Developing anticancer ferric prodrugs based on the N-donor residues of human serum albumin carrier IIA subdomain

Article abstract of DOI:10.1021/acs.jmedchem.6b00509

To improve the selectivity, delivery, and activity of ferric (Fe) anticancer agents, we design prodrugs based on N-donor residues of the human serum albumin (HSA) carrier IIA subdomain. We synthesized six Fe(III) compounds derived from 2-hydroxy-1-naphtha

Full text of DOI:10.1021/acs.jmedchem.6b00509

Subscriber access provided by La Trobe University Library
Article
Developing Anticancer Ferric Pro-drugs Based on the N-donor
Residues of Human Serum Albumin Carrier IIA Subdomain
Jinxu Qi, Yi Gou, Yao Zhang, Kun Yang, Shifang Chen, Li
Liu, Xiaoyang Wu, Tao Wang, Wei Zhang, and Feng Yang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00509 • Publication Date (Web): 21 Jul 2016
Downloaded from http://pubs.acs.org on July 23, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Developing Anticancer Ferric Proꢀdrugs Based on the Nꢀdonor
Residues of Human Serum Albumin Carrier IIA Subdomain
1
1
1
1
1
2
3
Jinxu Qi , Yi Gou , Yao Zhang , Kun Yang , Shifang Chen , Li Liu , Xiaoyang Wu ,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
1
1
Tao Wang *, Wei Zhang *, Feng Yang *
1
School of pharmacy, Nantong University, Nantong, Jiangsu, China.
2
Department of Biology, Southern University of Science and Technology,
Shenzhen, Guangdong, China.
3
Ben May Department for Cancer Research, University of Chicago,
Chicago, IL, USA.
*
Corresponding author: Feng Yang
Email : fyang@mailbox.gxnu.edu.cn
Phone/Fax : 86ꢀ773ꢀ584ꢀ8836
Address: 9 Seyuan Road, Nantong, Jiangsu, China. Zip code: 226019
KEYWORDS: human serum albumin; ferric compound; anticancer mechanism;
tumor targeting; therapeutic effect.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
To improve the selectivity, delivery and activity of ferric (Fe) anticancer agents, we
design proꢀdrugs based on Nꢀdonor residues of the human serum albumin (HSA)
carrier IIA subdomain. We synthesized six Fe(III) compounds derived from
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
ꢀHydroxyꢀ1ꢀnaphthaldehyde thiosemicarbazone (7ꢀ12). HSA complex structure
revealed that Fe compound binds to the hydrophobic cavity in the HSA IIA subdomain.
Lys199 and His242 of HSA replace the two Cl atoms of Fe compound, coordinating
3
+
with Fe . In vivo data revealed that compound 12 and HSAꢀ12 complex inhibit the
growth of the liver tumor, and the HSAꢀ12 complex has stronger targeting ability and
therapeutic efficacy than compound 12 alone. In addition, our results have shown that
compound 12 and HSAꢀ12 complex induce Belꢀ7402 cell death possible by several
mechanisms.
2
ACS Paragon Plus Environment

Page 3 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
Many studies have shown that metal compounds have a biological and chemical
1
,2
diversity that organic drugs simply do not. Certain metalꢀorganic ligand complexes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
have increased biological activity over ligands alone, especially in antitumor
3,4
activity. Therefore, a different strategy for developing anticancer drugs would be to
5
ꢀ7
design and synthesize metal complexes derived from organic ligands.
Current studies reveal that Fe and Cu chelators are promising novel anticancer
8,9
agents, as cancer cells can absorb more Fe and Cu than the normal cells. For
example, both desferrioxamine (DFO) (through in vitro and in vivo studies) and
10ꢀ12
2
ꢀhydroxyꢀ1ꢀnaphthylaldehyde isonicotinoyl have limited tumor activity.
In fact,
3ꢀaminopyridineꢀ2ꢀcarboxaldehyde thiosemicarbazone (3ꢀAP) has reached clinical
12
trials as an anticancer therapeutic. In addition, a series of novel chelators, such as
thiosemicarbazones, have pronounced anticancer activity and selectivity, and
8,9
overcome resistance to established chemotherapeutics in vivo. However, these
chelators may produce low antitumor activity; in addition, doseꢀlimiting side effects,
including methemoglobinemia and hypoxia, have limited their potential in the
1
2ꢀ14
clinic.
Richardson et al. have investigated the effects of Cu(II) complexes with
thiosemicarbazones and determined that they have the potential for antipoliferative
3
,12
activity along with redox in an accessible range for natural reductants.
These
pioneering studies have shown novel anticancer metal agents, derived from
2
thiosemicarbazone ligands can be developed. Currently, metal compounds derived
with a great variety of ligands are under extensive investigation as anticancer
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
5,16
agents.
Although a large number of metal compounds have been evaluated in vitro and in
vivo, and some have even reached clinical trials, researchers still need to determine
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
how to increase targeting success and decrease in vivo side effects; the use of
1
7ꢀ20
proꢀdrugs and drug carriers may help to tackle these issues.
Human serum
albumin (HSA) is a promising nonꢀtoxic, nonꢀantigenic, biocompatible, and
biodegradable endogenous protein drug delivery system that lacks
21ꢀ24
immunogenicity.
HSA, the most abundant protein in plasma with many active
25ꢀ28
residues, can bind to a diverse group of exogenous compounds.
Therefore, to
improve the targeting of anticancer agents, these properties can be utilized to design
2
0,30
novel anticancer proꢀdrugs.
There are two major ways for design proꢀdrugs based
on HSA: through chemically coupling the metal proꢀdrug to the residues of HSA; or
30,31
through direct complexation of the proꢀdrug with HSA.
Unfortunately, there can
be problems with the release of a drug when metal compounds are delivered to cancer
cells using complexed HSA carriers in vivo due to the strength of the binding between
30
HSA and the proꢀdrug. For example, the metal proꢀdrug can be released from HSA
prematurely if it is not bound strongly or it may not be released at all if the bond is too
tight. To address these issues, Yang et al. proposed the use of a metal proꢀdrug whose
3
0,32ꢀ34
design is based on the natural HSA IIA subdomain and the known cancer cell.
The metal proꢀdrug can contain potential leaving group(s) with no antiꢀcancer activity
that are then displaced by Lys199 and/or His242 residue(s) of HSA. This is outlined
below.
4
ACS Paragon Plus Environment

