Chemical and Pharmaceutical Bulletin Advance Publication by J-STAGE Advance Publication DOI:10.1248/cpb.c17-00824 December 22, 2017

Chemical and Pharmaceutical Bulletin

Regular Article

Supramolecular Complex of Methyl-β-cyclodextrin with - grafted Hyaluronic Acid as a Novel Antitumor Agent

Khaled M. Elamina, Yuki Yamashitaa, Taishi Higashia, Keiichi Motoyamaa,

Hidetoshi Arimaa,b,*

a Department of Physical Pharmaceutics, Graduate School of Pharmaceutical Sciences,

Kumamoto University, 5-1 Oe-honmachi, Chuo-ku, Kumamoto 862-0973, Japan

b Program for Leading Graduate Schools “HIGO (Health life science: Interdisciplinary and

Glocal Oriented) Program”, Kumamoto University

*Corresponding author: Hidetoshi Arima

E-mail address: [email protected]

Ⓒ 2017 The Pharmaceutical Society of Japan

ABSTRACT

Methyl-β-cyclodextrin (M-β-CyD) exhibits cytotoxic activity, and has the potentials as an

antitumor agent. However, a tumor-selectivity of M-β-CyD is low, leading to low antitumor

activity and the adverse effects. Meanwhile, hyaluronic acid (HA) is known as a promising tumor targeting ligand, because various cancer cells overexpress CD44, a HA-binding glycoprotein. In the present study, to develop a tumor-selective delivery system for M-β-

CyD, we designed a supramolecular complex of M-β-CyD with adamantane-grafted HA (Ad-

HA/M-β-CyD) and evaluated it as a tumor-selective antitumor agent. M-β-CyD formed a

4 -1 stable complex with Ad-HA (Kc > 10 M ). In addition, Ad-HA/M-β-CyD formed slightly a

negative-charged nanoparticle with ca. 140 nm of a particle size, indicating the favorable physicochemical properties for antitumor agents. Ad-HA/M-β-CyD showed the superior

cytotoxic activity via CD44-mediated endosomal pathways in HCT116 cells (CD44 (+)), a

human colon cancer cell line. In addition, cytotoxic activity of Ad-HA/M-β-CyD was induced by apoptosis. These results suggest that Ad-HA/M-β-CyD has the potentials as a

tumor-selective supramolecular antitumor agent.

Keywords:

Methyl-β-cyclodextrin; Hyaluronic acid; Adamantine; Supramolecular complex; Antitumor

agent

Chemical and Pharmaceutical Bulletin Advance Publication

Introduction

In recent years, various antitumor agents, such as small molecules, proteins, antibodies, genes

and small interfering RNA (siRNA), have been developed.1, 2) In addition, drug delivery

systems (DDS), such as an active targeting based on ligands modification and passive

targeting based on the enhanced permeability and retention (EPR) effect, are also utilized to

improve the antitumor effects and to reduce adverse effects of antitumor agents.3, 4) However,

in spite of evolution of cancer chemotherapy, low therapeutic effects, serious side effects,

drug resistance, and high cost are important issues in the clinical field.5) Therefore,

development of new kinds of antitumor agents is strongly required.

Recently, cyclodextrins (CyDs), typical pharmaceutical excipients, have attracted

considerable attention because they show therapeutic effects by themselves against Niemann-

Pick disease type C (NPC),6) familial amyloid polyneuropathy (FAP),7) Alzheimer's disease,8) atherosclerosis,9) septic shock,10, 11) and so on. Notably, CyDs are also acknowledged to

show antitumor activity. For instance, Grosse et al. reported that methyl-β-CyD (M-β-CyD)

shows antitumor activity in human tumor xenografted athymic nude mice after intraperitoneal

administration.12) In addition, we previously reported that M-β-CyD induces apoptosis in KB

cells, a human oral squamous carcinoma cell line, Ihara cells, a highly pigmented human

melanoma cell line, and M213 cells, a human cholangiocarcinoma cell line, through

cholesterol depletion in plasma membranes, and drastically inhibits the tumor growth after

intratumoral administration to Colon-26 cells-bearing mice.13) We also demonstrated that 2-

hydroxypropyl-β-CyD (HP-β-CyD) inhibits leukemic cell proliferation at physiologically

available doses without significant adverse effects.14) More recently, we prepared tumor

targeting ligand-appended M-β-CyD, namely folate-appended M-β-CyD (FA-M-β-CyD), to

achieve tumor-selective antitumor activity.15) FA-M-β-CyD induced strong antitumor

