Novel α-substituted tropolones promote potent and selective caspase-dependent leukemia cell apoptosis

Article abstract of DOI:10.1016/j.phrs.2016.09.020

Tropolones, such as β-thujaplicin, are small lead-like natural products that possess a variety of biological activities. While the β-substituted natural products and their synthetic analogs are potent inhibitors of human cancer cell growth, less is known about their α-substituted counterparts. Recently, we synthesized a series of α-substituted tropolones including 2-hydroxy-7-(naphthalen-2-yl)cyclohepta-2,4,6-trien-1-one (α-naphthyl tropolone). Here, we evaluate the antiproliferative mechanisms of α-naphthyl tropolone and the related α-benzodioxinyl analog. The α-substituted tropolones inhibit growth of lymphocytic leukemia cells, but not healthy blood cells, with nanomolar potency. Treatment of leukemia cell lines with the tropolone dose-dependently induces apoptosis as judged by staining with annexin V and propidium iodide and Western blot analysis of cleaved caspase 3 and 7. Moreover, pre-treatment of cells with the caspase inhibitor Z-VAD-FMK inhibited the apoptotic effects of the tropolone in two lymphocytic lines. Caspase inhibition also blocked elevated histone acetylation caused by the tropolone, indicating that its effects on histone acetylation are potentiated by caspases. In contrast, α-naphthyl tropolone upregulated p53 expression and phosphorylation of Akt and mTOR in a manner that was not rescued by caspase inhibition. The effects of tropolone were blocked by co-incubation with high levels of free extracellular iron but not by pre-loading with iron. Additionally, dose and time dependent reduction in ex vivo viability of cells from leukemia patients was observed. Taken together, we demonstrate that α-substituted tropolones upregulate DNA damage repair pathways leading to caspase-dependent apoptosis in malignant lymphocytes.

Full text of DOI:10.1016/j.phrs.2016.09.020

Pharmacological Research 113 (2016) 438–448
Contents lists available at ScienceDirect
Pharmacological Research
journal homepage: www.elsevier.com/locate/yphrs
Novel ␣-substituted tropolones promote potent and selective
caspase-dependent leukemia cell apoptosis
Jin Lia, Eric R. Falconea, Sarah A. Holsteinb, Amy C. Andersona, Dennis L. Wrighta,
Andrew J. Wiemera,c,∗
a Department of Pharmaceutical Sciences, University of Connecticut, Storrs, CT, United States
b Department of Medicine, Roswell Park Cancer Institute, Buffalo, NY, United States
c Institute for Systems Genomics, University of Connecticut, Storrs, CT, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 15 June 2016
Received in revised form 7 September 2016
Accepted 19 September 2016
Available online 20 September 2016
Tropolones, such as ␤-thujaplicin, are small lead-like natural products that possess a variety of biological
activities. While the ␤-substituted natural products and their synthetic analogs are potent inhibitors of
human cancer cell growth, less is known about their ␣-substituted counterparts. Recently, we synthesized
a series of ␣-substituted tropolones including 2-hydroxy-7-(naphthalen-2-yl)cyclohepta-2,4,6-trien-1-
one (␣-naphthyl tropolone). Here, we evaluate the antiproliferative mechanisms of ␣-naphthyl tropolone
and the related ␣-benzodioxinyl analog. The ␣-substituted tropolones inhibit growth of lymphocytic
leukemia cells, but not healthy blood cells, with nanomolar potency. Treatment of leukemia cell lines with
the tropolone dose-dependently induces apoptosis as judged by staining with annexin V and propidium
iodide and Western blot analysis of cleaved caspase 3 and 7. Moreover, pre-treatment of cells with
the caspase inhibitor Z-VAD-FMK inhibited the apoptotic effects of the tropolone in two lymphocytic
lines. Caspase inhibition also blocked elevated histone acetylation caused by the tropolone, indicating
that its effects on histone acetylation are potentiated by caspases. In contrast, ␣-naphthyl tropolone
upregulated p53 expression and phosphorylation of Akt and mTOR in a manner that was not rescued
by caspase inhibition. The effects of tropolone were blocked by co-incubation with high levels of free
extracellular iron but not by pre-loading with iron. Additionally, dose and time dependent reduction in
ex vivo viability of cells from leukemia patients was observed. Taken together, we demonstrate that ␣-
substituted tropolones upregulate DNA damage repair pathways leading to caspase-dependent apoptosis
in malignant lymphocytes.
Chemical compounds studied in this article:
Vorinostat (PubChem CID: 5311)
alpha-naphthyl tropolone (PubChem CID:
72163687)
Keywords:
Tropolone
Leukemia
Apoptosis
Vorinostat
Iron
© 2016 Published by Elsevier Ltd.
1. Introduction
involve cell proliferation, such as cancer. Indeed, agents such as
3,7-dihydroxytropolone exhibit potent inhibition of melanoma cell
Tropolones are seven-membered non-benzenoid aromatic com-
pounds with a propensity for metal binding [1–8]. A representative
tropolone, ␤-thujaplicin (Fig. 1) is produced from monoterpenes [9]
by trees such as the Western red cedar where it functions to resist
fungal decay [10], and similar ␤-substituted tropolones may also
exhibit insecticidal properties [11]. These potent anti-proliferative
activities underlie the strong resistance of the wood to damage
and raise the question of whether tropolones or their derivatives
may find therapeutic benefit to human health in diseases that
growth and extend lifespan in mouse models of melanoma [12,13].
From a medicinal chemistry standpoint, simple low molecular
weight tropolones such as ␤-thujaplicin exhibit low hydropho-
bicity, contain ample sites for diversification, and are readily
synthesized. Additionally, the hydroxyketone functionality embed-
ded in the aromatic system is a distinct pharmacophore that
allows these compounds to coordinate metal ions [14,15]. As such,
tropolones have therapeutic potential both as direct metal chela-
tors [14–17] and metalloenzyme inhibitors [18–21]. These features
have led us to characterize these molecules as lead-like natural
products, an important distinction relative to many more complex
naturally-derived non-benzenoid aromatics such as colchicine.
Derivatization of the tropolone ring with lipophilic substituents
may provide an opportunity to enhance intracellular delivery [22]
of the compound while retaining the beneficial pharmacophore.
∗
Corresponding author at: Department of Pharmaceutical Sciences, University of
Connecticut, 69 North Eagleville Road Unit 3092, Storrs, Connecticut, 06269, United
States.
E-mail address: andrew.wiemer@uconn.edu (A.J. Wiemer).
http://dx.doi.org/10.1016/j.phrs.2016.09.020
1043-6618/© 2016 Published by Elsevier Ltd.

J. Li et al. / Pharmacological Research 113 (2016) 438–448
439
In order to further explore the utility of tropolone derivatives
as anticancer agents, we recently synthesized a library of novel
tropolone derivatives and tested them for inhibition of malig-
nant cell growth [23]. Some ␤-substituted tropolones, such as
the ␤-phenyl analog, exhibited nanomolar growth inhibition. At
a molecular level, we have observed tropolones may reduce the
in vitro activity of purified metalloenzymes that regulate epige-
netic modifications including histone deacetylases [23] and sirtuins
[24]. However, the biological activity of tropolones is not well-
understood and the importance of their in vitro metalloenzyme
inhibition to the cellular mechanism of action is unclear. Like-
wise, while the ␤-substituted natural products and their synthetic
analogs are inhibitors of human cancer cell growth, less is known
about their ␣-substituted analogs.
Our underlying hypothesis is that ␣-substituted tropolones
will function as potent and specific anti-leukemia agents. Here,
we evaluate the ability of ␣-substituted tropolones including
2-hydroxy-7-(naphthalen-2-yl)cyclohepta-2,4,6-trien-1-one
(␣-naphthyl tropolone) and its novel counterpart 2-(2,3-dihydro-
1,4-benzodioxin-6-yl)-7-hydroxycyclohepta-2,4,6-trien-1-one
(␣-benzodioxinyl tropolone) to inhibit leukemia cell proliferation
and induce apoptosis.
