Interconvertible geometric isomers of Plasmodium falciparum dihydroorotate dehydrogenase inhibitors exhibit multiple binding modes

Article abstract of DOI:10.1016/j.bmcl.2017.06.052

Two new tricyclic β-aminoacrylate derivatives (2e and 3e) have been found to be inhibitors of Plasmodium falciparum dihydroorotate dehydrogenase (PfDHODH) with Ki 0.037 and 0.15?μM respectively. ¹H and ¹³C NMR spectroscopic data show that these compounds undergo ready cis-trans isomerisation at room temperature in polar solvents. In silico docking studies indicate that for both molecules there is neither conformation nor double bond configuration which bind preferentially to PfDHODH. This flexibility is favourable for inhibitors of this channel that require extensive positioning to reach their binding site.

Full text of DOI:10.1016/j.bmcl.2017.06.052

Accepted Manuscript
Interconvertible geometric isomers of Plasmodium falciparum dihydroorotate
dehydrogenase inhibitors exhibit multiple binding modes
Glenn A. McConkey, Paul T.P. Bedingfield, David R. Burrell, Nicholas C.
Chambers, FraserCunningham, Timothy J. Prior, Colin W.G. Fishwick, Andrew
N. Boa
PII:
S0960-894X(17)30646-7
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.06.052
BMCL 25086
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
3 May 2017
15 June 2017
17 June 2017
Please cite this article as: McConkey, G.A., Bedingfield, P.T.P., Burrell, D.R., Chambers, N.C., Cunningham, F.,
Prior, T.J., Fishwick, C.W.G., Boa, A.N., Interconvertible geometric isomers of Plasmodium falciparum
dihydroorotate dehydrogenase inhibitors exhibit multiple binding modes, Bioorganic & Medicinal Chemistry
Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.2017.06.052
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Interconvertible geometric isomers of Plasmodium falciparum
dihydroorotate dehydrogenase inhibitors exhibit multiple
binding modes
Glenn A. McConkeya, Paul T. P. Bedingfielda, David R. Burrellb, Nicholas C. Chambersb,
Fraser Cunninghamc, Timothy J. Priorb, Colin W.G. Fishwickc, Andrew N. Boab*
a School of Biology, Faculty of Biological Sciences, University of Leeds, Leeds LS2 9JT, UK
b School of Mathematics & Physical Sciences, Faculty of Science and Engineering, University
of Hull, Hull HU6 7RX, UK
c School of Chemistry, Faculty of Mathematics and Physical Sciences, University of Leeds,
Leeds LS2 9JT, UK
Abstract
Two new tricyclic β-aminoacrylate derivatives (2e and 3e) have been found to be inhibitors
of Plasmodium falciparum dihydroorotate dehydrogenase (PfDHODH) with Ki 0.037 and
0.15 µM respectively. 1H and 13C NMR spectroscopic data show that these compounds
undergo ready cis-trans isomerisation at room temperature in polar solvents. In silico docking
studies indicate that for both molecules there is neither conformation nor double bond
configuration which bind preferentially to PfDHODH. This flexibility is favourable for
inhibitors of this channel that require extensive positioning to reach their binding site.
Keywords: dihydroorotate dehydrogenase; DHODH; Plasmodium falciparum; inhibitor
*Corresponding author. Tel.: +44 (0)1482 465022; E-mail address: a.n.boa@hull.ac.uk