Chemical and Pharmaceutical Bulletin Advance Publication

activity beyond M-β-CyD and doxorubicin without significant adverse effects.16)

Importantly, FA-M-β-CyD was selectively recognized by folate receptor-α overexpressing cancer cells, and induced mitophagy-mediated cell death, not apoptosis.17)

Thus, FA-M-β-CyD has the great potentials as novel antitumor agents. However, few reports are available on tumor targeting ligand-appended M-β-CyD except for FA-M-β-

CyD.

Hyaluronic acid (HA) is one of the promising tumor targeting ligands, because it is recognized by CD44, a cell surface HA-binding glycoprotein that is overexpressed on various cancer cells, such as pancreatic cancer cells, lung cancer cells, breast cancer cells, and tumor initiating cancer stem like cells. 18, 19) HA shows high binding affinity with

-12 CD44 (Kd ≈ 10 M) and high safety, thus various HA-based drug carriers for antitumor

agents have been developed.20) The combination systems of HA and CyDs are also

developed as advanced drug carriers for antitumor agents. For instance, Yang et al. developed a supramolecular nanoparticle consisting of adamantane (Ad)-grafted HA (Ad-

HA) and camptothecin/β-CyD conjugate.21) Moreover, Badwaik et al. reported a plasmid

DNA carrier consisting of a supramolecular complex of Ad-HA with cationic β-CyD.22)

Yin et al. prepared HA conjugate with oligoethylenimine-grafted β-CyD as a tumor- selective gene carrier.23) In these cases, CyDs are used as the building block of drug carriers; however, little is known about the use of HA for delivery of CyDs to tumor cells, namely the use of combination systems of HA and CyDs as antitumor agents.

Based on these backgrounds, we herein prepared a supramolecular complex of M-β-

CyD with Ad-HA (Ad-HA/M-β-CyD), and evaluated its in vitro antitumor activity.

Firstly, Ad-HA was synthesized, and then Ad-HA/M-β-CyD was prepared by mixing the both components. Next, physicochemical properties of Ad-HA/M-β-CyD were examined

Chemical and Pharmaceutical Bulletin Advance Publication

by 1H-NMR and dynamic light scattering. In addition, cytotoxic activity, cellular association, and mechanism for cytotoxic activity of Ad-HA/M-β-CyD were also examined in HCT116 cells (CD44 (+)), a human colon cancer cell line, and NIH3T3 cells (CD44 (-)), a mouse fibroblast cell line.

Experimental

Materials

HA (M.W. 50 kDa) was supplied by Kewpie Corporation (Tokyo, Japan). 1-Adamantane methyl amine was purchased from Wako Pure Chemical Industries (Osaka, Japan). 1-Ethyl-

3-(3- dimethylaminopropyl)carbodiimide hydrochloride (EDC), 1-hydroxybenzotriazole

(HoBt), and Annexin V-FITC Apoptosis Detection Kit were purchased from Nacalai Tesque

(Kyoto, Japan). M-β-CyD (degree of substitution (DS) of methyl group of 12.2) was obtained from Tokyo Kasei (Tokyo, Japan). Dulbecco’s modified Eagle’s medium (DMEM) and fetal bovine serum (FBS) were purchased from Nissui Pharmaceuticals (Tokyo, Japan) and Nichirei (Tokyo, Japan), respectively. CD44 siRNA was purchased from Hokkaido

System Science, Co., Ltd. (Hokkaido, Japan). LipofectamineTM2000, cytochrome c monoclonal antibody-FITC conjugate, and 5-(4,6-dichlorotriazinyl)aminofluorescein (5-

DTAF) were obtained from Thermo Fisher Scientific Inc (Tokyo, Japan). Sephadex® G-15 was obtained from GE Healthcare UK Ltd. (Buckinghamshire, UK). All other chemicals and solvents were of analytical reagent grade, and deionized double-distilled water was used throughout the study.

