Dual-functional conjugates improving cancer immunochemotherapy by inhibiting tubulin polymerization and indoleamine-2,3-dioxygenase

Article abstract of DOI:10.1016/j.ejmech.2020.112041

A series of novel conjugates comprising tublin and IDO inhibitors were designed, synthesized and evaluated for their antiproliferative activity. Among them, HI5, composed of combretastatin A-4 (CA-4) and (D)-1-methyltryptophan (D-MT) by a linker, exhibited the most potent antitumor activity, in particular with higher IC₅₀ value (0.07 μM) than CA-4 (0.21 μM) against HeLa cancer cell line. Mechanism studies indicated that HI5 can inhibit tubulin polymerization and cell migration, cause G2/M phase arrest, concurrent induce apoptosis via the mitochondrial dependent apoptosis pathway and cause reactive oxidative stress generation in HeLa cells. Furthermore, HI5 can inhibit IDO expression and decrease kynurenine production, leading to stimulating T cells activation and proliferation to enhance antitumor immunity in vitro. Interestingly, HI5 can effectively limit the tumor growth in the HeLa xenograft mice models without causing significant loss of body weight. Consequently, such a conjugation can be a potent and safe immunochemotherapeutic method for improving cancer therapy.

Full text of DOI:10.1016/j.ejmech.2020.112041

Journal Pre-proof
Dual-functional conjugates improving cancer immunochemotherapy by inhibiting
tubulin polymerization and indoleamine-2,3-dioxygenase
Shixian Hua, Feihong Chen, Xinyi Wang, Shaohua Gou
PII:
S0223-5234(20)30008-8
DOI:
https://doi.org/10.1016/j.ejmech.2020.112041
EJMECH 112041
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 26 September 2019
Revised Date: 5 January 2020
Accepted Date: 6 January 2020
Please cite this article as: S. Hua, F. Chen, X. Wang, S. Gou, Dual-functional conjugates improving
cancer immunochemotherapy by inhibiting tubulin polymerization and indoleamine-2,3-dioxygenase,
European Journal of Medicinal Chemistry (2020), doi: https://doi.org/10.1016/j.ejmech.2020.112041.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Masson SAS.

Graphical abstract
Dual-functional conjugates improving cancer immunochemotherapy
by inhibiting tubulin polymerization and indoleamine-2,3-dioxygenase
Shixian Hua a, b, Feihong Chen a, b, Xinyi Wang a, b, Shaohua Gou a, b,
*
An immunochemical strategy consisting of both IDO and microtubule inhibitors was reported,
which could hydrolyze to release a tubulin inhibitor (CA-4) to inhibit tubulin polymerization, cause
cancer cells death, and concurrently release a IDO inhibitor to block IDO protein expression to
promote T-cell proliferation and kill cancer cells. In vivo study indicated HI5 could remarkably
promote the antitumor activity of CA-4 in HeLa tumor xenograft mice models.
1

Dual-functional
conjugates
improving
cancer
immunochemotherapy by inhibiting tubulin polymerization
and indoleamine-2,3-dioxygenase
Shixian Hua,a,b Feihong Chen,a,b Xinyi Wang,a,b Shaohua Gou,a,b,*
a
Pharmaceutical Research Center and School of Chemistry and Chemical
Engineering and b Jiangsu Province Hi-Tech Key Laboratory for Biomedical Research,
Southeast University, Nanjing 211189, China.
ABSTRCT: A series of novel conjugates comprising tublin and IDO inhibitors were
designed, synthesized and evaluated for their antiproliferative activity. Among them,
HI5, composed of combretastatin A-4 (CA-4) and (D)-1-methyltryptophan (D-MT)
by a linker, exhibited the most potent antitumor activity, in particular with higher IC50
value (0.07 µM) than CA-4 (0.21 µM) against HeLa cancer cell line. Mechanism
studies indicated that HI5 can inhibit tubulin polymerization and cell migration, cause
G2/M phase arrest, concurrent induce apoptosis via the mitochondrial dependent
apoptosis pathway and cause reactive oxidative stress generation in HeLa cells.
Furthermore, HI5 can inhibit IDO expression and decrease kynurenine production,
leading to stimulating T cells activation and proliferation to enhance antitumor
immunity in vitro. Interestingly, HI5 can effectively limit the tumor growth in the
HeLa xenograft mice models without causing significant loss of body weight.
Consequently, such a conjugation can be a potent and safe immunochemotherapeutic
method for improving cancer therapy.
Keywords: combretastatin-A4, indoleamine-2,3-Dioxygenase, immunomodulator,
anticancer, immunochemotherapy
* Corresponding author. Jiangsu Province Hi-Tech Key Laboratory for Biomedical
Research, Southeast University, Nanjing 211189, China
E-mail address: sgou@seu.edu.cn (S. Gou).

1. Introduction
Immunotherapy is one of the most promising and exciting development
strategies currently in treatment with cancer by exploiting the host’s immunity upon T
cells [1-4]. Many studies indicated that T cells play an important role in antitumor
immune response, which are sensitive to the levels of the essential amino acid
tryptophan and its metabolites [5-7]. Indoleamine-2,3-dioxygenase (IDO), a
monomeric and extrahepatic heme-containing enzyme, is encoded by the INDO gene
and induced by tumor necrosis factor alpha (TNF-α), interferon-γ and other
inflammatory mediators [8-10]. As a negative feedback enzyme belonging to immune
checkpoint, IDO can catalyze tryptophan to produce kynurenine, leading to depletion
of tryptophan and accumulation of kynurenine, and finally give rise to T cells anergy
and/or apoptosis [6-12]. In addition, IDO is overexpressed in many tumor cells and
tumor-draining lymph nodes that correlate with a less favorable prognosis for survival
in several types of cancer [9,13-15]. All available facts demonstrated that IDO can
involve in T-cell-mediated antitumor immune response by the kynurenine pathway
that catalyzes tryptophan metabolism to produce biologically metabolites and natural
ligands for the aryl hydrocarbon receptor [16,17]. Thus, inhibition of IDO can
improve antitumor immunity by blocking the production of tryptophan metabolites to
provide more tryptophan for T cells proliferation, and at last enhance the therapeutic
effect of tumors. It has been found that some IDO inhibitors including D-MT [18],
INCB24360 [19], NLG919 [20] and PF-0684003 [21] (Fig. 1) can suppress
tryptophan degradation to produce kynurenine, which were selected for clinical trials
as immunoregulatory drugs, but only D-MT is ongoing. But some of these IDO
inhibitors as mono agents usually display insufficient activation of antitumor
immunity or require a high dosage for effectiveness [22,23]. Fortunately, a growing
body of evidences suggested that administration of IDO1 inhibitor with vaccination,
chemotherapy, or radiationtherapy can improve tumor therapeutic efficacy both in
vitro and in vivo [20, 24-28].

Microtubule/tubulin is a well-known target in cancer therapy, and some of the
microtubule targeting agents had been successfully used in clinical for the treatment
of cancer [29-31]. Microtubule, an essential element of the cytoskeleton consisting of
α and β-tubulin heterodimers, plays an important role in a variety of cellular processes
including cell shape organization, regulation of motility, transportation of vesicles and
organelles, cell division, and mitosis [32-33]. Meanwhile, microtubule targeting
agents can promote microtubule polymerization or induce microtubule
depolymerization, resulting in mitotic blockade and cell apoptosis [34]. Among the
microtubule targeting agents, CA-4 (Fig. 1), one of the most active compounds, is a
natural cis-stilbene product isolated from the African willow tree (Combretum
caffrum) of bark [35]. CA-4 has extremely strong effects on the inhibition of tubulin
polymerization by binding to the colchicine-binding site, and in particular strong
cytotoxicity against the majority of human cancer cells [36-37]. Interestingly, CA-4
was found to have a potent effect on provoking tumor vasculature collapse and then
reducing tumor blood flow, leading to tumor cell death, and thus also acting as a
vascular disrupting agent [38-39]. Importantly, the water-soluble phosphate prodrug
CA-4P [40] and CA-1P [41] of CA-4 are currently under clinical trials as a single
drug or in combination for anticancer therapy. In addition, chalcones, another
important colchicine site binders bearing an α,β-unsaturated ketone moiety, display a
broad range of biological activities including anti-tumor, anti-inflammatory and
anti-mitotic properties [42,43]. As the representative of anti-tubulin chalcones,
compounds 14 and 14a (Fig. 1) exhibit remarkable antiproliferative activities and
concurrently inhibit the polymerization of microtubule [44,45]. As expected, some
studies indicated that combination of platinum-based antitumor agents with CA-4 or
chalcone analogue could improve antitumor activity [46,47].
Nowadays, the combination strategy of antitumor agents with immune
checkpoint inhibitors has been demonstrated as an effective way for cancer therapy
[48-53]. Herein, a series of novel conjugates consisting of IDO and tubulin inhibitors
are reported to surmount the limitations of chemotherapy and immunotherapy, which
are designed to kill cancer cells directly by inhibiting tubulin polymerization and