Fig. 1. Chemical structures of compounds used in this study. Tropolones are seven
membered aromatic rings which may be functionalized. Prior known tropolones
include the natural product ␤-thujaplicin (aka Hinokitiol) and the synthetic ␤-
phenyl tropolone. We have developed methodologies for synthesis of ␣-substituted
tropolones such as the naphthyl and benzodioxinyl tropolones included in this
study.
2. Materials and methods
2.1. Test compounds
The ␣-naphthyl tropolone was synthesized as previously
described [23].
The ␣-benzodioxinyl tropolone was synthesized as follows.
Tropolone (3.24 mmol) was dissolved in 16 mL acetonitrile and
allowed to react with MeI (16.2 mmol) in the presence of K2CO3
(9.72 mmol) and 18-crown-6 (0.324 mmol) for 18 h at 85 ◦C to
form the tropolone methyl ether. The mixture was cooled and
filtered through a fritted funnel. The filtrate was evaporated to
dryness, re-dissolved in dichloromethane and washed with base
and brine. The organic layer was then evaporated to dryness and
analyzed by NMR for purity. The methyl ether (1.87 mmol) was dis-
solved in 9.3 mL carbon tetrachloride and allowed to react with NBS
(1.05 eq) for 2 h at 80 ◦C, forming the ␣-bromo tropolone methyl
ether. The mixture was cooled and filtered through a fritted fun-
nel. The solid was washed several times with dichloromethane
and the filtrate was evaporated to dryness. The sample was then
re-dissolved in dichloromethane and washed (3x). The combined
organic layers were washed with brine, dried over Na2SO4, filtered
and concentrated in vacuo. Flash chromatography of the crude
residue (SiO2, EtOAc in hexanes) provided the desired compound.
Suzuki coupling was performed between the ␣-bromo tropolone
methyl ether (0.5 mmol) and 1,4-benzodioxane-6-boronic acid
(0.75 mmol) (1.5 eq) to give the alpha-benzodioxinyl tropolone
methyl ether. The coupling was performed in 1.7 mL dioxane
and 0.16 mL water containing cesium carbonate (3.0 eq) and
bis(triphenylphosphine)palladium(II) dichloride (0.1 eq) that had
been degassed with argon and was allowed to proceed at 90 ◦C for
18 h. Water was added and the mixture was extracted with EtOAc
(3x). The combined organic layers were washed with brine, dried
over Na2SO4, filtered and concentrated in vacuo. Flash chromatog-
raphy of the crude residue (SiO2, EtOAc in hexanes) provided the
desired compound. The methyl ether (0.36 mmol) was hydrolyzed
in the presence of lithium chloride (3 eq) at 120 ◦C for 2–3 h. The
alpha-benzodioxinyl tropolone acid was purified by C-18 column
chromatography. The crude product was added to the column in
4:1 water: methanol and eluted with 50% methanol in water. Over-
all yield: 64%. Rf = 0.42 (1:1 EtOAc: hexanes). Mp = 102.8-103.6. IR
(KBr): ꢀ: 3170, 2976, 2932, 2874, 1590, 1578, 1551, 1504, 1470,
Fig. 2. Molt-4 cell proliferation is reduced by ␣-naphthyl tropolone in a dose and
time dependent manner. (A) 72 h proliferation assay of ␣-naphthyl tropolone and
vorinostat in Molt-4 cells (n = 6), (B) Proliferation assay of ␣-naphthyl tropolone in
Molt-4 cells at 12, 24, 48 and 72 h (n ≥ 3), (C) 72 h proliferation assay of ␣-naphthyl
tropolone in different cell lines (n = 3), (D) 72 h proliferation assay of vorinostat and
tropolone derivatives in human peripheral blood mononuclear cells (hPBMC) (n = 3).
(E) 72 h proliferation assay of vorinostat in different cell lines (n = 4).

440
J. Li et al. / Pharmacological Research 113 (2016) 438–448
Fig. 4. Treatment with ␣-naphthyl tropolone induces apoptosis in Molt-4 cells.
(A-D) Cell apoptosis assay of vorinostat and tropolone derivatives in Molt-4 cells.
Cells were treated with various concentrations of compounds for 24 h, stained with
Annexin V/FITC and propidium iodide and tested by Flow Cytometry. (A) ␣-naphthyl
tropolone, (B) vorinostat, (C) Percentage of early apoptotic cells (Annexin V positive-
PI negative) and (D) late apoptotic cells (double positive). (E, F) Western blot of
cleaved caspase 3 and cleaved caspase 7.
Both tropolones were dissolved in dimethyl sulfoxide at 10 mM
and stored at −20 ◦C. They were diluted in culture medium and
used at concentrations as indicated. Vorinostat was purchased from
(Thermo Fisher Scientific, Waltham, MA). Z-VAD-FMK was pur-
chased from ApexBio (Houston, TX).
2.2. Cell culture
Cell lines were purchased from American Type Culture Collec-
tion (ATCC, Manassas, VA, USA) and cultured as described [25]. Cells
were grown at 37 ◦C with 5% CO2 in RPMI-1640 medium (Thermo
Fisher Scientific) with 1x Penicillin-Streptomycin (Thermo Fisher
Scientific). 10% fetal clone III (Hyclone, South Logan, UT, USA) was
added to the medium of the K562, Molt-4, Ba/F3 and HL-60 cell
lines whereas 10% fetal bovine serum (Hyclone) was added to the
medium of the Daudi and HuT-78 cell lines.
Human Peripheral Blood Mononuclear Cells (hPBMCs) were
purified from human blood (purchased from Research Blood Com-
ponents, Brighton, MA) by density centrifugation as described
[26]. In brief, hPBMCs were frozen in freezing media (80% media,
10% DMSO, 10% FBS) in liquid nitrogen until needed. Cells were
thawed and resuspended prior to use at 1 million/mL in 10%
heat-inactivated fetal bovine serum in RPMI-1640 medium (sup-
plemented with 1x HEPES buffer, pyruvate, non-essential amino
acids, and 2-mercaptoethanol) and immediately used for the cell
viability assay.
Fig. 3. Iron loading inhibits the tropolone-induced anti-proliferative effect. (A-D)
Molt-4 cells were preloaded with or without 500 ␮M Fe3+ for 24 h and then treated
by beta-thujaplicin (HKT), vorinostat (VOR), ␣-naphthyl tropolone (NAP) and ␣-
benzodioxinyl tropolone (BEN) for 72 h. (E–H) Molt-4 cells were treated by different
concentrations of HKT, VOR, NAP and BEN with the absence (UN) or presence of indi-
cated concentrations of Fe3+ for 72 h. (I–J) Molt-4 cells were treated by 10 ␮M HKT
or NAP with the presence of different concentrations of Fe3+ for 72 h. Cell viability
was assessed by Cell Quanti-Blue assay. Statistics were determined by ANOVA with
Tukey’s post-hoc analysis and with p < 0.05.
1423, 1409, 1356, 1299, 1288, 1250, 1231, 1194, 1158, 1127, 1063,
1045, 1004, 924.1HNMR (500 MHz, CD2Cl2): ı 7.60 (d, J = 10.2 Hz,
1H), 7.40 (dt, J = 19.9, 10.3 Hz, 2H), 7.13–7.02 (m, 3H), 6.97 (d,
J = 8.4 Hz, 1H), 4.33 (s, 4H). 13CNMR (125 MHz, CD2Cl2): ␦. 170.92,
170.60, 143.79, 143.22, 140.44, 136.58, 133.02, 127.29, 122.66,
122.20, 118.44, 117.13, 64.54, 64.38, 30.99. HRMS (ESI) calculated
for C15H12O4 [M + H]+: 257.0814; found:. 257.0780.

J. Li et al. / Pharmacological Research 113 (2016) 438–448
441
Fig. 6. Treatment with ␣-naphthyl tropolone enhances acetylation of histone H3 in
Molt-4 cells which is blocked by pre-treatment with caspase inhibitor. (A) West-
ern blot of acetylated H3K23, H3K9 and H4K12 from Molt-4 cells treated with
␣-naphthyl tropolone (NAP, 10 ␮M) and ␣-benzodioxinyl tropolone (BEN, 10 ␮M)
for 24 h. (B) Quantification of histone acetylation levels after treatment in Molt-
4 cells. (C) Cells were pre-incubated with 100 ␮M Z-VAD-FMK for 1 h and treated
with ␣-naphthyl tropolone for 24 h. (D) Quantification of histone acetylation levels
after tropolone treatment in the presence of Z-VAD-FMK in Molt-4 cells. *P < 0.05
by ANOVA with Tukey’s post-hoc analysis, compared to DMSO (n = 3). #indicates
significant difference compared to each other.