Malaria is one of the main infectious diseases in the world with an estimated 212
million cases in 2015, centred to a great extent on the countries in central Africa and south
east Asia.1 The numbers are even starker when one considers it is estimated these cases led to
an estimated 429,000 deaths and that approximately 70% of these deaths were of children
under five years old. Malaria is considered a ‘preventable’ disease, largely because of the use
of chemotherapeutic agents,2 however widespread drug resistance3 has rendered these agents
ineffective in many regions of the world. This problem has led to much effort being expended
into validating novel drug targets, as well as discovering small molecule enzyme inhibitors to
act as leads for development of new antimalarial drugs.4 One such target which has attracted
recent attention is P. falciparum dihydroorotate dehydrogenase (PfDHODH),5 a key enzyme
in the obligate de novo pyrimidine pathway for uridine monophosphate (UMP) biosynthesis.
Early work aimed towards the discovery of PfDHODH inhibitors focussed on modification of
the human DHODH (hDHODH) inhibitors known at the time, such as brequinar (Figure 1).6
This work was naturally followed by high throughput screening7 and de novo design
approaches.8 During the progress of this work, co-crystallisation9 and in silico docking8,10
experiments revealed that the inhibitors discovered targeted the ubiquinone binding channel,
and also revealed important features for the design of selective and potent inhibitors. Firstly
amino acid residues H185 and / or R265, at the head of the ubiquinone-binding channel, were
found to be essential for hydrogen bonding interactions with the inhibitors. The ubiquinone
channel also contained a hydrophobic region which contributes to the binding of the more
potent inhibitors reported. Indeed many of the inhibitors reported to date6-12 reveal broadly
this amphiphilic nature, and possess a polar ‘head group’ and hydrophobic ‘tail’.
Triazolopyrimidine DSM265 (Figure 1) is currently the most advanced PfalDHODH
inhibitor, and Phase 1 clinical trials for this candidate have recently been reported.13

We have reported previously that β-aminoacrylate derivatives of the general structure 1
(Figure 1) are inhibitors of PfDHODH.10 In particular, tricyclic derivatives 2a and 3b
displayed sub-micromolar IC50 values against PfDHODH, and in the case of 2a there was a
1000-fold greater potency compared to the human enzyme (hDHODH). An extensive series
of experiments in this previous work with this class of inhibitors10a showed that compounds
2a and 3b, and related structures, were competitive inhibitors with the enzyme cofactor CoQ
and reversible. Compounds such as 2a may in principle exist as a pair of geometric isomers
and we considered that different H-bonding groups on either side of the double bond, and
their position relative to the aromatic portion of the structure through rotation about the CAr-N
bond, would be key in determining their binding to the enzyme. It is perhaps noteworthy to
mention that alternative binding modes have been reported with certain hDHODH
inhibitors,14 and also differential binding modes for A77 1726, the active metabolite of
leflunomide (Figure 1), when binding to PfDHODH is compared to its principal target
hDHODH.9a We therefore sought to examine whether alterative polar ‘head groups’ to those
investigated would lead to improved activity of members in this class of compound, and the
results obtained are reported herein.

Figure 1. Example structures of known PfalDHODH (brequinar, DSM265, 2a and 3b) and
hDHODH (A77 1726) inhibitors.
Ten new derivatives (2b-f, 3a,c-f) based on the 3-aminocarbazole (4) and 3-
aminodibenzothiophene (5) core unit were synthesised using the method reported previously
(Scheme 1).10a Derivatives 2b and 3a were prepared by simply heating the aromatic amine
with the appropriate commercially available β-ethoxy-α,β-unsaturated carbonyl compound
6b and 6a respectively. For 2c-f and 3c-f the relevant active methylene compounds 7c-f were
first heated with an excess of triethylorthoformate (or acetate) in toluene, forming the β-
ethoxy-α,β-unsaturated carbonyl compounds 6c-f in situ, followed by addition of either the
aromatic amine 4 or 5 and heating for a further short period. After cooling, the products
precipitated upon standing, were collected by filtration and then recrystallised.15