simultaneously stimulate antitumor immunity via inhibiting IDO expression to
enhance therapeutic efficacy. Such a dual function strategy is anticipated to provide an
effective immunochemotherapeutic method for cancer treatment.
2. Results and discussion
2.1. Chemistry
The synthetic routes were outlined in Schemes S1-S4. Compounds 6 (CA-4) and
14 were synthesized according to the reported literatures [45,46]. Compound 6 or 14
was first treated with 8 in the presence of EDCI and DMAP in DMF to give
compound 9 or 15. Subsequently, compound 9 or 15 was used to interact with
3-butyn-1-ol to produce compound 10 or 16 by copper-catalyzed click reactions. The
IDO inhibitor 19 was prepared by a reported method [21b]. Compound 20 was
obtained by treatment of 19 with succinic anhydride in CH2Cl2. Then, 20 was used to
react with 6 or its derivative 10 in the presence of EDCI/DMAP to produce the
targeted compounds HI1 and HI2，and similar reaction of 20 with 14 or its derivative
16 generated HI3 and HI4. Commercially available D-MT was utilized to react with
(BOC)2O to afford 22. HI5-boc and HI6-boc, achieved via esterification between 22
and 10 or 16, were deprotected with TFA in CH2Cl2 to afford the targeted compounds
HI5 and HI6. HI7 and HI8 were achieved by esterification between 22 and 6 or 23,
which was similar to HI5 and HI6, while 23 was prepared by treatment of 6 and
bromo-1-propanol in the presence of K2CO3. The IDO inhibitor 24 (IDO1 IC50=67 nm)
was prepared by a reported method [54], the synthetic route was outlined in Scheme
1
13
S4. All resulting compounds (Fig. 2) were characterized by H NMR and C NMR
spectroscopy, and high resolution mass spectrometry (HR-MS) (Supporting
Information)
2.2. Biological evaluation.
2.2.1. In vitro cytotoxicity
The in vitro cytotoxicity of the targeted compounds was evaluated by MTT assays
against three human cancer cell lines including A549 (lung), PC-3 (prostatic) and
HeLa (cervical) and a human normal cell line (HUVEC) with CA-4, D-MT,

compounds 14 and 19 as the positive controls. The IC50 values were obtained after 72
h treatment and given in Table 1. As expected, D-MT displayed insignificant
antiproliferative activity, and compound 19 showed lower cytotoxicity against the
tested cancer cell lines. However, the targeted compounds as well as CA-4 and 14
exhibited potent anticancer activity toward the three cancer cell lines but low toxicity
against HUVEC cells. Compounds HI1-HI4, containing 19 and a tubulin inhibitor
(CA-4 or 14), exhibited comparable cytotoxicity to CA-4 and 14 against the tested
cancer cell lines. Notably, compounds HI1 and HI2 consisting of a CA-4 unit
displayed a little better antiproliferative activity than HI3 and HI4 composed of a
chalcone analogue. In addition, both HI5-HI8, consisting of D-MT and CA-4 or 14,
also exhibited potent cytotoxicity comparable to their parent compounds against the
tested cell lines. Exceptionally, HI5 possessed the most potent cytotoxicity against
HeLa cancer cells with an IC50 value of 0.07 µM, 3-fold as potent as CA-4 (0.21 µM).
The superior cytotoxicity of HI5 may be due to the fact that HI5 was able to yield
CA-4 and D-MT via hydrolysis at a weakly acidic environment in PBS (pH=5.0) or
HeLa cells(Fig. S1). Hence, HI5 was selected for further study on antitumor and
immune mechanism in HeLa cells (IDO expressed) [28].
2.2.2. Immunofluorescent staining assay.
It is well-known that CA-4 can depolymerize cellular microtubules as a tubulin
inhibitor [36,37], while HI5 was a conjugate that contained a CA-4 unit. Thus, it was
interesting to investigate the inhibitory effect of HI5 on cellular microtubule network
by immunofluorescent assays. As shown in Fig. 3, confocal images exhibited normal
arrangement of filamentous microtubules in HeLa cells in the control group. However,
cells displayed the disruption of the microtubule networks after exposure to CA-4 at
0.5 µM compared with the control group. As anticipated, HeLa cells treated with HI5
also exhibited markedly disruption of microtubule organization in comparison with
the control and CA-4 groups, in a concentration-dependent manner. The result
demonstrated that HI5 can inhibit tubulin polymerization, which may finally lead to
cell apoptosis.

2.2.3. Analysis of cell cycle.
Due to the inhibition of tubulin polymerization may disrupt cell mitosis and
affect cell cycle distribution [55], we evaluated the arrest effect of HI5 on the cell
cycle distribution in HeLa cells. As depicted in Fig. 4, the treatment group of CA-4
and the mixture of CA-4 and D-MT caused the cell cycle arrest at the phase of G2/M
with the percentage of 66.57% and 58.47%, respectively, in comparison with the
percentage of 10.7% in the control group. Similarly, HI5 also arrested the cell cycle at
the phase of G2/M but the percentage increased from 10.7% to 76.27% compared to
the control cells. These results indicated that HI5 can induce the HeLa cells cycle
arrest at the G2/M phase.
2.2.4. Cell apoptosis analysis.
The promotion or inhibition of microtubule polymerization can lead to mitotic
blockade and cell apoptosis [34]. Here, apoptosis induction by HI5 on HeLa cells was
evaluated by flow cytometry via double staining with FITC-Annexin V and propidium
iodide (PI). As shown in Fig. 5, the treatment groups showed significant apoptosis
rates compared with the control group. After treatment with CA-4 and the mixture of
CA-4 and D-MT, apoptosis cells were increased to 35.3% and 29.6%, respectively, in
contrast to the control group of 3.77%. Among the treatment groups, incubation with
HI5 exhibited a much higher apoptotic rate of 52.1%, which was the optimal efficacy
in comparison with other groups. These results indicated that HI5 can induce the most
efficiently HeLa cells apoptosis, which was in agreement with its in vitro
antiproliferative activity.
2.2.5. Acridine orange/ethidium bromide (AO/EB) dual staining assay.
Acridine orange (AO) can emit green fluorescence and stain the intact cell
membrane of viable cells as a cationic fluorescent dye. On the contrary, ethidium
bromide (EB) can stain the cells of lose membrane integrity and emit orange
fluorescence. In order to study the effect of HI5 on apoptosis further, the AO/EB
staining assay was carried out to identify the live and apoptotic HeLa cells by using
fluorescence microscopy. As shown in Fig. 6, the living cells in control group showed

bright green fluorescence because of the normal cells morphology. On the contrary,
after treated with CA-4 and HI5, the orange fluorescence as characteristic of
apoptotic cells was observably strengthened with the reduction of green fluorescence,
indicating that the cells lost their membrane integrity and formed apoptotic body. All
of these results confirmed that HI5 can induce cell death on HeLa cells.
2.2.6. Measurement of reactive oxygen species (ROS).
Increasing evidences have indicated that cellular damage and apoptosis can be
triggered by intracellular ROS, particularly at high levels [56,57]. In order to explore
whether HeLa cells apoptosis induced by HI5 was associated with the level of ROS,
the production of intracellular ROS was measured. The results were showed in Fig. 7,
in comparison with the negative untreated control, both CA-4 and HI5 induced a
significant elevation of ROS production in HeLa cells. Especially, HI5 (49.7%)
boosted an increase of 2.3 fold as much as CA-4 (21.8%), which may be attributed to
the excellent cytotoxic activity of HI5. These results indicated that HI5 can cause
oxidative imbalance and lead to HeLa cells apoptosis.
2.2.7. Mitochondrial membrane potential measurement (JC-1 assay).
To further study the apoptosis-inducing mechanism of HI5, we measured the
alterations of mitochondrial membrane potential (MMP) by using the fluorescent
lipophilic cationic dye (JC-1). After treatment with the indicated compounds and
stained with JC-1, the cells were observed and showed in Fig. 8A. Both CA-4 and
HI5 can induce the dissipation of MMP in comparison with the control group (6.6%),
in particular, HI5 was superior to CA-4 with an induction percentage of 68.8%
compared to that of CA4 (54.2%). Furthermore, the same results were also detected
by fluorescent microscope because JC-1 can form red-emitting aggregates in the
mitochondrial matrix in normal condition but form monomers emitting green
fluorescence in cytoplasm after the disruption of MMP [58]. As depicted in Fig. 8B,
the control group without treatment displayed normally red, but the green
fluorescence increased after incubation with CA-4 or HI5. Therefore, it was clearly
demonstrated that HI5 can induce HeLa cells apoptosis, which was associated with

the mitochondrial pathway.
2.2.8. Western blotting analysis on the apoptosis-related proteins expression.
The above experiment demonstrated that the induction of cell apoptosis by HI5
was closely related to the mitochondrial pathway. Here the mitochondrial apoptotic
pathway related proteins of Bax, Bcl-2, caspase-3, and PARP in HeLa cells were
detected by western blotting assay. As shown in Fig. 9, after treated with the
measured compounds at 0.5 µM for 24 h, the expression level of Bax protein was
significantly up-regulated, while the protein of Bcl-2 was down-regulated compared
to the untreated group. Furthermore, the protein expressions of caspase-3 and PARP
were markedly up-regulated. As a result, the activation of caspase-3 and the cleavage
of downstream PARP can lead to cell death finally. Interestingly, HI5 exhibited the
most effective in regulating apoptosis-related proteins expression. In summary, HI5
can induce HeLa cells apoptosis through a mitochondrial apoptosis-related pathway.
2.2.9. Cells migration.
As migration of cancer cells plays an important role in the tumor progression and
metastatic cascade [59], a wound healing assay was carried out to investigate the
inhibition effect of HI5 on tumor cells migration. As shown in Fig. 10, after treatment
by CA-4 for 24 h, the inhibition ratio of HeLa cells migration was 79%, which had a
stronger effect than the control of 61%. However, HI5 showed much higher inhibition
ratio (88%) in comparison to the CA4 group, indicating that HI5 can markedly
suppress the migration of cancer cells.
2.2.10. Inhibition of IDO and kynurenine assay.
As an IDO inhibitor, D-MT can suppress the expression of IDO protein that
block the degradation of tryptophan to produce kynurenine [18]. Hence, the IDO
inhibitory effect by HI5 to block the production of kynurenine was evaluated. Firstly,
the IDO expression level in HeLa cells with hIFN-γ stimulation was detected by
western blotting. As depicted in Fig. 11A and 11B, the expression of IDO1 protein
showed insignificantly regulation in contrast to the control group after exposure to