2.3. Cell viability assays
Cell viability assays were performed using the Quanti-Blue (QB)
Cell Viability Assay Kit (BioAssay Systems, Hayward, CA). Briefly,
100 ␮L of cells were seeded in 96 well plates at various concen-
trations (105 cells/mL for K562, Molt-4, HL-60 and HuT-78 cells;
5 × 104 cells/mL for Daudi cells; 2.5 × 104 cells/mL for Ba/F3 cells
and 106 cells/mL for hPBMCs). Cells were treated with various con-
centrations of the compounds and incubated at 37 ◦C with 5% CO2
for 24–72 h. 10 ␮L of the QB reagent was added into each well for
the final 2 h incubation. The plates were read by a Perkin-Elmer
VICTOR X multilabel plate reader at ␭ex 530 nm/␭em 590 nm. IC50
values were obtained by regression analysis. All experiments were
repeated at least three independent times (n = 3).
For the iron experiments, ferric chloride was freshly dissolved
in water and diluted to the final concentration in media as indi-
cated. In pre-loading experiments, cells were cultured for 24 h in
the presence of 500 ␮M ferric chloride, washed, and exposed to
test compounds. In co-treatment experiments, both iron and com-
pounds were added at the same time.
For the clinical isolates, de-identified peripheral blood bio-
specimens of patients with newly diagnosed acute leukemias were
obtained from the Roswell Park Cancer Hematologic Procurement
Shared Resource. Cells were treated with test compounds at indi-
cated concentrations and their viability was assessed by both
CellToxGreen assay (Promega) and MTT assay (Sigma) according
to manufacturer protocols.
Fig. 5. Caspase inhibitors block the tropolone-induced anti-proliferative effect. Cells
were pre-incubated with 100 ␮M Z-VAD-FMK for 1 h and treated with ␣-naphthyl
tropolone for (A) 24 h or (B) 48 h or ␣-benzodioxinyl tropolone for (C) 24 h or (D)
48 h. The same experiments were also performed in HuT-78 cells (E–H) and HL60
cells (I–L). Cell viability was assessed by Cell Quanti-Blue assay. Statistics were
determined by ANOVA with Tukey’s post-hoc analysis and with p < 0.05. #indicates
significant difference compared to each other.
2.4. Apoptosis analysis
Molt-4 cells were resuspended at 105 cells/mL in fresh media
and added to 12 well plates (2 mL/well). Cells were treated with
compounds at the indicated concentrations for 24 h. To detect
the apoptotic cells and discriminate live cells from dead cells

442
J. Li et al. / Pharmacological Research 113 (2016) 438–448
Fig. 7. Treatment of Molt-4 cells with ␣-naphthyl tropolone or ␣-benzodioxinyl tropolone leads to changes in gene expression as determined by microarray analysis. Molt-4
cells were treated with ␣-naphthyl tropolone (NAP) and ␣-benzodioxinyl tropolone and gene expression was analyzed using an Illumina HumanHT-12 Expression Beadchip.
(A) Numbers of genes that were upregulated or down-regulated by treatment with either tropolone. The overlapping region represents genes that were differentially regulated
by both compounds. (B) List of genes that were upregulated by more than 10 fold by both compounds. (C) List of genes that were downregulated by more than 5 fold by both
compounds.
respectively, Molt-4 cells were stained with Annexin V/FITC and
propidium iodide (PI) using annexin V-FITC Apoptosis Detection
Kit I (BD Biosciences Pharmingen, SAN Diego, CA). Briefly, cells
were resuspended in 100 ␮L of 1× binding buffer, to which 2 ␮L
of annexin V-FITC and 2 ␮L of propidium iodide were added and
mixed well. After incubation at room temperature for 30 min in the
dark, 400 ␮L of 1× binding buffer was added. Cells were analyzed
by flow cytometry within 30 min.
in RIPA buffer (25 mM Tris–HCl pH 7.6, 150 mM NaCl, 1% NP-40,
1% sodium deoxycholate, 0.1% SDS) or histone lysis buffer (50 mM
Tris–HCl pH 7.5, 150 mM NaCl, 2% SDS) with protease and phos-
phatase inhibitors. Total protein was quantified using BCA assay
(Thermo Fisher).
Equivalent masses of proteins were loaded and separated by
SDS–PAGE using 7.5% or 15% gels and proteins were transferred
to nitrocellulose, and blotted with antibodies to acetyl-histone H3
(K9) (clone C5B11), acetyl-histone H3 (K23) (polyclonal, #8848),
and acetyl-histone H4 (K8) (polyclonal, #2594) purchased from
Cell Signaling Technologies (Danvers, MA). Antibodies to AKT
(clone C67E7), phospho-AKT (S473) (D9E), mTOR (clone 7C10), and
phospho-mTOR (S2448) (clone D9C2) were also purchased from
Cell Signaling. JLA20 actin antibody (University of Iowa Develop-
2.5. Western blot analysis
Molt-4 cells were resuspended and added to 6 well plates at
6 × 106 cells/4 mL in fresh media. Cells were treated with the indi-
cated concentrations of compounds for 24 h. Cells were then lysed

J. Li et al. / Pharmacological Research 113 (2016) 438–448
443
Fig. 8. Treatment with ␣-naphthyl tropolone enhances phosphorylation of Akt and
mTOR in Molt-4 cells which is not dependent upon caspase activation. Molt-4 cells
were treated with DMSO, ␣-naphthyl tropolone (NAP, 10 ␮M) and vorinostat (VOR,
2.5 ␮M) for 24 h. (A) Western blot of p-AKT, AKT, p-mTOR and mTOR. (B-E) Quanti-
fied data. (F) p-AKT/AKT ratio and (G) p-mTOR/mTOR ratio were quantified from the
Western results. (H) Western blot of p-AKT, p-mTOR, AKT and mTOR from Molt-4
cells. Cells were pre-incubated with Z-VAD-FMK for 1 h and then treated by DMSO or
␣-naphthyl tropolone (NAP, 10 ␮M) for 24 h. Statistics were determined by ANOVA
with Tukey’s post-hoc analysis and with p < 0.05. *indicates significant difference
compared to condition with control. #indicates significant difference compared to
each other.
Fig. 9. Treatment with ␣-naphthyl tropolone dose dependently increases p53
expression and nuclear localization. (A) Western blot of p53, p-AKT and AKT from
the Molt-4 cells treated with DMSO and ␣-naphthyl tropolone (NAP, 2.5 ␮M, 10 ␮M,
25 ␮M) or vorinostat (VOR, 1 ␮M, 2.5 ␮M, 10 ␮M) for 24 h. (B) Western blot of p53 in
whole cell lysate (WC), cytoplasm (CYTO) and nucleus (NU) from Molt-4 cells treated
with solvent, ␣-naphthyl tropolone (NAP, 10 ␮M) or vorinostat (VOR, 2.5 ␮M) for
24 h. (C) quantified p53 expression level from 3 replicates. (D) Western blot of p53
from Molt-4 cells pretreated with Z-VAD-FMK. (E) Quantified data from 3 repli-
cates. Statistics were determined by ANOVA with Tukey’s post-hoc analysis and
with p < 0.05. *indicates significant difference compared to condition with control.
#indicates significant difference compared to each other.
mental Studies Hybridoma Bank), RIAM (Santa Cruz Biotechnology)
and vinculin (Cell Signaling) were used as loading controls for dif-
ferent blots. The p53 antibody (DO-1) was purchased from Santa
Cruz.
imposed at 1.5-fold. In general, ␣-naphthyl tropolone gave lower
q-values than ␣- benzodioxinyl tropolone, but many of the “hits”
were similar between the two treatments. With 47,323 genes/gene
variants analyzed, 2740 were up-regulated 1.5-fold and greater in
both treatments and 1751 were down-regulated.
