Novobiocin–ferrocene conjugates possessing anticancer and antiplasmodial activity independent of HSP90 inhibition

Article abstract of DOI:10.1007/s00775-018-1634-9

Abstract: A series of tailored novobiocin–ferrocene conjugates was prepared in moderate yields and investigated for in vitro anticancer and antiplasmodial activity against the MDA-MB-231 breast cancer line and Plasmodium falciparum 3D7 strain, respectively. While the target compounds displayed moderate anticancer activity against the breast cancer cell line with IC₅₀ values in the mid-micromolar range, compounds 10a–c displayed promising antiplasmodial activity as low as 0.889?μM. Furthermore, the most promising compounds were tested for inhibitory effects against a postulated target, heat shock protein 90 (Hsp90). Graphical abstract: A selection of tailored novobiocin derivatives bearing the organometallic ferrocene unit were synthesized and characterized by common spectroscopic techniques. The target compounds were investigated for in vitro anticancer and antimalarial activity against the MDA-MB-231 breast cancer cell line and Plasmodium falciparum 3D7 strain, respectively.[Figure not available: see fulltext.].

Full text of DOI:10.1007/s00775-018-1634-9

JBIC Journal of Biological Inorganic Chemistry
https://doi.org/10.1007/s00775-018-1634-9
ORIGINAL PAPER
Novobiocin–ferrocene conjugates possessing anticancer
and antiplasmodial activity independent of HSP90 inhibition
Mziyanda Mbaba1 · Jo‑Anne de la Mare2,3,4 · Jason N. Sterrenberg2,3 · Deborah Kajewole2,3
·
Shantal Maharaj2,3 · Adrienne L. Edkins2,3,4 · Michelle Isaacs4 · Heinrich C. Hoppe2,3,4
Setshaba D. Khanye1,4,5
·
Received: 28 September 2018 / Accepted: 4 December 2018
© Society for Biological Inorganic Chemistry (SBIC) 2018
Abstract
A series of tailored novobiocin–ferrocene conjugates was prepared in moderate yields and investigated for in vitro anticancer
and antiplasmodial activity against the MDA-MB-231 breast cancer line and Plasmodium falciparum 3D7 strain, respectively.
While the target compounds displayed moderate anticancer activity against the breast cancer cell line with IC50 values in the
mid-micromolar range, compounds 10a–c displayed promising antiplasmodial activity as low as 0.889 µM. Furthermore,
the most promising compounds were tested for inhibitory efects against a postulated target, heat shock protein 90 (Hsp90).
Graphical abstract A selection of tailored novobiocin derivatives bearing the organometallic ferrocene unit were synthesized
and characterized by common spectroscopic techniques. The target compounds were investigated for in vitro anticancer and
antimalarial activity against the MDA-MB-231 breast cancer cell line and Plasmodium falciparum 3D7 strain, respectively.
Keywords Novobiocin · Coumarin · Ferrocene · Heat shock protein 90 · Hsp90 · Breast cancer · Plasmodium falciparum
Introduction
Electronic supplementary material The online version of this
The development of resistance to clinically approved anti-
article (https://doi.org/10.1007/s00775-018-1634-9) contains
cancer and antimalarial drugs has intensified research
supplementary material, which is available to authorized users.
endeavours towards the search for novel drug targets and
* Mziyanda Mbaba
bioactive compounds. Novobiocin (1) is an antibiotic iso-
lated from Streptomyces bacteria and is a weak inhibitor of
the chaperone, heat shock protein 90 (Hsp90) [1]. Novo-
biocin derivatives have shown promise as anticancer agents
mziyanda.mbaba@gmail.com
* Setshaba D. Khanye
s.khanye@ru.ac.za
Extended author information available on the last page of the article
Vol.:(0123456789)
1 3

JBIC Journal of Biological Inorganic Chemistry
by inhibiting the function of this chaperone [2–4]. Within
the cell, Hsp90 is responsible for the stabilisation and con-
formational regulation of polypeptides to support the attain-
ment of their functional forms [5]. Numerous proteins sta-
bilised by Hsp90, i.e., Hsp90 client proteins including cell
signalling proteins, steroid hormone receptors and regula-
tory kinases, are involved in essential cellular processes [6].
Hence, Hsp90 is crucial for normal cell function as it medi-
ates these processes by maturing the proteins involved. As
a result, the chaperone bears therapeutic potential as a drug
target [7, 8].
potentiating drug resistance [12, 18]. PfHsp90 is consid-
ered to be an appealing drug target in malaria, and the anti-
malarial activity of a group of aminoalcohol-carbazoles
and 7-azaindole compounds is due to their ability to inhibit
this protein [19, 20]. In addition, selected analogues of
the beta-carboline alkaloid harmine that bound PfHsp90
inhibited the growth of P. falciparum W2 strain in vitro
and were active against Plasmodium berghei in vivo
[21]. These compounds signifcantly increased the sen-
sitivity of Plasmodium parasites to the antimalarial agent
dihydroartemisinin [21]. A signifcant number of studies
have demonstrated 1 to modestly suppress the growth of
selected strains of P. falciparum in vitro [22, 23]. Despite
the evidence underpinning the importance of Hsp90 in P.
falciparum growth and the antimalarial activity of 1, there
are no reports in literature regarding the exploration of
PfHsp90 as a therapeutic target of compounds based on
the chemical scafold of 1.
In cancer, Hsp90 is associated with the manifestation of
the oncogenic phenotype [9–12]. It is central to the hall-
marks pertaining the oncogenic phenotype as several of the
corresponding client proteins are associated with each of
these traits [11]. The research groups of Blagg and Alami
have extensively explored the inhibition of the C-terminal
domain of the Hsp90 by organic novobiocin derivatives
and have demonstrated that these induce antiproliferative
efects in cancer cell lines [13–16]. More importantly, these
studies revealed that the replacement of the benzamide
(RHS=right-hand side) and noviose (LHS=left-hand side)
moieties of 1 with simpler structural motifs retains the anti-
cancer efcacy of the resulting compounds [13–16]. Some
examples of tailored novobiocin derivatives with promising
anticancer activity are displayed in Fig. 1 [13–17].
The incorporation of organometallic units into biologi-
cally active scafolds has been hailed as an attractive strategy
in the feld of medicinal chemistry in the quest for novel
compounds with efcacy against various diseases [24–26].
Most auspicious in this application is the organometallic
unit, ferrocene, which has attracted great interest in drug
discovery owing to its appealing medicinal attributes such
as high aromaticity, lipophilicity, chemical stability, favour-
able redox behaviour, ability to form reactive oxygen species
(ROS) and absence of cytotoxicity in human cells [27, 28].
The success of ferrocene in drug discovery is best repre-
sented by ferroquine (FQ, SSR97193), an efcacious chlo-
roquine–ferrocene conjugate currently in phase II clinical
trials for treatment of malaria [29]. Moreover, the pioneering
work of Jaouen and co-workers on anti-tumour ferrocifens
which are modelled on the tamoxifen scafold has led to the
development of a substantial number of ferrocifen type anti-
cancer drugs possessing diferent modes of action against
Similarly, Hsp90 has been shown to play a vital role
in the life-cycle of the parasite, Plasmodium falciparum,
which causes cerebral malaria [9, 10, 12]. Previous stud-
ies by Banumathy and co-workers revealed the signif-
cance of P. falciparum Hsp90 (PfHsp90) in the develop-
ment of the parasite by demonstrating that the N-terminal
Hsp90 inhibitor geldanamycin suppressed the growth of
the parasite in vitro [10]. It has also been postulated that
PfHsp90 mediates pro-survival mechanisms of the parasite
in stressful environments like drug-induced stress, thereby
Fig. 1 Chemical structures of
RHS
novobiocin (1) and its simplifed
OH
OMe
derivatives 2–4 possessing anti-
OH
H
N
H
N
cancer activity. LHS left-hand
side, RHS right-hand side
O
O
O
OMe
O
O
O
O
O
O
O
HO
O
O
O
OH
OH
O
H2N
O
1
Novobiocin ( )
2
KU174 ( )
LHS
O
OH
H
H
N
N
O
O
Fe
S
O
O
O
O
O
O
O
O
3
7TCNA ( )
4
1 3

JBIC Journal of Biological Inorganic Chemistry
various cancer cells and, thus, bear potential of circumvent-
ing resistance [30, 31].