CA-4 for 24 h. On the contrary, compounds D-MT and HI5 as well as the mixture of
CA-4 and D-MT can effectively down-regulate the protein expression of IDO1.
Among them, HI5 exhibited the most significant inhibition of IDO1 expression,
particularly the inhibitory effect was similar to the IDO inhibitor of compound 24 (Fig.
S2). In addition, HI7 and HI8, consisting of D-MT and CA-4, also displayed
significant inhibition of IDO1 expression (Fig. S2). Subsequently, the levels of
kynurenine production in HeLa cells were detected by HPLC, and the results were
showed in Fig. 11C. All the tested compounds displayed the inhibitory efficacy on the
production of kynurenine, but D-MT and the mixture of CA-4/D-MT showed weaker
inhibitory effects than HI5, which was similar to the efficacy of IDO inhibition.
Collectively, our findings indicated that HI5 can block the kynurenine pathway by
suppressing IDO expression.
2.2.11. Measurement of T cells proliferation.
With the inhibition of IDO to reduce kynurenine production, the antitumor
immunity will be boosted by stimulating T cells activation and proliferation. In order
to assess the effect of HI5 on improving T cell immune responses in vitro, a mixed
leukocyte reaction (MLR) assay was performed. HeLa cells were treated with the
indicated compounds. Human peripheral blood mononuclear cells (PBMCs) from
unrelated healthy donors were used in this experiment. After stimulated by using
phytohemagglutinin and staining with a fluorescent cell dye of carboxyfluorescein
succinimidyl ester (CFSE), the cells were co-cultured with HeLa cells. The mixed
cells were collected by centrifugation, and stained with anti-CD3 antibody (CD3 is a
marker of the development of T cells and required for T cells activation [60]). The
changes of T cells were analyzed by flow cytometry. As shown in Fig. 12, the control
group without any treatment showed as low as 6.2% T cells proliferation. In the
mixture group of D-MT and CA-4, the percentage of T cells proliferation was
significantly increased to 28.5% as compared with that of the D-MT group (17.8%).
Remarkably, the percentage of T cells proliferation of HI5 reached 40.3%, over
two-fold higher than those of the mixture and D-MT groups. Collectively, our data

indicated that HI5 can decrease the production of kynurenine by inhibiting IDO
expression, leading to the enhancement of T cells proliferation.
2.2.12. Antitumor activity of HI5 on the tumor growth in vivo.
Upon the in vitro cytotoxicity and mechanism observations, the therapeutic
efficacy of HI5 was investigated in nude mice bearing HeLa xenograft. After the
tumor model was established by subcutaneous injection and tumor volume arrived to
150 mm3, 20 mice were randomly divided into four groups. HI5 was intravenously
administered at two doses of 15 and 30 mg/kg with CA-4 dosed at 30 mg/kg per day.
The tumor size and body weights of the mice were observed and recorded every other
day. As shown in Fig. 13 and S3, both HI5 (two groups) and CA-4 significantly
decreased the tumor volume after 21 days treatment as compared with the control
group. Notably, HI5 exhibited significant inhibition of tumor growth at a
dose-dependent manner with inhibitory rates of 50.73% and 65.76% at doses of 15
and 30 mg/kg, respectively. Importantly, the inhibitory effect of HI5 was more potent
than the CA-4 group at the same dose of 30 mg/kg. Furthermore, the body weight and
the toxicity of the main organs (liver, heart, lung, kidney and spleen) by H&E staining
were observed. As depicted in Fig. 13D and 14, the body weight did not show
significant loss but slightly increased in the treatment groups. There was no obvious
damage to the main organs after treatments either. Altogether, HI5 can act as an
efficient anticancer agent with high antitumor activity and low toxicity for further
investigation.
3. Conclusions
In summary, a series of novel conjugates bearing both IDO and tubulin inhibitors
were designed and prepared as immunochemotherapeutic agents to improve antitumor
immune response. All the resulting conjugates exhibited comparable cytotoxicity to
their parent tubulin inhibitors toward the tested cancer cell lines and lower toxicity
against a human normal cell. Among them, HI5 displayed excellent cytotoxicity with
an IC50 value of 0.07 µM against HeLa cancer cells in comparison with those of CA-4

(0.21 µM) and the mixture of CA-4/D-MT (0.40 µM). HI5, as a conjugate derived
from CA-4, can effectively inhibit tubulin polymerization and suppress the migration
of HeLa cells. In addition, mechanism study revealed that HI5 can cause cell cycle
arrest in the G2/M phase, and induce apoptosis by the mitochondrial mediated
pathway via upregulating the expression of Bax, downregulating the expression of
Bcl-2, further activating downstream caspase-3 and upregulating PARP protein, and
simultaneously improve the generation of ROS in HeLa cells. Meanwhile, HI5 can
block tryptophan degradation to produce kynurenine by suppressing IDO protein
expression, leading to T cells activation and proliferation in vitro. More importantly,
HI5 effectively inhibited tumor growth with a higher inhibition rate than CA-4 at the
same dose in HeLa xenograft mice without causing obvious major organ-related
toxicity and body weight loss. Consequently, HI5 has the potential to be further
developed as an immunomodulator for cancer therapy.
4. Experimental section
4.1. Chemistry.
Chemicals and reagents were obtained from Energy Chemical (Shanghai of
China). All the solvents and reagents were analytical pure that using without further
purification unless otherwise specified. TLC was displayed on silica gel precoated
GF-254 plates and column chromatography was carried out by using silica gel
(200-300 mesh). 1H NMR and 13C NMR spectra were recorded on a Bruker 600 MHz
spectrometer (Bruker Company, Germany) in CDCl3 or DMSO-d6 (TMS as internal
standard). Mass spectroscopy was characterized on the instrument of Agilent 6224
TOF LC/MS. All of human cell lines (cancer and normal cells) were obtained from
the Shanghai Institute for Biological Science (Shanghai, China). The antibodies of
β-Acitn, Bcl-2, Bax, PAPR, caspase-3 and IDO1 were obtained from Cell Signaling
Technology (Danvers, MA, USA). FITC-conjugated rabbit anti-tubulin antibody was
purchased from Jackson Immuno Research Laboratories (USA). Peripheral blood
mononuclear cells (PBMCs) were from unrelated healthy donators, anti-CD3 antibody
(Abcam) was purchased for flow cytometry.
Compound 19 and 24, an IDO inhibitor, was prepared according to a literature

report [21b,54]. D-MT was purchased from Shanghai Aladdin Bio-Chem Technology
Co., LTD. The tubulin inhibitors of CA-4 [46] and chalcone derivative compound 14
[45] were prepared in agreement with the methods reported previously.
4.1.1. Synthesis of CA-4.
Compound 6 (CA-4) was obtained as a white solid. 1H NMR (600 MHz, CDCl3)
δ 6.92 (d, J = 2.0 Hz, 1H), 6.80 (dd, J = 8.3, 1.9 Hz, 1H), 6.73 (d, J = 8.3 Hz, 1H),
6.53 (s, 2H), 6.47 (d, J = 12.2 Hz, 1H), 6.41 (d, J = 12.2 Hz, 1H), 5.52 (s, 1H), 3.87 (s,
3H), 3.84 (s, 3H), 3.70 (s, 6H).
4.1.2. Synthesis of compound 14.
1
Compound 14 was prepared as a yellow solid (80.1%). H NMR (600 MHz,
CDCl3) δ 7.75 (d, J = 15.5 Hz, 1H), 7.36 (d, J = 15.5 Hz, 1H), 7.31 (d, J = 2.1 Hz,
1H), 7.28 (s, 2H), 7.14 (dd, J = 8.3, 2.1 Hz, 1H), 6.89 (d, J = 8.3 Hz, 1H), 5.72 (s, 1H),
3.96 (s, 6H), 3.95 (s, 3H), 3.94 (s, 3H).
4.1.3. Synthesis of compounds 10 and 16.
Compound 8 was obtained from 7 and sodium azide in the presence of dimethyl
sulfoxide and extracted with diethyl ether. Compound 8 was stirred in DMF, and then
EDCI, DMAP and CA-4 or 14 were added into the solution. Then the resulting
mixture was stirred at the room temperature for 24 h, CH2Cl2 (200 mL) was added
and washed with water. After removal of solvents, the remaining was purified by
chromatography to give compound 9 or 15. Compound 9 or 15 and 3-butyn-1-ol were
dissolved in a mixture of CH3OH/H2O (V:V=2:1), then CuSO4.5H2O and sodium
erythorbate were added into the mixture. After the reaction was stirred for 4 h under
N2 atmosphere, the mixture was evaporated in vacuo and extracted with CH2Cl2, the
organic layers were combined and concentrated under reduce pressure. The residue
was purified by chromatography to afford the white product 10 in yield of 89.2% or
the yellow product 16 in yield of 75.7%. Compound 10: HR-MS (m/z) (ESI): calcd
for C30H31N3O7 [M+H]+: 546.22348; found: 546.23692. Compound 16: HR-MS (m/z)
(ESI): calcd for C31H31N3O8 [M+H]+: 574.21839; found: 574.22134.
4.1.4. Synthesis of compound 19.
1
The product was prepared as a yellow solid (80.1%). H NMR (600 MHz,