Page 5 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Among the next generation of metalꢀbased anticancer compounds, Fe anticancer
1
compounds may be promising. In humans, Fe is an essential bioactive element with
30
an oxidative nature. HSA has active Nꢀdonor residues (Lys199 and His242). These
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
residues, which can substitute for metal compound ligands that bind to the HSA IIA
33,34
subdomain, can converge to the metal center.
Therefore, to enhance the delivery
efficiency, selectivity, and anticancer activity of the Fe agent, Fe antiꢀcancer proꢀdrug
with two leaving groups that initially binds to the HSA IIA subꢀdomain was created. It
then binds to Lys199 and His242 of HSA, displacing the Fe proꢀdrug’s leaving groups.
Subsequently, Lys199 and His242 of HSA are protonated in the cancer cell’s
lysosomal acidic environment, which decreases its coordination ability with Fe ion.
This allows the Fe agent to be released from the HSA carrier (Figure 1). Furthermore,
liver cancer, which has become very common across the globe, is responsible for a
35
large number of cancer deaths. Taking into consideration the above factors, a model
was created using Fe agents derived from thiosemicarbazones and liver cancer
(Belꢀ7402) cells to design anticancer Fe proꢀdrugs based on the Nꢀdonor residues
(
Lys199 and His242) of HSA. The following studies were carried out: (1) design and
synthesis of six 2ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀthiosemicarbazone Fe(III)ꢀSchiff base
anticancer compounds (7ꢀ12) (Figure 2), followed by investigation of their
structureꢀactivity relationships on Belꢀ7402 cells; (2) provided feasibility evidence of
developing Fe proꢀdrugs based on Nꢀdonor residues of the HSA carrier IIA subdomain;
(3) compared in vivo targeting ability, therapeutic efficacy, and side effects of the HSA
complex to the Fe compound; (4) investigated the release behavior of the Fe
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
compound from HSA and the possible mechanism for HSA complex penetrating
cancer cells; (5) determined the potential anticancer mechanism of the Fe
compound/HSA complex.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RESULTS
This study focuses on Fe compounds containing the tridentate
2
ꢀHydroxyꢀ1ꢀnaphthaldehyde thiosemicarbazone Schiffꢀbase ligands (TsS) for the
3
6
following reasons: (1) TsS ligands themselves are promising anticancer agents; (2)
hydrophobic properties of the benzyl and alkyl components can facilitate targeting of
the hydrophobic binding sites of the HSA IIA subdomain by the Fe compound.
Therefore, the TsS ligand was designed to complex with the Fe compounds
(pharmacophore), and to then allow for the two Cl atoms to be potential leaving
groups (Figure 2).
Design and structure of Fe compounds
It was determined by Richardson et al. that modification at the N4 position of a
thiosemicarbazone can effectively increase lipophilicity and produce superior
12
antiproliferative activity. Thus, we performed this modification with alkyl or phenyl
groups to investigate the structureꢀactivity relationships of Fe (III) compounds (Figure
2
). The following six Fe (III) compounds were synthesized according to the same
method. The single crystals of compounds (compound 7ꢀ12), suitable for Xꢀray
diffraction were obtained.
Singleꢀcrystal structure analysis showed that all compounds are isomorphous. They
all have the same skeleton (1 tridentate Schiff base ligand, 1 Fe(III) center, and two
6
ACS Paragon Plus Environment

Page 7 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
coordinated Cl atoms), and differ only in the modified group of the Schiff base ligand.
As shown in Figure 2, the coordination polyhedron around all of the Fe(III) centers
forms a distorted square pyramid (τ = 0.43 for compound 7, 0.27 for compound 8,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
.34 for compound 9, 0.11 for compound 10, 0.12 for compound 11, and 0.38 for
compound 12). The metal was displaced from the O1/N1/S1/Cl1 basal plane, and the
Cl appears at the apex. The ligand and metal ion complex forms a five and a
2
sixꢀmembered closed loop ring, this coordination can enhance the stability of
complexes. We examined the antitumor activity (see Biological Studies) and
considered the probability of its involvement in the thiosemicarbazone activity to
determine the crystal structure of each Fe complex.
Feasibility of developing Fe proꢀdrug based on the Nꢀdonor residues of HSA IIA
subdomain
HSA fluorescence (~347 nm) was gradually suppressed with the increase of
compound 12 concentrations, at pH 4.7 and 7.4 (Figure S1A and B). This indicates
that the Fe compounds bind close to the HSA IIA subdomain. The MALDIꢀTOFꢀMS
spectrum shows an increase of 366 Da for the HSA complexes, relative to pure HSA.
This is equivalent to the molecular weight of ca. one compound 12 (Figure S1C).
Furthermore, the electrospray ionization mass spectrometry (ESIꢀMS) of products
released from the HSAꢀ12 complex at pH 4.7, showed an intense signal at m/z =
3
68.04 (Figure S2). This was identified with isotopic envelopes corresponding to
2+
[compound 12] (fit: 368.05), implying that two Cl atoms are lost from compound
2.
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
We needed to resolve the structure of the HSAꢀPAꢀ12 complex in order to gather
feasibility evidence for developing Fe proꢀdrugs based on Lys199 and His242
residues of the HSA IIA subdomain. Electron density maps of the compound 12
compound complexed with HSA reveal the presence of one compound 12 molecule at
the IIA subdomain (Figure 3A). The compound 12 bound to the HSAꢀPA complex is
heartꢀshaped (Figure 3B) and in the HSA IIA subdomain, the Fe compound binds to a
large hydrophobic pocket, which is surrounded by residues, including Arg218,
Arg222，Lys199, Trp214, Leu219, Phe223, Leu238, His242, Arg257, Leu260, Ile264,
Ser287, Ile290, and Ala291 (Figure 3C). Lys199 and His242 replaced two Cl atoms of
compound 12, within the HSAꢀPAꢀ12 complex structure and they coordinate with
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
+
Fe (Figures 3A and C). Comparing binding site and binding mode of compound 12
2ꢀ
to HSA with cisplatin and [RuCl
5
(ind)] , we found that they bind to HSA by residue(s)
3
2, 37
of HSA displacing the ligand(s) in metal compounds, coordinated to metal.
2ꢀ
Interestingly, cisplatin can not bind to IIA subdomain, but [RuCl (ind)] and
5
compound 12 could bind to IIA subdomain. Obviously, the molecular structure and
geometry of metal compounds play an important role in their binding site and binding
mode in HSA.
Structureꢀactivity relationships of anticancer Fe compounds in vitro
Fe compounds have higher cytotoxicity against Belꢀ7402 cells than the ligands and
3+
3+
Fe alone (Table 2), implying that the chelation of Fe to ligands is the reason that
the Fe compounds have such high cytotoxicity. The cytotoxicity of Fe compounds
against Belꢀ7402 cells (≤ 13.36 ± 1.35 µM) is slightly higher than that of cisplatin
8
ACS Paragon Plus Environment