Scheme 1. Synthesis of β-aminoacrylate derivatives of 9-ethyl-3-aminocarbazole and 3-
aminodibenzothiophene in this study.
Table 1 shows the IC50 values obtained for the compounds discussed in this work. These
values were obtained using the screening methodology reported previously.10a The IC50
values of derivatives 2c,d,f and 3c,d,f were above the arbitrary cut-off value of 0.1 mM and
thus not further examined. Derivatives 2d and 3d, structures based upon Meldrum’s acid, are
conformationally restricted analogues. We assigned their low activity to the restricted
freedom imposed by the six-membered ring to position the polar head group for maximal
hydrogen bonding to H-bond donors on either side the co-factor channel.
Entry PfalDHODH
IC50 (µM)
hDHODH
IC50 (µM)
0.28 0.05
2a
0.44 0.06 10a
491 42 10a
n.d.
n.d.
1.2
4.4
3a
2b
0.16 0.05
0.16 0.05 10a
>100
>100
>100
3b
30 5.9 10a
n.d.
n.d.
n.d.
n.d.
>100
>100
n.d.
2c
3c
2d
3d
2e
3e
2f
>100
0.094 0.031 †
0.38 0.06 ‡
>100
>100
n.d.
3f

Table 1. IC50 values of selected compounds against P. falciparum and human DHODH. The
IC50 values were determined according to methodology reported previously.10a n.d. = not
determined; † Ki vs PfalDHODH) for 2e was 0.037 0.012 µM; ‡ Ki (PfalDHODH) for 3e
was 0.150 0.001 µM (cf. Ki = 0.05 and 0.02 µM for 2a and 3b respectively10a)
The activity of the cyanoacetate (2a, 3a) and malonate derivatives (2b, 3b) are not too
dissimilar, and the IC50 values are probably subtly altered depending on the influence of the
second ethyl ester and/or N-ethyl group. The lowest activity is seen with 2b, where both ethyl
groups are present, may mean that the molecule is slightly too large to fit well into the
ubiquinone binding site. Similarly, derivatives 2f and 3f, which only differ from 2a and 3a by
a single methyl group, showed that substitution at this position cannot be tolerated sterically.
The lack of activity in 2c and 3c was somewhat surprising, given the similar electron
withdrawing effect of the methyl ketone compared to the ethyl ester and cannot so easily be
explained.
Nitroacrylate derivatives 2e and 3e were as active, if not more so, than cyanoacrylate
and malonate derivatives 2a,b and 3a,b. As discussed above, we considered that the
conformational preferences in the polar ‘head group’ would be important in determining the
binding mode of these inhibitors to DHODH, and so it was interesting in particular to
compare the isomerism in carbazole derivatives 2a, 2e and 2f, and the corresponding
analogues of the dibenzothiophene series 3a, 3e and 3f. In the case of 2a and 3a, evidence of
only one isomer is seen in the 1H NMR. The heavily deshielded NH (δ 10.96 ppm for 2a and
11.26 ppm for 3a) and large coupling to the -CH= (J 13.7 Hz) indicated the presence of
intramolecular hydrogen bonding to the ester carbonyl group, and the anti- displacement of
the hydrogen atoms in the Ar-NH-CH= unit showed that the (Z)-isomer was preferred. In the
case of 2f and 3f analysis of the products similarly revealed single isomers [δNH 11.56 (s) for
2f, and δNH 11.65 (s) for 3f] in spite of the extra methyl group (R3). The nitroacrylate
derivatives 2e and 3e have the possibility to exhibit intramolecular hydrogen bonding in

