The ferroquine antimalarial conundrum: Redox activation and reinvasion inhibition

Article abstract of DOI:10.1002/anie.201303690

Metal health: Ferroquine is a ferrocene-based analogue of the antimalarial drug chloroquine. In addition to the primary mechanism of quinoline action, fluorescent probe studies in infected red blood cells show another mechanism is at work. It is based on the production of HO^. in the acidic and oxidizing environment of the digestive vacuole of the malaria parasite and implies that, with ferroquine, reinvasion can be inhibited. Copyright

Full text of DOI:10.1002/anie.201303690

Angewandte
Chemie
Communications
Bioorganometallics
Metal health: Ferroquine is a ferrocene-
based analogue of the antimalarial drug
chloroquine. In addition to the primary
mechanism of quinoline action, fluores-
cent probe studies in infected red blood
cells show another mechanism is at work.
It is based on the production of HOC in
the acidic and oxidizing environment of
the digestive vacuole of the malaria
parasite and implies that with ferroquine
reinvasion can be inhibited.
F. Dubar, C. Slomianny, J. Khalife, D. Dive,
H. Kalamou, Y. Guꢀrardel, P. Grellier,
C. Biot*
&&&& — &&&&
The Ferroquine Antimalarial Conundrum:
Redox Activation and Reinvasion
Inhibition
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
DOI: 10.1002/anie.201303690
Bioorganometallics
The Ferroquine Antimalarial Conundrum: Redox Activation and
Reinvasion Inhibition**
Faustine Dubar, Christian Slomianny, Jamal Khalife, Daniel Dive, Hadidjatou Kalamou,
Yann Guꢀrardel, Philippe Grellier, and Christophe Biot*
The idea of using organometallic complexes in drug discovery
is becoming more and more popular among the medicinal
[
1–8]
chemistry community.
Indeed, a number of organometallic
[9]
compounds are presently undergoing evaluation in cancer
and infectious disease
[
10]
[11]
clinical trials.
Out of those,
ferrocene–pharmacophore conjugates have paved the way
for overcoming the problem of resistance. Ferrocenyl com-
pounds are stable, lipophilic and redox active, and are usually
[
7]
devoid of serious toxicity.
One of the most advanced projects is ferroquine (FQ,
SSR97193, Figure 1), a ferrocenyl analogue of the antima-
larial chloroquine (CQ, Figure 1), in the Phase II Sanofi
portfolio for uncomplicated Plasmodium falciparum (P.
Figure 1. Structural formula and antimalarial activities of chloroquine
CQ) and Ferroquine (FQ) on CQ-susceptible (3D7 and HB3) and CQ-
resistant (FcB1) strains. (See Table S1 for more details)
(
[
12–14]
falciparum) malaria.
As no drug produces a single effect, there are several
impediments to understanding the mechanisms of action
challenge is the identification and localization of reactive
oxygen species production when treated by FQ in comparison
with CQ, and the impact on the parasite reinvasion.
Herein, we investigate the oxidative conditions inside
P. falciparum and the enhancement or inhibition of P. falci-
parum erythrocyte reinvasion, so as to understand the
mechanism of action of FQ.
(
MOA) of the drug (or drug candidate) and the elucidation of
the site of action or the nature of the drug–cell interaction.
Previously, we established that FQ selectively targets the
digestive vacuole (DV) of the P. falciparum parasite through
other transport mechanism(s) than those involved for the
[15]
organic parent drug CQ. Later, we found that CQ-resistant
parasites treated by FQ accumulate a sulfur-containing
compound, likely glutathion (GSH), in their DV suggesting
During its intra-erythrocytic phase, the parasite digests
[
17,18]
hemoglobin as its main source of amino acids
and to
[
16]
[19]
the production of reactive oxygen species (ROS). The next
preserve the osmotic integrity of the host. The process of
degradation of hemoglobin generates not only a high amount
of ferriprotoporphyrin IX (FP) but also a variety of reactive
oxygen species, such as hydrogen peroxide, superoxide anion,
singlet oxygen, and hydroxyl radical (HOC). Several analytical
approaches have been used to estimate oxidative stress in
[*] Dr. Y. Guꢀrardel, Prof. C. Biot
Universitꢀ Lille1, Unitꢀ de Glycobiologie Structurale et Fonction-
nelle, CNRS UMR 8576
IFR 147, 59650 Villeneuve d’Ascq Cedex (France)
E-mail: christophe.biot@univ-lille1.fr
[
20,21]
different stages of P. falciparum.