Page 9 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
14.54 ± 0.85 µM). Among the six Fe compounds studied, compound 7 has the lowest
cytotoxic activity against Belꢀ7402 cells (13.36 ± 1.35 µM). When, compared to
compound 7, where one H atom attached to the N4 of thiosemicarbazone is replaced
by a methyl or aryl group, as in compound 8 and compound 9, the cytotoxic activity is
enhanced (11.87 or 2.22 µM). The Fe compounds derived from thiosemicarbazone
where both of the H atoms at the N4 position are altered, i.e., in (compound 10,
compound 11, and compound 12), even higher cytotoxicity is observed (0.86−0.65
µM) (Table 2). The Fe compounds in complexation with HSA increase their
cytotoxicity in Belꢀ7402 cells by approximately 2–6 fold, relative to the Fe
compounds alone; however, in normal cells [immortalized human hepatocyte cells
(HL7702)], they do not raise cytotoxicity levels (Table 2). As compound 12 and
HSAꢀ12 complex have high cytotoxic activity on Belꢀ7402 cells, relative to other Fe
compounds in vitro, these were selected for further investigation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In vivo animal studies of compound 12 and HSAꢀ12 complex
To further evaluate combining the HSA carrier with proꢀdrug strategies for the
therapeutic efficacy of Fe compound in vivo, the liver cancer Belꢀ7402 xenograft
mouse model was established.
Acute toxicity of compound 12 and HSAꢀ12 complex in vivo
We measured levels of creatine kinase (CK), blood urea nitrogen (BUN), aspartate
aminotransferase (AST), and alkaline phosphatase (ALT) 3 days after intravenous
injection of the drugs in order to assess acute toxicity of compound 12 and HSAꢀ12
complex in the heart, kidney, and liver of normal mice. Levels of CK in compound
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
2ꢀtreated mice were greater than levels in the control (NaCl) group. However, the
levels of CK in the HSAꢀ12 complexꢀtreated mice were similar to those of the control.
This indicates that HSAꢀ12 complex has low cardiotoxicity. High BUN levels
correspond to higher toxicity in the kidney. compound 12 caused significant
nephrotoxicity, as evidenced by a higher level of BUN than that observed in the
control group. In contrast, BUN levels significantly decreased in the HSAꢀ12
complexꢀtreated groups, indicating lower nephrotoxicity. Serum AST and ALT levels
were significantly elevated in the group treated with compound 12; however, the
levels induced by HSAꢀ12 complex were almost the same as the control group, which
indicates lower levels of liver damage in those treated with HSAꢀ12 complex.
Antiꢀtumor activity of compound 12 and HSAꢀ12 complex in vivo
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Belꢀ7402 tumorꢀbearing mice were injected with compound 12 and HSAꢀ12
complex and saline was injected as a control to evaluate therapeutic efficacy.
Tumorꢀbearing mice had their body weight and tumor volume monitored every 2 days
for 28 days. At the end of the experiment, the tumor volumes in the HSAꢀ12
complexꢀtreated mice were much smaller than those in salineꢀtreated and compound
12ꢀtreated mice (Figure 4A). Compared to the control group, the tumor volume after
2
8 days treatment was 53.2 ± 4.5% for the compound 12 group and 24.6 ± 2.8% for
the HSAꢀ12 complex group, demonstrating that HSAꢀ12 complex resulted in greater
tumor growth inhibitory efficacy than compound 12. Compared to the NaCl group,
the tumor inhibition rate (TIR) of HSAꢀ12 complex was 75.6% (P < 0.001), which
was significantly higher than that of compound 12 (46.8%, P < 0.01).
10
ACS Paragon Plus Environment

Page 11 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To further evaluate the antitumor effects of compound 12 and HSAꢀ12 complex in
animals, the tumor tissues were excised for pathology. The TUNEL stained tissue
sections indicate obvious differences in tumor tissue morphology between treated
groups and the NaCl group (Figure 4B). As shown in Figure 4B, tumor samples from
animals treated with compound 12 and HSAꢀ12 complex showed increased apoptosis
compared with tumor samples from the control mice. In particular, HSAꢀ12 complex
was more effective in promoting cell necrosis, relative to compound 12.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Comparison of the side effects of compound 12 and HSAꢀ12 complex in vivo
To evaluate the toxicology of compound 12 and HSAꢀ12 complex in mice, mouse
body weight was monitored throughout the intravenous treatment. At the end of the
study, specific organs were collected and weighed (Table S3). When compared with
the control mice, those intravenously treated with HSAꢀ12 complex, did not show
significant weight loss (p > 0.05). In contrast, mice treated with compound 12 lost
approximately 10% of their body weight (Figure 5A). The weight of the liver and
heart of mice treated with compound 12 altered slightly, but no change was observed
in mice treated with HSAꢀ12 complex (Table S3) and there was not a significant
difference in the weight of the other organs when comparing the HSAꢀ12
complexꢀtreated and the control groups (Table S3). Pathological sections stained with
H&E revealed drugꢀrelated side effects to major organs (Figure 5B). Serious damage
to the heart (vacuolation), liver (vacuolation), and kidneys (focal abnormalities) was
observed in mice treated with compound 12. This damage was significantly decreased
in mice treated with the HSAꢀ12 complex. Therefore, increased liver and heart
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
weights are likely attributed to increased vacuolation.
Selectivity of HSA complex in vivo
To evaluate whether the HSA complex selectively accumulates in the Belꢀ7402
tumor in vivo, the Fe content in tumors of mice treated with compound 12 and
HSAꢀ12 complex, was measured. Using inductively coupled plasma atomic emission
spectrometry (ICPꢀAES), the iron content in tumors treated with HSAꢀ12 complex is
significantly (p<0.05) higher compared to those treated with compound 12. This data
suggests that the HSA complex selectively accumulates in tumors. Furthermore, the
data also indicates that the HSA complex decreased Fe compound accumulation in the
liver and kidney (Figure 6A).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Fe proꢀdrug release from HSA
To examine the release behavior of Fe from HSA in different environments, the
environment inside the cancer cell was stimulated, and the compound 12 released
from the HSA carrier was measured in buffers of pH 4.7 and pH 7.4. Approximately
5% of the Fe compound was released from the HSA complex within 48 h in pH 7.4
buffer, whereas up to 80% of Fe compound was released from HSA in pH 4.7 buffer
(Figure 6B). In addition, the binding affinity of Fe compound to HSA in pH 7.4 buffer
6
−1
[
K = 6.88 (±0.08) × 10 M ] was significantly greater than that of HSA to Fe
4
−1
compound in pH 4.7 buffer [K = 4.75 (±0.05) × 10 M ]. These results indicate that
Fe compound bind weakly to, and are more easily released from, HSA in an acidic
environment.
Possible mechanism of HSAꢀ12 complex absorbed by cancer cells
12
ACS Paragon Plus Environment