either isomeric form, and the balance between these isomeric forms was expected to have an
important effect on the overall isomeric preferences and therefore binding to the cofactor
channel of DHODH. The 1H NMR of both 2e and 3e showed a 1:1 mixture (in CDCl3)
however many times it was recrystallised under a range conditions (solvent, rate of cooling
etc), indicating no clear preference in solution for one isomer over the other.16 The 13C NMR
of 2e in CDCl3 showed doubling of almost all signals. However for 3e in the more polar d6-
DMSO only one set of the aromatic carbon signals and ethyl group were observed, and also
the signals for the acrylate unit were completely absent. This was accompanied by changes in
the 1H NMR where in d6-DMSO the NH-CH= signals were broadened significantly and
almost reduced to baseline level, and the ethyl CH2 quartet was greatly broadened. This
indicates clearly that the geometric forms of 3e are readily interconvertible in a polar solvent
on the NMR timescale.
Carbazole 2e could be recrystallised from ethyl acetate to provide bright red crystals
suitable for single crystal X-ray diffraction (see Supplementary Information).17 The first crop
showed 2e had crystallised preferentially as the (Z)-isomer (Figure 2A). Examining the single
crystal X-ray data of 2e it was interesting to find that the N2-C15 bond was much shorter
than the C15-C16 bond and C9-N2 bonds [1.3201(17) Å vs. 1.3914(18) Å vs. 1.4242(17) Å
respectively]. These data are consistent with extensive delocalisation of the nitrogen lone pair
into the enamine unit in the (Z)-isomer (Figure 2B), and could explain its deep red colour.
After evaporation of the mother liquors and another recrystallization, the quality of the red
crystals diminished and an orange powder was obtained upon evaporation of the mother
liquors. Optical microscopy of this material showed two microcrystalline forms, one red and
one orange, with the latter in large excess. We surmised the orange material may be the more
soluble (E)-isomer. A portion of the orange powder was examined by X-ray powder
diffraction using a PANAlytical Empyrean diffractometer operating with Cu Kα1 radiation. It

was shown to be highly crystalline, but the diffraction pattern of the orange powder did not
match that calculated from the structure of the red crystals obtained (Supplementary
Information Figure S1). It was possible to index the powder diffraction pattern of the orange
powder using a large triclinic cell. The volume of this cell (2506 Å3) is approximately three
times that of the cell found for the red crystals. It seems likely that the bulk of the orange
sample contains the (E) isomer of 2e, and the tripling of the volume suggests that there are
three molecules in the asymmetric unit, perhaps two of one isomeric form and one of the
other. Importantly, 1H NMR of either the red or orange crystalline forms showed, in CDCl3
solution, the same 1:1 mixture of the two geometric isomers.
We were able to select a small single crystal of the orange material from among the
original red crystals. The small crystal only diffracted weakly but we were able to obtain a
crystal structure, albeit of lower quality than that of the red form.18 The structure is not too
dissimilar to that of the red crystals, however it had a different orientation of both the N-ethyl
and ethyl ester groups (Supplementary Information Figure S2). It also showed that the nitro
group was not co-planar with the NH, as in the red crystals, but was inclined at an angle of
25(1) °. The crystal structure of this orange single crystal did not, however, match the powder
diffraction pattern of the bulk orange material (Supplementary Information Figure S3).

A
O
O
N
H
N
O
O
N
B
Figure 2A: Crystal structure of 2e with atoms drawn as 50% thermal ellipsoids The more
crystalline form of 2e reveals a shortened N2-C15 bond, lengthened C15-C16 bond and an
intermolecular hydrogen bond between the nitro group and the NH. Figure 2B: The bond
lengths from the crystal structure of the deep red coloured 2e is consistent with large degree
of delocalisation of the nitrogen lone pair.
In order to probe the details of the interaction of these inhibitors with PfDHODH,
docking (with both eHiTS and Autodock) of the E- and Z-forms of 2e and 3e into the putative
ubiquinione binding channel of PfDHODH was performed using the PfDHODH crystal
structure (PDB code; 1TV5). The data produced suggest several possibilities for the binding
of each isomer of 2e and 3e into PfDHODH. However all of the suggested poses are
predicted to bind in the same general space of the enzyme, with the aromatic ‘tail’ group of
each ligand binding deep within the large hydrophobic pocket of the binding site (Figures 3A
and 3B). For each of the four structures (i.e. both E- and Z- isomers of both 2e and 3e) two
equally favourable rotamers were observed with the tail group seen to adopt the orientations
shown in Figures 3A and 3B.