Synthesis of Ad-HA

HA (200 mg) was dissolved in 50 mL of water by agitation for 3 h. Then, EDC (122 mg),

HOBt (4 mg) and 1-adamantne methyl amine (18 mg) were dissolved in 50 mL of DMSO,

Chemical and Pharmaceutical Bulletin Advance Publication

and mixed with the HA solution. After agitation for 24 h at 45°C, 500 mL of acetone was

added to yield the white precipitates, and the precipitates were collected by centrifugation

(12,000 rpm, 10 min). After washing 3 times with n-hexane (50 mL), the product was dried overnight under the reduced pressure. The resulting Ad-HA was dissolved in deuterium

1 oxide (D2O), and characterized by H-NMR (JEOL JNM-R 500 instrument, Tokyo, Japan),

operating at 500 MHz for protons at 25°C.

Stability Constant

The stability constant (Kc) of Ad-HA/M-β-CyD was determined by the analysis of peak

1 shifting in H-NMR at different concentration of M-β-CyD (0-5 mM) in D2O. The Kc value

was obtained from the following Benesi-Hildebrand Equation, assuming the 1:1 guest/host

interaction (Ad moiety of Ad-HA/M-β-CyD).

= + ∆ ∆ ∆

1 Where Δδ is the change in the H-NMR shifting, Δδ max is the maximum possible change in

1 H-NMR shift, [H]0 is the total M-β-CyD concentration, and Kc is the stability constant.

Physicochemical Properties

The particle size, polydispersity index and ζ-potential of Ad-HA/M-β-CyD were measured

using a Zetasizer Nano analyzer (Malvern Instruments, Worcestershire, UK). The complex

was prepared in 5% mannitol solution at a molar ratio of 1:1 (Ad moiety of Ad-HA/M-β-

CyD). The concentrations of Ad-HA and M-β-CyD were 4.2 nM and 102 nM, respectively.

Cell Culture

Chemical and Pharmaceutical Bulletin Advance Publication

HCT116 cells (CD44 (+)), a human colon cancer cell line, and NIH3T3 cells (CD44 (-)), a

mouse fibroblast cell line, were grown in a DMEM containing penicillin and streptomycin

supplemented with 10% FBS at 37ºC in a humidified 5% CO2 and 95% air atmosphere.

In Vitro Cytotoxic Activity

In vitro cytotoxic activity was assayed by the WST-1 method (a cell counting kit, Wako

Pure Chemical Industries, Osaka, Japan), as reported previously.24) Briefly, HCT116 cells

and NIH3T3 cells were seeded at 2 × 104 cells onto 96-well microplate (Iwaki, Tokyo,

Japan), and then incubated for 24 h in a humidified atmosphere of 5% CO2 and 95% air at

37°C. Cells were washed twice with DMEM, and then incubated with 150 μL of DMEM containing 0-5 mM M-β-CyD with or without Ad-HA for 2 h at 37°C. For the competition study, cells were pretreated for 1 h with DMEM containing 10 mg/mL of HA, and then incubated for 2 h with DMEM containing and 5 mM M-β-CyD with Ad-HA at 37°C. After

washing twice with phosphate-buffered saline (PBS, pH 7.4), cells were incubated with

10% WST-1 in HBSS for 30 min at 37°C, and then the absorbance was measured at 450 nm

against the reference wave length 620 nm using a microplate reader (Bio-Rad Model 550,

Tokyo, Japan).

In the case of CD44 knockdown HCT116 cells, the cells were seeded at 5 × 104 cells onto 24-well microplate (Iwaki, Tokyo, Japan), and then the cells were transfected for 1 h with Lipofectamine 2000TM containing siCD44 or siGL3 (siControl) as a control siRNA.

After adding 10% FBS, the cells were incubated for 24 h in a humidified atmosphere of 5%

CO2 and 95% air at 37°C. Cells were washed twice with DMEM, and then incubated for 2

h with 300 μL of DMEM containing 5 mM M-β-CyD with Ad-HA at 37°C. After that, the

cytotoxic activity was measured by the WST-1 method.