DMSO) δ 11.29 (s, 1H), 11.14 (s, 1H), 7.41 (d, J = 2.5 Hz, 1H), 7.36 (dd, J = 8.8, 4.6
Hz, 1H), 7.20 (dd, J = 10.0, 2.5 Hz, 1H), 6.95 (td, J = 9.2, 2.5 Hz, 1H), 5.75 (s, 1H),
4.33 (dd, J = 9.4, 5.5 Hz, 1H), 3.17 (dd, J = 18.0, 9.5 Hz, 1H), 2.79 (dd, J = 18.0, 5.5
Hz, 1H).
4.1.5. Synthesis of compound 20.
To a CH2Cl2 solution of 19 (0.6 g, 2.6 mmol) and succinic anhydride (0.4 g, 4.0
mmol), Et3N (0.52 g, 5.1 mmol) and DMAP (0.032 g, 0.26 mmol) were added at room
temperature, then the reaction solution was kept refluxing overnight. After completion
of the reaction, the mixture was concentrated in vacuo and the residue was purified by
flash chromatography to afford a pale gray solid of compound 20 (0.61 g, 70.1%). 1H
NMR (600 MHz, DMSO) δ 12.34 (s, 1H), 11.41 (s, 1H), 8.34 (dd, J = 9.0, 4.8 Hz,
1H), 8.12 (s, 1H), 7.41 (dd, J = 9.2, 2.6 Hz, 1H), 7.21 (td, J = 9.2, 2.6 Hz, 1H), 4.45
(dd, J = 9.1, 6.1 Hz, 1H), 3.24 (dd, J = 12.0, 6.1 Hz, 2H), 3.16 (dd, J = 12.6, 5.3 Hz,
1H), 3.02 (dd, J = 18.0, 6.0 Hz, 1H), 2.66 (t, J = 6.3 Hz, 2H).
4.1.6. Synthesis of HI1.
A mixture of compound 20 (0.1 g, 0.3 mmol) and EDCI (0.087 g, 0.45 mmol) in
dry DMF was stirred at room temperature, then DMAP (0.037 g, 0.3 mmol) and CA-4
(0.095 g, 0.3 mmol) were added. After the resulting solution was kept stirring at room
temperature for one day, it was poured into water (500 mL) and extracted with CH2Cl2
(2 × 150 mL). The combined organic layer was washed with saturated brine, dried
over anhydrous Na2SO4, and concentrated under reduced pressure to afford the crude
product. At last, the crude product was purified by column chromatography to give
the desired product of HI1 as a white solid (0.12 g, yield 61.5%). 1H NMR (600 MHz,
DMSO) δ 11.41 (s, 1H), 8.33 (dd, J = 9.0, 4.8 Hz, 1H), 8.11 (s, 1H), 7.41 (dd, J = 9.2,
2.6 Hz, 1H), 7.23 – 7.19 (m, 1H), 7.15 (dd, J = 8.6, 2.0 Hz, 1H), 7.07 (d, J = 8.6 Hz,
1H), 7.03 (d, J = 2.1 Hz, 1H), 6.55 (s, 2H), 6.49 (t, J = 11.1 Hz, 2H), 4.45 (dd, J = 9.1,
6.0 Hz, 1H), 3.73 (s, 3H), 3.63 (s, 3H), 3.59 (s, 6H), 3.38 (dd, J = 13.7, 6.4 Hz, 2H),
3.16 (dd, J = 18.0, 9.5 Hz, 1H), 3.01 (dd, J = 18.0, 6.0 Hz, 1H), 2.97 (d, J = 6.5 Hz,
2H). 13C NMR (150 MHz, DMSO) δ 179.13, 178.07, 170.84, 170.77, 159.99, 158.41,
153.09, 150.50, 139.41, 137.25, 132.35, 132.29, 131.15, 131.09, 129.96, 129.85,

128.66, 127.95, 125.98, 123.02, 118.39, 117.75, 113.23, 113.16, 112.99, 106.36,
105.88, 60.51, 56.35, 56.01, 38.61, 36.43, 30.50, 28.41. HR-MS (m/z) (ESI): calcd for
C34H31FN2O9 [M+H]+: 631.20919; found: 631.20905.
4.1.7. Synthesis of HI2.
HI2 was prepared according to a similar method to the preparation of HI1 in
which compound 10 was treated with 20 to give HI2 as a white solid (0.15 g, 62.2%).
1H NMR (600 MHz, CDCl3) δ 8.54 (s, 1H), 8.37 (dd, J = 8.5, 4.3 Hz, 1H), 8.14 – 8.11
(m, 1H), 8.08 (s, 1H), 7.57 (s, 1H), 7.49 – 7.43 (m, 3H), 7.17 (dd, J = 8.5, 2.1 Hz, 1H),
7.12 – 7.07 (m, 3H), 6.89 (d, J = 8.5 Hz, 1H), 6.52 (d, J = 4.0 Hz, 2H), 6.47 (d, J =
3.6 Hz, 2H), 5.54 (s, 2H), 4.40 (t, J = 6.4 Hz, 2H), 4.28 (dd, J = 9.6, 5.6 Hz, 1H), 3.81
(s, 3H), 3.77 (s, 3H), 3.72 (s, 6H), 3.30 (dd, J = 18.4, 9.7 Hz, 1H), 3.19 (t, J = 6.4 Hz,
2H), 3.09 (t, J = 6.4 Hz, 2H), 2.92 (dd, J = 18.4, 5.6 Hz, 1H), 2.77 (t, J = 6.4 Hz, 2H).
13C NMR (150 MHz, CDCl3) δ 176.49, 175.05, 171.98, 169.55, 163.94, 158.87,
152.96, 150.28, 144.62, 139.37, 137.09, 135.36, 132.90, 132.44, 130.55, 130.17,
130.11, 129.63, 129.48, 128.50, 128.00, 123.65, 123.15, 121.92, 118.13, 117.41,
113.83, 113.67, 112.12, 105.82, 104.82, 104.66, 63.55, 60.89, 55.97, 55.94, 53.61,
38.90, 36.57, 30.45, 28.49, 25.40. HR-MS (m/z) (ESI): calcd for C46H42FN5O11
[M+H]+: 860.29431; found: 860.29658.
4.1.8. Synthesis of HI3.
HI3 was obtained in an analogous way to prepare HI1 in which compound 14
1
was treated with 20 to give HI3 as a yellow solid (0.15 g, 65.0%). H NMR (600
MHz, DMSO) δ 11.42 (s, 1H), 8.38 (dd, J = 9.0, 4.8 Hz, 1H), 8.15 (s, 1H), 7.85 (d, J
= 15.5 Hz, 1H), 7.81 (d, J = 2.1 Hz, 1H), 7.76 (dd, J = 8.6, 2.0 Hz, 1H), 7.71 (d, J =
15.4 Hz, 1H), 7.43 (s, 2H), 7.42 (d, J = 2.6 Hz, 1H), 7.24 – 7.19 (m, 2H), 4.46 (dd, J =
9.2, 5.9 Hz, 1H), 3.90 (s, 6H), 3.82 (s, 3H), 3.77 (s, 3H), 3.43 (dd, J = 12.6, 6.3 Hz,
2H), 3.17 (dd, J = 18.0, 9.5 Hz, 1H), 3.07 (t, J = 6.4 Hz, 2H), 3.02 (dd, J = 18.0, 5.9
Hz, 1H). 13C NMR (150 MHz, DMSO) δ 188.17, 179.14, 178.08, 170.86, 170.83,
160.00, 158.42, 153.38, 153.32, 143.54, 142.37, 140.07, 133.59, 132.32, 131.13,
129.90, 128.20, 126.02, 122.75, 120.83, 118.46, 117.74, 113.39, 113.20, 113.03,
106.64, 106.08, 100.00, 60.66, 56.72, 56.59, 38.62, 36.44, 30.44, 28.42. HR-MS (m/z)