2.6. Microarray analysis
Molt-4 cells were resuspended and added to 6 well plates at
6 × 106 cells/4 mL in fresh media. Cells were treated with 1 ␮M of
compounds for 24 h. The RNA was extracted using the PureLink RNA
Mini Kit (Thermo Fisher Scientific). Samples were sent to the Boston
Children’s Hospital Molecular Genetics Core Facility where they
were analyzed using an Illumina HumanHT-12 Expression Bead-
chip. The cubic spline data from the microarray was analyzed using
the Stanford Tools plug-in for Excel. This plug-in analyzes the data
and determines fold changes in gene expression as well as estimat-
ing the false discovery rate by giving a q-value. A fold threshold was
2.7. Statistical analysis
All the data are mean values of at least
3 independent
experiments and expressed as means ( SD). For multiple-group
comparison, one-way ANOVA was used followed by a Tukey’s post-
hoc analysis, using statistical software Minitab 16. A p value less
than 0.05 was considered to be statistically significant.

444
J. Li et al. / Pharmacological Research 113 (2016) 438–448
Fig. 10. Viability of primary leukemia cells is reduced by ␣-substituted tropolones in a dose and time dependent manner. Peripheral blood mononuclear cells from leukemia
patients were treated with ␣-substituted tropolones for indicated times. (A) CellToxGreen assay of B-ALL cells treated with 10 ␮M of tropolone for 24, 48, or 72 h. (B)
CellToxGreen assay of AML cells treated with 10 ␮M of tropolone for 24 h. (C) MTT assay of primary AML cells treated with various concentrations of tropolone for 48 h.
Table 1
Table 2
IC50 values in different cell lines (n ≥ 3).
Selectivity of ␣-naphthyl tropolone and vorinostat for malignant cells versus pri-
mary cells (n = 6).
Cell line
Time (h)
IC50 [␮M]
IC50 [␮M]
Selectivitya
␣-naphthyl tropolone
vorinostat
hPBMC
Molt-4
Molt-4
Daudi
K562
HuT-78
Ba/F3 WT
HL-60
Molt-4
Molt-4
Molt-4
72
72
72
72
72
72
12
24
48
0.35
0.92
3.25
1.36
2.75
1.97
>25
0.33
0.32
0.71
0.16
0.32
0.74
ND
␣-naphthyl tropolone
vorinostat
Fold difference
4.89
0.86
5.66
0.35
0.33
1.08
13.8
2.6
a
Selectivity is the fold difference of IC50 values (hPBMC/Molt-4).
0.99
1.05
ND
ND
HuT-78). Importantly, ␣-naphthyl tropolone only weakly inhibited
the viability of healthy human peripheral blood mononuclear cells
(hPBMC) (Fig. 2D). Here, hPBMC were treated with different con-
centrations of ␣-naphthyl tropolone, ␣-benzodioxinyl tropolone
and vorinostat for 72 h. All three compounds exhibited dose-
dependent toxicity towards the hPBMC, but ␣-naphthyl tropolone
was significantly less toxic towards the primary cells than was
vorinostat. Vorinostat also strongly inhibited growth of all cell
lines tested (Fig. 2E). As shown in Table 2, ␣-naphthyl tropolone
has a better selectivity for malignant cells versus primary cells. To
confirm that the mild hPBMC activity was not compound-specific,
we evaluated a second ␣-substituted tropolone, ␣-benzodioxinyl
tropolone, in this assay. Both tropolones showed only mild cyto-
toxicity in the primary cells. Taken together, these data show that
␣-substituted tropolones potently inhibit growth of some lympho-
cytic leukemia cell lines with less cytotoxicity to healthy blood
cells.
3. Results
3.1. Molt-4 cell proliferation is reduced by treatment with
˛-naphthyl tropolone in a dose and time dependent manner
A prior study suggested leukemia and lymphoma cells were
particularly sensitive to growth inhibition caused by tropolones
[23]. To validate these results, we treated Molt-4 cells (an acute
T cell leukemia line) with various concentrations of ␣-naphthyl
tropolone (Fig. 1) and monitored proliferation. The pan-histone
deacetylase inhibitor vorinostat was used as a positive control
in these assays because the tropolones were suspected to inhibit
metalloenzymes and because vorinostat is used clinically for
treatment of T cell malignancies [27,28]. As shown in Fig. 2A,
␣-naphthyl tropolone inhibits Molt- 4 proliferation at 72 h in a
dose-dependent manner (IC50 = 0.35 ␮M, Table 1). The potency of
␣-naphthyl tropolone was similar to that of the clinical agent
vorinostat (IC50 = 0.33 ␮M). The activity of ␣-naphthyl tropolone
was dependent upon the incubation time, as the anti-proliferative
effect was reduced with shorter incubation times (Fig. 2B).
3.3. Co-incubation with iron reduces the tropolone activity
Because some tropolones are known to form strong interac-
tions with iron [14–17], we next wanted to assess whether or
not the growth inhibition caused by treatment with the novel ␣-
substituted tropolones could be antagonized by exogenous iron.
We measured proliferation of Molt-4 cells in response to var-
ious concentrations of ␣-naphthyl tropolone, ␣-benzodioxinyl
tropolone, or vorinostat in the presence or absence of pre-
treatment with Fe3+ (Fig. 3). The natural product ␤-thujaplicin was
used as an additional positive control in these assays. Iron pre-
loading had no effect on the toxicity of any of the four compounds
(Fig. 3A–D). Next, we assessed co-treatment of Molt-4 cells with
iron and the four compounds. As expected, in the absence of iron a
dose-dependent decrease in proliferation was observed (Fig. 3E–H).
Iron mitigated the antiproliferative effects of both of ␣-naphthyl
tropolone (Fig. 3G) and ␣-benzodioxinyl tropolone (Fig. 3H) as
determined by the IC50 values in the absence or presence of iron.
Iron had no effect on the activity of vorinostat. Interestingly, only
3.2. Proliferation of malignant hematological cell lines is reduced
by treatment with ˛-naphthyl tropolone, while healthy blood cells
are resistant
Given the potent growth inhibition observed in the Molt-4 cells,
we tested ␣-naphthyl tropolone in a small panel of other estab-
lished blood cell lines. While the potency towards Molt-4 cells was
the strongest, the compound also shows a dose-dependent inhibi-
tion of the leukemia cell lines in the panel including Daudi, K562,
HuT-78, Ba/F3 and HL-60 cells (Fig. 2C, see Table 1 for IC50 val-
ues). This suggests that the therapeutic potential of tropolones may
extend to myeloid cells (HL-60 and K562) and B-cell malignan-
cies (Daudi), in addition to the two T cell lines tested (Molt-4 and

J. Li et al. / Pharmacological Research 113 (2016) 438–448
445
Table 3
Gene set enrichment analysis of microarray data from untreated Molt-4 cells versus cells treated for 24 h with 1 ␮M ␣-naphthyl tropolone. GO-Gene Ontology, KEGG-Kyoto
Encyclopedia of Genes and Genomes.
GO-decreased by tropolone
Protein expression
KEGG-decreased by tropolone
Protein expression
TRANSLATION
PROTEASOME
RIBOSOME
GO-increased by tropolone
Metabolism
DNA/cell cycle
PORPHYRIN AND CHLOROPHYLL METABOLISM
FRUCTOSE AND MANNOSE METABOLISM
Signaling
DNA INTEGRITY CHECKPOINT
G1S TRANSITION OF MITOTIC CELL CYCLE
DNA DAMAGE RESPONSE SIGNAL TRANSDUCTION
MITOTIC CELL CYCLE CHECKPOINT
Immune function
MYELOID CELL DIFFERENTIATION
T CELL ACTIVATION
REGULATION OF LYMPHOCYTE ACTIVATION
Metabolism
PEPTIDYL AMINO ACID MODIFICATION
GLYCOPROTEIN BIOSYNTHETIC PROCESS
CARBOHYDRATE BIOSYNTHETIC PROCESS
Signaling
CYTOKINE/CYTOKINE RECEPTOR INTERACTION
KEGG-increased by tropolone
DNA/cell cycle
P53 SIGNALING PATHWAY
Immune function
LEUKOCYTE TRANSENDOTHELIAL MIGRATION
T CELL RECEPTOR SIGNALING PATHWAY
Structural
ADHERENS JUNCTION
GAP JUNCTION
TRANSMEMBRANE RECEPTOR PROTEIN TYROSINE
KINASE SIGNALING PATHWAY
REGULATION OF ACTIN CYTOSKELETON
FOCAL ADHESION
the lowest concentration of iron mitigated the effect of the natural
product ␤-thujaplicin (HKT) (Fig. 3E). Surprised by this result, we
performed a dose response experiment (Fig. 3I, J). Here, we noted
partial rescue of the natural product only when iron was used at 2.1
or 6.2 ␮M. In contrast, an obvious dose-dependent relationship was
observed for the ability of iron to block the effect of the ␣-naphthyl
tropolone. Taken together, co-treatment but not pre-loading with
iron blocks the growth inhibitory effect of ␣-naphthyl tropolone
and to a lesser extent, ␤-thujaplicin.
