Identification of an anti-MRSA dihydrofolate reductase inhibitor from a diversity-oriented synthesis

Article abstract of DOI:10.1039/b812901k

The screening of a diversity-oriented synthesis library followed by structure-activity relationship investigations have led to the discovery of an anti-MRSA agent which operates as an inhibitor of Staphylococcus aureus dihydrofolate reductase. The Royal Society of Chemistry.

Full text of DOI:10.1039/b812901k

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Identification of an anti-MRSA dihydrofolate reductase inhibitor
from a diversity-oriented synthesisw
Emma E. Wyatt,a Warren R. J. D. Galloway,a Gemma L. Thomas,a
Martin Welch,b Olivier Loiseleur,c Alleyn T. Plowrightd and David R. Spring*a
Received (in Cambridge, UK) 25th July 2008, Accepted 12th August 2008
First published as an Advance Article on the web 10th September 2008
DOI: 10.1039/b812901k
The screening of a diversity-oriented synthesis library followed
by structure–activity relationship investigations have led to the
discovery of an anti-MRSA agent which operates as an inhibitor
of Staphylococcus aureus dihydrofolate reductase.
The discovery and development of antibacterial agents is
widely regarded as one of the greatest successes of 20th
century medicine.1 However, bacteria have quickly become
resistant to the most commonly prescribed antibiotics.2,3
Combined with the lack of fundamental antibacterial research
carried out by pharmaceutical companies over recent decades
we are left with a legacy of few new drugs with an ever-
decreasing efficacy.4–7 Thus bacterial infection, particularly
from multi-drug resistant strains, such as methicillin-resistant
Staphylococcus aureus (MRSA), remains a serious threat to
human lives.3,4,8–10 Consequently the identification and devel-
opment of novel antibacterial agents is of paramount impor-
tance for human healthcare.11
Fig. 1 Nitrogen-based molecular frameworks present in the majority
of the DOS library compounds that exhibited anti-MRSA activity.
We have previously reported the synthesis of a structurally
diverse small molecule library totalling 223 members via a
DOS approach from a simple fluorous-tagged diazoacetate
starting material.14 Initial inhibition of proliferation pheno-
typic experiments identified 64 compounds that modulated the
growth of EMRSA-15 and EMRSA-16 strains over a concen-
tration range of 100 mM to 10 mM. Compounds based around
four types of nitrogen-containing heterocyclic frameworks
(1–4) were found to dominate as the active species (Fig. 1).
Substituents generally associated with increased levels of
antibacterial activity were identified; for example, pyrimidine
derivatives (2) were typically more active when R3 = phenyl,
thiophene or iso-butyl and R1 = ethyl or aryl. In addition, it
was found that heteroatom, in particular halogen, substitution
on aryl ring substituents frequently produced compounds with
Small molecules that exhibit antibacterial activity (so-called
‘hits’) can be identified through the phenotypic screening of
structurally diverse small molecule collections.12 However, a
formidable challenge associated with the further development
of these hits is the identification of the small molecule’s
biological target, which in turn provides information regard-
ing the molecule’s mode of action.13 Herein, we report the
results of screening experiments carried out on a small mole-
cule library produced in a previous diversity-oriented synthesis
(DOS) campaign, together with analogue synthesis, SAR
analyses and target identification studies. This work has
culminated in the discovery of a structurally novel antibacter-
ial agent that displays activity against epidemic strains of
MRSA (EMRSA) in cellular assays, and has been shown to
act as a prokaryotic-selective uncompetitive reversible inhibi-
tor of the EMRSA-16 dihydrofolate reductase enzyme DfrB
(DfrBEMRSA16) in vitro.
Table 1 Structures and activities (MIC50) of emmacin, gemmacin,
erythromycin and oxacillin, which display growth inhibitory activity
against methicillin-susceptible and -resistant strains of S. aureus
a Department of Chemistry, University of Cambridge, Lensfield Road,
Cambridge, UK. E-mail: drspring@ch.cam.ac.uk;
Fax: +44 (0) 1223-336362; Tel: +44 (0) 1223-336498
b Department of Biochemistry, University of Cambridge, Cambridge,
UK
c Syngenta Crop Production AG, Schwarzwaldallee 215, 4002 Basel,
Switzerland
d AstraZeneca, Pepparedsleden 1, 431 83 Molndal, Sweden
¨
w Electronic supplementary information (ESI) available: Full experi-
mental details and spectra for emmacin together with biological assay
data. See DOI: 10.1039/b812901k
ꢀc
This journal is The Royal Society of Chemistry 2008
4962 | Chem. Commun., 2008, 4962–4964

View Online
Scheme 1 One-pot synthesis of emmacin.
improved biological activity.15 Thus various combinations of
substituted b-keto ester, aldehyde and 3-formyl chromone
‘building-blocks’ were used to synthesize a focused collection
of 35 compounds.16
Fig. 3 Lineweaver–Burk plot of enzymatic reaction parameters for
the conversion of dihydrofolate (DHF; 12.5–50 mM) to tetrahydrofolic
acid by DfrBEMRSA16 with constant NADPH concentration (60 mM).