Page 13 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
It was determined by Desai et al. that after interacting with SPARC (Secreted
Protein, Acid, and Rich in Cysteine) on tumor cell surfaces, HSA penetrates the cell
3
8,39
membrane by the process of caveolae protein mediated endocytosis.
To determine
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
if a similar mechanism for HSAꢀ12 complex penetration into Belꢀ7402 cells is
observed here, the organization of caveolae was disturbed by treating Belꢀ7402 cells
40
with methylꢀβꢀcyclodextrin, a known endocytosis inhibitor. The HSA and Fe
permeability were then measured. Compared with the control (0 mM of
methylꢀβꢀcyclodextrin), the amount of intracellular HSA and Fe in cells incubated
with methylꢀβꢀcyclodextrin was reduced (Figure 7). In addition, the amount of
intracellular HSA and Fe in cells incubated with l mM methylꢀβꢀcyclodextrin is more
than that of cells incubated with 2 mM methylꢀβꢀcyclodextrin (Figure 7). These
results affirm that the Fe cell content is explicitly proportional to the HSA cell content.
This indicates that methylꢀβꢀcyclodextrin prevents the albuminꢀinduced increase in
transendothelial HSA and Fe permeability, which results in reduction of the amount of
HSA and Fe in cells. In turn, this suggests that the HSAꢀ12 complex possible
penetrates Belꢀ7402 cells by endocytosis.
Potential anticancer mechanism of compound 12/HSAꢀ12 complex
DNA transcription and replication require separation of the superꢀcoiled DNA
doubleꢀhelix. TopoꢀIiα is responsible for unlinking DNA during replication, which
then allows DNA relaxation during transcription and supports chromatin
41
remodeling. Disruption of TopoꢀIIα leads to doubleꢀstrand DNA breaks, ultimately
causing cell death. This has led to the identification of TopoꢀIIα as one of the most
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
important clinical targets for modern cancer chemotherapy, and its inhibitors are the
42
central components of many therapeutic regimens. Recently, many reports have
4
2ꢀ45
acknowledged that thiosemicarbazone derivatives likely target TopoꢀIIα
so we
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
decided to evaluate the use of human TopoꢀIIαꢀmediated pBR322 relaxation and
kDNA decatenation assays to determine if the compound 12 compound can indeed
inhibit TopoꢀIiα. The results show that the compound 12 significantly inhibits
TopoꢀIIα activity, suggesting compound 12 is a potent TopoꢀIIα inhibitor (Figure 8A).
41
TopoꢀIIα inhibition is thought to affect cell cycle. Thus, we investigated the
influence of compound 12 and HSAꢀ12 complex on the cell cycle distribution in
human liver cancer cells using flow cytometric DNA content analysis. We noted that
in the S phase, when treated with compound 12, the percentage of cells increased
from 27.78% to 42.10% than when treated with HSAꢀ12 complex. The control cells,
with no drug treatment, had 19.99% of cells in the same phase (Figure 8B). The Fe(III)
compound and HSA complex may cause cell accumulation in the S phase of the cell
cycle by either delaying or inhibiting progression to the next phase. Cell cycle
regulatory protein expression, including that of cyclins and cyclinꢀdependent kinases,
was investigated to elucidate the molecular mechanism underlying the S phase arrest
induced by Fe(III) compound and HSA complex in Belꢀ7402 cells. compound 12
increased the expression of cyclins E and A (Figure 8D), which associate with Cdk2;
the resulting complexes, cyclin E/Cdk2 and cyclin A/Cdk2, regulate the initiation and
4
6,47
progression of the S phase, respectively.
The expression of cyclin E and cyclin A
was high after 24 h, which suggests that compound 12 persevered in enhanced S
14
ACS Paragon Plus Environment

Page 15 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
phase progression and compound 12 caused increased E2Fꢀ1 expression (which
48
promotes cyclin E and A transcription) at 24 h (Figure 8D). The enhanced
expression of cyclin E and A and E2Fꢀ1 appear to be directly responsible for the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
compound 12 induced S phase arrest.
ꢀ
Many studies have affirmed that metal compounds can introduce reactive oxygen
4
9
species (ROS) into cancer cells. Increased ROS can damage macromolecules and
5
0
alter cell function or induce apoptosis. To investigate if compound 12 and HSAꢀ12
complex can produce ROS in Belꢀ7402 cells, a fluorescent DCF probe and flow
cytometry were utilized. The cells treated with compound 6 and compound 12 or
HSAꢀ12 complex have greater DCF fluorescence intensity and a right shifted peak
relative to control cells (Figure 9A). compound 6 does not obviously increase ROS
compared to control cells after 24 h. However, compound 12 significantly (p < 0.001)
2
increased H DCF oxidation 223 ± 20% after 24 h incubation. Thus, compound 12
produces significantly (p < 0.001) more ROS than compound 6 (105 ± 20%), but did
not produce as much as HSAꢀ12 complex (355 ± 20%). Prior studies have confirmed
that Fe compounds introduce reactive oxygen species via redoxꢀcycling. In their
II
reduced (Fe ) form, these complexes can react with molecular oxygen; the resulting
ferric complexes may subsequently interact with cellular reductants. Voltammograms
were recorded to determine if compound compound 12 can generate reactive oxygen
III
II
species through the reduction of Fe to Fe (Figure 9B). Redox potentials of the
III
compound 12 compound are relatively high, suggesting that the Fe compound can
be easily reduced and then able to take part in redox cycling and ROS generation.
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
0,51
Increased ROS and mitochondrial dysfunction are related apoptosis events.
The apoptosis process is mediated by the mitochondrial/apoptotic cascade and is often
linked with mitochondrial membrane changes caused by the leakage of proꢀapoptotic
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
2
factors. The lipophilic fluorescent probe JCꢀ1 (5, 5, 6, 6′ꢀtetrachloroꢀ1, 1′, 3,
3
′ꢀtetraethylꢀimidacarbocyanine iodide) was used to measure mitochondrial
membrane potential changes (ꢁψm). JCꢀ1 staining indicates decreased ꢁψm, as
evidenced by less red fluorescence (JCꢀ1 aggregates) and increased green
49
fluorescence (JCꢀ1 monomers). Belꢀ7402 cells were then treated with 1 ꢂM of FeCl
3
,
compound 6, compound 12, or HSAꢀ12 complex for 24 h and ꢁψm was decreased
Figure 9C). Mitochondrialꢀmediated apoptosis activation was confirmed and each
compound’s ability to decrease ꢁψm was consistent with in vitro cytotoxicity.
BeLꢀ7402 was incubated with FeCl and compound 6 for 24 h, which decreased ꢁψm
(3.46% and 5.52%, respectively) (p < 0.05), and a significant (p < 0.001) decrease in
ψm was observed after incubation with compound 12 (12.0%) or HSAꢀ12 complex
(
3
ꢁ
(22.2%) (Figure 9C).
TopoꢀIIα inhibition could potentially lead to an attack on components of DNA,
41
DNA strand breaks, DNA intraꢀstrand adducts, and DNA–protein crosslinks.
Transcription factor p53 is another major player in apoptosis or cell cycle arrest
response of cells to DNA damage. Western blotting was used to detect changes of p53
to determine if compound 12 and HSAꢀ12 complex had caused DNA damage.
Pretreatment with HSAꢀ12 complex enhanced radiationꢀinduced DNA damage more
than compound 12, as evidenced by increased p53 phosphorylation levels as shown in
16
ACS Paragon Plus Environment