A
B
Figure 3. Docking studies reveal that the hydrophobic aromatic tail group of 3e (forest green)
and 2e (latter not shown) may adopt one of two equally preferred conformations (3A and 3B,
rotated 180° around the Ar-NH bond) when docked into the ubiquinone co-factor channel of
PfDHODH. In both cases hydrogen bonding networks with Tyr528, His185 and Arg265 are
maintained.
A
B
Figure 4. Docking studies reveal that the hydrophilic head group of 2e and 3e (latter
not shown) may bind in two equally preferred conformations (4A and 4B, rotated 180°
around the Ar-NH bond) when docked into the ubiquinone co-factor channel of PfDHODH.
For both molecules 2e and 3e the polar ‘head group’ of the molecule is predicted to
bind in the polar region of the binding site. Both isomers of each molecule can adopt two
distinct hydrogen bonding networks (Figures 4A and 4B). This is again due to a rotation of

180° around Ar-NH bond causing the polar ‘head’ of the molecule to flip. Although this
‘flipping’ forces a change in the hydrogen bonds made by the molecule and the enzyme, the
residues involved in binding in PfDHODH are preserved, i.e. histidine185 (H185), tyrosine528
(Y528) and arginine265 (R265). All suggested binding poses have hydrogen bonds to at least
two of these residues. The results from the docking studies suggest that the ligands may bind
with multiple poses to the ubquinone channel and that there is no clear preference for any one
arrangement.
In conclusion, we have discovered two new derivatives (2e and 3e) that are potent
inhibitors of PfDHODH. 1H and 13C NMR spectroscopic data revealed that these compounds
undergo ready isomerisation at room temperature in d6-DMSO, but the docking studies
indicate that there is neither conformation nor configuration which binds preferentially to
PfDHODH. This flexibility is favourable for inhibitors of this channel that require extensive
positioning to reach their binding site. X-ray crystallographic study of the red crystal form of
(Z)-2e reveals shortened N2-C15 and lengthened C15-C16 bonds. The delocalisation of the
charge onto the nitro group must promote the hydrogen bonding interactions between the
nitro group and the N9-H, but also with any H-bond donor group on the enzyme. This is
consistent with the observations made by Phillips and co-workers where an intrinsic dipole
contribution9b in their triazolopyrimidine class of compounds is also believed to add to the
hydrogen bonding and thus binding of the inhibitors to the enzyme. Overall, these results
further add to the knowledge of inhibitor binding to the co-factor binding site of PfDHODH
and may assist in the design and optimisation of other, more drug-like inhibitors of P.
falciparum dihydroorotate dehydrogenase.
Acknowledgements

We thank the University of Hull for funding, BBSRC for support (GM), and the EPSRC
National Mass Spectrometry Service (Swansea) for accurate mass determination of 2e and 3e.
References and notes
1.
World Malaria Report 2016, World Health Organisation, Geneva
(http://www.who.int/malaria/publications/world-malaria-report-2016/en/).
Schlitzer, M. ChemMedChem 2007, 2, 944–986.
2.
3.
(a) Parija, S. C.; Praharaj, I. Indian J. Med. Microbiol. 2011, 29, 243-248; (b) White, N.
J. J. Clin. Invest. 2004, 113, 1084–1092.
4.
(a) MMV Global portfolio of antimalarial medicines 2016, https://www.mmv.org/sites
/default/files/uploads/docs/RandD/4Q_2016_Global_Portfolio.pptx (accessed
19/4/2017); (b) Arora, N.; Banerjee, A.K. Mini-Rev. Med. Chem. 2012, 12, 210-226;
(c) Burrows, J. N.; Waterson, D. Top. Med. Chem. 2011, 7, 125-180; (d) Burrows, J.
N.; Chibale, K.; Wells, T. N. C. Curr. Top. Med. Chem. 2011, 11, 1226-1254; (e)
Schlitzer, M.; Ortmann, R. ChemMedChem 2010, 5, 1837-1840.
5.
6.
(a) McRobert, L.; McConkey, G. A. Mol. Biochem. Parasitol. 2002, 119, 273-8; (b)
Phillips, M. A.; Rathod, P. K. Infect. Dis. Drug Targ. 2010, 10, 226-239.
(a) Boa, A. N.; Canavan, S. P.; Hirst, P. R.; Ramsey, C.; Stead, A. M. W.; McConkey,
G. A. Bioorg. Med. Chem. 2005, 13, 1945-1967; (b) Hurt, D. E.; Sutton, A. E.; Clardy,
J. Bioorg. Med. Chem. Lett. 2006, 16, 1610-1615.
7.
(a) Baldwin, J., Michnoff, C. H.; Malmquist, N. A.; White, J.; Roth. M. G.; Rathod, P.
K.; Phillips, M. A. J. Biol. Chem. 2005, 280, 21847-21853; (b) Patel, V.; Booker, M.;
Kramer, M.; Ross, L.; Celatka, C. A.; Kennedy, L. M.; Dvorin, J. D.; Duraisingh, M.
T.; Sliz, P.; Wirth, D. F.; Clardy, J. J. Biol. Chem. 2008, 283, 35078-35085 (c) Booker,
M. L.; Bastos, C. M.; Kramer, M. L.; Barker, R. H.; Skerlj, R.; Sidhu, A. B.; Deng, X.;