Chemical and Pharmaceutical Bulletin Advance Publication

Cellular Association

To obtain 5-DTAF-labelled M-β-CyD (5-DTAF-M-β-CyD), M-β-CyD (15 μM) and 5-

DTAF (7.6 μM) were dissolved in 0.2 M NaOH aqueous solution, and then stirred at room temperature for 24 h. The crude product was fractionized by Sepadex®-G15 to remove the unreacted 5-DTAF, and then lyophilized.

Next, HCT116 cells were seeded at 1 × 105 cells onto 24-well microplate, and then incubated for 24 h in a humidified atmosphere of 5% CO2 and 95% air at 37°C. Cells were washed twice with DMEM, and then incubated for 2 h with 300 μL of DMEM containing 5-DTAF-M-β-CyD (15 μM) with or without Ad-HA. For the competition study, cells were pretreated for 1 h with DMEM containing 10 mg/mL of HA, and then incubated for 2 h with 300 μL of DMEM containing 5-DTAF-M-β-CyD (15 μM) with or without Ad-HA. For the endocytosis inhibition, the cells were pretreated for 1 h with

DMEM containing amiloride (25 μg/mL), chloropromzine (10 μg/mL), genistein (200

μg/mL) or filipin III (5 μg/mL). Then, the cells were incubated for 2 h with DMEM containing 5-DTAF-M-β-CyD (15 μM) with or without Ad-HA. The cells were washed twice with PBS, detached, suspended in PBS, and kept on ice. The intensity was measured using FACS Calibur flow cytometer with

CellQuest software (Becton-Dickinson, Mountain View, CA, USA).

Intracellular Distribution

HCT116 cells were seeded at 2 × 105 cells onto 35 mm glass bottom dish, and then incubated for 24 h in a humidified atmosphere of 5% CO2 and 95% air at 37°C. Cells were washed twice with DMEM, and then incubated for 2 h with 150 μL of DMEM containing

5-DTAF-M-β-CyD (15 μM) with or without Ad-HA. For the competition study, cells were

Chemical and Pharmaceutical Bulletin Advance Publication

pretreated for 1 h with DMEM containing 10 mg/mL of HA, and then incubated for 2 h

with 200 μL of DMEM containing 5-DTAF-M-β-CyD (15 μM) with Ad-HA. Then, the

cells were washed with PBS, and incubated with 10 μM Hoechest 33342 for 10 min. To

observe the colocalization with mitochondria, 10 μM 123, a mitochondria

marker, was incubated for 30 min before the treatment with Hoechest 33342, a nucleus

marker. The cells were fixed with 4% paraformaldehyde in PBS for 15 min, and observed

by a Biozero BZ-8000 fluorescent microscope (KEYENCE, Osaka, Japan).

Cholesterol Efflux from the Cells

HCT116 cells were seeded in 12-well plate (5 × 105/well) for 24 h. The cells were

incubated with 500 μL of M-β-CyD (5 mM) with Ad-HA at 37°C for 1 h. After

centrifugation (10,000 rpm, 5 min), the supernatant was recovered, and total cholesterol

level was determined using a Cholesterol E test Wako (Wako Pure Chemical Industries,

Osaka, Japan).

Mechanism for Cytotoxic Activity

HCT116 cells were seeded at 2 × 105 cells onto 24-well microplate and then incubated for

24 h in a humidified atmosphere of 5% CO2 and 95% air at 37°C. Cells were washed twice

with DMEM, and then incubated for 2 h at 37°C with 300 μL of DMEM containing 5 mM

M-β-CyD with Ad-HA in the absence and presence of 10 mg/mL of HA. The cells were trypsinized, centrifuged and washed 2 times with PBS. The cells were incubated with

Annexin V-FITC and propidium iodide (PI) solution according to the Annexin V Kit protocol, and the fluorescence intensity was measured using a FACS Calibur flow cytometer with CellQuest software (Becton-Dickinson, Mountain View, CA, USA). For

Chemical and Pharmaceutical Bulletin Advance Publication

cytochrome c assay, the cells collected by centrifugation were suspended in 100 μL of PBS containing 1 μg of anti-cytochrome c, and the fluorescence intensity was measured using the FACS Calibur flow cytometer.