Page 17 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
3,54
Figure 10. The Bclꢀ2 protein family are thought to be key regulators of ꢁψm.
Thus,
Western blotting was used to determine if compound 12 or HSAꢀ12 complex
significantly (p < 0.001) upregulated Bax expression (proꢀapoptotic Bclꢀ2 protein
family) and suppressed expression of Bclꢀ2 and Bclꢀxl (proꢀsurvival Bclꢀ2 family
proteins). We determined that the ratio of Bclꢀ2/Bax was decreased, which indicates
that Bclꢀ2 proteins regulated the loss of ꢁψm. HSAꢀ12 complex significantly
upregulated the expression of Bax and suppressed Bclꢀ2 and Bclꢀxl expression in
BELꢀ7402 cells in comparison to compound 12 (p < 0.001). In addition, Western
blotting revealed that cytochrome c (Cyt C) and the caspase protein family (caspaseꢀ3
and ꢀ9) were greatly upregulated (p < 0.001), suggesting that Cyt C was the cause of
caspaseꢀ3 and ꢀ9 activation. Results show that HSAꢀ12 complex is significantly (p <
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
.001) more effective in upꢀregulating caspase family protein expression and Cyt C in
Belꢀ7402 cells, compared to compound 12.
We used Annexin VꢀFITC/PI staining to further valid compound 12 and HSAꢀ12
complex’s ability to bolster cell apoptosis; the data showed Belꢀ7402 cell apoptosis of
1
0.3% for compound 12 and 20.9% for HSAꢀ12 complex (Figure 10C). Results show
that HSAꢀ12 complex is significantly (p < 0.001) more effective at promoting
Belꢀ7402 cell apoptosis compared to compound 12.
DISCUSSION
The study demonstrated that the anticancer activity of ligands may be enhanced by
coordinating with complexes containing Fe(III). The nature of the Feꢀbound ligand
directly affected the antiꢀcancer properties of the Fe compound and lipophilic groups
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
attached to the ligand also played an important role in increasing the antiꢀcancer
activity of the potential Fe therapeutic. When we replaced the H atom in the N4
location of the ligand with a more lipophilic functional moiety, such as an alkyl or
aryl group, the other five Fe(III) compounds we investigated showed a oneꢀ to
twentyꢀfold increase in antiꢀcancer activity relative to compound 7 (Table 1), and the
Fe(III) compound is proportional to the degree of lipophilicity of the ligand. With the
basic pharmacophore of the ligand attached to Fe(III) left intact, lipophilic groups of
the ligand could be modified in an attempt to regulate the antiꢀcancer activity of the
Fe(III) compound.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
There is the potential to increase anticancer activity and metal drug selectivity by
altering the carrier; however, efficacy of in vivo drug delivery and the ability to
release in cancer cells need to be simultaneously taken into consideration. Yang et al.
has proposed a new strategy for designing metal proꢀdrug that is based on the nature
32ꢀ34
of the cancer cells and the HSA IIA subdomain.
The results of this study confirm
that it is feasible to design an Fe proꢀdrug according to Yang et al.’s proposal. By
examining the HSA complex structure, it was determined that Lys199 and His242
replaced two Cl atoms of the Fe compound, and coordinated with Fe(III), which
enables the Fe compound to forge a strong bond in the hydrophobic cavity of the HSA
IIA subdomain. As a result, minimal Fe compound (about 5%) is released from HSA
at pH 7.4, whereas approximately 80% of Fe compound are released from HSA in
acidic environments (pH 4.7). This is due to a dramatic decrease in the Fe compound
binding affinity in these two environments. Based on these release profiles, it is
18
ACS Paragon Plus Environment

Page 19 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
suggested that the HSA complex will be stable in blood during in vivo circulation, and
that the Fe compound will be released after there is a selective accumulation of acid
lysosome in the cancer cells.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The therapeutic efficiency of compound 12/HSAꢀ12 complex in vivo was
evaluated by considering three factors: TIR, side effects, and selectivity, in mice. The
TIR of the FAꢀHSAꢀ12 complex treated mice almost reached 75.6%, and was 1.6
times higher than that of the compound 12 treated group. FAꢀHSAꢀ12 complex is well
tolerated in mice and causes less weight loss relative to compound 12 alone. H&E
staining indicated that HSAꢀ12 complex can effectively reduce the heart, liver, and
kidney damage induced by free compound 12. Furthermore, the ICPꢀAES results
show that HSAꢀ12 complex facilitates Fe compound accumulation in the tumor in
vivo. The liver cancer xenograft experiments in mice further indicate that HSA
complexation is helpful for decreasing side effects, improving antiꢀtumor activity and
selectivity of Fe compound. Furthermore, the HSA complex has a lower toxicity
against normal cells and tissues than the Fe compound. This is due to a more selective
accumulation of the HSA complex in tumor cells, enhanced permeability, and the
3
7ꢀ39
retention (EPR) effect of the macromolecule in tumors (Figure 11).
These results confirm that the HSA complex has a better therapeutic efficacy and
lower systemic toxicity, than free compound 12. The principle of the classic approach
for developing metal anticancer drugs is for their them to quickly replicate and speed
5
up the mitotic processes of malignant cells. However, there are still some major
drawbacks concerning these metal drugs: they only work with certain types of cancers;
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
some tumors may have acquired or intrinsic resistance; and they often have severe
7
side effects. Therefore, new approaches are needed to develop cancer drugs that
circumvent these drawbacks. Among these, developing anticancer drugs that target
proteins related to cancer may be promising. Excitingly, the results outlined here show
that Fe compounds eradicate cancer cells using multiple anticancer mechanisms, such
as regulating the expression of proteins in cancer cells. Furthermore, the anticancer
capacity of the Fe compound is improved when complexed with HSA, relative to the
Fe compound alone. Therefore, it is necessary to create a multiꢀtarget anticancer Fe
proꢀdrug based on the HSA carrier that improves delivery efficiency and increases
anticancer activity and selectivity while overcoming resistance.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CONCLUSION
Due to His242 and Lys199 replacing the groups of Fe proꢀdrug that have left and
3
+
then coordinating with Fe , certain aspects of Fe compounds have improved,
including delivery efficiency, anticancer activity, and selectivity, since they bind
strongly to the IIA subdomain of HSA. Compared to the Fe compound alone, the HSA
complex showed better tolerance, higher drug accumulation in tumor tissues, and
lower toxicity, indicating that it had superior antiꢀtumor activity and was associated
with milder side effects. These results suggest that the HSA carrier proꢀdrug strategy
for intravenous administration of novel, active thiosemicarbazone containing Fe(III)
compounds, may be a promising approach for targeted cancer therapy.
Experimental Section
Chemicals and reagents. HSA (fatty acid content <0.05%) was purchased from
20
ACS Paragon Plus Environment