Preparation of compounds: general procedure
for synthesis of 10a–f [15]
Previously, we demonstrated that the incorporation of the
organometallic ferrocene unit into the novobiocin framework
led to compounds such as 4 (Fig. 1) with enhanced activity
against the HCC38 breast cancer cell line and chloroquine-
sensitive 3D7 strain of the P. falciparum parasite [17]. Fur-
thermore, the resultant ferrocenyl novobiocin derivatives
contained phenyl substituents in lieu of the structurally com-
plex noviose appendage on the LHS. In an efort to further
investigate the pharmacological profle of these derivatives,
in this work we explored the potential activity of ferrocenyl
derivatives of 1 containing simpler structural units (e.g.,
benzyl and piperidine) devoid of the complex noviose moi-
ety at C-7 of the coumarin core. We also sought to ascer-
tain if the observed anticancer and antiplasmodial efcacies
of pursued compounds are attributable to inhibition of the
chaperone Hsp90 by screening the inhibitory activity of the
most promising compounds against human and P. falcipa-
rum Hsp90 isoforms, namely HsHsp90β and PfHsp90.
An appropriate volume of diisopropryl azodicarboxy-
late (DIAD) (2.0 eq.) was added to a suspension of a rel-
evant phenolic substrate 8a–b (1.0 eq.), piperidinol (9a–d)
(1.0 eq.) and triphenylphosphine (PPh3) (2.0 eq.) in THF
(20 mL) on ice. The reaction mixture was stirred at 0°C
for 5 h followed by removal of the solvent under reduced
pressure. The resulting orange residue was purifed by silica
gel chromatography (5% MeOH/DCM) to aford the desired
product 10a–f.
7‑(N‑methyl‑4‑piperidinyl)‑O‑3‑ferrocenylcarboxamidocou‑
marin (10a)
7-Hydroxy-3-ferrocenylcarboxamidocoumarin 8a (0.100 g,
0.26 mmol), 9a (0.030 g, 0.26 mmol), PPh3 (0.135 g,
0.51 mmol) and DIAD (0.104 g, 0.51 mmol) aforded 10a as
a red solid (0.051 g, 41%). M.p.: 161–163°C. IR vmax/cm−1:
3355 (N–H, amide), 3107 (C–H, pyrone HC = C), 2935
(C–H, methyl), 1704 (C=O, lactone), 1659 (C=O, amide),
1
Materials and methods
1501 (C = C, aromatic), 1173 (C–O, aromatic ether); H
NMR (600 MHz, CDCl3): δ 8.68 (s, 1H, NH), 8.25 (s, 1H,
−HC=C−), 7.40 (d, J=8.6 Hz, 1H, ArH), 6.89 (dd, J=8.6,
2.4 Hz, 1H, ArH), 6.85 (d, J = 2.4 Hz, 1H, ArH), 4.82 (t,
J=1.9 Hz, 2H, FcH), 4.46 (t, J=1.9 Hz, 2H, FcH), 4.39 (br
s, 1H, CH), 4.26 (s, 5H, FcH), 2.73 (s, 2H, CH2), 2.34 (br
s, 5H, CH2, CH3), 2.09–2.04 (m, 2H, CH2), 1.92–1.87 (m,
2H, CH2); 13C NMR (150 MHz, CDCl3): δ 170.0, 159.5,
159.1, 151.5, 128.8, 123.4, 122.1, 114.6, 113.6, 102.8, 75.4,
71.5 (2C), 70.2 (5C), 68.6 (2C), 53.6 (2C), 46.1, 30.4 (2C),
29.9; HRMS (ESI+) m/z calcd for C26H26FeN2O4: 486.1242;
found: 487.1318 [M+H]+.
Materials
All chemical reagents and solvents used in this study were
sourced from Merck (Pty) Ltd. and were used without fur-
ther purifcation. The progress of each reaction was moni-
tored by analytical thin-layer chromatography (TLC) using
Merck F254 silica gel plates (supported on aluminium sheets)
and the plates were visualised under ultraviolet light (UV
254 and 366 nm) and in an iodine fask. Where necessary,
the crude compounds were purifed by silica gel column
chromatography using Merck Kieselgel 60 Å: 70–230
(0.068–0.2 mm) silica gel mesh.
7‑(N‑methyl‑3‑piperidinyl)‑O‑3‑ferrocenylcarboxamidocou‑
marin (10b)
Instrumentation
7-Hydroxy-3-ferrocenylcarboxamidocoumarin 8a (0.100 g,
0.26 mmol), 9b (0.030 g, 0.26 mmol), PPh3 (0.135 g,
0.51 mmol) and DIAD (0.104 g, 0.51 mmol) aforded 10b
The 1H and 13C NMR spectra were recorded on Bruker Bio-
spin 300, 400 or 600 MHz spectrometers, and were refer-
enced internally using residual solvent signals of DMSO-
as a red solid (0.046 g, 37%). M.p.: 161–163°C. IR vmax
/
1
d6: 2.50 ppm for H NMR and 39.5 ppm for 13C NMR,
cm−1: 3380 (N–H, amide), 3068 (C–H, pyrone HC = C),
2923 (C–H, methyl), 1706 (C = O, lactone), 1663 (C = O,
amide), 1518 (C=C, aromatic), 1237 (C–O, aromatic ether);
1H NMR (600 MHz, CDCl3): δ 8.68 (s, 1H, NH), 8.25 (s,
1H, –HC = C–), 7.40 (d, J = 8.6 Hz, 1H, ArH), 6.92 (dd,
J=8.6, 1.9 Hz, 1H, ArH), 6.89 (d, J=1.6, 1H, ArH), 4.82
(br s, 2H, FcH), 4.46–4.44 (m, 3H, FcH and CH), 4.26 (s,
5H, FcH), 2.91–2.58 (m, 2H, CH2), 2.32–2.17 (m, 5H, CH3
and CH2), 2.05–1.84 (m, 2H, CH2), 1.68–1.54 (m, 2H, CH2);
13C NMR (150 MHz, CDCl3): δ 170.3, 159.8, 159.6, 151.7,
129.1, 123.8, 122.3, 114.8, 113.9, 103.1, 75.7, 73.7, 71.8
1
or CDCl3: 7.26 ppm for H NMR and 77.2 ppm for 13C
NMR at ambient temperature. The high-resolution mass
spectra (HRMS) were recorded on Waters Synapt G2 Mass
Spectrometer (Central Analytical Facility, University of
Stellenbosch) using electron spray ionization (ESI) in the
positive ionization mode, and the IR spectra were recorded
on PerkinElmer Spectrum 100 FT-IR Spectrometer in the
mid-IR range (640–4000 cm−1). The melting points were
determined using the Reichert melting point apparatus and
were uncorrected.