‘+’ and ‘À’ emmacin indicates reaction in the presence (20 mM) or
Purified analogues were screened in inhibition of prolifera-
tion phenotypic assays against EMRSA-15 and EMRSA-16
strains.17 The most potent compound identified by this study
was named emmacin, which compared favourably with two
clinical antibacterial agents and gemmacin, another anti-
MRSA agent discovered recently (Table 1).12a
absence of emmacin; v0 = steady-state reaction velocity (rate); KM
=
substrate concentration required for the enzyme to reach half max-
imum velocity; Vmax = maximum reaction velocity.
heterocyclic core which is reminiscent of known bacterial
DHFR inhibitors (Fig. 2).
Emmacin was synthesised by a one-pot, three-component
Biginelli-type reaction18 of 5, 6 and 7 (Scheme 1).19
The assay results, together with structural similarities with
existing bacterial DHFR inhibitors, raised the possibility that
emmacin may inhibit S. aureus proliferation by acting as a
prokaryote-selective DHFR inhibitor. Although there have
been reports of anti-MRSA compounds that act as DHFR
inhibitors,23,24 the mechanism of the inhibition has not been
well characterised. Therefore in this study we isolated the
specific EMRSA-16 DHFR enzyme ‘DfrBEMRSA16’ by the
expression of a cloned MRSA-16 DHFR gene (dfrB) in
E. coli DH5a-cells. A DHFR inhibition assay was subse-
quently performed using the purified DfrBEMRSA16 enzyme.25
Emmacin was found to inhibit its enzymatic activity with an
IC50 value of 5.4 mM (1.9 mg mlÀ1). Analysis of the observed
enzyme kinetics showed approximately parallel lines in the
Lineweaver–Burk double reciprocal plot, where both KM and
Vmax values decrease in the presence of emmacin (Fig. 3).
Therefore, emmacin could be defined kinetically as an un-
competitive inhibitor with respect to dihydrofolate,26 which
has been observed for other DHFR inhibitors.30 The inhibi-
tion appeared reversible as normal enzyme could be recovered.
Despite recent research into the discovery of new bacterial
DHFR inhibitors,22,23,27,28 the enzyme is still viewed as an
underexploited target in the antibacterial field.11 Interestingly,
we could find no examples of substituted dihydropyrimidine
compounds, of the type exemplified in emmacin, having been
applied to this therapeutic mode of action. The majority of
disclosed agents are based around a heteroaromatic core with
a 1,3-arrangement of amine groups,29 which is believed to fit
Emmacin was selected for preliminary mode of action
studies and subjected to a battery of around 25 assays designed
to identify antibacterial, fungicidal and herbicidal properties
in addition to common cytotoxic effects. The assays were
principally performed using insect or mammalian cell lines
and only very moderate fungicidal or herbicidal properties
were observed, providing an indication that emmacin is a
selective antibacterial agent. Significantly, emmacin did not
display activity in a range of cross-indication assays (100 mM
emmacin) designed to investigate common cytotoxic modes of
action, such as modulation of kinase activity, g-aminobutyric
acid receptors, protein synthesis, generation of reactive oxygen
species and ATP synthesis uncoupling. Due to emmacin’s
structural features we were particularly intrigued by its lack
of activity against bovine dihydrofolate reductase (DHFR),
which is often associated with mammalian toxicity. DHFR is
an enzyme present in all eukaryotic and prokaryotic cells
which catalyses the reduction of 7,8-dihydrofolate to 5,6,7,8-
tetrahydrofolate using NADPH as a cofactor.20,21 Tetra-
hydrofolate is involved in the biosynthesis of nucleotide bases
of DNA; thus inhibition of the DHFR enzyme blocks DNA
synthesis, thereby arresting cell growth.22 Marked differences
in the structures of mammalian and bacterial DHFR enzymes
have been exploited in the development of several potent and
selective bacterial DHFR inhibitors which have demonstrated
antibacterial effects.22,23 Emmacin contains a nitrogen-based
Fig. 2 Comparison of the structures of some known DHFR inhibitors (trimethoprim11 and iclaprim23) with emmacin. The common elements of
the nitrogen-based heterocyclic frameworks present in each structure are highlighted.