(ESI): calcd for C35H31FN2O10 [M+H]+: 659.20355; found: 659.20298.
4.1.9. Synthesis of HI4.
HI4 was achieved as a pale yellow solid (1.12 g, 75.0%) by an analogous way to
the synthesis of HI1 in which reaction of compound 20 with compound 16 afforded
HI4. 1H NMR (600 MHz, CDCl3) δ 8.81 (s, 1H), 8.34 (dd, J = 8.7, 4.4 Hz, 1H), 8.18
(dd, J = 5.9, 2.7 Hz, 1H), 8.12 (s, 1H), 7.74 (d, J = 15.5 Hz, 1H), 7.52 (ddd, J = 10.6,
9.4, 3.2 Hz, 5H), 7.46 (d, J = 1.9 Hz, 1H), 7.35 (d, J = 15.5 Hz, 1H), 7.26 (s, 2H),
7.07 (ddd, J = 20.3, 8.8, 2.4 Hz, 2H), 7.01 (d, J = 8.5 Hz, 1H), 5.57 (s, 2H), 4.40 (t, J
= 6.2 Hz, 2H), 4.27 (dd, J = 9.4, 5.5 Hz, 1H), 3.94 (s, 6H), 3.93 (s, 3H), 3.84 (s, 3H),
3.29 (dd, J = 18.4, 9.7 Hz, 1H), 3.15 (d, J = 5.9 Hz, 2H), 3.11 (t, J = 6.1 Hz, 2H), 2.90
(dd, J = 16.7, 3.7 Hz, 1H), 2.75 (t, J = 5.8 Hz, 2H). 13C NMR (150 MHz, CDCl3) δ
189.12, 176.73, 175.29, 172.07, 169.59, 164.02, 158.86, 153.21, 153.14, 144.63,
143.64, 142.44, 140.04, 135.51, 133.54, 133.19, 132.48, 130.66, 129.92, 129.65,
129.61, 128.93, 128.11, 123.64, 122.12, 121.96, 120.42, 117.57, 113.80, 113.64,
112.49, 106.08, 104.73, 63.51, 61.02, 56.45, 56.12, 53.67, 38.89, 36.59, 30.43, 29.72,
28.46, 25.37. HR-MS (m/z) (ESI): calcd for C47H42FN5O12 [M+H]+: 888.28932;
found: 888.29195.
4.1.10. Synthesis of HI5-boc.
To a solution of D-MT (0.25 g, 1.14 mmol) and NaHCO3 (0.29 g, 3.44 mmol) in
30 mL of THF/H2O (1:1) was added (BOC)2O (0.3 g, 1.37 mmol) at 0 °C. The
mixture was stirred for 10 min, and then kept stirring at room temperature for 24 h.
After completion, the mixture was evaporated in vacuo to remove THF and the
aqueous layer was acidified with 1 M HCl to pH=1. Subsequently, the aqueous layer
was extracted with ethyl acetate, washed with saturated brine, and dried over with
anhydrous Na2SO4, and finally concentrated in vacuo to give 22 (0.27 g, 74.1%) as a
primrose yellow solid. To a mixture of compound 22 (0.065 g, 0.2 mmol), EDCI
(0.053 g, 0.27 mmol) and DMAP (0.023 g, 0.18 mmol) in DMF, compound 10 (0.1 g,
0.18 mmol) was added. The mixture was stirred at room temperature for 24 h, then
HI5-boc was obtained through the general handling procedure of HI1, with a white
1
solid (0.098 g, 61.3%). H NMR (600 MHz, CDCl3) δ 8.11 (d, J = 7.4 Hz, 1H), 8.08

(s, 1H), 7.51 (d, J = 7.9 Hz, 1H), 7.42 (dt, J = 13.9, 6.0 Hz, 2H), 7.25 (s, 1H), 7.22 –
7.15 (m, 3H), 7.10 (dd, J = 13.1, 4.7 Hz, 2H), 6.90 (d, J = 8.5 Hz, 1H), 6.74 (s, 1H),
6.53 (s, 2H), 6.48 (q, J = 12.2 Hz, 2H), 5.45 (s, 2H), 5.06 (d, J = 7.8 Hz, 1H), 4.56 (d,
J = 7.0 Hz, 1H), 4.38 – 4.29 (m, 2H), 3.81 (s, 3H), 3.79 (s, 3H), 3.73 (s, 6H), 3.70 (s,
3H), 3.18 (d, J = 5.7 Hz, 2H), 3.04 – 2.91 (m, 2H), 1.41 (s, 9H). HR-MS (m/z) (ESI):
calcd for C47H51N5O10 [M+H]+: 846.37842; found: 846.37944.
4.1.11. Synthesis of HI5.
HI5-boc (0.1 g, 0.11mmol) was suspended in 10% TFA/CH2Cl2 (v/v, 10 mL) at
room temperature for 2 h. The resulting solution was diluted with water (100 mL) and
extracted with CH2Cl2 (2 × 50 mL), and then dried with anhydrous Na2SO4 and
concentrated under vacuum. The crude product was purified on silica to give a white
1
solid of HI5 (0.078 g, 89.0%). H NMR (600 MHz, CDCl3) δ 8.11 – 8.05 (m, 2H),
7.54 (d, J = 7.9 Hz, 1H), 7.39 (dd, J = 10.6, 8.4 Hz, 2H), 7.25 (s, 1H), 7.21 – 7.17 (m,
2H), 7.12 – 7.08 (m, 3H), 6.90 (d, J = 8.5 Hz, 1H), 6.83 (s, 1H), 6.53 (s, 2H), 6.48 (d,
J = 5.4 Hz, 2H), 5.45 (s, 2H), 4.40 – 4.31 (m, 2H), 3.81 (s, 3H), 3.78 (s, 3H), 3.72 (s,
6H), 3.71 (s, 1H), 3.70 (s, 3H), 3.16 (dd, J = 14.5, 5.5 Hz, 1H), 2.98 (ddd, J = 23.2,
11.8, 4.6 Hz, 3H). 13C NMR (150 MHz, DMSO) δ 175.24, 164.00, 153.11, 150.51,
144.02, 139.39, 137.65, 137.24, 136.93, 133.92, 132.44, 130.11, 129.99, 129.92,
129.64, 129.38, 128.71, 128.49, 128.18, 128.13, 123.52, 123.31, 121.46, 118.97,
118.87, 113.22, 109.97, 109.52, 106.39, 63.43, 60.50, 56.38, 56.04, 55.53, 52.56,
32.67, 30.68, 25.25. HR-MS (m/z) (ESI): calcd for C42H43N5O8 [M+H]+: 746.31844;
found: 746.31828.
4.1.12. Synthesis of HI6-boc.
HI6-boc was obtained according to the general procedure of preparing HI5-boc.
22 was treated with 16, and then purified by column chromatograph to give HI6-boc
as a yellow solid (91 mg, 60.4%). 1H NMR (600 MHz, CDCl3) δ 8.18 (d, J = 7.4 Hz,
1H), 8.14 (s, 1H), 7.78 (d, J = 15.5 Hz, 1H), 7.56 – 7.49 (m, 4H), 7.49 – 7.44 (m, 2H),
7.36 (d, J = 15.5 Hz, 1H), 7.25 (s, 2H), 7.20 (t, J = 7.5 Hz, 1H), 7.17 (s, 1H), 7.10 (t, J
= 7.3 Hz, 1H), 7.05 (d, J = 8.5 Hz, 1H), 6.74 (s, 1H), 5.47 (s, 2H), 5.06 (d, J = 7.9 Hz,
1H), 4.59 – 4.54 (m, 1H), 4.37 – 4.30 (m, 2H), 3.94 (s, 6H), 3.93 (s, 3H), 3.87 (s, 3H),

3.70 (s, 3H), 3.18 (d, J = 5.6 Hz, 2H), 2.97 (dd, J = 44.2, 6.2 Hz, 2H), 1.41 (s, 9H).
HR-MS (m/z) (ESI): calcd for C48H51N5O11 [M+H]+: 874.37579; found: 874.37121.
4.1.13. Synthesis of HI6.
HI6 was obtained by a method similar to that of HI5 to give HI6 as a yellow
solid (80 mg, 90.0%). 1H NMR (600 MHz, DMSO) δ 8.11 (s, 1H), 8.07 (d, J = 7.9 Hz,
1H), 7.97 (s, 1H), 7.87 (d, J = 15.5 Hz, 1H), 7.80 (dd, J = 8.6, 1.9 Hz, 1H), 7.75 (d, J
= 15.4 Hz, 1H), 7.65 (d, J = 7.8 Hz, 1H), 7.58 (t, J = 7.7 Hz, 1H), 7.46 – 7.42 (m, 3H),
7.34 (d, J = 8.2 Hz, 1H), 7.27 (d, J = 8.6 Hz, 1H), 7.22 (t, J = 6.4 Hz, 1H), 7.11 (t, J =
7.3 Hz, 1H), 7.01 – 6.97 (m, 2H), 5.69 (s, 2H), 4.21 – 4.17 (m, 2H), 3.89 (s, 6H), 3.83
(d, J = 2.8 Hz, 3H), 3.76 (s, 3H), 3.69 (s, 3H), 3.58 (d, J = 6.2 Hz, 1H), 2.96 – 2.92 (m,
1H), 2.87 (dd, J = 11.5, 5.1 Hz, 3H). 13C NMR (150 MHz, DMSO) δ 188.11, 175.49,
164.08, 153.38, 153.35, 153.16, 144.03, 143.51, 142.41, 140.04, 137.75, 136.92,
134.12, 133.57, 130.35, 130.09, 130.04, 129.73, 129.27, 128.45, 128.35, 128.20,
123.56, 122.79, 121.44, 120.81, 118.98, 118.86, 113.42, 109.96, 109.68, 106.64,
105.98, 63.36, 60.66, 56.73, 56.63, 55.64, 52.57, 32.67, 30.87, 26.81, 25.26. HR-MS
(m/z) (ESI): calcd for C43H43N5O9 [M+H]+: 774.34316; found: 774.34332.
4.1.14. Synthesis of HI7.
HI7 was obtained by an analogous way to the synthesis of HI5 in which reaction
of compound 22 with compound 6 to give HI7 (89 mg, 83.1%). 1H NMR (600 MHz,
CDCl3) δ 7.67 (d, J = 7.9 Hz, 1H), 7.29 (d, J = 8.2 Hz, 1H), 7.24 – 7.21 (m, 1H), 7.12
(td, J = 8.5, 1.5 Hz, 2H), 6.99 (s, 1H), 6.90 (d, J = 2.1 Hz, 1H), 6.85 (d, J = 8.5 Hz,
1H), 6.50 (s, 2H), 6.45 (s, 2H), 5.31 (d, J = 19.9 Hz, 1H), 4.05 (dd, J = 7.9, 4.8 Hz,
1H), 3.82 (s, 3H), 3.79 (s, 3H), 3.75 (s, 3H), 3.71 (s, 6H), 3.46 (dd, J = 14.5, 4.7 Hz,
13
1H), 3.13 (dd, J = 14.5, 8.0 Hz, 1H). C NMR (150 MHz, CDCl3) δ 173.47, 152.99,
150.17, 139.43, 137.24, 137.05, 132.42, 130.20, 129.65, 128.57, 128.06, 127.87,
127.68, 123.23, 121.78, 121.12, 119.08, 115.09, 112.04, 109.59, 109.30, 106.07,
105.94, 60.92, 55.96, 55.20, 32.70, 30.56, 29.72. HR-MS (m/z) (ESI): calcd for
C30H32N2O6 [M+H]+: 517.23331; found: 517.22940.
4.1.15. Synthesis of HI8.
To a solution of 6 (0.90 g, 2.85 mmol) and K2CO3 (0.79 g, 5.71 mmol) in 20 mL

