Design, synthesis and biological evaluation of simplified analogues of the RNA polymerase inhibitor etnangien

Article abstract of DOI:10.1016/j.bmcl.2009.12.066

Novel simplified analogues of the potent RNA polymerase inhibitor etnangien were obtained by total synthesis and evaluated for antibacterial activity against Gram-positive bacteria and one Gram-negative bacterium.

Full text of DOI:10.1016/j.bmcl.2009.12.066

Bioorganic & Medicinal Chemistry Letters 20 (2010) 939–941
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis and biological evaluation of simplified analogues
of the RNA polymerase inhibitor etnangien
a
b
Dirk Menche a, , Pengfei Li , Herbert Irschik
*
a Institut für Organische Chemie, Ruprecht-Karls-Universität Heidelberg, INF 270, D-69120 Heidelberg, Germany
b Helmholtz-Zentrum für Infektionsforschung, Research Group Microbial Drugs, Inhoffenstr. 7, D-38124 Braunschweig, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Novel simplified analogues of the potent RNA polymerase inhibitor etnangien were obtained by total syn-
thesis and evaluated for antibacterial activity against Gram-positive bacteria and one Gram-negative
bacterium.
Received 13 November 2009
Revised 15 December 2009
Accepted 15 December 2009
Available online 23 December 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Etnangien
Analogues
RNA-polymerase
Structure–activity data
Antibiotics
Various strains of the myxobacterium Sorangium cellulosum are
the natural sources of etnangien (1, Fig. 1), a poly-unsaturated 22-
membered polyketide macrolide.1–3 It presents a highly potent
antibiotic, which inhibits the growth of Gram-positive bacteria
with average IC50 values in the submicromolar range and shows
only low cytotoxicity against eukaryotic cell lines.1,2 At the molec-
ular level, it inhibits bacterial RNA polymerase in various strains,
an attractive drug target for antibiotic development.4–6 So far, the
rifamycins are the only class of clinically used RNA polymerase
inhibitors. However, resistance has been increasingly emerging7
necessitating the need of structurally novel RNA polymerase inhib-
itors. Importantly, etnangien shows no cross-resistance to rifampi-
cin, and also retains a certain activity against retroviral DNA
polymerase which adds to its attractiveness for further develop-
ment.1,2 Preclinical advancement, however, is severely hampered
by its notorious instability, which is intrinsically associated with
the polyene side chain. This renders the development of stable
and more readily available analogues an important research goal.
So far, only one derivative of etnangien has been reported, the
methyl ester 2, which has been shown to retain the biological po-
tential of the parent natural product.8 The relative and absolute
configuration of etnangien has been determined in our group in
cooperation with Rolf Müller by extensive high-field NMR-studies,
modelling, chemical derivatization and an innovative bioinformat-
ics approach.8 This assignment was recently confirmed by the first
total synthesis of these macrolide antibiotics, accomplished in our
group.9,10
Herein we describe the synthesis of a first set of carefully se-
lected simplified etnangien analogues, which lack the labile poly-
ene side chain. The antibacterial activity of these compounds
against a range of Gram-positive and -negative bacteria relative
to etnangien (1) and etnangien methyl ester (2) is reported.
HO
38
OH
O
36
OH
O
OH
RO
O
OH
OMe
OH
1
2
: Etnangien (R = H)
: Etnangien methylester
(R = Me)
=
36
38
solution conformation
(MMFS calculations)
* Corresponding author. Tel.: +49 6221 546207; fax: +49 6221 544205.
E-mail address: dirk.menche@oci.uni-heidelberg.de (D. Menche).