Celatka, C.; Cortese, J. F.; Guerrero Bravo, J. E.; Crespo Llado, K. N.; Serrano, A. E.;
Angulo-Barturen, I.; Jiménez-Díaz, M. B.; Viera, S.; Garuti, H.; Wittlin, S.;
Papastogiannidis, P.; Lin, J. W.; Janse, C. J.; Khan, S. M.; Duraisingh, M.; Coleman,
B.; Goldsmith, E. J.; Phillips, M. A.; Munoz, B.; Wirth, D. F.; Klinger, J. D.; Wiegand,
R.; Sybertz, E. J. Biol. Chem. 2010, 285, 33054-33064.
8.
9.
(a) Heikkilä, T.; Thirumalairajan, S.; Davies, M.; Parsons, M. R.; McConkey, G. A.;
Fishwick, C. W. G.; Johnson, A. P. Bioorg. Med. Chem. Lett. 2006, 16, 88-92; (b)
Davies, M.; Heikkilä, T.; McConkey, G.A.; Fishwick, C. W. G.; Parsons, M. R.;
Johnson, A.P. J. Med. Chem. 2009, 52, 2683-2693.
(a) Hurt, D. E.; Widom, J.; Clardy, J.; Acta Crystallogr., Sect. D: Biol. Crystallogr.
2006, 62, 312-23; (b) Deng, X.; Gujjar, R.; El Mazouni, F.; Kaminsky, W.; Malmquist,
N. A.; Goldsmith, E. J.; Rathod, P. K.; Phillips, M. A. J. Biol. Chem. 2009, 284,
26999-27009.
10. (a) Heikkilä, T.; Ramsey, C.; Davies, M.; Galtier, C.; Stead, A. M. W.; Johnson, A. P.;
Fishwick, C. W. G.; Boa, A. N.; McConkey, G. A. J. Med. Chem. 2007, 50, 186–191;
(b) Cowen, D.; Bedingfield, P.; McConkey, G. A.; Fishwick, C. W. G.; Johnson, A. P.
Bioorg. Med. Chem. Lett. 2010, 20, 1284–1287.
11. Fritzson, I.; Bedingfield, P. T. P.; Sundin, A. P.; McConkey, G.; Nilsson, U. J. Med.
Chem. Comm. 2011, 2, 895-898.
12. (a) Coteron, J. M.; Marco, M.; Esquivias, J.; Deng, X.; White, K. L.; White, J.; Koltun,
M.; El Mazouni, F.; Kokkonda, S.; Katneni, K.; Bhamidipati, R.; Shackleford, D. M.;
Angulo-Barturen, I.; Ferrer, S. B.; Jiménez-Díaz, M. B.; Gamo, F. J.; Goldsmith, E. J.;
Charman, W. N.; Bathurst, I.; Floyd, D.; Matthews, D.; Burrows, J. N.; Rathod, P. K.;
Charman, SA.; Phillips, M. A. J. Med. Chem. 2011, 54, 5540-5561; (b) Gujjar, R.;
Marwaha, A.; El Mazouni, F.; White, J.; White, K. L.; Creason, S.; Shackleford, D. M.;