of DMF was added bromo-1-propanol (0.48 g, 3.42 mmol), the mixture was kept
stirring at 45 °C for 24 h. After completion, the mixture was poured into water and
extracted with water and CH2Cl2. The aqueous layer was washed with saturated brine,
and dried over with anhydrous Na2SO4, and finally the crude product was purified by
column chromatography to give 23 as a white solid (1.02 g, yield 78.5%). HI8 was
prepared by a way similar to that of HI5, in which 22 was treated with 23 to give HI8
1
as a white solid (91 mg, 89.1%). H NMR (600 MHz, CDCl3) δ 7.57 (d, J = 8.0 Hz,
1H), 7.24 (s, 1H), 7.19 (t, J = 7.5 Hz, 1H), 7.09 (t, J = 7.6 Hz, 1H), 6.90 (s, 1H), 6.88
(dd, J = 8.3, 1.4 Hz, 1H), 6.77 (dd, J = 8.9, 5.0 Hz, 2H), 6.51 (s, 2H), 6.49 – 6.42 (m,
2H), 4.24 (t, J = 6.2 Hz, 2H), 4.11 – 3.99 (m, 1H), 3.82 (d, J = 3.7 Hz, 6H), 3.79 (dd,
J = 6.2, 3.2 Hz, 2H), 3.70 (s, 3H), 3.68 (s, 6H), 3.23 (dd, J = 14.4, 5.2 Hz, 1H), 3.04
(dd, J = 14.4, 7.3 Hz, 1H), 1.99 (dd, J = 12.8, 6.4 Hz, 2H). 13C NMR (150 MHz,
CDCl3) δ 175.36, 152.97, 148.73, 147.65, 137.12, 137.01, 132.93, 129.92, 129.66,
128.94, 128.00, 127.65, 122.27, 121.72, 119.02, 118.90, 113.88, 111.42, 109.58,
109.28, 105.98, 103.35, 65.24, 61.81, 60.91, 55.96, 55.17, 32.62, 30.63, 29.72, 28.45.
HR-MS (m/z) (ESI): calcd for C33H38N2O7 [M+H]+: 575.27098; found: 575.27518.
4.1.15. Synthesis of 24.
Compound 24 was prepared according to a literature report [54] (Scheme S4). To
a solution of compound 25 (1.0 g, 5.3 mmol) in DMF (30 mL), compound 26 (0.94 g,
0.8 mmol) was added at 0 °C and stirred at room temperature overninght. After
completion, the reaction mixture was poured into water and extracted with EtOAC.
The combined organic layer was washed with sat. NaCl solution, dried over
anhydrous Na2SO4, concentrated and purified by flash chromatography
1
(CH2Cl2/MeOH=1:1) to give 24 as yellow brown solid (0.2 g, 12% yield). H NMR
(600 MHz, DMSO) δ 9.80 (s, 1H), 8.19 (dd, J = 29.4, 8.6 Hz, 2H), 7.94 (d, J = 8.4 Hz,
2H), 7.87 (s, 1H), 7.26 (d, J = 8.8 Hz, 2H), 6.75 (d, J = 8.8 Hz, 2H). HR-MS (m/z)
(ESI): calcd for C13H11N2O4S [M+H]+: 370.9623; found: 370.96029.
4.1.16. Analysis of hydrolysis behavior of HI5 by HPLC.
The compounds were dissolved in PBS (pH=5.0) or a solvent consisting of
acetonitrile and water (60:40, v/v) to prepare as standard samples, and then analyzed

by HPLC. Reversed phase HPLC was implemented on a 250×4.5 mm ODS column
and the HPLC profiles were recorded on UV detection at 245 nm. Mobile phase
consisted of acetonitrile/water (60:40, v/v), and flow rate was 1.0 mL/min. The
samples were filtrated by 0.45 mm filter before taken for HPLC analysis.
The release of HI5 in HeLa cells was also analyzed. Briefly, after co-incubated
with HI5 for 24 h, HeLa cells were collected and washed with PBS, and then lysed to
obtain cell lysates. Subsequently, the supernatant fraction was collected by
centrifugation, and filtered with a microporous filter membrane (0.22 µm). Finally, the
supernatant fraction was analyzed by reversed phase HPLC. The experimental
conditions of HPLC were the same as description above.
4.2. Biology
4.2.1. In vitro cytotoxicity
Cell culture. HeLa (Human cervix carcinoma), A549 (human lung cancer), PC-3
(human prostate cancer) and HUVEC (human umbilical vein endothelial cell) cell
lines were cultured at 37 °C in a humidified atmosphere with 5% CO2 in monolayer
culture in DMEM medium containing 100 mg/mL of penicillin, 10% fetal bovine
serum (FBS) and 100 mg/ mL of streptomycin.
MTT assay. Briefly, about a density of 5 × 104 cells/well cells in DMEM medium
with 10% FBS in each well of 96-well plates and incubated at 37 °C in a humidified
5% CO2 for 24 h. All the tested compounds were dissolved in DMSO and prepared for
different concentrations, subsequently added to each well and incubated at 37 °C in a
5% CO2 atmosphere for 72 h. The cells were collected and stained with MTT and
incubated for an additional 4 h, the suspension was thrown away and dissolved in
dimethyl sulfoxide. The UV absorption intensity was detected with an ELISA reader
at 490 nm. Cytotoxicity was determined on the percentage of cell survival in
comparison with the negative control. The IC50 values were calculated by the Bliss
method and repeated in three times.
4.2.2. Immunofluorescent staining.
For laser fluorescence confocal microscopy, this assay method was according to
the literature reported [61]. HeLa cells were seed into each well of six-well plates and

treated with the tested compounds at the indicated concentrations for 24 h. After that
cells were fixed in 4% paraformaldehyde at 37 °C for 15 min and washed three times
with ice-PBS, then permeabilized with 0.5% Triton X-100 for 15 min and blocked for
30 min by 10% goat serum at room temperature. After incubated overnight with
primary antibody (a-tubulin) and washed three times by ice-PBS, the cells were
incubated with fluorescence conjugated secondary antibody for 1 h and stained nuclei
of cells with DAPI in the dark at room temperature. Finally, Cells were visualized
using a confocal microscope.
4.2.3. Wound healing assays.
A wound-healing assay was carried out to assess the effect of HI5 on cell
migration. HeLa cells were grown in each well of 6-well plates for one day at 37 °C
in a humidified 5% CO2. The wounds were made perpendicular to the lines in
confluent monolayers by 200 µL tips, and nonadherent cells were removed by
washing with ice-PBS three times. The DMEM with 2.5 mL of 10% FBS containing
the tested compounds at the indicated concentrations were added and incubated at
37 °C in 5% CO2. After 24 h, the cells were dealt with ice-PBS two times, and then
photographed to mark the final scratched tracks at 0 and 24 h. The distance of cells
migrated to the wound area was determined manually.
4.2.4. Cell cycle measurement.
HeLa cells were grown in 6-well plates for 24 h at 37 °C in a humidified 5% CO2,
and then treated with or without the tested compounds at indicated concentrations.
After co-incubated for 24 h, cells were collected and washed with ice-cold PBS two
times, then fixed with 70% ethanol at 20 °C overnight. Subsequently, the cells were
washed with ice-cold PBS and resuspended in PBS containing 100 mg/mL RNase for
30 min, and then stained with propidium iodide (PI) at 4 °C for 30 min. Cell cycle
was finally measured by flow cytometry (FAC Scan, Becton Dickenson) using Cell
Quest software and recording propidium iodide (PI) in the FL2 channel.
4.2.5. Cell apoptosis analysis.
Apoptosis was evaluated by flow cytometry of cells double stained with annexin
V/FITC and PI. HeLa cells were treated with or without the tested compounds at