Figure 1. Etnangien and its methyl ester: potent macrolide antibiotics.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.12.066

940
D. Menche et al. / Bioorg. Med. Chem. Lett. 20 (2010) 939–941
Our starting point for analogue studies were conformational
between OH-36 and -38 was correctly predicted, as shown for
etnangien in Figure 1. These findings together with lability of the
polyene fragment prompted us to evaluate the importance of this
polyene portion on biological activity, by preparing and evaluating
truncated macrocyclic analogues.
analyses (MMFS, MacroModel 8.5) of etnangien and related struc-
tures with simplified side chains, which were carried out both in
vacuo and in solution (water).11 These calculations revealed that
the side chain only had a minor impact on the respective 3D-struc-
tures. Importantly, in all cases the characteristic hydrogen bond
As a first target, we decided to remove the side chain com-
pletely and target truncated analogue 13. As shown in Scheme 1,
the synthesis was based an Ipc-boron mediated aldol reaction of
Roche ester derived methyl ketone 4 with aldehyde 3,12 which pro-
ceeded with excellent yield and selectivity. Subsequent 1,3-anti
reduction13 of the derived b-hydroxy ketone 5 and step-wise selec-
tive protection of the less hindered hydroxyl at C-24 as TBS ether
and the more congested 22-OH as TES ether gave rise to 6. Homol-
ogation to the southern ring fragment 7 was effected by removal of
the benzoate under basic conditions, oxidation of the primary
alcohol, Wittig–Stork reaction of the derived aldehyde and selec-
tive removal of the TES ether (49% yield, 4 steps). Union with our
previously reported poly-propionate fragment 810 relied on a
Yamaguchi esterification, followed by a Heck-macrocyclization of
derived ester 9, which proceeded with high yield and stereoselec-
tivity. Final deprotection proved challenging, due to a pronounced
tendency of lactone 10 towards translactonizations under basic or
acidic conditions. Various degrees of transesterification by attack
of OH-24, as well as OH-38 leading to irreversible d-lactone
formation, were observed. Finally, the desired compound 13 was
obtained in reliable fashion by a two step-procedure. This involved
first removing all TBS-groups with TBAF/AcOH under neutral con-
ditions, giving 11 and 12 as the main products. Secondly, removal
of the acetonide was achieved with aq. AcOH, giving the desired
macrolactone 13 as well as the contracted congener 14 starting
from 11 and 12, respectively, with acceptable yields.14
OBz OMe
O
OPMB
O
a
+
H
3
4
OBz OMe
OH
O
OPMB
b, c, d
5
TBS TES
O
OBz OMe
O
OPMB
e, f, g, h
TBS
24
22
6
TBS
O
I
O
OMe
O
OH OPMB
O
O
+
OH
7
8
i
O
OTBS
38
O
OTBS
O
O
O
O
j
O
O
I
OPMB
OPMB
OPMB
24
OPMB
k
TBS
O
OMe
OTBS
OMe
OTBS
10
9
O
O
O
OH
+
HO
8
I
OTBS
O
OH
O
O
OH
O
O
TBS
O
O
OMe
OTBS
(ref. 10)
O
OPMB
O
15
OH
OMe
OH
OMe
O
OTBS
O
11
12
O
l
m
O
OTBS
HO
OH
O
HO
OH
O
O
OH
OH
TBS
OMe
OTBS
O
OPMB
O
16
a
OH
OMe
OH
OMe
14
O
OH
O
13
O
Scheme 1. Synthesis of simplified analogues 13 and 14. Reagents and conditions:
(a) (+)-Ipc2BCl,TEA, Et2O, À78 °C 4.5 h then À20 °C 14 h, 75%, dr = 19:1; (b)
Me4NBH(OAc)3, CH3CN/AcOH, À40 °C, 99%, dr > 20:1; (c) TBSCl, imidazole, DMAP,
DMF, 0 °C to rt, 2 h, 47%, 79% brsm; (d) TESOTf, 2,6-lutidine, DCM, À78 °C, 30 min;
(e) K2CO3, 95% MeOH, rt, 3 h, 77% over 2 steps; (f) Dess–Martin periodinane,
NaHCO3, DCM, rt, 40 min, 81%; (g) [Ph3PCH2I]+IÀ, NaHMDS, THF/HMPA, 20 °C, 1 min
then À78 °C, aldehyde, 20 min, 83%, Z/E = 19:1; (h) PPTS, EtOH, rt, 1 h, 85%; (i) TEA,
DMAP, 2,4,6-trichlorobenzoyl chloride, toluene, rt, 30 min; (j) Pd(OAc)2, Bu4NCl,
K2CO3, DMF, 60 °C, 90 min, 47%; (k) 2.0 M TBAF and 0.2 M AcOH in DMF, 0 °C, 2 h,
35% for 11 and 53% for 12; (l) 65% AcOH, rt, 2 h, 41%; (m) 65% AcOH, 0 °C, 2 h, 44%.