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 4962–4964 | 4963

View Online
ideally into a narrow pocket in the active site of the enzyme.11
Therefore, to the best of our knowledge, emmacin represents
the first member of a new structural sub-class of bacterial-
selective DHFR inhibitors.31
D. A. Smith and B. C. Jones, J. Comput. Aided Mol. Des., 2001, 15,
273–286.
16 Details of the synthetic routes employed for analogue generation
are available in the supplementary information and in our original
report on the DOS library synthesis (see ref. 14); structures shown
in supporting information appendix.
In conclusion, we have exploited a DOS small molecule
collection in the discovery of a structurally novel antibacterial
agent called emmacin. Emmacin inhibited the growth of two
epidemic strains of MRSA in vitro, and crucially, was found to be
inactive in a variety of mammalian cytotoxicity assays. It was
shown to act as a prokaryote-selective, uncompetitive and rever-
sible inhibitor of EMRSA-16 DHFR (DfrBEMRSA16). Due to its
dihydropyrimidine core, emmacin represents a new structural sub-
class of DHFR inhibitors, which could potentially be exploited in
the development of critically-needed, new antibacterial agents.
We gratefully acknowledge financial support from the
EPSRC, AstraZeneca and Pfizer. We also acknowledge the
EPSRC National Mass Spectrometry Service Centre,
Swansea, for providing mass spectrometric data and Syngenta
for carrying out preliminary mode of action studies.
17 Analysis of this data provided some insight into possible SAR
exhibited by these nitrogen-based scaffolds; bicyclic species 3 were
typically more active with halogen atoms (Br or Cl) at the R4
position, bicyclic compounds 4 generally displayed higher levels of
activity when R1 was an aryl halide rather than an alkyl substituent
and the replacement of an iso-butyl group with an aryl halide at the
R3 position of 2 also appeared to confer a higher level of activity.
For selected biological screening data consult the supplementary
information.
18 R. Milcent, J. C. Malanda, G. Barbier and J. Vaissermann,
J. Heterocycl. Chem., 1997, 34, 329–336.
19 Emmacin is a chiral molecule; the data presented in this report
refer to the use of a racemic sample of emmacin synthesised by the
route outlined in Scheme 1. Studies towards the isolation and
independent biological screening of the two enantiomers of
emmacin are ongoing.
20 N. V. Kovalevskaya, E. D. Smurnyi, B. Birdsall, J. Feeney and V.
I. Polshakov, Pharm. Chem. J., 2007, 41, 350–353.
21 J. M. Blaney, C. Hansch, C. Silipo and A. Vittoria, Chem. Rev.,
1984, 84, 333–407.
22 S. Hawser, S. Lociuro and K. Islam, Biochem. Pharmacol., 2006,
71, 941–948.
23 The IC50 values for inhibition of S. aureus DHFR enzyme
by trimethoprim and iclaprim are both 7 nM; P. Schneider,
S. Hawser and K. Islam, Bioorg. Med. Chem. Lett., 2003, 13,
4217–4221.
24 W. J. Peppard and C. D. Schuenke, Curr. Opin. Invest. Drugs,
2008, 9, 210–225.
25 The effect of emmacin on DfrBEMRSA16 activity was determined
using a commercially available DHFR assay kit (Sigma Aldrich,
Product Code CS0340). Human DHFR was not inhibited by
emmacin. See supplementary information for more detail.
26 D. Voet, J. G. Voet and C. W. Pratt, Principles of Biochemistry,
John Wiley & Sons, Inc., New Jersey, 2008.
Notes and references
1 C. Walsh and G. Wright, Chem. Rev., 2005, 105, 391–394.
2 P. Fernandes, Nat. Biotechnol., 2006, 24, 1497–1503.
3 E. D. Brown and G. D. Wright, Chem. Rev., 2005, 105, 759–774.
4 R. E. W. Hancock, Nat. Rev. Drug Discovery, 2007, 6, 28.
5 D. J. Payne, M. N. Gwynn, D. J. Holmes and D. L. Pompliano,
Nat. Rev. Drug Discovery, 2007, 6, 29–40.
6 M. L. Katz, L. V. Mueller, M. Polyakov and S. F. Weinstock, Nat.
Biotechnol., 2006, 24, 1529–1531.
7 L. M. Jarvis, Chem. Eng. News, 2008, 86, 15–20.
8 D. Hopwood, Nat. Rev. Drug Discovery, 2007, 6, 8–12.
9 Editorial, Nat. Biotechnol., 2006, 24, 1489.
10 W. R. Jarvis, J. Schlosser, R. Y. Chinn, S. Tweeten and M.
Jackson, Am. J. Infect. Control, 2007, 35, 631–637.