1 3

JBIC Journal of Biological Inorganic Chemistry
(2C), 70.5 (5C), 68.9 (2C), 59.9, 55.8, 46.8, 29.3, 23.2;
HRMS (ESI+) m/z calcd for C26H26FeN2O4: 486.1242; found
487.1328 [M+H]+.
0.51 mmol) and DIAD (0.104 g, 0.51 mmol) aforded 10e
as a red solid (0.055 g, 37%). M.p.: 159–161°C. IR vmax
/
cm−1: 3399 (N–H, amide), 3089 (C–H, pyrone HC = C),
2934 (C–H, methyl), 1692 (C = O, lactone), 1662 (C = O,
amide), 1520 (C = C, aromatic), 1234 (C–O, aromatic
7‑(N‑methyl‑4‑piperidinyl)‑O‑8‑methyl‑3‑ferrocenylcarbox‑
amidocoumarin (10c)
1
ether); H NMR (600 MHz, CDCl3): δ 8.69 (s, 1H, NH),
8.26 (s, 1H, –HC = C–), 7.42 (d, J = 8.6 Hz, 1H, ArH),
6.89 (dd, J = 8.6, 2.4 Hz, 1H, ArH), 6.86 (d, J = 2.3 Hz,
1H, ArH), 4.82 (t, J = 1.8 Hz, 2H, FcH), 4.55–4.51 (m,
1H, CH), 4.47 (t, J = 1.8 Hz, 2H, FcH), 4.26 (s, 5H,
FcH), 3.73–3.69 (m, 2H, CH2), 3.39–3.35 (m, 2H, CH2),
1.97–1.95 (m, 2H, CH2), 1.81–1.77 (m, 2H, CH2), 1.47
(s, 9H, tert-butyl); 13C NMR (150 MHz, CDCl3): δ 170.0,
167.6, 159.4, 158.9, 155.0, 151.4, 128.9, 123.4, 122.1,
114.7, 113.7, 102.8, 80.0 (2C), 77.5, 73.1 (2C), 71.6
(2C), 70.3 (5C), 68.6 (2C), 28.6 (3C), 21.9; HRMS (ESI+)
m/z calcd for C30H32FeN2O6: 572.1610; found 494.1045
[C25H23FeN2O4 + Na]+.
7-Hydroxy-8-methyl-3-ferrocenylcarboxamidocoumarin
8b (0.060 g, 0.15 mmol), 9a (0.017 g, 0.15 mmol), PPh3
(0.078 g, 0.30 mmol) and DIAD (0.060 g, 0.30 mmol)
aforded 10c as a red solid (0.016 g, 22%). M.p.: 161–163°C.
IR vmax/cm−1: 3363 (N–H, amide), 3079 (C–H, pyrone
HC=C), 2921 (C–H, methyl), 1704 (C=O, lactone), 1658
(C = O, amide), 1521 (C = C, aromatic), 1107 (C–O, aro-
1
matic ether); H NMR (600 MHz, CDCl3): δ 8.67 (s, 1H,
NH), 8.27 (s, 1H, –HC=C–), 7.29 (d, J=8.6 Hz, 1H, ArH),
6.87 (d, J = 8.6 Hz, 1H, ArH), 4.85–4.82 (m, 2H, FcH),
4.49–4.43 (m, 3H, FcH and CH), 4.26 (s, 5H, FcH), 2.66 (s,
2H, CH2), 2.38 (s, 2H, CH2) 2.35 (s, 3H, CH3), 2.33 (s, 3H,
CH3), 2.07–1.99 (m, 2H, CH2), 1.95–1.87 (m, 2H, CH2);
13C NMR (150 MHz, CDCl3): δ 169.9, 159.7, 156.9, 149.5,
125.5, 123.9, 121.7, 115.4, 113.6, 110.7, 75.4, 71.5 (2C),
70.2 (5C), 68.5 (2C), 52.5 (2C), 46.4, 30.9 (2C), 29.8, 8.5;
HRMS (ESI+) m/z calcd for C27H28FeN2O4: 500.1398; found
501.1471 [M+H]+.
7‑(N‑Boc‑3‑piperidinyl)‑O‑3‑ferrocenylcarboxamidocou‑
marin (10f)
7-Hydroxy-3-ferrocenylcarboxamidocoumarin 8a (0.100 g,
0.26 mmol), 9d (0.052 g, 0.26 mmol), PPh3 (0.135 g,
0.51 mmol) and DIAD (0.104 g, 0.51 mmol) aforded 10f
7‑(N‑methyl‑3‑piperidinyl)‑O‑8‑methyl‑3‑ferrocenylcarbox‑
amidocoumarin (10d)
as a red solid (0.060 g, 40%). M.p.: 112–114°C. IR vmax/
cm−1: 3384 (N–H, amide), 3056 (C–H, pyrone HC = C),
2926 (C–H, methyl), 1698 (C = O, lactone), 1663 (C = O,
amide), 1518 (C=C, aromatic), 1232 (C–O, aromatic ether);
1H NMR (400 MHz, CDCl3): δ 8.68 (s, 1H, NH), 8.27 (s,
1H, –HC = C–), 7.41 (d, J = 8.6 Hz, 1H, ArH), 7.05 (d,
J=2.2 Hz, 1H, ArH), 6.99 (dd, J=8.6, 2.2 Hz, 1H, ArH),
5.14–4.94 (m, 1H, CH), 4.82 (t, J=1.29 Hz, 2H, FcH), 4.46
(t, J = 1.26 Hz, 2H, FcH), 4.26 (s, 5H, FcH), 4.08–3.95
(m, 2H, CH2), 2.31–2.09 (m, 4H, 2 ×CH2), 2.04–1.94 (m,
2H, CH2), 1.47 (s, 9H, tert-butyl); 13C NMR (100 MHz,
CDCl3): δ 169.8, 159.3, 158.6, 151.0, 128.4, 123.3, 122.0,
114.7, 113.9, 103.9, 102.6, 75.2, 71.3 (2C), 70.1 (5C), 68.5
(2C), 68.4 (3C), 32.7, 28.4 (3C), 23.3 (2C); HRMS (ESI+)
m/z calcd for C30H32FeN2O6: 572.1610; found 373.1765
[C20H15FeNO3 +H]+.
7-Hydroxy-8-methyl-3-ferrocenylcarboxamidocoumarin
8b (0.060 g, 0.15 mmol), 9b (0.017 g, 0.15 mmol), PPh3
(0.078 g, 0.30 mmol) and DIAD (0.060 g, 0.30 mmol)
aforded 10d as a red solid (0.016 g, 38%). M.p.: 100–102°C.
IR vmax/cm−1: 3361 (N–H, amide), 3088 (C–H, pyrone
HC=C), 2925 (C–H, methyl), 1702 (C=O, lactone), 1660
(C = O, amide), 1517 (C = C, aromatic), 1107 (C–O, aro-
1
matic ether); H NMR (300 MHz, CDCl3): δ 8.67 (s, 1H,
NH), 8.27 (s, 1H, –HC=C–), 7.31 (d, J=8.6 Hz, 1H, ArH),
6.97 (d, J=8.7 Hz, 1H, ArH), 4.83 (t, J=1.95 Hz, 2H, FcH),
4.56 (br s, 1H, CH), 4.46 (t, J=1.95 Hz, 2H, FcH), 4.2 (s,
5H, FcH), 3.19–2.80 (m, 2H, CH2), 2.42–2.32 (m, 3H, CH3),
2.30 (s, 3H, CH3), 2.17 (s, 3H, CH3), 1.90–1.80 (m, 3H,
CH3); 13C NMR (150 MHz, CDCl3): δ 170.0, 159.7, 156.7,
149.4, 125.7, 123.8, 121.8, 115.5, 114.0, 111.1, 75.4, 71.5
(2C), 70.2 (5C), 68.6 (2C), 55.3, 53.6, 46.0, 31.1, 29.9, 22.1,
8.6; HRMS (ESI+) m/z calcd for C27H28FeN2O4: 500.1398;
found 501.1474 [M+H]+.
