br. Med. Bull. 2010, 93, 179–200.
[17] K. T. Howitz, K. J. Bitterman, H. Y. Cohen, D. W. Lamming,
S. Lavu, J. G. Wood, R. E. Zipkin, P. Chung, A. Kisielewski,
L. L. Zhang, B. Scherer, D. A. Sinclair, Nature 2003, 425, 191–
196.
Conclusions
[18] C. Abad-Zapatero, J. T. Metz, Drug Discovery Today 2005, 10,
464–469.
In summary, our simple and general synthesis will allow
the future generation of more derivatives that can then be
tested in additional assays, which will uncover new bio-
activities of this already very promising and potent class of
small molecules.
[19] B. Vergnes, D. Sereno, J. Tavares, A. Cordeiro-da-Silva, L.
Vanhille, N. Madjidian-Sereno, D. Depoix, A. Monte-Alegre,
A. Ouaissi, Gene 2005, 363, 85–96.
[20] M. Lakshminarasimhan, D. Rauh, M. Schutkowski, C. Steeg-
born, Aging (NY) 2013, 5, 151–154.
[21] S. C. Owen, A. K. Doak, P. Wassam, M. S. Soichet, B. K. So-
ichet, ACS Chem. Biol. 2012, 7, 1429–1435.
Received: October 14, 2014
Supporting Information (see footnote on the first page of this arti-
cle): Experimental conditions and NMR spectroscopy data, includ-
ing assignment tables and spectra.
Published Online: November 13, 2014
8028
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 8026–8028