Page 21 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
SigmaꢀAldrich (Shanghai, China) and used without requiring further purification. All
other chemicals and solvents were of high purity and are available from commercial
sources. Distilled water was used in the reactions. Elemental analyses (C, N, H, and S)
were carried out on a PerkinꢀElmer 2400 analyzer. We use the xꢀray diffraction to
determine the compounds structure, and element analysis (C, H, N, S) to determine
the purity of compounds are ≥ 95%.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis and characterization of ligands. The ligands compound 1ꢀ6 were
4,36
prepared according to the following procedures.
In brief; the thiosemicarabazide
(
10 mmol) and 2ꢀHydroxyꢀ1ꢀnaphthaldehyde (10 mmol) were combined in methanol
20 mL) with acetic acid (1ꢀ2 drops). While the mixture was boiled, a white or
(
offꢀwhite precipitate appeared; after 4 h, it was allowed to cool at room temperature.
The mixture was subsequently filtered, washed with cold water, and dried in vacuo.
2ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀthiosemicarbazide (1). Yield: 8.24 mmol (82.4%).
Anal. Calcd (%) for C H N OS: C, 59.00; H, 4.13; N, 17.20; O, 6.55; S, 13.13.
12
10
3
ꢀ1
Found: C, 59.05; H, 4.10; N, 17.21; O, 6.51; S, 13.13. IR, cm : 3698 (m, OH), 3449
(s, amide), 3256 (s, NH), 3165 (m, aromatic hydrogen), 1607 (s), 1573 (s), 1510 (s),
1471 (s, aromatic), 1394 (s, C=N), 1328 (s, thioamide), 1238 (s), 1118 (s), 880 (m),
+
8
19 (m, CꢀH), 748 (m, C=S), 647 (m). ESI m/z: calcd for C12
H
10
N
3
OS, 244.02 [M ꢀ
H].
ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀ4ꢀmethylthiosemicarbazide (2). Yield: 7.96 mmol
2
(79.6%). Anal. Calcd (%) for C13
H
13
N
3
OS: C, 60.44; H, 4.68; N, 16.27; O, 6.19; S,
ꢀ1
1
2.41. Found: C, 60.40; H, 4.65; N, 16.30; O, 6.16; S, 12.48. IR, cm : 3702 (m, OH),
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
1
419 (s, amide), 3201 (s, NH), 3007 (m, aromatic hydrogen), 1620 (m), 1596 (s),
532 (s), 1428 (s, aromatic), 1397 (m, C=N), 1330 (s, thioamide), 947 (s), 888 (s,
+
CꢀH), 776 (m), 744 (m), 716 (m, C=S), 684 (m). ESI m/z: calcd for C13
H
13
N
3
OS,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
58.06 [M ꢀ H].
ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀ4ꢀphenylthiosemicarbazide (3). Yield: 85.5 mmol
OS: C, 67.27; H, 4.70; N, 13.07; O, 4.98; S,
2
(85.5%). Anal. Calcd (%) for C18
H
15
N
3
ꢀ1
9
3
1
.98. Found: C, 67.22; H, 4.73; N, 13.03; O, 4.99; S, 10.13. IR, cm : 3711 (m, OH),
385 (s, amide), 3151 (s, NH), 3000 (m, aromatic hydrogen), 1621 (m), 1595 (s),
542 (s), 1442 (m, aromatic), 1415 (m, C=N), 1326 (s, thioamide), 1094 (s), 939 (s,
+
CꢀH), 902 (m), 856 (m), 766 (m, C=S), 719 (m). ESI m/z: calcd for C18
15 3
H N OS,
320.10 [M ꢀ H].
2
ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀ4,4ꢀdimethylthiosemicarbazide (4). Yield: 9.01
mmol (90.1%). Anal. Calcd (%) for C14
H
15
N
3
OS: C, 61.51; H, 5.53; N, 15.37; O, 5.85;
ꢀ1
S, 11.73. Found: C, 61.51; H, 5.53; N, 15.37; O, 5.85; S, 11.73. IR, cm : 3739 (m,
OH), 3279 (s, amide), 3039 (s, NH), 3000 (m, aromatic hydrogen), 1619 (s), 1548 (m),
1
520 (s), 1462 (s, aromatic), 1357 (m, C=N), 1238 (s, thioamide), 1181 (s), 1141 (s),
+
906 (m), 884 (m, CꢀH), 650 (m, C=S), 594 (m). ESI m/z: calcd for C14
H
15
N
3
OS,
2
72.09 [M ꢀ H].
2ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀ4,4ꢀdiethylthiosemicarbazide (5). Yield: 7.90 mmol
79.0%). Anal. Calcd (%) for C16 OS: C, 63.76; H, 6.35; N, 13.94; O, 5.31; S,
10.64. Found: C, 63.74; H, 6.36; N, 13.90; O, 5.33; S, 10.67. IR, cm : 3715 (m, OH),
(
19 3
H N
ꢀ1
3
293 (s, amide), 2982 (s, NH), 2934 (m, aromatic hydrogen), 1620 (m), 1590 (m),
22
ACS Paragon Plus Environment