Data Analysis

The data are presented as the mean ± standard error (S.E.). The statistical significance of the mean coefficients of the experimental data was performed using an analysis of variance

(ANOVA) followed by Scheffe's test. In addition, p-values for significance were set at

0.05.

Result and Discussion

Preparation of Ad-HA

To fabricate Ad-HA/M-β-CyD, Ad-HA was prepared by the condensation reaction of 1- adamantane methyl amine and HA (Fig. 1). In 1H-NMR spectrum of Ad-HA, peaks of HA and Ad were observed, indicating the successful preparation of Ad-HA (Fig. 2). In addition, the average degree of substitution (DS) of Ad per one HA molecule was calculated as 19, i.e. 18% of carboxyl group of HA was substituted with Ad moiety. Oh et al. reported that

HA derivatives with less than 25% of degree of HA modification could be entered the cells through HA receptor-mediated endocytosis.25) Therefore, Ad-HA prepared in this study probably possesses CD44 binding ability.

Physicochemical Properties of Ad-HA/M-β-CyD

To confirm the interaction between Ad moiety of Ad-HA and M-β-CyD, namely the formation of Ad-HA/M-β-CyD, 1H-NMR spectra were measured at various concentrations

Chemical and Pharmaceutical Bulletin Advance Publication

of M-β-CyD (Fig. 3a). The peak of ethylene group of Ad in Ad-HA (ca. 1.58 ppm) was shifted by the addition of M-β-CyD in the concentration-dependent manner, indicating the interaction of the both components. Therefore, the Kc value of Ad-HA/M-β-CyD was determined by the Benesi-Hildebrand plot (Fig. 3b). As a result, the Kc value of Ad-HA/M-

4 -1 4 −1 β-CyD was calculated as 1.04 × 10 M . Herein, Stella et al. reported that Kc > 10 M for

a CyD complex with a guest molecule is required to exist as an inclusion complex in

26) 4 -1 vivo. Additionally, Kurkov et al. demonstrated that more than 4 × 10 M of Kc of

CyD/drug complexes is needed to alter the drug pharmacokinetics after the parenteral administration.27) Furthermore, Leong et al. reported that sulfobutylether β-CyD is able to

alter the pharmacokinetics of Ad derivatives in intravenous administration, when the Kc of

the complex is more than ca. 2 × 104 M−1.28) Therefore, the interaction between Ad-HA

and M-β-CyD may be somewhat weak for the systemic administrations. However, a

competitive interaction of CyDs with endogenous compounds, guest molecule binding to

plasma and tissue components, guest molecule uptaken into tissues, and an elimination rate

of the CyDs are also important factors for the dissociation of the CyD complexes.26)

Therefore, we should investigate these factors and further in vivo pharmacokinetics parameters. Anyhow, these results suggest that M-β-CyD formed a supramolecular complex with Ad-HA.

Next, the particle size, polydispersity index and ζ-potential of Ad-HA/M-β-CyD were examined by a Zetasizer Nano (Table 1). Particle size, polydispersity index and ζ-potential of Ad-HA/M-β-CyD were 140.3 nm, 0.25 and -4.2 mV, respectively, and no significant difference with Ad-HA was observed, suggesting their nanoparticle formations. The size of

Ad-HA/M-β-CyD is favored for the EPR effect.29, 30) Moreover, He et al. demonstrated

that the nanoparticles with a slight negative charge and particle size of 150 nm efficiently

Chemical and Pharmaceutical Bulletin Advance Publication

accumulate in tumor tissue.31) Thus, Ad-HA/M-β-CyD may possess favorable physicochemical properties for antitumor agents.