indicated concentrations for 24 h. After 24 h, the cells were collected and washed with
PBS for two times and then resuspend cells in Binding Buffer. The cells were
transferred to 5 mL culture tube, and co-incubated with FITC Annexin V (BD,
Pharmingen) and propidium iodide (PI) to stain using annexin-V FITC apoptosis kit
followed. Gently vortex the cells and incubated in the darkness for 30 min at room
temperature. The apoptosis experiment was analyzed using the system software (Cell
Quest; BD Biosciences).
4.2.6. AO/EB double staining assay.
The apoptotic cells were further detected by using AO/EB dual staining assay. In
brief, HeLa cells at the density of 2 ×106/well were seeded into 6-well plates in 10%
FBS-DMEM and cultured at 37 °C in 5% CO2 atmosphere overnight. After the
medium was replaced with fresh medium plus 10% fetal bovine serum, the cells were
treated with the tested compounds at indicated concentrations. After treatment for 24
h, the cells were collected and suspended in PBS, then stained with 20 µL of AO/EB
stain (100 mg/mL) for 20 min at room temperature. Fluorescence was observed on a
Nikon ECLIPSETE2000-S fluorescence microscope (OLYMPUS Co., Japan).
4.2.7. Measurement of the generation of intracellular ROS.
The intracellular ROS production was detected via using the peroxide-sensitive
fluorescent probe DCFH-DA (Beyotime, Haimen, China) by flow cytometry. In brief,
after HeLa cells treated with the tested compounds for 24 h, the cells were centrifuged
(2000 rpm) and collected, and subsequently cells were washed with PBS for twice and
incubated with 10 mM DCFHDA dissolved in cell free medium at 37 °C for 30 min.
The cells were washed with PBS for twice, cellular fluorescence was finally measured
by flow cytometry at an emission of 538 nm and an excitation of 485 nm.
4.2.8. Analysis of mitochondrial membrane potential by JC-1 staining.
Briefly, after treatment with the tested compounds for 24 h in 6 well plates, HeLa
cells were collected by centrifugation, and subsequently resuspended in a solution of
PBS containing JC-1 (5mg/mL) and incubated at 37 °C for 30 min. After that, the
cells were washed with PBS for twice, and analyzed immediately by using flow
cytometry. Furthermore, Visualization of JC-1aggregates (red fluorescence) and JC-1

monomers (green fluorescence) in red and green channels was done by using a
fluorescence microscope (10 × magnification).
4.2.8. Western blotting analysis.
HeLa cells were treated with or without the test compounds in 6-well plates at
37 °C for 24 h. After 24 h, cells were collected by centrifugation, and lysed in cell
lysis buffer containing phenylmethane sulfonyl fluoride (PMSF) for 30 min, the
lysates were then centrifuged and collected at 4 °C for 10 min. The concentration of
protein was determined by using the BCA protein assay reagents. Equal amounts of
protein per line was separated by sodium dodecyl sulfate polyacrylamide gel
electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes.
Membranes were blocked with 5% skim milk for 1 h after washed with tris-buffered
saline Tween-20 (TBST) buffer and then the membranes were then incubated with
primary antibodies in appropriate dilutions at 4 °C for overnight. Next, Membranes
were washed three times with TBST and incubated with peroxidase labeled secondary
antibodies for 2 h at RT (25 °C). The protein blots were measured by
chemiluminescence reagent (Thermo Fischer Scientifics Ltd.). Anti-β-actin antibody
was served as loading control.
4.2.9. Determination of kynurenine level assays.
The kynurenine level was determined according to the reported literature method
[28]. HeLa cells were seeded in a 6-well plate with 2 mL medium (containing 10%
FBS, 100 U/mL penicillin, and 100 mg/mL streptomycin). After replaced with fresh
medium containing L-tryptophan, the cells were incubated with hIFN-γ (25 ng/mL)
and the tested compounds for 48 h. After 48 h of incubation, the cells were collected
by centrifugation, and incubated with the 100 ml 20% trichloroacetic acid for protein
precipitation. Subsequently, the sample was centrifuged to remove the sediments, and
the supernatant was analyzed by HPLC. Reversed phase HPLC was implemented on a
250×4.5 mm ODS column and the HPLC profiles were recorded on UV detection at
480 nm. Mobile phase consisted of acetonitrile/water, and flow rate was 1.0 mL/min.
The samples were analyzed by HPLC after filtration with 0.45 mm filter.
4.2.10. Analysis of the proliferation of T cells.

A mixed leukocyte reaction (MLR) was performed to analyze the proliferation of
T cells as described previously [28]. HeLa cells were cultured in 6-well plates for 24 h.
After 24 h, the medium was replaced with fresh complete DMEM containing 50 ng/ml
IFN-γ and 0.1 mM L-tryptophan, and simultaneously treated with or without the
tested compounds. The HeLa cells were treated with compounds for 2 days at 37 °C.
PBMCs were stimulated with PHA-M and stained with Cell TraceTM Far Red Cell
Proliferation Kit (CFSE), and subsequently inserted into plates. After co-incubation at
37 °C for 6 days, the PBMCs were centrifuged and collected, and then the cell pellets
were resuspended in PBS and stained with anti-CD3 antibody. The activation and
proliferation of T-cell was analyzed by flow cytometry.
4.2.11. In vivo antitumor efficacy of HI5 in HeLa xenograft mice model.
The in vivo antitumor efficacy was evaluated in HeLa xenograft BALB/c
background nude mice model. Five-week-old male BALB/c nude mice (18-22 g) were
purchased from Shanghai slack laboratory animal co. LTD (China). The HeLa
cervical cancer cells (1 × 107 cells/0.1 mL/mouse) were subcutaneously inoculated
into the right armpit region of mice to establish tumors model. After the tumors
achieved a volume of 120-150 mm3 in each mouse, all mice were randomly divided
into four groups with five mice per group. The groups were injected with CA-4 at
doses of 30 mg/kg and HI5 at doses of 15 mg/kg and 30 mg/kg, respectively. The
negative control group was administered with an equivalent volume of physiological
saline. All groups were intravenously injected via a tail vein one dose per day in a
period of 21 days. The body weight and tumor volumes of mice were monitored every
2 days. The compounds were dissolved in solution containing Tween 80 and
physiological saline. After 3 weeks of treatment, all mice were sacrificed and weighed,
and tumor volumes were measured with electronic digital calipers and determined by
measuring length (A) and width (B) to calculate volume (V = AB2/2).
4.2.12. Immunohistochemistry assay.
The excised tumor and organs from one of each group (liver, heart, lung and
kidney) were collected and fixed in 4% paraformaldehyde, and subsequently
embedded in paraffin by using tissue embedding machine. The samples were sliced

into 5 mm sections by microtome and stained with H&E staining. In brief, sections
were prepared by dewaxing, and stained with Harris hematoxylin solution and
Eosin-phloxine solution, and then dehydrated and mounted with neutral resin. The
histopathologic changes were finally determined under fluorescence microscopy using
an Eclipse E800 Nikon (Nikon, Tokyo, Japan).
Notes
The authors declare no competing financial interest.
Acknowledgments
We are grateful to the National Natural Science Foundation of China (Grant Nos.
21571033 and 81503099) for financial aids to this work. The research was also
supported by the Zhishan Youth Scholar Program of SEU (2242019R40045). The
authors would also like to thank the Fundamental Research Funds for the Central
Universities (Project 2242017K41024) for supplying basic facilities to our key
laboratory.
References
[1] G.P. Dunn, L.J. Old, R.D. Schreiber, The immunobiology of cancer
immunosurveillance and immunoediting, Immunity 21 (2004) 137-148.
[2] M.H. Kershaw, J.A. Westwood, C.Y. Slaney, P.K. Darcy, Clinical application of
genetically modified T cells in cancer therapy, Clin. Transl. Immunol. 3 (2014)
1-16.
[3] W. Zou, Immunosuppressive networks in the tumour environment and their
therapeutic relevance, Nat. Rev. Cancer 5 (2005) 263-274.
[4] a) I. Mellman, G. Coukos, G. Dranoff, Cancer immunotherapy comes of age,
Nature 480 (2011) 480-489; b) C.A. Klebanoff, S.A. Rosenberg, N.P. Restifo,
Prospects for gene-engineered T cell immunotherapy for solid cancers, Nat. Med.
22 (2016) 26-36.
[5] A.L. Mellor, D.H. Munn, IDO expression by dendritic cells: tolerance and

tryptophan catabolism, Nat. Rev. Immunol. 4 (2004) 762-774.
[6] J.R. Moffett, M.A. Namboodiri, Tryptophan and the immune response, Immunol.
Cell Biol. 81 (2003) 247-265.
[7] P. Hwu, M.X. Du, R. Lapointe, M. Do, M.W. Taylor, H.A. Young, Indoleamine
2,3-dioxygenase production by human dendritic cells results in the inhibition of T
cell proliferation, J. Immunol. 164 (2000) 3596-3599.
[8] M. Sono, M.P. Roach, E.D. Coulter, J.H. Dawson, Hemecontaining oxygenases,
Chem. Rev. 96 (1996) 2841-2888.
[9] J.B. Katz, A.J. Muller, G.C. Prendergast, Indoleamine 2,3-dioxygenase in T-cell
tolerance and tumoral immune escape, Immunol. Rev. 222 (2008) 206-221.
[10] A. Coluccia, S. Passacantilli, V. Famiglini, M. Sabatino, A. Patsilinakos, R.
Ragno, C. Mazzoccoli, L. Sisinni, A. Okuno, O. Takikawa, R. Silvestri, G.L.
Regina, New inhibitors of indoleamine 2,3-dioxygenase 1: molecular modeling
studies, synthesis, and biological evaluation, J. Med. Chem. 59 (2016) 9760-9773.
[11] a) E. Gilboa, The promise of cancer vaccines, Nat. Rev. Cancer 4 (2004)
401-411. b) D.H. Munn, E. Shafizadeh, J.T. Attwood, I. Bondarev, A. Pashine, A.L.
Mellor, Inhibition of T cell proliferation by macrophage tryptophan catabolism, J.
Exp. Med. 189 (1999) 1363-1372.
[12] A.L. Mellor, D.H. Munn, Tryptophan catabolism and T-cell tolerance:
immunosuppression by starvation? Immunol. Today 20 (1999) 469-473.
[13] G. Brandacher, A. Perathoner, R. Ladurner, S. Schneeberger, P. Obrist, C. Winkler,
E.R. Werner, G. Werner-Felmayer, H.G. Weiss, G. Gobel, R. Margreiter, A.
Konigsrainer, D. Fuchs, A. Amberger, Prognostic value of indoleamine
2,3-dioxygenase expression in colorectal cancer: effect on tumor-infiltrating T
cells, Clin. Cancer Res. 12 (2006) 1144-1151.
[14] D.H. Munn, A.L. Mellor, Indoleamine 2,3-dioxygenase and tumor-induced
tolerance, J. Clin. Invest. 117 (2007) 1147-1154.
[15] A. Okamoto, T. Nikaido, K. Ochiai, S. Takakura, M. Saito, Y. Aoki, N. Ishii, N.
Yanaihara, K. Yamada, O. Takikawa, R. Kawaguchi, S. Isonishi, T. Tanaka, M.
Urashima, Indoleamine 2,3-dioxygenase serves as a marker of poor prognosis in