However, to our knowl-
Dr. F. Dubar
edge, ratiometric fluorescent probes which can be used for
specific (and not general) intracellular reactive oxygen
species detection have never been reported. The limitation
was mainly due to the necessary requirements of such
responsive probes: concentration, excitation intensity,
absence of interaction with the cellular environment.
Universitꢀ Lille1, Unitꢀ de Catalyse et Chimie du Solide—UMR
CNRS 8181
B.P. 90108, 59652 Villeneuve d’Ascq Cedex (France)
Dr. C. Slomianny
Universitꢀ Lille1, Inserm U1003—Laboratoire de Physiologie Cel-
lulaire
Bꢁtiment SN3, 59655 Villeneuve d’Ascq Cedex (France)
Recently, a ratiometric fluorescent probe 1 (Figure 2) has
been successfully developed for detection of intracellular
Dr. J. Khalife, Dr. D. Dive, H. Kalamou
CIIL, Inserm U 1019, UMR CNRS 8024 Universitꢀ Lille Nord de 15
France, Institut Pasteur de Lille
1
rue du Pr Calmette, 59019 Lille Cedex (France)
Prof. P. Grellier
National Museum of Natural History, UMR 7245 CNRS-MNHN,
Team APE, CP 52
57 Rue Cuvier, Paris 75005 (France)
[**] This study was funded by a grant from the Ministꢂre de
l’Enseignement Supꢀrieur to F.D.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201303690.
Figure 2. Ratiometric fluorescent probe with high selectivity for hydrox-
yl radicals.
2
www.angewandte.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
[
22]
HOC. The hybrid coumarine–cyanine dye 1 has
a maximum emission at 495 nm in its reductive
form whereas it shifts to 651 nm in its oxidative
form 2 (Figure 2). Furthermore, a 210-fold varia-
tion in the emission ratios (I651/I495) at two different
wavelengths offers
a
very sensitive probe.
Although this probe has already been used in
cellular imaging in the relatively large HeLa cells
(
10–30 mm widths and 40–70 mm lengths), its use in
malaria infected red blood cells (RBCs, approx-
imately 6–8 mm) is unprecedented.
We have used probe 1 to assess HOC production
at the trophozoite and schizont stages in CQ-
susceptible (3D7) and CQ-resistant (FcB1) P. fal-
ciparum strains, and to examine the influence of
FQ (in comparison to CQ) in generating HOC.
DAPI (4’,6-diamidino-2-phenylindole) was also
used to stain the parasite nuclei.
As expected, without probe 1 uninfected red
blood cells gave no fluorescence (see Figure S2 in
the Supporting Information). A similar result was
obtained when uninfected red blood cells were
incubated with 1 indicating that uninfected red
blood cells do not produce a detectable amount of
HOC. We then proceeded to examine the potential
usefulness of 1 for HOC imaging in P. falciparum
infected red blood cells (iRBCs). In the absence of
Figure 3. a)–d) Red blood cells infected by the CQ-susceptible strain (3D7) pre-
treated by FQ (40 nm, 30 min), fixed and stained with DAPI; e)–h) red blood cells
infected by the CQ-susceptible strain (3D7) pre-treated by FQ (40 nm, 30 min),
incubated with 1 (40 mm for 5 min), fixed and stained with DAPI; i)–l) red blood cells
1, no fluorescence was detected in both strains (see
Figure S3). After incubation with 1, both strains
exhibited fluorescence at the green and red wave- infected by the CQ-resistant strain (FcB1) pre-treated by FQ (40 nm, 30 min), fixed
lengths. Microscopic observations (see Figure S4) and stained with DAPI; m)–p) red blood cells infected by the CQ-resistant strains
(
FcB1) pre-treated by FQ (40 nm, 30 min), incubated with 1 (40 mm for 5 min), fixed
not only demonstrated the transport of the dye
across the red blood cell and parasite membranes,
but also enabled for the first time the visualization
of HOC in the cytosol and the digestive vacuole of
trophozoite and schizont stages. As suspected, HOC
and stained with DAPI. Bright-field images (a, e, I, m) and corresponding DAPI
fluorescence emission (b, f, j, n), green fluorescence emission (c, g, k, o), and red
fluorescence emission (d, h, l, p).