Cytotoxic Activity of Ad-HA/M-β-CyD

To evaluate the potential of Ad-HA/M-β-CyD as an antitumor agent, we next examined its cytotoxic activity in HCT116 cells (CD44 (+)) (Fig. 4a). Ad-HA/M-β-CyD showed significantly higher cytotoxic activity than M-β-CyD. Importantly, the cytotoxic activity of

Ad-HA/M-β-CyD was decreased by the addition of HA, as a competitor of CD44 (Fig. 4a), although the inhibitory effects were somewhat weak. The complexation of Ad-HA with M-

β-CyD is actually based on a noncovalent host-guest interaction through inclusion phenomena between M-β-CyD and Ad. Therefore, it is highly possible that Ad-HA/M-β-

CyD dissociated outside the cells in the presence of HA, because cellular uptake of Ad-

HA/M-β-CyD was inhibited by HA. Meanwhile, M-β-CyD released from Ad-HA/M-β-

CyD should extract cholesterol from the plasma membranes, leading to moderate cytotoxic activity (Fig. 4a). Hence, we next examined the cytotoxic activity of Ad-HA/M-β-CyD at

4°C (Fig. 4b), because inclusion complex is stable at low temperature and endocytic trafficking is prevented at 4°C. As a result, the cytotoxic activity of Ad-HA/M-β-CyD was markedly attenuated at 4°C. These results suggest the superior cytotoxic activity of Ad-

HA/M-β-CyD through CD44-mediated endosomal pathways.

Next, to make sure the importance of CD44 for cytotoxic activity of Ad-HA/M-β-CyD, we compared its cytotoxic activity in HCT116 cells (CD44 (+)) with that in NIH3T3 cells (CD44 (-))32) (Fig. 5a, b). Ad-HA/M-β-CyD showed higher cytotoxic activity than

M-β-CyD in HCT116 cells (CD44 (+)) (Fig. 5a). In contrast, a negligible difference of the cytotoxic activity between Ad-HA/M-β-CyD and M-β-CyD was observed in NIH3T3

Chemical and Pharmaceutical Bulletin Advance Publication

cells (CD44 (-)) (Fig. 5b). Moreover, the activity of Ad-HA/M-β-CyD was also decreased

by the CD44 knockdown in HCT116 cells (Fig. 5c), strongly suggesting that CD44-

mediated endosomal pathways are involved in the cytotoxic activity of Ad-HA/M-β-CyD.

To exhibit cytotoxic activity, Ad-HA/M-β-CyD should dissociate in the cells. The

detailed mechanism for dissociation of Ad-HA/M-β-CyD is unclear. However, HA is

degraded to small fragments by hyaluronidase in lysosomes. This degradation may lead to

the dissociation of Ad-HA/M-β-CyD, along with disruption of the nanoparticle structure.

Cellular Association of Ad-HA/M-β-CyD

To gain insight into the mechanism for the CD44-overexpressing tumor cell-selective cytotoxic activity of Ad-HA/M-β-CyD, we examined the cellular association of the complex of fluorescent labeled M-β-CyD (5-DTAF-M-β-CyD) with Ad-HA after the

incubation for 2 h in HCT116 cells in the absence and presence of HA as a competitor of

CD44 (Fig. 6a). The cellular association of Ad-HA/5-DTAF-M-β-CyD was significantly

higher than that of 5-DTAF-M-β-CyD, and was decreased by the addition of HA. These

results suggest the CD44-mediated cellular association of Ad-HA/M-β-CyD.

There are several pathways for cellular uptake of macromolecules, such as

macropinocytosis, clathrin-dependent endocytosis, clathrin-independent endocytosis, and

caveolae-dependent endocytosis.33) In order to clarify the endocytosis mechanism of Ad-

HA/M-β-CyD into HCT116 cells, we examined the cellular association of Ad-HA/M-β-

CyD in the presence of some endocytosis inhibitors, including amiloride as a

macropinocytosis inhibitor, chlorpromazine as a clathrin-dependent endocytosis inhibitor,

genistein as a clathrin-independent endocytosis inhibitor, and filipin III as a caveolae- dependent endocytosis inhibitor (Fig. 6b).33, 34) The cellular association of Ad-HA/M-β-

Chemical and Pharmaceutical Bulletin Advance Publication

CyD was significantly decreased by the treatment with genistein and filipin III, but not with amiloride or chlorpromazine, suggesting the clathrin-independent pathway and caveolae-dependent pathway for the cellular association of Ad-HA/M-β-CyD.