effects of ␣-substituted tropolones. To that end, we pre-treated
Molt-4 cells with 100 ␮M of the caspase inhibitor Z-VAD-FMK
for 1 h and then exposed the cells to the indicated concentra-
tions of ␣-naphthyl tropolone (Fig. 5A/B) or ␣-benzodioxinyl
tropolone (Fig. 5C/D) for 24 or 48 h and monitored cell proliferation.
Importantly, pre-treatment with Z-VAD-FMK strongly reduced
the anti-proliferative effects of both compounds. Z-VAD-FMK also
reduced the effect of the tropolones in a second T lymphocyte line
(HuT-78) (Fig. 5E–H) but not in the myeloid HL-60 line (Fig. 5I–L).
Pre-treatment with the caspase inhibitor similarly reduced total
apoptosis that was induced by ␣-naphthyl tropolone and late apo-
ptosis induced by ␣-benzodioxinyl tropolone (not shown). Thus,
the loss in malignant cell proliferation is almost entirely due to
the ability of the compounds to induce activation of pro-apoptotic
caspases in lymphocytic cell lines.
3.4. Treatment of leukemia cells with ˛-substituted tropolones
induces apoptosis
To further assess the mechanisms of growth inhibition, we
asked whether or not ␣-substituted tropolones induced apoptosis
of Molt-4 cells. Here, cells were treated with indicated concentra-
tions of ␣-naphthyl tropolone or vorinostat for 24 h. Cells were
then stained with Annexin V/FITC and propidium iodide (PI) and
analyzed by flow cytometry (Fig. 4). Both compounds displayed
a dose-dependent increase in the percentage of early apoptotic
cells (Annexin V positive-PI negative) and late apoptotic cells
(Annexin V positive-PI positive) (Fig. 4A, B). At concentrations of
2 ␮M and 10 ␮M, vorinostat significantly induces Molt-4 cell apo-
ptosis compared to the untreated cells, while ␣-naphthyl tropolone
significantly induces Molt-4 cell apoptosis at 10 ␮M. The effects
were statistically significant with respect to both early (Fig. 4C)
and late (Fig. 4D) apoptosis.
In order to confirm these results in a second assay, we treated
cells with the compounds and analyzed cleavage of caspase-3 by
Western blot analysis. The Western blot results show that lev-
els of cleaved caspase 3 and cleaved caspase 7 were significantly
increased by treatment with ␣-naphthyl tropolone at 10 ␮M and
vorinostat at 2, 10 ␮M (Fig. 4E, 4F). Similar to its anti-proliferative
effect, the activity of ␣-naphthyl tropolone was stronger than that
of ␣-benzodioxinyl tropolone. Surprisingly, ␣-naphthyl tropolone
was weaker in this assay than vorinostat, which may be attributed
to its activity being less cytotoxic and more growth inhibitory rel-
ative to vorinostat.
3.6. Tropolones increase levels of histone acetylation, which is
dependent upon caspase activation
As had been previously reported [23], tropolones can inhibit
some metalloenzymes including histone deacetylases. To deter-
mine whether or not ␣-substituted tropolones could affect histone
acetylation in Molt-4 cells, we treated cells with ␣-naphthyl
tropolone or ␣-benzodioxinyl tropolone and measured levels of
acetylated histone H3 lysine 23, H3 lysine 9, and H4 lysine 12 by
Western blot analysis (Fig. 6). In congruency with the prior study,
both tropolones did indeed cause increased H3 acetylation at lysine
23 (Fig. 6A, B). We next wanted to determine whether or not these
elevated histone acetylation could be blocked by pre-treatment
with Z-VAD-FMK. We expected that if histone acetylation was
a direct result of the tropolone mechanism, then the effect on
histone acetylation would not be prevented by pre-treatment
with the caspase inhibition. However, unexpectedly the caspase
inhibitor clearly prevented the ability of ␣-naphthyl tropolone to
induce histone acetylation (Fig. 6C, D). Therefore, while tropolones
inhibit purified histone deacetylases in vitro and they increase his-
tone acetylation in cells, the increased cellular histone acetylation
appears secondary to the ability of the tropolones to induce apo-
ptosis.
3.5. Caspase activation is required for antiproliferative and
pro-apoptotic effects of ˛-substituted tropolones
We next sought to determine whether or not the caspase acti-
vation was required for the antiproliferative and pro-apoptotic

446
J. Li et al. / Pharmacological Research 113 (2016) 438–448
Table 4
were similar in the HuT-78 line, but not in the HL-60 line (Figs.
S2-S4).
Leading edge analysis of microarray data from untreated Molt-4 cells versus cells
treated for 24 h with 1 ␮M ␣-naphthyl tropolone. Data show proteins likely to
influence the effects of tropolone treatment on down regulation of affected KEGG
pathways.
3.9. Tropolones affect p53
Structural
Because it has been suggested that levels of p53 increase fol-
lowing iron chelation due to increased binding of HIF-1␣ [36],
we tested whether or not tropolone would affect p53 expression
(Fig. 9). We treated Molt-4 cells with ␣-naphthyl tropolone and
measured p53 expression by Western blot analysis (Fig. 9A). We
observed a dose-dependent and statistically significant increase
in p53 expression following treatment with ␣-naphthyl tropolone.
Additionally, treatment with tropolone, but not vorinostat, led to
an accumulation of nuclear p53 (Fig. 9B, C). In the presence of the
caspase inhibitor (Fig. 9D, E), tropolone treatment also significantly
increased p53 expression from the baseline, though both conditions
were slightly lower in the presence of the caspase inhibitor versus
the absence of the caspase inhibitor.
ACTB (7), ACTG1 (7), ACTN1 (5), ITGB1 (7), ITGAV (4), ITGA4 (4), CTNNB1 (8),
CTNNA1 (4), VCL (4), MYL12A (3), WASL (3), WAS (3), NCK2 (3)
Small GTPase signaling
NRAS (11), RHOA (5), Rock1 (5), VAV3 (4), RAC1 (3), KRAS (3), GNAI3 (5), PAK2
(4)
Phospholipid signaling
PIK3R3 (12), PIK3R1 (12), PIK3CB (12), PIK3CG (12), PTEN (7), PLCG1 (5),
PDPK1 (4)
Non-receptor kinases
PRKACB (3), PRKCB (3), ADCY3 (3), MAPK14 (3)
DNA/Cell cycle
TP53 (9), RB1 (6), CDK6 (6), BCL2 (4), CCNE2 (3)
Transcription
LEF1 (5), E2F2 (5), CHUK (4)
3.10. Tropolones reduce viability of leukemia blasts from acute
leukemia patients
3.7. Microarray analysis of ˛-substituted tropolones in Molt-4
cells reveals a DNA damage response mechanism
In order to assess the clinical relevance of the ␣-substituted
tropolones, we tested peripheral blood samples that were obtained
from newly diagnosed acute leukemia patients for their sensitivity
to tropolone treatment (Fig. 10). We observed a time dependent
loss in viability of cells from a patient with B cell acute lym-
phocytic leukemia (B-ALL) (Fig. 10A) following ex vivo tropolone
exposure. Additionally, exposure of acute myeloid leukemia cells
to tropolones led to loss of viability at 24 h (Fig. 10B). The loss of
viability observed in the cells from the AML patient was dose depen-
dent (Fig. 10C). Although the benzodioxinyl compound appears to
be more potent against the primary AML cells at 24 h, by 48 h there
does not appear to be a significant difference between benzodi-
oxinyl and naphthyl treatments.