Page 23 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
3
2
538 (s), 1462 (s, aromatic), 1414 (s, C=N), 1356 (s, thioamide), 1271 (m), 1133 (s),
+
074 (m), 955 (m, CꢀH), 803 (m, C=S), 694 (m). ESI m/z: calcd for C16
19 3
H N OS,
00.13 [M ꢀ H].
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀ3ꢀpiperidinethiosemicarbazide (6). Yield: 7.43
OS: C, 65.15; H, 6.11; N, 13.41; O, 5.10;
mmol (74.3%). Anal. Calcd (%) for C17
S, 10.23. Found: C, 65.10; H, 6.13; N, 13.42; O, 5.9; S, 10.26. IR, cm : 3640 (m, OH),
323 (s, amide), 3204 (s, NH), 2964 (m, aromatic hydrogen), 1620 (s), 1594 (s), 1537
19 3
H N
ꢀ1
3
(
(
[
s), 1465 (s, aromatic), 1403 (s, C=N), 1326 (s, thioamide), 1252 (s), 1180 (s), 1084
+
m), 957 (m, CꢀH), 822 (m, C=S), 746 (m). ESI m/z: calcd for C H N OS, 312.13
17 19
3
M ꢀ H].
Synthesis and characterization of Fe compounds. The compounds compound 7ꢀ12
were synthesized as the following methods. The relevant ligands (1 mmol) were
dissolved in MeOH (10 mL) over gentle heat and stirred. A solution of FeCl
3
(0.20 g,
1
mmol) in MeOH (10 mL) was added dropwise and stirred, and the ligand solution
immediately turned dark brown. Fine dark brown crystals formed upon standing,
which were then filtered off, washed with EtOH (5 mL) and diethyl ether (5 mL) was
added.
2
ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀthiosemicarbazideꢀiron(III)ꢀbischlorin (7). Yield:
0.65 mmol (65%). CCDC NO. 1442009. Anal. Calcd (%) for C H Cl FeN OS: C,
12
9
2
3
3
8.95; H, 2.45; N, 11.36; O, 4.32; S, 8.67. Found: C, 38.98; H, 2.40; N, 11.32; O, 4.36;
ꢀ1
S, 8.69. IR, cm : 3246 (s, amide), 3038 (s, NH), 2961 (m, aromatic hydrogen), 1615
(s), 1595 (s), 1573 (s), 1457 (s, aromatic), 1388 (s, C=N), 1337 (s, thioamide), 1196
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+
(
s), 1165 (s), 1146 (m), 974 (m, CꢀH), 821 (m, C=S), 749 (s). ESI m/z: calcd for
Cl FeN OS, 369.03 [M ꢀ H].
ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀ4ꢀmethylthiosemicarbazideꢀiron(III)ꢀbischlorin (8).
Yield: 0.57 mmol (57%). CCDC NO. 1442010. Anal. Calcd (%) for
FeN OS: C, 40.65; H, 2.89; N, 10.94; O, 4.17; S, 8.35. Found: C, 40.59; H,
.85; N, 10.96; O, 4.19; S, 8.41. IR, cm : 3288 (s, amide), 3005 (s, NH), 2974 (m,
aromatic hydrogen), 1608 (s), 1534 (s), 1477 (s), 1447 (s, aromatic), 1359 (s, C=N),
C
12
H
9
2
3
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
C
13
H11Cl
2
3
ꢀ
1
2
1
5
2
337 (s, thioamide), 1194 (s), 1142 (m), 970 (m, CꢀH), 823 (m, C=S), 756 (s), 603 (s),
+
61 (s). ESI m/z: calcd for C H Cl FeN OS, 383.06 [M ꢀ H].
13 11
2
3
ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀ4ꢀphenylthiosemicarbazideꢀiron(III)ꢀbischlorin (9).
Yield: 0.54 mmol (54%). CCDC NO. 1442011. Anal. Calcd (%) for
FeN OS: C, 48.46; H, 2.94; N, 9.42; O, 3.59; S, 7.19. Found: C, 48.43; H,
C
18
H
13Cl
2
3
ꢀ
1
2.95; N, 9.40; O, 3.58; S, 7.24. IR, cm : 3173 (s, amide), 3014 (s, NH), 2874 (m,
aromatic hydrogen), 1595 (s), 1571 (s), 1531 (s), 1456 (m, aromatic), 1389 (s, C=N),
1358 (s, thioamide), 1190 (s), 1139 (s), 1087 (m), 974 (m, CꢀH), 824 (m, C=S), 747
+
(s), 699 (s), 651 (m). ESI m/z: calcd for C18
H
13Cl
2
FeN
3
OS, 444.06 [M ꢀ H].
2ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀ4,4ꢀdimethylthiosemicarbazideꢀiron(III)ꢀbischlori
n (10). Yield: 0.48 mmol (48%). CCDC NO. 1442012. Anal. Calcd (%) for
C H Cl FeN OS: C, 42.24; H, 3.29; N, 10.56; O, 4.02; S, 8.05. Found: C, 42.21; H,
14
13
2
3
ꢀ1
3
.26; N, 10.59; O, 4.04; S, 8.06. IR, cm : 3248 (s, amide), 2986 (s), 2934 (m,
aromatic hydrogen), 1592 (s), 1536 (s), 1507 (s), 1454 (s, aromatic), 1393 (s, C=N),
1
362 (s, thioamide), 1241 (s), 1197 (s), 1165 (m), 976 (m, CꢀH), 830 (m, C=S), 749
24
ACS Paragon Plus Environment

Page 25 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
(
s). ESI m/z: calcd for C14
H13Cl
2
FeN
3
OS, 396.06 [M ꢀ H].
2
ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀ4,4ꢀdiethylthiosemicarbazideꢀiron(III)ꢀbischlorin
(
11). Yield: 0.39 mmol (39%). CCDC NO. 1442013. Anal. Calcd (%) for
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
C H Cl FeN OS: C, 45.10; H, 4.02; N, 9.86; O, 3.75; S, 7.52. Found: C, 45.13; H,
16
17
2
3
ꢀ1
4
.04; N, 9.80; O, 3.71; S, 7.57. IR, cm : 3256 (s, amide), 3055 (s), 2974 (m, aromatic
hydrogen), 1570 (s), 1507 (s), 1434 (s, aromatic), 1389 (s, C=N), 1353 (s, thioamide),
+
1
197 (s), 1143 (s), 1086 (m), 975 (m, CꢀH), 827 (m, C=S), 746 (s). ESI m/z: calcd
for C16
H
17Cl
2
FeN
ꢀHydroxyꢀ1ꢀnaphthaldehydeꢀ3ꢀpiperidinethiosemicarbazideꢀiron(III)ꢀbischlorin
12). Yield: 0.55 mmol (55%). CCDC NO. 1442014. Anal. Calcd (%) for
3
OS, 424.09 [M ꢀ H].
2
(
C
17
H
17Cl
2
FeN
3
OS: C, 46.60; H, 3.91; N, 9.59; O, 3.65; S, 7.32. Found: C, 46.56; H,
ꢀ1
3
.93; N, 9.57; O, 3.68; S, 7.33. IR, cm : 2924 (s, amide), 2807 (s), 2758 (m, aromatic
hydrogen), 1595 (s), 1576 (s), 1536 (s), 1464 (s, aromatic), 1356 (s, C=N), 1319 (s,
+
thioamide), 1243 (s), 1185 (s), 969 (m, CꢀH), 826 (m, C=S), 749 (s). ESI m/z: calcd
2 3
for C17H17Cl FeN OS, 437.14 [M ꢀ H].
Crystal structures determination of Fe compounds. Xꢀray crystallographic data
were collected on a Bruker SMART Apex II CCD diffractometer using
graphiteꢀmonochromated MoꢀKα (λ = 0.71073 Å) radiation. Empirical adsorption
corrections were applied to all data using SADABS. Direct methods were used to
2
solve the structures and they were then refined against F by fullꢀmatrix leastꢀsquares
55
methods using SHELXTL (version 5.1). All of the nonꢀhydrogen atoms were refined
anisotropically while all of the other hydrogen atoms were placed in ideal geometric
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
positions and constrained to ride on their parent atoms. The crystallographic data for
compound 7ꢀ12 are summarized in Table S1 and selected bond lengths and angles are
given in Table S2. Crystallographic data for the structural analyses were deposited at
the Cambridge Crystallographic Data Centre and assigned reference numbers
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
442009ꢀ1442014. The crystallographic data can be obtained free of charge from the
Cambridge Crystallographic Data Centre via
http://www.ccdc.cam.ac.uk/data_request/cif.
Xꢀray crystallography of HSA complex. Fatty acid (FA) free HSA was purified by
5
6
removing HSA dimers and multimers as published. Palmitic acid (PA) was dissolved
in alcohol and diluted to 2.5 mM with 20 mM potassium phosphate (pH 7.5). HSA
complexes were prepared by mixing 100 µL HSA (100 mg/mL), 380 µL PA (2.5 mM),
and 90 µL of each Fe compound (5 mM) overnight. Mixtures were then concentrated
to 100 mg/ml with a Millipore spin filter (10,000 Dalton Cutꢀoff). Crystallization was
induced using the sitting drop vapor diffusion method at room temperature. An equal
volume of the HSA complex was mixed with the reservoir solution, which contained
2
8–32% (w/v) polyethylene glycol 3350, 50 mM potassium phosphate (pH 7.5), 5%
glycerol, and 4% DMSO. Crystals were directly selected from the drop solution and
then frozen in liquid nitrogen.
Xꢀray diffraction data were collected under Cryoꢀconditions (100 K) at BL17U
beamline of Shanghai Synchrotron Radiation Facility and then integrated and scaled
5
7
with HKL2000. The HSA complex structure were solved by molecular replacement
using PHASER in PHENIX suites with initial model of HSAꢀMYR structure
26
ACS Paragon Plus Environment