O
OH
OH
OMe
OH
17
Scheme 2. Synthesis of analogue 17. Reagents and conditions: (a) 2.0 M TBAF and
0.2 M AcOH in DMF, 0 °C to rt, 160 min.

D. Menche et al. / Bioorg. Med. Chem. Lett. 20 (2010) 939–941
941
Table 1
observed. Truncation of the side chain leads to significant loss of
activity, which suggests this must be part of the pharmacophore
region and the complete loss of activity for a contracted macrocy-
clic analogue (14) indicates a significance of the authentic core for
biological potency. In total, these results will be helpful in design-
ing further SAR-studies and the development of potent but more
stable as well as simplified analogues, to further advance the
development of these macrolide antibiotics.
Antimicrobial activity of simplified etnangien analogues (13, 14, 17) in comparison to
etnangien (1) and its methyl ester (2)
Test organism
MICa
(lg/mL)
1
2
13
14
17
Staph. aureus
Micrococcus luteus
Escherichia coli
Corynebact. glutamicum
Mycobact. phlei
Saccharomyces cerevisiae
1
2.5
>20
>20
>20
10
20
>20
>20
>20
>20
20
>20
>20
>20
>20
>20
10
20
>20
0.06
>80
0.03
0.12
>80
0.39
>20
0.24
n.d.
>40
Acknowledgements
a
Experiments were run in duplicates or triplicates.
Financial support from the DFG and the Wild-Stiftung is most
gratefully acknowledged.
Our recently reported route10 to 16 from acid 8 and vinylic io-
References and notes
dide 15 opened the possibility of also elaborating a more closely
related analogue (see Scheme 2). Selective deprotection proved
again challenging, even more so due to the lability of the liberated
terminal homoallylic alcohol, which was prone to various decom-
position pathways. As before, TBS-deprotection could only be
effectuated in a practical and reliable fashion by use of buffered
TBAF solution, giving the desired analogue with acceptable
yields.15
1. Höfle, G.; Reichenbach, H.; Irschik, H.; Schummer, D. German Patent DE 196 30
980 A1: 1-7 (5.2.1998).
2. Irschik, H.; Schummer, D.; Höfle, G.; Reichenbach, H.; Steinmetz, H.; Jansen, R. J.
Nat. Prod. 2007, 70, 1060.
3. For a review on myxobacterial polyketides, see: Menche, D. Nat. Prod. Rep.
2008, 25, 905.
4. O’Neill, A.; Oliva, B.; Storey, C.; Hoyle, A.; Fishwick, C.; Chopra, I. Antimicrob.
Agents Chemother. 2000, 44, 3163.
5. Haebich, D. V.; Nussbaum, F. Angew. Chem., Int. Ed. 2009, 48, 3397.
6. Campbell, E. A.; Pavlova, O.; Zenkin, N.; Leon, F.; Irschik, H.; Jansen, R. EMBO
2005, 24, 674.
The potent antibiotic activity of etnangien based on RNA-poly-
merase inhibition prompted us to likewise analyze the foregoing
analogues for their antimicrobial potential. Table 1 summarizes
their inhibitory activities against different microorganisms, in di-
rect comparison to etnangien (1) and its methyl ester (2). As ex-
pected, bacteria belonging to the Corynebacterineae, such as
Nocardia corallina and some Mycobacteria, were particularly sensi-
tive to 1 and 2, while yeast and Gram-negative Escherichia coli
proved to be rather resistant. Possibly, this may be correlated with
the Gram-negative nature of the producing myxobacterium.1,2 In
agreement with these data, the novel analogues showed likewise
no antimicrobial activity against Gram-negative E. coli and the
yeast Saccharomyces cerevisiae. However, a slight activity against
Corynebact. glutamicum and Mycobact. phlei. was observed for 13
and 17. In both cases, the inhibition was 40–300-fold smaller as
compared to 1 and 2, which suggests that the side side-chain is
part of the pharmacophoric region. No activity was observed for
analogue 14 bearing a contracted macrocycle, indicating that the
authentic ring size also is of importance for biological potency.