To understand the cellular effects of tropolones we performed
gene expression microarrays on Molt-4 cells in the presence or
absence of the ␣-substituted tropolones. Molt-4 cells were treated
for 24 h in the presence or absence of 1 ␮M ␣-naphthyl tropolone
or ␣-benzodioxinyl tropolone. We found several genes to be dif-
ferentially regulated and were able to identify 562 genes that were
upregulated and 48 genes that were down regulated in response
to two different tropolones (Fig. 7). Gene set enrichment anal-
ysis [29,30] also revealed several affected pathways (Table 3).
Tropolones decrease pathways related to protein expression and
increase pathways related to DNA/damage repair, cell cycle pro-
gression, immune cell signaling, and cell adhesion/structure.
Leading edge analysis of these pathways determined proteins likely
to be involved (Table 4). Both tropolones exhibited the ability to
affect genes related to DNA damage repair and cell cycle progres-
sion, which was also observed in microarray analysis of pan-HDAC
inhibitors [31,32]. The tropolones also alter expression of genes
related to cell adhesion and the cytoskeleton also seen previously
with pan-HDAC inhibitors [33–35].
4. Discussion
Here, we have demonstrated that ␣-substituted tropolones
potently inhibit the growth of leukemia cells with mid to low
nanomolar IC50 values. The tropolones we examined are 13.8-fold
more active against the malignant Molt-4 cell line than against
healthy primary blood cells, which may be beneficial for their
potential therapeutic applications and is in contrast to the clinical
agent vorinostat, which is more weakly selective. The ␣-substituted
tropolones function by inducing caspase dependent apoptosis and
importantly their anti-proliferative effects can be fully prevented
by caspase inhibition in lymphocytic cell lines. The ␣-substituted
tropolones display effects on gene expression, protein expres-
sion, protein activation, and localization of a number of important
mediators of cell survival and proliferation which are common anti-
cancer targets, including histones, Akt, mTor, and p53. However,
the effects of tropolones on histone acetylation are downstream of
caspase activity, indicating this is a secondary effect.
It should be noted that these current ␣-substituted tropolones
including the ␣-naphthyl tropolone exhibit excellent potency and
selectivity relative to other agents. The current data demonstrates
both ␣-substituted tropolones exhibit stronger potency towards
Molt-4 cells than we observed in any cell line/␤-substituted
tropolone combination [23]. Additionally, the potency of these
compounds is stronger than that observed by ourselves (Fig. 3) and
other groups for ␤-thujaplicin, which is active in the low to mid ␮M
range [37], while maintaining good selectivity. The current com-
pounds exhibit impressive potency even relative to vorinostat, a
clinically used chemotherapy agent, while at the same time show
less toxicity.
3.8. Treatment with ˛-substituted tropolones, but not vorinostat,
activates AKT and mTor in a caspase independent manner
Notably, we experimentally confirmed with Western blot anal-
ysis that some of these leading edge proteins are indeed affected
by treatment with ␣-naphthyl tropolone. Based on the microarray
findings, we hypothesized that the Akt and mTor signaling path-
ways may be altered. We treated Molt-4 cells with ␣-naphthyl
tropolone and assessed levels of total and phosphorylated Akt and
mTor (Fig. 8). Tropolone exposure led to a 3–4 fold increase in the
phosphorylation of both Akt and mTor relative to solvent treated
controls (Fig. 8B, C). At the same time, the total expression levels
of both proteins were decreased (Fig. 8D, E). The ratio of p-Akt to
Akt and p-mTor to mTor also were increased (Fig. 8F, G). To assess
whether this effect was upstream or downstream of caspase activa-
tion, we performed Western blots on cells that had been pre-treated
with Z-VAD-FMK prior to tropolone exposure (Fig. 8H). Here, we
clearly observed that, in contrast to histone acetylation, the phos-
phorylation of Akt and mTor that was induced by ␣-naphthyl
tropolone was not dependent upon caspase activation. Likewise,
inhibitors of PI3K (LY294002) and mTOR (rapamycin) did not res-
cue the antiproliferative effect of the tropolone (Fig. S1). Results

J. Li et al. / Pharmacological Research 113 (2016) 438–448
447
Previously, Ononye et al. had reported that tropolones inhibit
both cellular proliferation as well as in vitro activity of histone
deacetylases [23], yet the causal relationship between the two
effects was not clear. Our data reveals a complex interplay between
the two effects. On the one hand, tropolone treatment does lead
to increased levels of acetylated histones in cells. On the other
hand, the ability of tropolones to increase histone acetylation is
dependent upon activation of caspases. This suggests that histone
deacetylases are not the direct target of these particular tropolones.
The discrepancy could be due to an inability of the tropolone to
interact with histone deacetylases in a cellular context, lack of
identifying a specific acetylated histone amino acid residue, or
alternatively due to the metal chelating properties of the tropolone
affecting the in vitro enzyme assay in a non-specific manner. Either
way, in our hands all of the histone acetylation sites we examined
were either not affected by tropolone or were affected downstream
of caspase activation.
Our studies also demonstrate the activity of tropolones can be
mitigated by co-treatment with extracellular iron. Notably, the
novel ␣-substituted tropolones differ from ␤-thujaplicin in their
response to extracellular iron. ␤-thujaplicin has been reported to
form a 3:1 (tropolone:iron) complex at low iron concentrations and
a 1:1 complex at higher concentrations [17]. Our data is consistent
with that finding because iron only blocks ␤-thujaplicin activity
at concentrations near the 3:1 ratio. At higher iron concentra-
tions, activity of ␤-thujaplicin is regained. In contrast, iron becomes
inhibitory to the novel ␣-substituted tropolones once it surpasses
a 3:1 ratio. Therefore, while the ␣-substituted tropolones have
significantly stronger activity in the absence of iron, their activ-
ity is diminished more when iron concentrations are high. These
differences may be important with respect to potential in vivo
antagonism. Plasma iron binding is a concern for potential in vivo
applications, as some tropolones appear to bind iron more tightly
than transferrin [38]. While the bulk of plasma iron exists in com-
plex with transferrin and does not reach our experimental levels
[39], it is possible that total plasma iron would be sufficient to block
tropolone distribution in vivo. However, in light of recent findings
that ␤-thujaplicin can block growth of lung adenocarcinoma [40]
and oral squamous cell carcinoma [41] in mouse xenograft mod-
els, the potential for tropolone binding to plasma iron does not
appear to limit in vivo activity, though it remains unclear whether
the similar in vivo differences would be observed in the ␣- versus ␤-
substituted compounds. Furthermore, even at high concentrations,
potent siderophores appear to require nearly an hour to remove
iron from transferrin [42].
which may be detrimental to cellular proliferation [45]. One such
effect may be DNA-damage, as it has been suggested other known
iron chelators may similarly induce a DNA-damage response
[46,47]. In support of the latter model, metalloenzyme inhibitors
are generally thought to be specific to their targets and only mildly
affected by metal chelation [48]. Tropolones have been reported
to inhibit a variety of metalloenzymes in vitro [18–20,23]. In addi-
tional to our observed zinc-dependent HDAC inhibition described
above, tropolones can inhibit the zinc-dependent elastase [18]
and the copper-dependent tryosinase [19], and iron-dependent
prolyl-4-hydroxylases [49]. Although we have now ruled out HDAC
inhibition as a cause of tropolone-induced cellular growth inhibi-
tion, we cannot rule out involvement of another metalloenzyme at
this time.
In conclusion, metal-binding ␣-substituted tropolones demon-
strate more potent and selective antiproliferative effects relative to
the clinical agent vorinostat, and do so through a mechanism that
involves the p53 DNA damage repair pathway leading to induc-
tion of apoptosis. Ultimately, these compounds may offer a unique
alternative to existing agents to targeting metal metabolism or
metalloenzyme activity for treatment of cancers [50].
Conflict of interest statement
A.J.W. is a co-founder of Terpenoid Therapeutics. The current
work did not involve the company.