General procedure for synthesis of compounds
10g–h [15]
A relevant NBoc-protected piperidinyl derivative 10e–f
(1.0 eq.) was stirred in 10% TFA/DCM solution (2 mL)
for 12 h at room temperature. The solvents were removed
under reduced pressure to obtain a crude product, which was
purifed by silica gel column chromatography (10% MeOH/
DCM) to aford the desired product 10g–h.
7‑(N‑Boc‑4‑piperidinyl)‑O‑3‑ferrocenylcarboxamidocou‑
marin (10e)
7-Hydroxy-3-ferrocenylcarboxamidocoumarin 8a (0.100 g,
0.26 mmol), 9c (0.052 g, 0.26 mmol), PPh3 (0.135 g,
1 3

JBIC Journal of Biological Inorganic Chemistry
7‑(4‑Piperidinyl)‑O‑3‑ferrocenylcarboxamidocoumarin
(10g)
The toxicity of novobiocin derivatives and the reference
compounds (1 and Paclitaxel) was assessed in this cell line
using a WST-1 cell proliferation kit (Roche, South Africa)
conducted according to the manufacturer’s instructions.
Cells were seeded at equal density (1.2 × 105 cells/mL at
50 μL/well) in a 96-well plate and treated the following
day with a range of compound concentrations (0, 0.32,
1.6, 8, 40, 200, 1000 μM) in triplicate. After 96 h, media-
containing compounds were removed from the wells and
5 μL of a 5-mg/mL WST-1 reagent in 100 μL of medium
was added to each of the wells and incubated for 4 h prior
to reading absorbance at 450 nm in a Powerwave spectro-
photometer (Biotek, South Africa). The dose response and
half-maximal inhibitory concentrations (IC50) were deter-
mined by non-linear regression using GraphPad Prism 4
software.
7-(N-Boc-4-piperininyl)-O-3-ferrocenylcarboxamidoc-
oumarin 10e (0.045 g, 0.079 mmol) and 10% TFA/DCM
solution (2 mL) aforded 10g as a light red solid (0.035 g,
95%). Mp.: 76–80°C. IR vmax/cm−1: 3397 (N–H, amide
and piperidinyl amine), 2931 (C–H, pyrone HC=C), 2708
(C–H, methyl), 1658 (C=O, lactone), 1610 (C=O, amide),
1
1530 (C = C, aromatic), 1131 (C–O, aromatic ether); H
NMR (300 MHz, CDCl3): δ 8.69 (s, 1H, NH), 8.26 (s, 1H,
–HC=C–), 7.45 (d, J=8.6 Hz, 1H, ArH), 6.90 (dd, J=8.6,
2.2 Hz, 1H, ArH), 6.85 (d, J = 1.7 Hz, 1H, ArH), 4.82 (t,
J=1.8 Hz, 2H, FcH), 4.72 (br s, 1H, CH), 4.47 (t, J=1.8 Hz,
2H, FcH), 4.26 (s, 5H, FcH), 3.31 (br s, 4H, 2×CH2), 2.19
(br s, 4H, 2×CH2); 13C NMR (75 MHz, CDCl3): δ 170.0,
159.5, 159.3, 151.4, 128.8, 123.5, 122.1, 114.5, 113.6,
102.8, 75.4, 73.4, 71.5 (2C), 70.2 (5C), 68.5 (2C), 59.6
(2C), 55.5 (2C); HRMS (ESI+) m/z calcd for C25H24FeN2O4:
472.1085; found 473.1156 [M+H]+.
For the mammalian cell line Hsp90 inhibition assay,
MDA-MB-231 cells (5 × 105) were treated with vehicle
control (0.1% DMSO), compounds 10a and 10g (100 µM)
or Hsp90 inhibitors: 17-dimethylaminoethylamino-17-dem-
ethoxygeldanamycin (17-DMAG; Alvespimycin, 10 µM)
or novobiocin (NOV, 100 µM) for 24 h. Equal amounts of
cell lysates were resolved by SDS-PAGE and the levels of
Hsp90, Hsp70 or cyclin-dependent kinase 4 (CDK4) deter-
mined by Western blot analysis according to standard pro-
tocols [32, 33]. Glyceraldehyde 3-phosphate dehydrogenase
(GAPDH) was used as a loading control.
7‑(3‑Piperidinyl)‑O‑3‑ferrocenylcarboxamidocoumarin
(10h)
7-(N-Boc-3-piperininyl)-O-3-ferrocenylcarboxamidoc-
oumarin 10f (0.045 g, 0.079 mmol) and 10% TFA/DCM
solution (2 mL) aforded 10h as a light red solid (0.036 g,
97%). Mp.:>300°C. IR vmax/cm−1: 3385 (N–H, amide and
piperidinyl amine), 2920 (C–H, pyrone HC = C), 1702
(C = O, lactone), 1607 (C = O, amide), 1607 (C = C, aro-
In vitro antiplasmodial assay
1
matic), 1533 (C–O, ether); H NMR (300 MHz, CDCl3):
δ 8.68 (s, 1H, NH), 8.26 (s, 1H, –HC = C–), 7.40 (d,
J = 8.7 Hz, 1H, ArH), 7.04 (d, J= 2.3 Hz, 1H, ArH), 6.98
(dd, J = 8.6, 2.4 Hz, 1H, ArH), 5.89–5.80 (m, 1H, NH),
4.82 (t, J=1.92 Hz, 2H, FcH), 4.46 (t, J=1.92, 2H, FcH),
4.34–4.31 (m, 1H, CH), 4.26 (s, 5H, FcH), 4.06–3.92 (m,
2H, CH2), 2.52–1.92 (m, 6H, 3×CH2); 13C NMR (75 MHz,
CDCl3): δ 170.0, 159.5, 158.8, 151.1, 128.6, 123.5, 122.1,
114.8, 114.0, 102.7, 75.3, 71.5 (2C), 70.2 (5C), 68.6,
68.5 (3C), 32.9, 28.0, 23.5; HRMS (ESI+) m/z calcd for
C25H24FeN2O4: 472.1085; found 473.3404 [M+H]+.
The P. falciparum parasites (3D7 strain) were routinely
cultured in a medium consisting of RPMI1640 containing
25 mM HEPES (Lonza, South Africa), 0.5% (w/v) Albu-
max II (Thermo Fisher Scientifc, South Africa), 22 mM
glucose, 0.65 mM hypoxanthine, 0.05 mg/mL gentamicin
and 2–4% (v/v) human erythrocytes and were maintained at
37°C under an atmosphere of 5% CO2, 5% O2 and 90% N2.
The antiplasmodial activity of test compounds was assessed
using a colorimetric assay as previously described [34].
Biological assays
In vitro HeLa cell cytotoxicity assay
MDA‑MB‑231 breast cancer cell line toxicity and analysis
of Hsp90 inhibition
HeLa cells (Cellonex, South Africa) were cultured in Dul-
becco’s Modifed Eagle’s medium (Lonza, South Africa)
supplemented with 10% foetal calf serum and antibiotics
(penicillin, streptomycin and amphotericin B) in a 5% CO2
incubator maintained at 37°C. Following seeding of HeLa
cells into 96-well plates and incubation for 24 h, the test
compounds were added to a fnal concentration of 20 μM
and cell viability was surveyed using a previously described
resazurin fuorescence assay [35].