To further evaluate these data, the stability of etnangien (1) and
its methyl ester (2) was studied. Firstly, the stability of the com-
pounds under various pH-values in solution was evaluated. While
considerable degrees of decomposition were observed for 1 and 2
at very low (pH >3) and high pH-values (pH >10), both compounds
were stable under netural conditions and can be conveniently
stored at pH 7 in dark vessels in solution. Furthermore, the integ-
rity of etnangien methyl ester was evaluated under assay-type
conditions, by HPLC–MS analysis, revealing no signs of conversion
to the parent natural product, demonstrating the observed potency
of 2 was not caused by intermediate conversion to the parent nat-
ural product by ester cleavage. Furthermore, only very low degrees
of isomerisation of 2 were detected (<5%) even after prolonged
times (48 h). These results demonstrate that it is indeed possible
to modify and stabilize the etnangien structure, yet still retain
activity.
7. Parenti, F.; Lancini, G. In Antibiotic and chemotherapy; O’Grady, F., Lambert, H.
P., Finch, R. G., Greenwood, D., Eds.; Churchill Livingstone: New York, 1997; pp
453–459.
8. Menche, D.; Arikan, F.; Perlova, O.; Horstmann, N.; Ahlbrecht, W.; Wenzel, S. C.;
Jansen, R.; Irschik, H.; Müller, R. J. Am. Chem. Soc. 2008, 130, 14234.
9. For synthetic studies, see: (a) Arikan, F.; Li, J.; Menche, D. Org. Lett. 2008, 10,
3521; (b) Li, J.; Li, P.; Menche, D. Synlett 2009, 2417; (c) Li, J.; Menche, D.
Synthesis 2009, 11, 1904.
10. Total synthesis: Li, P.; Li, J.; Arikan, F.; Ahlbrecht, W.; Dieckmann, M.; Menche,
D. J. Am. Chem. Soc. 2009, 131, 11678.
11. 10,000 step conformational searches were carried out with the generalized
Born/surface area (CB/SA) solvent model and solution conformation data 18 as
input geometries: (a) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Kiskamp,
R.; Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J. Comput.
Chem. 1990, 11, 440; (b) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson J.
Am. Chem. Soc. 1990, 112, 6127.
12. Aldehyde 3 was obtained in four steps from a known alkene, compound 18 in
Ref. 10, by ozonolysis with reductive work-up, Bz-protection, TBS-deprotection
and oxidation to the aldehyde.
13. Evans, D. A.; Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc. 1988, 110, 3560.
14. All new compounds had spectroscopic data in full support of the assigned
structures. 1H NMR spectroscopic data:
Compound 13: 1H NMR (600 MHz, acetone-d6) d = ppm 0.97 (d, J = 7.0 Hz, 3H),
0.97 (d, J = 6.8 Hz, 3H), 1.00 (d, J = 6.8 Hz, 3H), 1.03 (d, J = 6.6 Hz, 3H), 1.40–1.60
(m, 6H), 1.64 (m, 3H), 1.83 (m, 1H), 1.87 (m, 3H), 2.25 (m, 1H), 2.31 (m, 1H),
2.36 (m, 2H), 2.46 (m, 1H), 3.23 (m, 1H), 3.28 (s, 3H), 3.32 (dd, J = 9.3, 6.2 Hz,