Baldwin, J.; Charman, W. N.; Buckner, F. S.; Charman, S.; Rathod, P. K.; Phillips, M.
A. J. Med. Chem. 2009, 52, 1864-1872; (c) Phillips, M. A.; Gujjar, R.; Malmquist, N.
A.; White, J.; El Mazouni, F.; Baldwin, J.; Rathod, P. K. J. Med. Chem. 2008, 51,
3649-3653.
13. (a) Ashley, E. Lancet Infect. Dis. 2017, dx.doi.org/10.1016/S1473-3099(17)30172-X;
(b) McCarthy, J. S.; Lotharius, J.; Rückle, T.; Chalon, S.; Phillips, M. A.; Elliott, S.;
Sekuloski, S.; Griffin, P.; Ng, C. L.; Fidock, D. A.; Marquart, L.; Williams, N. S.;
Gobeau, N.; Bebrevska, L.; Rosario, M.; Marsh, K.; Möhrle, J. J. Lancet Infect. Dis.
2017, dx.doi.org/10.1016/S1473-3099(17)30172-X; (c) Sulyok, M. ; Rückle, T.; Roth,
A.; Mürbeth, R. E.; Chalon, S.; Kerr, N.; Schnieper Samec, S.; Gobeau, N.; Lamsfus
Calle, C.; Ibáñez, J.; Sulyok, Z.; Held, J.; Gebru, T.; Granados, P.; Brückner, S.;
Nguetse, C.; Mengue, J.; Lalremruata, A.; Sim, K. L.; Hoffman, S. L.; Möhrle, J. J.;
Kremsner, P. G.; Mordmüller, B., Lancet Infect. Dis. 2017,
http://dx.doi.org/10.1016/S1473-3099(17)30139-1.
14. Baumgartner, R.; Walloschek, M.; Kralik, M.; Gotschlich, A.; Tasler, S.; Mies, J.;
Leban, J. J. Med. Chem. 2006, 49, 1239-1247.
15. Typical experimental procedure: Preparation of (E/Z)-ethyl 3-(9'-ethyl-9H-carbazol-3'-
ylamino)-2-nitroacrylate (2e). A mixture of ethyl nitroacetate (3.99 g, 30 mmol) and
triethylorthoformate (30 cm3) in toluene (40 cm3) was heated under reflux in a nitrogen
atmosphere for 2.5 h. After cooling 3-amino-9-ethylcarbazole (5.00 g, 24 mmol) was
added to this mixture, which was stirred for 10 minutes, to ensure thorough mixing,
before being heated under reflux for a further period of 1 hour. The reaction mixture
was then cooled and left to stand overnight. The resulting solid was collected by
filtration and was washed carefully with toluene (2 × 20 cm3). The resulting orange-red
solid was recrystallised from ethanol to yield the title compound as bright red crystals