The MDA-MB-231 human triple negative breast adeno-
carcinoma cell line (ATCC HTB-26) was maintained
in culture in L-15 media supplemented with 10% (v/v)
heat-inactivated foetal bovine serum (FBS), 1 mM l-glu-
tamine, 100 U/mL penicillin, 100 µg/mL streptomycin and
12.5 µg/mL amphotericin (PSA) at 37 °C and 9% CO2.
1 3

JBIC Journal of Biological Inorganic Chemistry
Yeast assay for Hsp90 inhibition
Results and discussion
Yeast strains lacking Saccharomyces cerevisiae Hsp82
and expressing Homo sapiens Hsp90β (DP584) and P. fal-
ciparum Hsp90 (DP553) were donated by the laboratory
of Prof. Didier Picard (University of Geneva, Switzer-
land) [36]. Strains were grown in YEP broth [10 g/L yeast
extract, 20 g/L peptone, 2% (w/v) glucose and 40 µg/mL
adenine] and on YEP agar [YEP broth supplemented with
2% (w/v) agar] and maintained using standard techniques.
For growth inhibition assays, YEP agar was supplemented
with 1 mM of geldanamycin (GA), novobiocin (1), com-
pound 10a, compound 10g or dimethylsulfoxide (DMSO)
as the vehicle control. Four tenfold dilutions were made
with strains growing at mid-logarithmic phase (OD600
of 0.8). Cell dilutions (undiluted, and 10−1, 10−2, 10−3
and 10−4 dilutions) were spotted onto each of the com-
pound-containing YEP agar plates and incubated for 48 h
(DP584) and 72 h (DP553) at 28 °C before the images
were captured.
Chemistry
As outlined in Scheme 1, the target compounds were efec-
tively prepared from hydroxycoumarins 5a–b, which were
synthesized using methods reported in literature [37, 38].
The benzylated 3-aminocoumarins 6a–b were achieved by
conventional benzylation of the hydroxyl group on C-7 of
hydroxycoumarins 5a–b followed by acidic boc-deprotec-
tion of the benzylated intermediates to liberate the NH2 at
position 3 of the coumarin nucleus [38, 39]. Subsequently,
ferrocenecarboxylic acid was appended to the RHS of the
coumarin core via standard amide coupling reaction condi-
tions to generate amides 7a–b.
Initially, the amidation reaction of 6a–b was performed
with the common coupling agent, 1-ethyl-3-(3-dimethylami-
nopropyl)carbodiimide (EDCI); however, the resulting yields
were too low (10–18%) to progress to subsequent steps [3].
In attempts to improve the yields, we employed a modifed
procedure reported by Dunetz et al. that utilises a peptide
coupling agent, n-propylphosphonic anhydride (T3P®), for
H
N
NH2
O
NHBoc
i, ii
Fe
iii
O
O
O
O
O
O
HO
O
O
R1
R1
R1
5a-b
6a-b
7a-b
iv
H
H
N
N
v
R2
R3
Fe
Fe
1'
O
O
HO
O
O
R2
R3
O
O
O
R1
R1
OH
8a-b
10a-h
R2
9a-d
R1
R3
a
b
H
H
NMe
CH2
CH2
NMe
c
d
e
Me NMe
CH2
NMe
CH2
Me
H
CH2
NBoc
f
g
h
H
H
H
CH2 NBoc
vi
NH
CH2
NH
CH2
Scheme 1 Reagents and conditions: (i) benzyl bromide, K2CO3,
acetone, 56°C, 3 h; (ii) 4 M HCl in dioxane, r.t., 12 h; (iii) Ferro-
cene carboxylic acid, 50% T3P®/EtOAc, 1:2 pyridine/CHCl3, 75°C,
24–48 h; (iv) 10 mol % Pd/C, THF, r.t., 12 h; (v) 9a–d, DIAD, PPh3,
THF, 0°C, 5 h; (vi) 10% TFA/DCM, r.t., 12 h
1 3

JBIC Journal of Biological Inorganic Chemistry
the amidation of epimerization-prone substrates [40]. Thus,
refuxing of equimolar amounts of amines (6a–b) and fer-
rocene carboxylic acid in 1:2 pyridine/chloroform solution
in the presence of two equivalents of the commercially avail-
able 50% T3P®/EtOAc solution yielded compounds 7a–b
in yields of 68–80%, which was a fourfold improvement
compared to the initial EDCI amidation protocol. Subse-
quently, the target benzylated compounds 7a–b were char-
in MDA-MB-231 cells using Western blot analysis [32, 33].
The parent compound novobiocin (1) is not a potent Hsp90
inhibitor, but was included in the biological assays together
with the potent inhibitors 17-dimethylaminoethylamino-
17-demethoxygeldanamycin (17-DMAG; alvespimycin)
and geldanamycin (GA) to determine whether the biologi-
cal responses of the derivatives are similar to novobiocin,
or whether they gain activity more similar to higher afnity
Hsp90 inhibitors.
1
acterized by H and 13C NMR spectroscopic techniques.
From the 1H NMR data, the ferrocene protons of compounds
7a–b appeared in the aliphatic region as two triplets and
one intense singlet, respectively, at δ 4.84–4.47 ppm and
δ 4.26 ppm, confrming successful coupling of ferrocene
carboxylic acid to the coumarin nucleus to form the target
compounds.
Anticancer activity
In the assay for anticancer activity, paclitaxel, a known anti-
cancer drug, was used as a positive control whilst novobiocin
(1) was included in the screening assays for reference pur-
poses. The activity of each compound against the MDA-
MB-231 breast cell line is presented as a concentration of
the compound that suppressed the viability of the cells by
half of their initial population, i.e., half-maximal inhibitory
concentration (IC50) value. The results are summarised in
Table 1.
To access the 7-hydroxy novobiocin derivatives 8a–b,
compounds 7a–b were stirred under hydrogen atmosphere in
THF to remove the benzyl group using 10 mol % palladium
on carbon (Pd/C) as a catalyst [39]. Successful benzyl depro-
tection was confrmed by the disappearance of the signals
corresponding to the benzyl group in both 1H and 13C NMR
spectra of the products, which were obtained as white solids
in 33–96% yields. Having successfully achieved the phenolic
derivatives 8a–d, the next step was to pursue the piperidi-
nyl variants 10a–f which were synthesized by coupling
piperidinols 9a–d to C-7 of the coumarin nucleus under
Mitsunobu reaction conditions [15]. The 3′- and 4′-NMe
and -NBoc piperidinyl derivatives (10a–f) were achieved in
one step and in modest yields (22–41%). Boc-deprotection
of 10e–f by stirring in 10% TFA/DCM solution generated
the corresponding NH variants (10g–h) in excellent yields
(95–97%). The methylene protons of the piperidine skeleton
were observed as multiplets in the aliphatic region of the 1H
NMR spectra of 10a–h, whereas the proton on C-1′ appeared
as a broad singlet or multiplet, thus confrming successful
attachment of the piperidine unit.