1H), 3.43 (dd, J = 9.2, 5.6 Hz, 1H), 3.44 (m, 1H), 3.52 (m, 1H), 3.60 (m, 2H), 3.78
(s, 3H), 4.16 (m, 1H), 4.40 (s, 2H), 5.23 (m, 1H), 5.40 (ddd, J = 10.4, 7.9, 7.9 Hz,
1H), 5.77 (m, 1H), 6.08 (pseudo-t, J = 11.5 Hz, 1H), 6.40 (dd, J = 14.3, 10.6 Hz,
1H), 6.90 (dd, J = 8.6, 2.4 Hz, 2H), 7.27 (dd, J = 8.8, 2.2 Hz, 2H); ESI-HRMS calcd
for C36H58NaO9 [M+Ma]+: 657.3979, found: 657.3984. Compound 14: 1H NMR
(600 MHz, acetone-d6) d = 0.81 (d, J = 7.0 Hz, 3H), 0.91 (d, J = 7.0 Hz, 3H), 0.98
(d, J = 6.9 Hz, 3H), 1.10 (d, J = 6.8 Hz, 3H), 1.47–1.64 (m, 7H), 1.74 (m, 2H), 1.83
(m, 2H), 2.13 (m, 1H), 2.19 (m, 2H), 2.34 (dd, J = 13.9, 10.8 Hz, 1H), 2.39 (m, 1H),
2.47 (m, 1H), 2.48 (dd, J = 13.9, 2.6 Hz, 1H), 3.14 (m, 1H), 3.26 (s, 3H), 3.40 (dd,
J = 9.1, 6.1 Hz, 1H), 3.51 (dd, J = 9.1, 6.1 Hz, 1H), 3.60 (m, 1H), 3.63 (dd, J = 10.5,
2.3 Hz, 1H), 3.79 (s, 3H), 3.81 (dd, J = 9.4, 2.3 Hz, 1H), 3.97 (m, 1H), 4.42 (s, 2H),
5.29 (tt, J = 9.9, 2.8 Hz, 1H), 5.41 (ddd, J = 10.4, 7.5, 7.5 Hz, 1H), 5.75 (ddd,
J = 15.4, 9.2, 6.0 Hz, 1H), 5.98 (pseudo-t, J = 10.8 Hz, 1H), 6.34 (dd, J = 14.5,
10.7 Hz, 1H), 6.91 (d, J = 8.7 Hz, 2H), 7.27 (d, J = 8.8 Hz, 2H); ESI-HRMS calcd for
C36H58NaO9 [M+Ma]+: 657.3979. found: 657.3986. Compound 17: 1H NMR
(600 MHz, acetone-d6) d = 0.83 (d, J = 7.1 Hz, 3H), 0.89 (d, J = 7.0 Hz, 3H), 0.94
(d, J = 6.6 Hz, 6H), 1.31 (s, 3H), 1.39 (s, 3H), 1.30–1.46 (m, 6H), 1.63 (m, 1H),
1.66 (m, 3H), 1.78 (s, 3H), 1.81 (m, 1H), 1.90 (m, 1H), 2.01 (m, 1H), 2.24 (m, 1H),
2.30 (m, 1H), 2.36 (dd, J = 15.5, 10.1 Hz, 1H), 2.43 (dd, J = 15.5, 1.2 Hz, 1H), 2.48
(m, 1H), 3.23 (m, 1H), 3.28 (s, 3H), 3.50 (d, J = 10.2 Hz, 1H), 3.58 (m, 1H), 3.61
(d, J = 4.6 Hz, 1H), 3.64 (dd, J = 9.2, 1.4 Hz, 1H), 3.83 (d, J = 4.44 Hz, 1H), 3.99 (d,
J = 3.7 Hz, 1H), 4.13 (t, J = 5.5 Hz, 1H), 4.15 (m, 1H), 4.37 (ddd, J = 9.0, 8.3,
4.6 Hz, 1H), 5.40 (ddd, J = 9.9, 8.3, 8.3 Hz, 1H), 5.44 (d, J = 9.5 Hz, 1H), 5.44 (m,
1H), 5.77 (m, 1H), 5.80 (dt, J = 15.6, 5.6 Hz, 1H), 6.09 (dd, J = 10.8, 10.8 Hz, 1H),
6.27 (d, J = 15.6 Hz, 1H), 6.40 (dd, J = 15.2, 10.8 Hz, 1H); LC–ESIMS calcd for
C37H62NaO9 [M+Na]+: 673.4. found: 673.5.
In conclusion, based on molecular modeling and stabilization
considerations, we have developed a first series of simplified ana-
logues of the etnangiens with simplified side-chains and/or con-
tracted macrolactone ring. The synthesis of these compounds
was enabled in a convergent manner by late stage-diversification
using different southern ring fragments. During preparation of
these compounds a strong tendency of these macrolides towards
translactonization processes under basic or adidic conditions was
15. Removal of the acetonide could not be effected without extensive
decomposition.