1
is thus one of the reactive oxygen species generated by the
mitochondrial respiratory chain and hemoglobin digestion. To
avoid any misinterpretation owing to the absorption proper-
ties of hemozoin (Hz) crystals, calculations of the intensity
ratios (I651/I495) has been performed close to Hz but in an area
of the digestive vacuole devoid of Hz so as to minimize this
phenomenon (see Figure S7). I651/I495 ratios were approxi-
mately 10-fold higher in the cytosol than in the digestive
vacuole at the trophozoite stage but this difference is much
less at the schizont stage for which global OHC production
decreases significantly (see Table S6). We obtained the first
complete map of HOC compartmentalization in P. falciparum.
After drug treatment (FQ 40 nm, 30 min, Figure 3 or CQ,
see Figure S5), we observed only small changes of the value of
the I651/I495 ratios (see Tables S6 and S11) at the trophozoite
stage in the cytosol for both strains suggesting that redox
properties of FQ has no influence over the HOC production
within this cellular compartment (Figure 4). On the contrary,
a significant difference between FQ and CQ was observed in
the parasite digestive vacuole. Indeed, the addition of CQ
decreased the I651/I495 ratios (in comparison to the I651/I495
ratios calculated in untreated red blood cells) whereas the
addition of FQ resulted in a 6 to 10-fold increase in the I651/I495
ratios according to the clone studied. In other words, the
purely organic drug CQ exerts an antioxidant effect by
quenching HOC whereas the organometallic FQ increases
the generation of HOC.
[23]
At the schizont stage (Figure 4), the addition of CQ led to
a 5 to 16-fold increase of the I651/I495 ratios (in comparison to
the I651/I495 ratios in untreated parasites) in the cytosol (see
Tables S6 and S11). Nevertheless, no significant difference
was observed in the digestive vacuole (see Tables S6 and S11)
So, CQ exerts no influence or, as in cytoplasm, a quenching
effect on HOC production in its site of action. Moreover,
recently, Egan et al. have shown that after CQ treatment,
a high level of iron was detected in the cytosol compared to in
the digestive vacuole, suggesting a free heme diffusion across
[
24]
the digestive vacuole membrane into the cytosol. These
data suggest that heme released, through the action of CQ, is
able to accumulate in the cytosol where it can react with H O
2
2
and form HOC, thus explaining the increase in the level of
these radicals in the cytosol compared to the digestive
vacuole. After FQ treatment, although the level of I651/I495
ratios are similar in the cytosol as those with to CQ, the FQ
gave a 17-fold increase of I651/I495 ratios in the digestive
vacuole (compared to the I651/I495 ratios in controls), contrary
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
for more than 120 h. Indeed, the effect of FQ on parasite
growth during the subsequent cycles when the parasite is no
longer exposed to FQ was not known.
Figure 5 clearly shows the absence of parasites after one
cycle when the schizonts were pretreated by FQ whereas
parasitaemia reached more than 2% when the cultures were
Figure 5. Evolution of the parasitaemia as a function of time of
cultures of synchronized parasites at the trophozoite-schizont stages
of the CQ-resistant strain FcB1 pretreated during 5 h with 25 and
Figure 4. Quantification of the I651/I495 ratio in the cytosol (top) and
the digestive vacuole (bottom) of P. falciparum parasites (CQ-suscep-
tible strain 3D7 and CQ-resistant strain FcB1) without drug (control)
or after drug treatment (CQ or FQ) at the trophozoite (T) and young
schizont (YS) stages.
100 nm FQ and CQ, or in the absence of drug (control).
pretreated by CQ under the same conditions. We hypothe-
sized that the oxidative stress due to FQ occurring in the
trophozoite and schizont stages also alters the merozoite
formation, thus inhibiting reinvasion. Similar results were
obtained with pretreated schizonts of the CQ-susceptible
strain HB3 (see Figure S8 and S9).