As illustrated in Table 1, the anticancer activities of
compounds in the hydroxyl series (8a–b) were the most
potent in the whole library. With the exception of 10h
(IC50 =108.9 µM), the piperidinyl derivatives (10a–g) dis-
played moderate activity with IC50 values ranging between
11 and 51 µM. Within this class, the 3′-N-methyl ana-
logues (10b and 10d) were almost twice as active as their
4′-N-methyl counterparts (10a and 10c), whilst the NBoc
derivatives (10e–f) displayed comparable activities of 29.8
and 30.0 µM, respectively. Zhao et al. noted similar trends
for organic novobiocin analogues bearing the benzamide
chain of 1, which were tested for antiproliferative efects
against SKBr3 and MCF-7 breast cancer cell lines [41]. The
NH group (10g) also appeared to be tolerated for activity.
Methylation of position 8 of the coumarin nucleus is known
to be favourable for anticancer activity of novobiocin ana-
logues [4]. This was similarly observed in our results as the
8-methyl derivatives had slightly lower IC50 values com-
pared to their variants devoid of the methyl group on C-8
(Table 1). Overall, the derivatives pursued in this study were
clearly superior to the parental compound 1 with approxi-
mately fourfold improvement in activity (Table 1). Addi-
tionally, the replacement of the noviose moiety at C-7 with
hydroxy, benzyl and piperidinyl units retained the anticancer
activity of the ferrocenyl novobiocin derivatives [16, 17].
Biological evaluation studies
The synthesized fnal novobiocin derivatives (10a–h) and
key intermediates, the benzyl (7a–d) and hydroxyl (8a–d)
series, were screened for in vitro anticancer and antimalar-
ial activity, respectively, against the breast cancer cell line
MDA-MB-231 and the chloroquine sensitive (CQS) P. fal-
ciparum strain, 3D7. These compounds were also evaluated
for general human cytotoxicity using the HeLa cell line. The
Hsp90 inhibitory activity of the most promising compounds
was investigated by means of a yeast assay employing Sac-
charomyces strains that express Hsp90 isoforms from human
and P. falciparum [36]. Additionally, the efects of these
compounds on the levels of Hsp90, Hsp70 (an indirect meas-
ure of Hsp90 inhibition) and an obligate Hsp90 client pro-
tein, cyclin-dependent kinase 4 (CDK4), were also assessed
Antiplasmodial assay
To establish their antiplasmodial activity, the target com-
pounds and the reference compound, novobiocin (1), were
subjected to the malaria parasite lactate dehydrogenase
(pLDH) assay that was performed in triplicate for each
1 3

JBIC Journal of Biological Inorganic Chemistry
Table 1 Anticancer and antiplasmodial activities of target compounds
against the MDA-MB-231 cancer cell line and 3D7 P. falciparum
strain, respectively
Entry
Compound
R1
R2
R3
R4
IC50 (µM)
MDA-MB-231
3D7
1
2
3
4
5
6
7
8
7a
H
Me
H
Me
H
H
Me
Me
H
H
H
OBn
OBn
OH
OH
piperidine
piperidine
piperidine
piperidine
piperidine
piperidine
piperidine
piperidine
‒
‒
‒
‒
‒
NMe
CH2
NMe
CH2
NBoc
CH2
NH
CH2
‒
‒
‒
‒
‒
CH2
NMe
CH2
NMe
CH2
NBoc
CH2
NH
‒
182.9
41.1
11.7
29.3
51.4
47.9
36.8
13.8
29.8
30.0
11.8
108.9
205.1
0.029
nd
na
na
na
na
7.06
9.16
0.889
na
na
na
na
na
na
nd
7b
8a
8b
10a
10b
10c
10d
9
10e
10
11
12
13
14
15
10f
10g
10h
H
‒
‒
‒
Novobiocin
Paclitaxel
CQ
‒
‒
‒
‒
‒
‒
0.0102
Activity of compounds is reported as IC50 values representing concentrations of the compounds that efected 50% inhibition of cellular growth.
Data are expressed as average µM concentration resulting from at least two independent experiments
nanot active, nd not determined
compound at a concentration of 20 μM. The antiplasmo-
dial screening assay data for the screened compounds is
presented in Fig. S1. The compounds were also investi-
gated for general human cytotoxicity using the HeLa cell
line to determine if the observed antiplasmodial activity
was independent of general cytotoxicity of the compounds.
From Fig. S1, three ferrocenyl derivatives 10a–c, which
all contain N-methyl substituents in the piperidine ring
displayed desirable antiplasmodial activity with P. falci-
parum viability reduced to below 25% without substantial
HeLa cell cytotoxicity. The data suggest that these com-
pounds were selective for the P. falciparum parasite. The
rest of the compounds in this series, including parental
novobiocin (1), were inactive with > 75% P. falciparum
percentage viability often observed (Fig. S1). In addition,
the N-methyl group seemed to be permissible for anti-
plasmodial activity since the three ferrocenyl analogues
(10a–c) with this group exhibited favourable activity.
Following the initial screening data, the most active
piperidinyl derivatives were further screened for antiplas-
modial activity at varying concentrations to determine their
corresponding IC50 values (Table 1, Fig. S2). Compounds
10a–c were screened in conjunction with chloroquine (CQ),
which was employed as a positive control. The compounds
showed moderate activity with IC50 values below 10 µM
(Table 1, Fig. S2). Most importantly, compound 10c dis-
played signifcant potency in the sub-micromolar range with
IC50 value of 0.889 µM.
Hsp90 inhibition studies
Having evaluated the compounds for in vitro anticancer
and antiplasmodial activity, the next step was to investi-
gate if these compounds act through inhibition of Hsp90.
To realize this, Saccharomyces cerevisiae (yeast) strains
expressing Homo sapiens Hsp90β (which is the essential
1 3

JBIC Journal of Biological Inorganic Chemistry
isoform in humans) [32] and P. falciparum Hsp90 isoform
were employed to probe the Hsp90 inhibitory activity of
the compounds showing the most promising anticancer
and antiplasmodial activities. The yeast strains employed
were DP584 (expressing HsHsp90β) and DP553 (express-
ing Pf Hsp90). Compounds 10a and 10g were selected as
they displayed superior activity against the P. falciparum
3D7 strain and the MDA-MB-231 breast cancer cell line,
respectively. The known Hsp90 inhibitor, geldanamycin
(GA), was included as a positive control and the parental
compound novobiocin (1) as a reference. For each com-
pound, diferent yeast densities were plated and incubated
with 1 mM of the test compound dissolved in DMSO. The
efects of the compounds on yeast growth were visually
monitored and the results are presented in Fig. 2a. The
white spots indicate the growth of the yeast, where density
is proportionate to the growth.
comparison less visible than that of the parental compound
1 on both strains.
In addition, the efect of these compounds on the lev-
els of Hsp90, Hsp70 and kinase CDK4 in MDA-MB-231
cells was compared to known Hsp90 inhibitors 17-DMAG
(Alvespimycin) and novobiocin (NOV) (Fig. 2b). CDK4 is
a validated Hsp90 client that undergoes degradation upon
Hsp90 inhibition [42]. Furthermore, Hsp70 levels can be
used as an indirect measure of Hsp90 inhibition due to
activation of the heat shock response by N-terminal Hsp90
inhibitors [43]. Treatment of cells with the Hsp90 inhibi-
tor 17-DMAG resulted in a signifcant loss of the obligate
Hsp90 client protein, CDK4, and resulted in upregulation
in Hsp70 levels without changing Hsp90 levels (Fig. 2b, c).
Treatment of the MDA-MB-231 cell line with compounds
10a and 10g did not signifcantly alter the levels of Hsp70
or CDK4, although 10a appeared to increase the levels of
CDK4, while 10g led to a minor reduction in CDK4 levels
(Fig. 2c).