gene expression profiles of serous ovarian cancer cells, Clin. Cancer Res. 11
(2005) 6030-6039.
[16] D.H. Munn, A.L. Mellor, Indoleamine 2,3-dioxygenase and metabolic control of
immune responses, Trends Immunol. 34 (2013) 137-143.
[17] N. Van Baren, B.J. Van den Eynde, Tryptophan-degrading enzymes in tumoral
immune resistance, Front. Immunol. 6 (2015) 1-9.
[18] a) C.A. Opitz, U.M. Litzenburger, U. Opitz, F. Sahm, K. Ochs, C. Lutz, W. Wick,
M.
Platten,
The
Indoleamine-2,3-dioxygenase
(IDO)
inhibitor
1-methyl-D-tryptophan upregulates IDO1 in human cancer cells, PLOS One 6
(2011) 19823; b) A.C. Moreno, R.O. Clara, J.B. Coimbra, A.R. Julio, R.C.
Albuquerque, E.M. Oliveira, S.S. Maria-Engler, A. Campa, The expanding roles
of 1-methyl-tryptophan (1-MT): in addition to inhibiting kynurenine production,
1-MT activates the synthesis of melatonin in skin cells, FEBS. J. 280 (2013)
4782-4792.
[19] E.W. Yue, B. Douty, B. Wayland, M. Bower, X. Liu, L. Leffet, Q. Wang, K.J.
Bowman, M.J. Hansbury, C. Liu, M. Wei, Y. Li, R. Wynn, T.C. Burn, H.K.
Koblish, J.S. Fridman, B. Metcalf, P.A. Scherle, A.P. Combs, Discovery of potent
competitive inhibitors of indoleamine 2,3-dioxygenase with in vivo
pharmacodynamic activity and efficacy in a mouse melanoma model, J. Med.
Chem. 52 (2009) 7364-7367.
[20] L. Zhai, S. Spranger, D.C. Binder, G. Gritsina, K.L. Lauing, F.J. Giles, D.A.
Wainwright, Molecular pathways: targeting IDO1 and other tryptophan
dioxygenases for cancer immunotherapy, Clin. Cancer Res. 21 (2015) 5427-5433.
[21] a) J. Tumang, B. Gomes, M. Wythes, S. Crosignani, P. Bingham, P. Bottemanne,
H. Cannelle, S. Cauwenberghs, J. Chaplin, D. Dalvie, S. Denies, C. De
Maeseneire, P. Folger, K. Frederix, J. Guo, PF-06840003: a highly selective
IDO-1 inhibitor that shows good in vivo efficacy in combination with immune
checkpoint inhibitors, in: 107th Annu Meet Am Assoc Cancer Res (AACR) (April
(2016)16-20, New Orleans), Abst 4863. b) S. Crosignani, P. Bingham, P.
Bottemanne, M. Wythes, and et al. Discovery of a novel and selective indoleamine

2,3-dioxygenase
(IDO-1)
inhibitor
3-(5-fluoro-1H-indol-3-yl)pyrrolidine-2,5-dione (EOS200271/PF-06840003) and
its characterization as a potential clinical candidate, J. Med. Chem. 60 (2017)
9617-9629.
[22] K.D. Lu, C.B. He, N.N. Guo, C. Chan, K.Y. Ni, R.R. Weichselbaum, W.B. Lin,
Chlorin-based nanoscale metal-organic framework systemically rejects colorectal
cancers via synergistic photodynamic therapy and checkpoint blockade
immunotherapy, J. Am. Chem. Soc. 138 (2016) 12502-12510.
[23] M.R. Mautino, F.A. Jaipuri, J. Waldo, S. Kumar, J. Adams, C. Van Allen, A.
Marcinowicz-Flick, D. Munn, N.N Vahanian, C.J. Link, In proceedings of the
104th annual meeting of the american association for cancer research; Apr (2013)
6-10; Washington, DC. Philadelphia (PA): AACR; Vol. 73, Abstract nr 491.
[24] Y.C. Chen, R. Xia, Y.X. Huang, W.C. Zhao, J. Li, X.L. Zhang, P.C. Wang, R.
Venkataramanan, J. Fan, W. Xie, X.C. Ma, B.F. Lu, S. Li, An immunostimulatory
dual-functional nanocarrier that improves cancer immunochemotherapy, Nat.
Commun. 7 (2016) 1-12.
[25] A.J. Muller, J.B. DuHadaway, P.S. Donover, E. Sutanto-Ward, G.C. Prendergast,
Inhibition of indoleamine 2,3-dioxygenase, an immunoregulatory target of the
cancer suppression gene bin1, potentiates cancer chemotherapy, Nat. Med. 11
(2005) 312-319.
[26] M. Li, A.R. Bolduc, M.N. Hoda, D.N. Gamble, S.B. Dolisca, A.K. Bolduc, K.
Hoang, C. Ashley, D. McCall, A.M. Rojiani, B.L. Maria, O. Rixe, T.J. MacDonald,
P.S. Heeger, A.L. Mellor, D.H. Munn, T.S. Johnson, The indoleamine
2,3-dioxygenase pathway controls complement-dependent enhancement of
chemoradiation therapy against murine glioblastoma, J. Immunother. Cancer 2
(2014) 21-35.
[27] C. Uyttenhove, L. Pilotte, I. Theate, V. Stroobant, D. Colau, N. Parmentier, T.
Boon, B.J. Van den Eynde, Evidence for a tumoral immune resistance mechanism
based on tryptophan degradation by indoleamine 2,3-dioxygenase, Nat. Med. 9
(2003) 1269-1274.

(28) N. Wang, Z.G. Wang, Z.F. Xu, X.F. Chen, G.Y. Zhu, A cisplatin-loaded
immunochemotherapeutic nanohybrid bearing immune checkpoint inhibitors for
enhanced cervical cancer therapy, Angew. Chem. Int. Ed. 130 (2018) 3484 -3488.
[29] M.A. Jordan, L. Wilson, Microtubules as a target for anticancer drugs, Nat. Rev.
Cancer 4 (2004) 253-265.
[30] Q. Li, H.L. Sham, Discovery and development of antimitotic agents that inhibit
tubulin polymerisation for the treatment of cancer, Expert Opin. Ther. Pat. 12
(2002) 1663-1702.
[31] E. Mukhtar, V.M. Adhami, H. Mukhtar, Targeting microtubules by natural agents
for cancer therapy, Mol. Cancer Ther. 13 (2014) 275- 284.
[32] M. Kavallaris, Microtubules and resistance to tubulin-binding agents, Nat. Rev.
Cancer 10 (2010) 194-204.
[33] F. Pellegrini, D.R. Budman, Review: tubulin function, action of antitubulin drugs,
and new drug development, Cancer Invest. 23 (2005) 264-273.
[34] A. Jordan, J.A. Hadfield, N.J. Lawrence, A.T. McGown, Tubulin as a target for
anticancer drugs: agents which interact with the mitotic spindle, Med. Res. Rev.
18 (1998) 259-296.
[35] G.R. Pettit, S.B. Singh, E. Hamel, C.M. Lin, D.S. Alberts, D. Garcia-Kendall,
Isolation and structure of the strong cell growth and tubulin inhibitor
combretastatin A-4, Experientia 45 (1989) 209-211.
[36] C.M. Lin, H.H. Ho, G.R. Pettit, E. Hamel, Antimitotic natural products
combretastatin A-4 and combretastatin A-2: studies on the mechanism of their
inhibition of the binding of colchicine to tubulin, Biochemistry 28 (1989)
6984-6991.
[37] G.G. Dark, S.A. Hill, V.E. Prise, G.M. Tozer, G.R. Pettit, D.J. Chaplin,
Combretastatin A-4, an agent that displays potent and selective toxicity toward
tumor vasculature, Cancer Res. 57 (1997) 1829-1834.
[38] D.M. Patterson, G.J.S. Rustin, vascular damaging agents, Clin. Oncol. 19 (2007)
443-456.
[39] G. Nagaiah, S.C. Remick, CA-4 phosphate: a novel vascular disrupting agent,