In addition to a drastic reduction of parasitemia, a signifi-
cant difference of parasite stage distribution is observed
between FQ and CQ or controls 24 h after treatment (Fig-
ure S9 and S10). CQ treatment does not affect parasite
reinvasion and new ring and young trophozoites stages are
observed as for the controls. In contrast, after FQ treatment,
the parasites able to reinvade are unable to develop as
demonstrated by the small number of parasites with a typical
young trophozoite morphology and the increasing number of
parasites with pycnotic appearance. All parasites were dead
48 h after the pretreatment.
to CQ. These data suggest that the level of free heme release
during the hemozoin inhibition is not sufficient to explain the
high levels of HOC in the digestive vacuole. Once again, only
the organometallic FQ is clearly able to induce an additional
oxidative stress in the parasite which in turn disrupts the
intracellular redox balance. The parasitic antioxidant defense
system is then overflowed which results in membrane
alterations leading to the death of the parasite.
Enhancement by FQ of the HOC formation in the digestive
vacuole is not yet documented, but it probably involves
a mechanism similar to that of the Fenton reaction. Note that
other ferrocene-based compounds have been shown to
generate H O by autooxidation, forming the ferricinium
2
2
[25,26]
cation.
FQ thus combines Hz inhibition activity, redox
activity, and antimalarial activity against both CQ-susceptible
and CQ-resistant strains. These results clearly demonstrate
that the HOC increase has a direct role in its antimalarial
action.
Along the Plasmodium erythrocytic life-cycle, the troph-
ozoite undergoes schizogony and develops into a schizont,
later releasing merozoites. Upon rupture of the schizont,
merozoites are released to rapidly invade new red blood cells
and perpetuate the cycle. So, we monitored the influence of
both drugs (FQ and CQ) upon the reinvasion step of the CQ-
resistant strain (FcB1). Highly synchronized parasite cultures
at the old-trophozoite-schizont stages were exposed for 5 h to
While further work will be required to better understand
the effects of FQ upon the reinvasion steps and the develop-
ment of new ring stages, these results provide the first
conclusive evidence for the specificity of an organometallic
drug candidate. Indeed, FQ and CQ are involved in the
[
27]
detoxification process of the parasite.
Therefore, their
action can lead to a free heme accumulation in the parasite,
and thus release of HOC. However, the level of these radicals
is much higher in the presence of FQ than CQ, demonstrating
that the ferrocene is involved in the production of HOC. We
have previously shown that the efflux phenomenon observed
with CQ in CQ-resistant strains, is not observed with FQ. The
concentration of FQ coupled to its redox properties therefore
explain the greater effectiveness of the FQ, including the CQ-
resistant strains of P. falciparum.
2
5 nm or 100 nm of FQ (or CQ). Exposed red blood cells were
then washed three times and put back in culture with new red
blood cells at a parasitemia of 0.8%. The ability of parasites
to develop and to reinvade was followed up at different times
4
www.angewandte.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
In summary, we report the first successful subcellular
mapping of HOC generation in P. falciparum red blood cells
treated with the drug candidate FQ using a ratiometric
fluorescent probe. The resulting oxidative damage is consis-
tent with the breakdown of the digestive vacuole membrane,
leading then to the death of the parasites. In addition to the
primary mechanism of quinoline action (inhibition of the
heme detoxification process), an important implication is that
the metallocenic moiety is part of the mechanisms of action.
Based on the evidence that different stages of the Plasmo-
dium erythrocytic life cycle (including rings) are targeted, we
conjecture that FQ could play a key role in the inhibition of
merozoites reinvasion. This finding clearly shows the advant-
age of an organometallic-based drug versus the purely organic
parent drug, and can provide the next generation of drugs.
[9] G. Gasser, I. Ott, N. Metzler-Nolte, J. Med. Chem. 2011, 54, 3 –
5.
10] M. Navarro, C. Gabbiani, L. Messori, D. Gambino, Drug
Discovery Today 2010, 15, 1070 – 1078.
11] E. A. Hillard, G. Jaouen, Organometallics 2011, 30, 20 – 27.
12] G. Mombo-Ngoma, C. Supan, M. P. Dal-Bianco, M. A. Missinou,
P.-B. Matsiegui, C. L. Ospina Salazar, S. Issifou, D. Ter-Minas-
sian, M. Ramharter, M. Kombila, et al., Malaria J. 2011, 10:53.
[13] C. Supan, G. Mombo-Ngoma, M. P. Dal-Bianco, C. L. Ospina
Salazar, S. Issifou, F. Mazuir, A. Filali-Ansary, C. Biot, D. Ter-
Minassian, M. Ramharter, et al., Antimicrob. Agents Chemother.