As can be observed from Fig. 2a, GA repressed the
growth of the yeast for both strains whereas 10a and 10g
exerted efects comparable to DMSO, which was used as a
negative control in the assay. As a modest Hsp90 inhibitor,
novobiocin (1) exhibited poor activity against these strains
(Fig. 2a). Although 10a might appear to be slightly efective
against DP854 and 10g against DP553, these efects were in
Taken together, and considering that these compounds
were signifcantly more potent than 1, against the MDA-
MB-231 breast cancer cell line and the 3D7 P. falciparum
strain, a similar trend would be expected against the pos-
tulated target Hsp90, as investigated in these assays, if the
Fig. 2 Hsp90 inhibition assay results for compounds 10a and 10g.
Analysis of anti-Hsp90 activity of compounds 10a and 10g assessing
a yeast growth assay in strains expressing either Plasmodium falci-
parum or human Hsp90 (DP553 or DP584) in comparison to Hsp90
inhibitors geldanamyin (GA) or novobiocin (NOV). b Western blot
and c densitometry relative to loading control (GAPDH) of levels of
Hsp90, Hsp70 or CDK4 in treated MDA-MB-231 cell lysates com-
pared to Hsp90 inhibitors 17-dimethylaminoethylamino-17-demeth-
oxygeldanamycin (17-DMAG; alvespimycin) or NOV. In all cases,
DMSO was used as the vehicle control. Data are representative of
two independent biological replicates. Statistical signifcance was
determined compared to the DMSO control by two-way ANOVA;
*p<0.05
1 3

JBIC Journal of Biological Inorganic Chemistry
observed activities of these compounds are primarily due
to the inhibition of this protein. The data presented (Fig. 2)
suggest that there is no correlation between the inhibition
of Hsp90 and the observed in vitro anticancer and anti-
plasmodial activities of the ferrocenyl novobiocin deriva-
tives. The increased toxicity of the ferrocenyl novobiocin
derivatives compared to novobiocin could not be explained
by increased anti-Hsp90 activity, suggesting that this class
of compounds may act on a target other than the Hsp90 to
induce the observed activities. In view of the fact that all
the organic novobiocin derivatives reported in literature are
known to exert anticancer activity through Hsp90 inhibition
[44], it is tempting to speculate from our fndings that the
presence of the ferrocene unit in the novobiocin scafold
switches the target of the derivatives from Hsp90.
References
1. Marcu MG, Schulte TW, Neckers L (2000) Novobiocin
and related coumarins and depletion of heat shock pro-
tein 90-dependent signaling proteins. J Natl Cancer Inst
92(3):242–248
2. Burlison JA, Neckers L, Smith AB, Maxwell A, Blagg BS (2006)
Novobiocin: redesigning a DNA gyrase inhibitor for selective
inhibition of hsp90. J Am Chem Soc 128(48):15529–15536
3. Burlison JA, Avila C, Vielhauer G, Lubbers DJ, Holzbeierlein
J, Blagg BS (2008) Development of novobiocin analogues that
manifest anti-proliferative activity against several cancer cell
lines. J Org Chem 73(6):2130–2137
4. Donnelly AC, Mays JR, Burlison JA, Nelson JT, Vielhauer G,
Holzbeierlein J, Blagg BS (2008) The design, synthesis, and eval-
uation of coumarin ring derivatives of the novobiocin scafold that
exhibit antiproliferative activity. J Org Chem 73(22):8901–8920
5. Young JC, Moaref I, Hartl FU (2001) Hsp90: a specialized but
essential protein-folding tool. J Cell Biol 154(2):267–274
6. Zuehlke A, Johnson JL (2010) Hsp90 and co-chaperones twist the
functions of diverse client proteins. Biopolymers 93(3):211–217
7. Holzbeierlein JM, Windsperger A, Vielhauer G (2010) Hsp90:
a drug target? Curr Oncol Rep 12(2):95–101
Conclusions
8. Pallavi R, Roy N, Nageshan RK, Talukdar P, Pavithra SR, Reddy
R, Venketesh S, Kumar R, Gupta AK, Singh RK (2010) Heat
shock protein 90 as a drug target against protozoan infections
biochemical characterization of Hsp90 from Plasmodium fal-
ciparum and trypanosoma evansi and evaluation of its inhibitor
as a candidate drug. J Biol Chem 285(49):37964–37975
9. Kumar R, Musiyenko A, Barik S (2003) The heat shock protein
90 of Plasmodium falciparum and antimalarial activity of its
inhibitor, geldanamycin. Malar J 2(1):30
Herein, we presented the synthesis of novobiocin–ferrocene
conjugates containing simplifed structural units (hydroxyl,
benzyl and piperidine) in place of the complex noviose
motif, which were obtained in moderate yields. The series
was investigated for anticancer and antimalarial activity,
respectively, using the MDA-MB-231 breast cancer cell
line and P. falciparum 3D7 strain. In general, the presence
of the ferrocene unit was found to favour high potency and
selectivity to the novobiocin derivatives evaluated against
the breast cancer cell line and the P. falciparum parasite.
These fndings appeared to assert our previous observa-
tion that the ferrocene unit imparts benefcial biological
efects to novobiocin derivatives. The replacement of the
noviose appendage of novobiocin with hydroxy, benzyl and
piperidine units seemed to retain the efcacy of the ferro-
cenyl derivatives. Furthermore, the in vitro anticancer and
antiplasmodial activity observed for these derivatives was
found to be independent of inhibition of the proposed target,
Hsp90.
10. Banumathy G, Singh V, Pavithra SR, Tatu U (2003) Heat shock
protein 90 function is essential for Plasmodium falciparum growth
in human erythrocytes. J Biol Chem 278(20):18336–18345
11. Peterson LB, Blagg BS (2009) To fold or not to fold: modula-
tion and consequences of Hsp90 inhibition. Future Med Chem
1(2):267–283
12. Shahinas D, Folefoc A, Pillai DR (2013) Targeting Plasmodium
falciparum Hsp90: towards reversing antimalarial resistance.
Pathogens 2(1):33–54
13. Le Bras G, Radanyi C, Peyrat J-F, Brion J-D, Alami M, Marsaud
V, Stella B, Renoir J-M (2007) New novobiocin analogues as anti-
proliferative agents in breast cancer cells and potential inhibitors
of heat shock protein 90. J Med Chem 50(24):6189–6200
14. Radanyi C, Le Bras G, Marsaud V, Peyrat J-F, Messaoudi S,
Catelli M-G, Brion J-D, Alami M, Renoir J-M (2009) Antipro-
liferative and apoptotic activities of tosylcyclonovobiocic acids
as potent heat shock protein 90 inhibitors in human cancer cells.
Cancer Lett 274(1):88–94
Acknowledgements The authors would like to acknowledge Rhodes
University Sandisa Imbewu (SDK, HCH and ALE) and the National
Research Foundation (MM, JNS, ALE and SDK) for fnancial support.
Financial assistance from South African Research Chairs Initiative of
the Department of Science and Technology and National Research
Foundation of South Africa (Grant No 98566) is also acknowledged
(ALE). The South African Medical Research Council (MRC) funded
the antiplasmodial bioassay component of the project with funds from
National Treasury under its Economic Competitiveness and Support
Package.