27 L. A. Sorbera, J. Castaner and X. Rabasseda, Drugs Future, 2004,
29, 220–225.
11 P. C. Wyss, P. Gerber, P. G. Hartman, C. Hubschwerlen, H. Locher,
H. P. Marty and M. Stahl, J. Med. Chem., 2003, 46, 2304–2312.
12 For a recent example of antibacterial discovery resulting from the
screening of a small molecule library see: (a) G. L. Thomas, R. J.
Spandl, F. G. Glansdorp, M. Welch, A. Bender, J. Cockfield, J. A.
Lindsay, C. Bryant, D. F. J. Brown, O. Loiseleur, H. Rudyk, M.
Ladlow and D. R. Spring, Angew. Chem., Int. Ed., 2008, 47,
2808–2812; (b) A. Robinson, G. L. Thomas, R. J. Spandl, M.
Welch and D. R. Spring, Org. Biomol. Chem., 2008, 6, 2978–2981.
13 For a discussion of target identification following the screening of
small molecule libraries, see: (a) J. Kotz, Nat. Chem. Biol., 2007, 3,
199; (b) I. A. Inverarity and A. N. Hulme, Org. Biomol. Chem.,
2007, 5, 636–643; (c) J. C. Yarrow, Y. Feng, Z. E. Perlman, T.
Kirchhausen and T. J. Mitchison, Comb. Chem. High Throughput
Screening, 2003, 6, 79–99; (d) C. P. Hart, Drug Discovery Today:
Targets, 2005, 10, 513–519; (e) U. S. Eggert and T. J. Mitchison,
Curr. Opin. Chem. Biol., 2006, 10, 232–237; (f) D. R. Spring, Chem.
Soc. Rev., 2005, 34, 472–482.
14 E. E. Wyatt, S. Fergus, W. R. J. D. Galloway, A. Bender, D. J.
Fox, A. T. Plowright, A. S. Jessiman, M. Welch and D. R. Spring,
Chem. Commun., 2006, 3296–3298.
15 The incorporation of halogen atoms into drug candidates has
generally been correlated with increased lipophilicity, which may
allow improved penetration through cell lipid membranes; see for
example: (a) A. A. Litvin, G. B. Kolyvanov, V. P. Zherdev and A.
P. Arzamastev, Pharm. Chem. J., 2004, 38, 581–586; (b) F. Hollosy,
J. Serprodi, L. Orfi, D. Eros, G. Keri and M. Idei, J. Chromatogr.,
B: Biomed. Appl., 2002, 780, 355–363; (c) H. van de Waterbeemd,
28 H. H. Locher, H. Schlunegger, P. G. Hartman, P. Angehrn and
R. L. Then, Antimicrob. Agents Chemother., 1996, 40, 1376–1381.
29 There are several examples of antibacterial compounds which are
presumed to act as DHFR inhibitors and are not based around a
1,3-diamine core, although information regarding their (potential)
toxicity in mammalian systems is not always available; see for
example: (a) K. N. Roa and S. R. Venkatachalam, Toxicol. in
Vitro, 2000, 14, 53–59; (b) M. D. Navarro-Martinez, E. Navarro-
Peran, J. Carbezas-Herrera, J. Ruiz-Gomez, F. Garcia-Canovas
and J. N. Rodriguez-Lopez, Antimicrob. Agents Chemother., 2005,
49, 2914–2920; (c) K. N. Rao and S. R. Venkatachalam, Bioorg.
Med. Chem., 1999, 7, 1105–1110.
30 For example see: (a) S. R. Stone and J. F. Morrison, Biochim.
Biophys. Acta, 1986, 869, 275–285; (b) E. Lebrun, Y. Tu, R. V.
Rapenbusch, A. R. Banijamali and W. O. Foye, Biochim. Biophys.
Acta, 1990, 1034, 81–85; (c) I.-H. Choi and C. Kim, Arch.
Pharmacal Res., 2002, 25, 807–816.
31 It is conceivable that emmacin’s dihydropyrimidine is oxidized to
give a heteroaromatic core under assay conditions. However, this is
not the case. Emmacin appears resistant to oxidation by air, and
even DDQ. Unlike Hantzsch-type dihydropyridines, where aroma-
tization to pyridines is typically facile, the dehydrogenation of
dihydropyrimidines is nontrivial; see: (a) C. O. Kappe, Tetrahedron,
1993, 49, 6937–6963; (b) J. J. V. Eynde, N. Audiart, V. Canonne, S.
Michel, Y. V. Haverbeke and C. O. Kappe, Heterocycles, 1997, 45,
1967–1978.
ꢀc
This journal is The Royal Society of Chemistry 2008
4964 | Chem. Commun., 2008, 4962–4964