Acknowledgements
We appreciate the help of Carol Norris at the University of
Connecticut Flow Cytometry and Confocal Microscopy Facility.
Research reported in this publication was supported by a grant
from the American Association of Colleges of Pharmacy (AG140125,
A.J.W., P.I.). The RPCI Hematologic Procurement Shared Resource is
supported by the NCIP30CA16056 core grant.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.phrs.2016.09.
020.
References
[1] R. Bentley, A fresh look at natural tropolonoids, Nat. Prod. Rep. 25 (2008)
118–138.
Determination of the role of iron binding in the mechanism
of tropolone cellular activity proves more difficult. At least two
clear models exist: (1) tropolones directly bind to iron resulting
in alteration of iron homeostasis and potential generation of reac-
tive oxygen through a Fenton reaction, or (2) tropolones inhibit
an unknown iron-dependent metalloenzyme. In support of the
former model, both un-substituted tropolone and ␤-thujaplicin
protected Jurkat cells from apoptosis that was dependent upon
their iron chelating potential [43]. It is possible that the ␣-
substituted tropolones that we examined allow the tropolone
to retain its iron chelating properties after delivery of the com-
pound to an intracellular compartment. This supports prior studies
which demonstrated that toxicity induced by ␤-thujaplicin was
also dependent upon iron chelation leading to increased produc-
tion of reactive oxygen [37,44]. Interestingly, in some cases the iron
chelating agent exhibits enhanced cytotoxicity in the presence of
iron, while others the cytotoxicity is diminished [37]- our novel
␣-substituted tropolones appear to fit the latter category. Presum-
ably disruption of intracellular free iron levels, which would have
immediate benefits in preventing apoptosis following short-term
treatments with reactive oxygen species, has longer term effects
[2] E.Z. Oblak, E.S. Bolstad, S.N. Ononye, N.D. Priestley, M.K. Hadden, D.L. Wright,
The furan route to tropolones: probing the antiproliferative effects of
beta-thujaplicin analogs, Org. Biomol. Chem. 10 (2012) 8597–8604.
[3] S. Liu, H. Yamauchi, Hinokitiol, a metal chelator derived from natural plants,
suppresses cell growth and disrupts androgen receptor signaling in prostate
carcinoma cell lines, Biochem. Biophys. Res. Commun. 351 (2006) 26–32.
[4] J. Zhao, K. Fujita, J. Yamada, K. Sakai, Improved beta-thujaplicin production in
Cupressus lusitanica suspension cultures by fungal elicitor and methyl
jasmonate, Appl. Microbiol. Biotechnol. 55 (2001) 301–305.
[5] S.R. Piettre, A. Ganzhorn, J. Hoflack, K. Islam, J.-M. Hornsperger,
␣-Hydroxytropolones: a new class of potent inhibitors of inositol
monophosphatase and other bimetallic enzymes, J. Am. Chem. Soc. 119
(1997) 3201–3204.
[6] S.R. Piettre, C. Andre, M.C. Chanal, J.B. Ducep, B. Lesur, F. Piriou, P. Raboisson,
J.M. Rondeau, C. Schelcher, P. Zimmermann, A.J. Ganzhorn, Monoaryl- and
bisaryldihydroxytropolones as potent inhibitors of inositol
monophosphatase, J. Med. Chem. 40 (1997) 4208–4221.
[7] D.L. Boger, Y. Zhu, Diels-Alder reactions of cyclopropenone ketals: a concise
tropolone annulation applicable to rubrolone C ring introduction, J. Org.
Chem. 59 (1994) 3453–3458.
[8] N.J. Leonard, G.C. Robinson, Tropolones by isomerization I. 3,
7-Dibenzyltropolone1, J. Am. Chem. Soc. 75 (1953) 2143–2147.
[9] K. Fujita, Y. Bunyu, K. i. Kuroda, T. Ashitani, J. Shigeto, Y. Tsutsumi, A novel
synthetic pathway for tropolone ring formation via the olefin monoterpene
intermediate terpinolene in cultured Cupressus lusitanica cells, J. Plant
Physiol. 171 (2014) 610–614.

448
J. Li et al. / Pharmacological Research 113 (2016) 438–448
[10] V. Hach, E.C. McDonald, Terpenes and terpenoids: IV. some esters and amides
of thujic acid, Can. J. Chem. 51 (1973) 3230–3235.
[31] C.S. Mitsiades, N.S. Mitsiades, C.J. McMullan, V. Poulaki, R. Shringarpure, T.
Hideshima, M. Akiyama, D. Chauhan, N. Munshi, X. Gu, C. Bailey, M. Joseph,
T.A. Libermann, V.M. Richon, P.A. Marks, K.C. Anderson, Transcriptional
signature of histone deacetylase inhibition in multiple myeloma: biological
and clinical implications, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 540–545.
[32] M.J. LaBonte, P.M. Wilson, W. Fazzone, S. Groshen, H.J. Lenz, R.D. Ladner, DNA
microarray profiling of genes differentially regulated by the histone
deacetylase inhibitors vorinostat and LBH589 in colon cancer cell lines, BMC
Med. Genomics 2 (2009) 67.
[33] K. Inoue, M. Kobayashi, K. Yano, M. Miura, A. Izumi, C. Mataki, T. Doi, T.
Hamakubo, P.C. Reid, D.A. Hume, M. Yoshida, W.C. Aird, T. Kodama, T. Minami,
Histone deacetylase inhibitor reduces monocyte adhesion to endothelium
through the suppression of vascular cell adhesion molecule-1 expression,
Arterioscler. Thromb. Vasc. Biol. 26 (2006) 2652–2659.
[34] T. Maeda, M. Towatari, H. Kosugi, H. Saito, Up-regulation of
costimulatory/adhesion molecules by histone deacetylase inhibitors in acute
myeloid leukemia cells, Blood 96 (2000) 3847–3856.
[35] S. Ropero, M. Esteller, The role of histone deacetylases (HDACs) in human
cancer, Mol. Oncol. 1 (2007) 19–25.
[36] N.T. Le, D.R. Richardson, The role of iron in cell cycle progression and the
proliferation of neoplastic cells, Biochim. Biophys. Acta 1603 (2002) 31–46.
[37] E. Yasumoto, K. Nakano, T. Nakayachi, S.R. Morshed, K. Hashimoto, H. Kikuchi,
H. Nishikawa, M. Kawase, H. Sakagami, Cytotoxic activity of deferiprone,
maltol and related hydroxyketones against human tumor cell lines,
Anticancer Res. 24 (2004) 755–762.
[38] L. Hendershott, R. Gentilcore, F. Ordway, J. Fletcher, R. Donati, Tropolone: a
lipid solubilizing agent for cationic metals, Eur. J. Nucl. Med. 7 (1982)
234–236.
[11] M. Baya, P. Soulounganga, E. Gelhaye, P. Gerardin, Fungicidal activity of
beta-thujaplicin analogues, Pest Manage. Sci. 57 (2001) 833–838.
[12] K. Sugawara, M. Ohbayashi, K. Shimizu, M. Hatori, H. Kamei, M. Konishi, T. Oki,
H. Kawaguchi, BMY-28438 (3,7-dihydroxytropolone), a new antitumor
antibiotic active against B16 melanoma I. Production, isolation, structure and
biological activity, J. Antibiot. (Tokyo) 41 (1988) 862–868.
[13] K. Tomita, Y. Hoshino, Y. Nakakita, S. Umezawa, T. Miyaki, T. Oki, H.
Kawaguchi, BMY-28438 (3,7-dihydroxytropolone), a new antitumor
antibiotic active against B16 melanoma II. Taxonomy of producing organism,
J. Antibiot. (Tokyo) 42 (1989) 317–321.
[14] B.E. Bryant, W.C. Fernelius, Formation constants of metal complexes of
tropolone and its derivatives IV. ␣-Bromotropolone, tribromotropolone and
2-aminotropone1, J. Am. Chem. Soc. 76 (1954) 4864–4865.
[15] B.E. Bryant, W.C. Fernelius, Formation constants of metal complexes of
tropolone and its derivatives II. Some alkyltropolones1, 2, J. Am. Chem. Soc. 76
(1954) 1696–1697.