(4.63 g, 55%). A sample suitable for X-ray diffraction analysis was prepared by
recrystallisation from ethyl acetate. 1H NMR (CDCl3, 400 MHz) for this 1:1 mixture of
geometric isomers, both chemical shifts are given when clearly resolved. δ 11.52
[11.24] (1H, bd, J ~14.5 [~14.7] Hz, =CH-NH), 9.21 [8.66] (1H, d, J 14.5 [14.7] Hz,
=CH-NH), 8.12-8.09 (1H, m, ArH), 7.95 (1H, dd, J 8.8, 2.2 Hz ArH), 7.57-7.52 (1H,
m, ArH), 7.47-7.43 (2H, m, ArH), 7.37-7.28 (2H, m, ArH), 4.45 (2H, q, J 7.1 Hz,
OCH2 or NCH2), 4.42-4.36 (2H, m, NCH2 or OCH2), 1.48-1.43 (3H, m, CH2CH3), 1.41
(3H, t, J 7.1 Hz, CH2CH3); 13C-NMR (CDCl3, 100 MHz) δ 164.99 [161.48], 149.82
[146.00], 140.73 [140.70], 138.60 [138.37], 130.07 [129.88], 126.86 [126.77], 123.74
[123.72], 122.09 [122.06], 120.71, 119.53 [119.45], 118.28 [117.38], 117.00 [116.69],
110.60 [110.06], 109.58 [109.54], 108.99 [108.94], 61.52 [61.35], 37.79 [37.76], 14.37
[14.28], 13.78; MS (ESI) m/z 354.1449, [C19H19N3O4+H]+ requires 354.1448. Also
+
+
observed were 724.3085 (M2NH4 , calc. 724.3089) and 1077.4459 (M3NH4 , calc.
1077.4465); Anal. (C19H19N3O4) Found: C, 64.74 %; H, 6.41 %; N, 11.70 %. Calcd. C,
64.58 %; H, 5.42 %; N, 11.89 %. In a similar fashion (E/Z)-ethyl 3-(dibenzothiophen-
2'-ylaminomethylene)-2-nitroacrylate (3e) was prepared from ethyl nitroacetate (0.45 g,
7.5 mol) and 3-aminodibenzothiophene (0.5 g, 2.5 mol) to give the title compound as a
lemon yellow, crystalline solid (0.61 g, 71%). 1H NMR (CDCl3, 400 MHz) δ 11.37
[11.18] (1H, bd, J ~14.3 [~14.3] Hz, =CH-NH), 9.22 [8.68] (1H, d, J 14.3 [14.3] Hz,
=CH-NH), 8.20-8.16 (1H, m, ArH), 7.99 (1H, dd, J 8.5, 2.4 Hz ArH), 7.93 (1H, d, J 8.5
Hz, ArH), 7.91-7.86 (2H, m, ArH), 7.56-7.51 (2H, m, ArH), 7.39-7.33 (1H, m, ArH),
4.46 [4.39] (2H, q, J 7.2 Hz, OCH2), 1.45 [1.41] (3H, t, J 7.2 Hz, CH2CH3); 13C-NMR
(d6-DMSO, 100 MHz, at 24.1 °C) 139.5, 136.6, 136.1, 136.0, 134.6, 127.6, 124.9,
123.9, 123.2, 122.4, 119.3, 112.5 (broad), 60.8, 14.2; MS (EI) m/z 342 (M+), 326, 252,

235, 222, 183, 139; Anal. (C17H14N2O4S) Found: C, 59.48 %; H, 4.25 %; N, 8.20%.
Calcd. C, 59.64 %; H, 4.12 %; N, 8.18 %.
16. Reaction of o-toluidine with 6a gave a product with an NMR indicating a 1:9 ratio of
doubled signals [δ 7.45 ppm vs 10.89 ppm for the NH, and δ 8.36 ppm vs 7.87 ppm for
the CH]. This however is due to the presence of rotamers about the N-Ar bond due to
the ortho methyl group. Crystallization of this derivative is induced by cooling a
hexane solution, rather than dissolution in boiling ethanol, as used for 2f and 3f. The
heat presumably encourages any interconversion of isomers in the latter case.
17. CCDC 857094 contains the supplementary crystallographic data for the red crystalline
form of (Z)-2e at 150 K. These data can be obtained free of charge from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44)
1223-336-033; e-mail: deposit@ccdc.cam.ac.uk or via
http://www.ccdc.cam.ac.uk/conts/retrieving.html.
18. CCDC 1550062 contains the supplementary crystallographic data for the orange
crystalline form of 2e at 150 K. These data can be obtained free of charge from the
Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK;
fax: (+44) 1223-336-033; e-mail: deposit@ccdc.cam.ac.uk or via
http://www.ccdc.cam.ac.uk/conts/retrieving.html.