15. Donnelly AC, Zhao H, Kusuma BR, Blagg BS (2010) Cyto-
toxic sugar analogues of an optimized novobiocin scafold. Med
Chem Commun 1(2):165–170
16. Zhao H, Donnelly AC, Kusuma BR, Brandt GE, Brown D,
Rajewski RA, Vielhauer G, Holzbeierlein J, Cohen MS, Blagg
BS (2011) Engineering an antibiotic to fght cancer: optimi-
zation of the novobiocin scafold to produce anti-proliferative
agents. J Med Chem 54(11):3839–3853
17. Mbaba M, Mabhula AN, Boel N, Edkins AL, Isaacs M, Hoppe
HC, Khanye SD (2017) Ferrocenyl and organic novobiocin
derivatives: synthesis and their in vitro biological activity. J
Inorg Biochem 172:88–93
Compliance with ethical standards
18. Raghavan PS, Kumar R, Tatu U (2007) Systems analysis of
chaperone networks in the malarial parasite Plasmodium falci-
parum. PLoS Comput Biol 3:e168
Conflict of interest The authors declare that they have no confict of
interest.
1 3

JBIC Journal of Biological Inorganic Chemistry
32. Laemmli UK (1970) Cleavage of structural proteins dur-
ing the assembly of the head of bacteriophage T4. Nature
227(5259):680–685
19. Wang T, Bisson WH, Mäser P, Scapozza L, Picard D (2014)
Diferences in conformational dynamics between Plasmodium
falciparum and human Hsp90 orthologues enable the structure-
based discovery of pathogen-selective inhibitors. J Med Chem
57(6):2524–2535
33. Towbin H, Staehelin T, Gordon J (1979) Electrophoretic trans-
fer of proteins from polyacrylamide gels to nitrocellulose sheets:
procedure and some applications. Proc Natl Acad Sci USA
76(9):4350–4354
20. Wang T, Mäser P, Picard D (2016) Inhibition of Plasmodium
falciparum Hsp90 contributes to the antimalarial activities of
aminoalcohol-carbazoles. J Med Chem 59(13):6344–6352
21. Bayih AG, Folefoc A, Mohon AN, Eagon S, Anderson M, Pillai
DR (2016) In vitro and in vivo anti-malarial activity of novel
harmine-analog heat shock protein 90 inhibitors: a possible part-
ner for artemisinin. Malar J 15(1):579
34. Lunga MJ, Chisango RL, Weyers C, Isaacs M, Taylor D, Edkins
AL, Khanye SD, Hoppe HC, Veale C (2018) Expanding the
SAR of non-toxic antiplasmodial indolyl-3-ethanone ethers and
thioethers. ChemMedChem 13:1353–1362
35. Oderinlo OO, Tukulula M, Isaacs M, Hoppe HC, Taylor D, Smith
VJ, Khanye SD (2018) New thiazolidine-2, 4-dione derivatives
combined with organometallic ferrocene: Synthesis, structure and
antiparasitic activity. Appl Organomet Chem 32(7):e4385
36. Wider D, Péli-Gulli M-P, Briand P-A, Tatu U, Picard D (2009)
The complementation of yeast with human or Plasmodium fal-
ciparum Hsp90 confers diferential inhibitor sensitivities. Mol
Biochem Parasitol 164(2):147–152
22. Divo A, Sartorelli A, Patton C, Bia F (1988) Activity of fuo-
roquinolone antibiotics against Plasmodium falciparum in vitro.
Antimicrob Agents Chemother 32(8):1182–1186
23. Ram ER, Kumar A, Biswas S, Kumar A, Chaubey S, Siddiqi MI,
Habib S (2007) Nuclear gyrB encodes a functional subunit of the
Plasmodium falciparum gyrase that is involved in apicoplast DNA
replication. Mol Biochem Parasitol 154(1):30–39
37. Mendelson WL, Hayden S (1996) Preparation of 2,4-dihydroxy-
benzaldehyde by the Vilsmeier-Haack Reaction. Synth Commun
26(3):603–610
24. Chavain N, Biot C (2010) Organometallic complexes: new tools
for chemotherapy. Curr Med Chem 17(25):2729–2745
25. Gasser G, Metzler-Nolte N (2012) The potential of organome-
tallic complexes in medicinal chemistry. Curr Opin Chem Biol
16(1–2):84–91
38. Kudale AA, Kendall J, Warford CC, Wilkins ND, Bodwell GJ
(2007) Hydrolysis-free synthesis of 3-aminocoumarins. Tetrahe-
dron Lett 48(29):5077–5080
26. Hartinger CG, Metzler-Nolte N, Dyson PJ (2012) Challenges and
opportunities in the development of organometallic anticancer
drugs. Organometallics 31(16):5677–5685
39. Nicolaou K, Bunnage ME, Koide K (1994) Total synthesis of
balanol. J Am Chem Soc 116(18):8402–8403
40. Dunetz JR, Xiang Y, Baldwin A, Ringling J (2011) General and
scalable amide bond formation with epimerization-prone sub-
strates using T3P and pyridine. Org Lett 13(19):5048–5051
41. Zhao H, Reddy Kusuma B, Blagg BS (2010) Synthesis and evalu-
ation of noviose replacements on novobiocin that manifest anti-
proliferative activity. ACS Med Chem Lett 1(7):311–315
42. Theodoraki MA, Caplan AJ (2012) Quality control and fate
determination of Hsp90 client proteins. BBA Mol Cell Res
1823(3):683–688
27. Patra M, Gasser G (2017) The medicinal chemistry of ferrocene
and its derivatives. Nat Rev Chem 1(9):0066
28. Kowalski K (2018) Recent developments in the chemistry of fer-
rocenyl secondary natural product conjugates. Coord Chem Rev
366:91–108
29. To evaluate the efcacy of a single dose regimen of ferroquine
and artefenomel in adults and children with uncomplicated Plas-
modium falciparum Malaria (FALCI) (2018) US National Library
of Medicine. https://clinicaltrials.gov/ct2/show/NCT02497612.
Accessed 28 October 2018
43. Sittler A, Lurz R, Lueder G, Priller J, Hayer-Hartl MK, Hartl
FU, Lehrach H, Wanker EE (2001) Geldanamycin activates
a heat shock response and inhibits huntingtin aggregation in
a cell culture model of Huntington’s disease. Hum Mol Genet
10(12):1307–1315
30. Top S, Tang J, Vessières A, Carrez D, Provot C, Jaouen G (1996)
Ferrocenyl hydroxytamoxifen: a prototype for a new range of
oestradiol receptor site-directed cytotoxics. Chem Commun
8:955–956
44. Dlugosz A, Janecka A (2017) Novobiocin analogs as potential
anticancer agents. Mini Rev Med Chem 17(9):728–733
31. Jaouen G, Vessières A, Top S (2015) Ferrocifen type anti cancer
drugs. Chem Soc Rev 44(24):8802–8817
Afliations
Mziyanda Mbaba1 · Jo‑Anne de la Mare2,3,4 · Jason N. Sterrenberg2,3 · Deborah Kajewole2,3
·
Shantal Maharaj2,3 · Adrienne L. Edkins2,3,4 · Michelle Isaacs4 · Heinrich C. Hoppe2,3,4
Setshaba D. Khanye1,4,5
·
1
4
Department of Chemistry, Faculty of Science, Rhodes
University, Grahamstown 6140, South Africa
Centre for Chemico- and Biomedical Research, Rhodes
University, Grahamstown 6140, South Africa
2
5
Department of Biochemistry and Microbiology, Faculty
of Science, Rhodes University, Grahamstown 6140,
South Africa
Faculty of Pharmacy, Rhodes University, Grahamstown 6140,
South Africa
3
Biomedical Biotechnology Research Unit (BioBRU), Rhodes
University, Grahamstown 6140, South Africa
1 3