[16] O. Menis, B.E. McClellan, D.S. Bright, Determination of the formation
constants of iron(III) and vanadium(V) with ␤-isopropyltropolone using the
extraction method, Anal. Chem. 43 (1971) 431–435.
[17] Y. Oka, S. Matsuo, Mutual determination of iron (III) and tropolone, and the
formation constant of their complex, Bunseki kagaku 7 (1958) 215–219.
[18] J.L. Fullagar, A.L. Garner, A.K. Struss, J.A. Day, D.P. Martin, J. Yu, X. Cai, K.D.
Janda, S.M. Cohen, Antagonism of a zinc metalloprotease using a unique
metal-chelating scaffold: tropolones as inhibitors of P. aeruginosa elastase,
Chem. Commun. (Camb.) 49 (2013) 3197–3199.
[19] W.T. Ismaya, H.J. Rozeboom, A. Weijn, J.J. Mes, F. Fusetti, H.J. Wichers, B.W.
Dijkstra, Crystal structure of Agaricus bisporus mushroom tyrosinase:
identity of the tetramer subunits and interaction with tropolone,
Biochemistry 50 (2011) 5477–5486.
[20] Y.J. Kim, H. Uyama, Tyrosinase inhibitors from natural and synthetic sources:
structure, inhibition mechanism and perspective for the future, Cell. Mol. Life
Sci 62 (2005) 1707–1723.
[39] R.E. Fleming, B.R. Bacon, Orchestration of iron homeostasis, N. Engl. J. Med.
352 (2005) 1741–1744.
[40] L.H. Li, P. Wu, J.Y. Lee, P.R. Li, W.Y. Hsieh, C.C. Ho, C.L. Ho, W.J. Chen, C.C. Wang,
M.Y. Yen, S.M. Yang, H.W. Chen, Hinokitiol induces DNA damage and
autophagy followed by cell cycle arrest and senescence in gefitinib-resistant
lung adenocarcinoma cells, PLoS One 9 (2014) e104203.
[21] D.R. Hirsch, G. Cox, M.P. D’Erasmo, T. Shakya, C. Meck, N. Mohd, G.D. Wright,
R.P. Murelli, Inhibition of the ANT(2)-Ia resistance enzyme and rescue of
aminoglycoside antibiotic activity by synthetic alpha-hydroxytropolones,
Bioorg. Med. Chem. Lett. 24 (2014) 4943–4947.
[41] Y.-H. Shih, K.-W. Chang, C.-C. Yu, M.-C. Kao, M.Y. Chen, T.-H. Wang, T.-Y. Chi,
Y.-L. Chen, T.-M. Shieh, Hinokitiol suppressed pan-histone expression and cell
growth in oral squamous cell carcinoma cells, J. Funct. Foods 15 (2015)
452–463.
[22] J.L. Buss, E. Arduini, K.C. Shephard, P. Ponka, Lipophilicity of analogs of
pyridoxal isonicotinoyl hydrazone (PIH) determines the efflux of iron
complexes and toxicity in K562 cells, Biochem. Pharmacol. 65 (2003)
349–360.
[23] S.N. Ononye, M.D. VanHeyst, E.Z. Oblak, W. Zhou, M. Ammar, A.C. Anderson,
D.L. Wright, Tropolones as lead-like natural products: the development of
potent and selective histone deacetylase inhibitors, ACS Med. Chem. Lett. 4
(2013) 757–761.
[24] L. Cao, A. Kuratnik, W. Xu, J.D. Gibson, F. t. Kolling, E.R. Falcone, M. Ammar,
M.D. Van Heyst, D.L. Wright, C.E. Nelson, C. Giardina, Development of
intestinal organoids as tissue surrogates: cell composition and the epigenetic
control of differentiation, Mol. Carcinog. 54 (2015) 189–202.
[25] E.Z. Oblak, M.D. VanHeyst, J. Li, A.J. Wiemer, D.L. Wright, Cyclopropene
cycloadditions with annulated furans: total synthesis of (+)- and
(−)-frondosin B and (+)-frondosin A, J. Am. Chem. Soc. 136 (2014) 4309–4315.
[26] C.H. Hsiao, X. Lin, R.J. Barney, R.R. Shippy, J. Li, O. Vinogradova, D.F. Wiemer,
A.J. Wiemer, Synthesis of a phosphoantigen prodrug that potently activates
Vgamma9Vdelta2 T-lymphocytes, Chem. Biol. 21 (2014) 945–954.
[27] P.A. Marks, R. Breslow, Dimethyl sulfoxide to vorinostat: development of this
histone deacetylase inhibitor as an anticancer drug, Nat. Biotechnol. 25 (2007)
84–90.
[42] D.H. Hamilton, I. Turcot, A. Stintzi, K.N. Raymond, Large cooperativity in the
removal of iron from transferrin at physiological temperature and chloride
ion concentration, J. Biol. Inorg. Chem. 9 (2004) 936–944.
[43] P.T. Doulias, L. Nousis, B.Z. Zhu, B. Frei, D. Galaris, Protection by tropolones
against H2O2-induced DNA damage and apoptosis in cultured Jurkat cells,
Free Radic. Res. 39 (2005) 125–135.
[44] K. Murakami, Y. Ohara, M. Haneda, R. Tsubouchi, M. Yoshino, Prooxidant
action of hinokitiol: hinokitiol-iron dependent generation of reactive oxygen
species, Basic Clin. Pharmacol. Toxicol. 97 (2005) 392–394.
[45] T. Simunek, M. Sterba, O. Popelova, M. Adamcova, R. Hrdina, V. Gersl,
Anthracycline-induced cardiotoxicity: overview of studies examining the
roles of oxidative stress and free cellular iron, Pharmacol. Rep. 61 (2009)
154–171.
[46] V.A. Rao, Iron chelators with topoisomerase-inhibitory activity and their
anticancer applications, Antioxid. Redox Signal. 18 (2013) 930–955.
[47] P.J. Jansson, D.S. Kalinowski, D.J. Lane, Z. Kovacevic, N.A. Seebacher, L. Fouani,
S. Sahni, A.M. Merlot, D.R. Richardson, The renaissance of polypharmacology
in the development of anti-cancer therapeutics: inhibition of the triad of
death in cancer by Di-2-pyridylketone thiosemicarbazones, Pharmacol. Res.
100 (2015) 255–260.
[48] J.A. Day, S.M. Cohen, Investigating the selectivity of metalloenzyme inhibitors,
J. Med. Chem. 56 (2013) 7997–8007.
[49] M.J. Lee, J.W. Kim, E.G. Yang, Hinokitiol activates the hypoxia-inducible factor
(HIF) pathway through inhibition of HIF hydroxylases, Biochem. Biophys. Res.
Commun. 396 (2010) 370–375.
[28] C. Mercurio, S. Minucci, P.G. Pelicci, Histone deacetylases and epigenetic
therapies of hematological malignancies, Pharmacol. Res. 62 (2010) 18–34.
[29] A. Subramanian, P. Tamayo, V.K. Mootha, S. Mukherjee, B.L. Ebert, M.A.
Gillette, A. Paulovich, S.L. Pomeroy, T.R. Golub, E.S. Lander, J.P. Mesirov, Gene
set enrichment analysis: a knowledge-based approach for interpreting
genome-wide expression profiles, Proc. Natl. Acad. Sci. U. S. A. 102 (2005)
15545–15550.
[50] Y. Yu, E. Gutierrez, Z. Kovacevic, F. Saletta, P. Obeidy, Y. Suryo Rahmanto, D.R.
Richardson, Iron chelators for the treatment of cancer, Curr. Med. Chem. 19
(2012) 2689–2702.
[30] V.K. Mootha, C.M. Lindgren, K.F. Eriksson, A. Subramanian, S. Sihag, J. Lehar, P.
Puigserver, E. Carlsson, M. Ridderstrale, E. Laurila, N. Houstis, M.J. Daly, N.
Patterson, J.P. Mesirov, T.R. Golub, P. Tamayo, B. Spiegelman, E.S. Lander, J.N.
Hirschhorn, D. Altshuler, L.C. Groop, PGC-1alpha-responsive genes involved in
oxidative phosphorylation are coordinately downregulated in human
diabetes, Nat. Genet. 34 (2003) 267–273.