2
[
[
[
2012, 56, 3165 – 3173.
[
[
14] D. Dive, C. Biot, ChemMedChem 2008, 3, 383 – 391.
15] F. Dubar, S. Bohic, C. Slomianny, J.-C. Morin, P. Thomas, H.
Kalamou, Y. Guꢂrardel, P. Cloetens, J. Khalife, C. Biot, Chem.
Commun. 2012, 48, 910 – 912.
16] F. Dubar, S. Bohic, D. Dive, Y. Guꢂrardel, P. Cloetens, J. Khalife,
C. Biot, ACS Med. Chem. Lett. 2012, 3, 480 – 483.
17] M. Aikawa, P. K. Hepler, C. G. Huff, H. Sprinz, J. Cell Biol. 1966,
[
[
[
Received: April 30, 2013
Published online: && &&, &&&&
28, 355 – 373.
18] J. Liu, E. S. Istvan, I. Y. Gluzman, J. Gross, D. E. Goldberg, Proc.
Natl. Acad. Sci. USA 2006, 103, 8840 – 8845.
19] V. L. Lew, T. Tiffert, H. Ginsburg, Blood 2003, 101, 4189 – 4194.
20] Y. Fu, L. Tilley, S. Kenny, N. Klonis, Cytometry Part A 2010, 77,
Keywords: bioorganometallic chemistry · ferroquine · malaria ·
[
[
.
reactive species · subcellular imaging
253 – 263.
[
21] Y. Fu, N. Klonis, C. Suarna, G. J. Maghzal, R. Stocker, L. Tilley,
[
[
[
[
[
1] G. Jaouen, N. Metzler-Nolte, R. Alberto, Medicinal Organome-
Cytometry Part A 2009, 75, 390 – 404.
tallic Chemistry, Springer, Heidelberg, 2010.
[22] L. Yuan, W. Lin, J. Song, Chem. Commun. 2010, 46, 7930.
[23] Y. Miyachi, A. Yoshioka, S. Imamura, Y. Niwa, Ann. Rheum.
Dis. 1986, 45, 244 – 248.
[24] J. M. Combrinck, T. E. Mabotha, K. K. Ncokazi, M. A. Ambele,
D. Taylor, P. J. Smith, H. C. Hoppe, T. J. Egan, ACS Chem. Biol.
2013, 8, 133 – 137.
[25] R. Kovjazin, T. Eldar, M. Patya, A. Vanichkin, H. M. Lander, A.
Novogrodsky, FASEB J. 2003, 17, 467 – 469.
[26] E. Hillard, A. Vessiꢃres, L. Thouin, G. Jaouen, C. Amatore,
Angew. Chem. 2006, 118, 291 – 296; Angew. Chem. Int. Ed. 2006,
45, 285 – 290.
2] E. Alessio, Bioinorganic Medicinal Chemistry, Wiley-VCH,
Weinheim, 2011.
3] M. Navarro, W. Castro, C. Biot, Organometallics 2012, 31, 5715 –
5727.
4] M. Adams, Y. Li, H. Khot, C. De Kock, P. J. Smith, K. Land, K.
Chibale, G. S. Smith, Dalton Trans. 2013, 42, 4677 – 4685.
5] L. Glans, W. Hu, C. Jçst, C. de Kock, P. J. Smith, M. Haukka, H.
Bruhn, U. Schatzschneider, E. Nordlander, Dalton Trans. 2012,
41, 6443 – 6450.
[
6] L. Glans, A. Ehnbom, C. de Kock, A. Martꢀnez, J. Estrada, P. J.
Smith, M. Haukka, R. A. Sꢁnchez-Delgado, E. Nordlander,
Dalton Trans. 2012, 41, 2764 – 2733.
[27] F. Dubar, T. J. Egan, B. Pradines, D. Kuter, K. K. Ncokazi, D.
Forge, J.-F. Paul, C. Pierrot, H. Kalamou, J. Khalife, et al., ACS
Chem. Biol. 2011, 6, 275 – 287.
[
7] D. R. van Staveren, N. Metzler-Nolte, Chem. Rev. 2004, 104,
5931 – 5985.
[
8] C. G. Hartinger, P. J. Dyson, Chem. Soc. Rev. 2009, 38, 391 – 401.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!