Page 27 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
PDB:1BJ5) but stripped the ligand as initial searching model, all ligands were built
into the model by LigandFit in PHENIX and manually modified and adjusted in
56,58,59
COOT.
The HSA structure was refined in PHENIX at resolution of 2.80 Å,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
with Rfactor=0.209 and R =0.267. All residues located in favored or allowed region
free
and no residues located in outlier region by Ramachandran plot in HSA structure, the
statistics of data collection and refinement see Table 3. Structure superimposition
were done by Chimera (http://www.cgl.ucsf.edu/chimera) and all crystallographic
60
figure were drawn by Pymol (http://www.pymol.org).
In vitro anticancer activity
L
Cell culture. Culture medium DMEM (with ꢀglutamin), fetal bovine serum (FBS),
phosphate buffered saline (PBS, pH = 7.2), and AntibioticeꢀAntimycotic came from
E.U. Gibco BRL. Human hepatocellular cell line Belꢀ7402 and normal liver cells
HLꢀ7702 were purchased from the American Type Culture Collection and the German
Collection of Microorganisms and Cell Cultures and maintained in DMEM
supplemented with 10% FBS, 50 U/mL of penicillin, and 50 mg/mL of streptomycin
2
at 37 ℃ and 5% CO .
Cytotoxicity assay (MTT). One hundred microliters of cell suspension at a density of
4
5
× 10 cells/mL was seeded in 96ꢀwell plates and incubated for 24 h at 37℃ in 5%
CO . Complexes at various concentrations were then added to the test well. The
2
2
resultant cell mixture was incubated at 37℃ in a 5% CO atmosphere for 48 h. An
enzyme labeling instrument was used to read absorbance with 570/630 nm double
wavelength measurement. Cytotoxicity was determined based on the percentage of
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 63
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
cell survival compared with the negative control. The final IC50 values were
calculated by the Bliss method (n = 5). All of the tests were repeated in triplicate.
In vivo animal studies
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Animal subject and tumor models. Athymic nude mice were obtained from Beijing
HFK Bioscience Co., Ltd. and used in the study at 6 weeks of age. All animal
experiments were carried out in compliance with the Animal Management Rules of
the People’s Republic of China’s Ministry of Health (document NO. 55, 2001) and
under the guidelines of the University of Jinan Ethics Committee’s Care and Use of
Animals. Athymic nude mice were obtained from Beijing HFK Bioscience Co., Ltd.
and used in the study at 6 weeks of age. We dissolved the complexes in DMSO
solution as mother liquid mixture (10 mM), and then diluted by NaCl solution to the
required concentration in vivo experiments.
Acute toxicity study. The acute toxicity of compound 12 and HSAꢀ12 complex were
61
assessed on normal mice using the method previously described. Briefly, 32 healthy
Kunming mice (aged 3～4 weeks and weighing 18～22 g with an equal number of
female and male subjects) were divided into four groups, with 8 mice in each group.
compound 12 and HSAꢀ12 complex were administrated into the different groups of
mice at doses of 25 µmol Fe/kg body weight. NaCl was injected into the control
group. Blood samples from each group of mice were drawn 3 days after intravenous
(IV) injection to prepare the serum samples. The serum biochemical parameters of the
blood samples were determined: aspartate aminotransferase (AST), alanine
aminotransferase (ALT), creatinine kinase (CK), and blood urea nitrogen (BUN).
28
ACS Paragon Plus Environment

Page 29 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Finally, the major organs heart, liver and kidney were sectioned for histopathological
analysis using hematoxylin and eosin (H&E) staining.
In vivo antiꢀtumor activity study. The nude mice were subcutaneously injected in
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
the right flank with 200 µL of cell suspension containing 4 × 10 Belꢀ7402 cells.
3
When the tumor volume was approximately 100 mm for the study, the Belꢀ7402
tumorꢀbearing mice were randomly divided into three groups, and mice in different
,
treatment groups were intravenously injected with NaCl, compound 12 and HSAꢀ12
complex at doses of 2.5 µmol Fe/kg body weight every 2 days. Each mouse of
different group was earmarked and followed individually throughout the whole
experiments. The width and length of the tumor and the body weight of mice were
measured before every injection by the end of experiment. Volume was determined
2
using the following equation: tumor volume (V) = 1/2 × length × width . Mice were
killed after 28 days of treatment and tumor tissues and major organs were excised for
histopathological analysis with H&E staining.
Hematoxylin and eosin (H&E) staining. Tumor samples were given a routine
histopathological examination using standard H&E staining. Small pieces were
collected in 4% paraformaldehyde for proper fixation and they were then processed
and embedded in paraffin wax. Sections were cut and stained with hematoxylin and
eosin and samples were then observed under light microscope using an Eclipse E800
Nikon (Nikon, Tokyo, Japan). Representative images were shown. Quantitative
analysis of eosin stained areas in treated tumors expressed as relative amount
compared to eosin necrotic area of untreated tumors (± SD), were assessed by area
29
ACS Paragon Plus Environment