Oommen et al. demonstrated that cellular uptake of -grafted HA is derived from clathrin-mediated endocytosis in HCT116 cells.35) Moreover, Contreras-Ruiz et al. reported that HA-chitosan oligomer-based nanoparticles enter HCE cells, a SV40- immortalized human corneal epithelial cell line, and IOBA-NHC cells, a spontaneously immortalized epithelial cell line from normal human conjunctiva, through caveolin- dependent endocytosis.36) These results suggest that Ad-HA/M-β-CyD entered

HCT116 cells via clathrin-independent pathway and caveolae-dependent pathway as well as some HA-grafted compounds. Meanwhile, the negligible effects of the inhibitors on cellular association of 5-DTAF-M-β-CyD were observed (Fig. 6c). These results suggest that cellular association of 5-DTAF-M-β-CyD is independent of endocytosis. Actually, 5-DTAF-M-β-CyD possesses the hydrophobic 5-DTAF moiety, therefore, it is likely that cellular association of 5-DTAF-M-β-CyD is dependent on passive transport and/or binding to the cell surface.

Intracellular Distribution of Ad-HA/M-β-CyD

For further investigation of the intracellular behavior of Ad-HA/M-β-CyD, we examined its intracellular distribution in HCT116 cells (Fig. 7). Ad-HA/M-β-CyD mainly localized in the cytoplasm rather than in nucleus 2 h after the treatment, and its accumulation was inhibited by the addition of HA. Therefore, cytotoxic activity of Ad-HA/M-β-CyD is induced probably due to the interaction of M-β-CyD with the components of cell membrane, such as cholesterol and phospholipids, in cytoplasm, not nucleus.

Chemical and Pharmaceutical Bulletin Advance Publication

We next examined intracellular distribution of Ad-HA/M-β-CyD (Fig. 7b). After Ad-

HA/M-β-CyD applied to the HCT116 cells, the 5-DTAF-M-β-CyD distributed in the cytoplasm, suggesting the endosomal escape of Ad-HA/M-β-CyD. Herein, we previously reported that

CyDs accelerate endosomal escape through the interaction of CyDs with phospholipids and cholesterol. 37-41) Therefore, Ad-HA/M-β-CyD may be also able to escape from endosomes

through the interaction between M-β-CyD released from Ad-HA/M-β-CyD and

phospholipids or cholesterol of endosomal membranes. Additionally, Ad-HA/M-β-CyD

colocalized with rhodamine 123, a mitochondria marker (Fig. 7b). Hence, these results

suggest that Ad-HA/M-β-CyD may distribute, at least in part, in mitochondria after cellular

uptake into HCT116 cells (CD44 (+)), and may affect its function.

Mechanism for Cytotoxic Activity of Ad-HA/M-β-CyD

A number of mechanisms of cell death, such as apoptosis, autophagy, necrosis and

programed necrosis, are acknowledged. We previously demonstrated that M-β-CyD

induces apoptotic cell death in KB cells through the efflux of the cholesterol from the lipid

raft of plasma membrane.13) Interestingly, we also demonstrated that FA-M-β-CyD induces

the cell death via mitophagy, one of the autophagy, in KB cells.17) In the latter case, FA-M-

β-CyD enters the cells and impairs the mitochondrial function. Thus, M-β-CyDs could

induce the different cell death mechanisms by acting outside or inside the cells.

Therefore, to gain insight into the mechanism for cytotoxic activity of Ad-HA/M-β-

CyD, we examined the efflux of the cholesterol from the plasma membrane after the

treatment with Ad-HA/M-β-CyD in HCT116 cells (Fig. 8a). The cholesterol level

extracted from the cells after treatment with Ad-HA/M-β-CyD was significantly lower than

that with M-β-CyD, although it was higher than that in the control. These results suggest

Chemical and Pharmaceutical Bulletin Advance Publication

that the effects of Ad-HA/M-β-CyD on the plasma membrane are lower than those of

M-β-CyD, probably due to occupation of the cavity of M-β-CyD by Ad-HA outside the cells.

Next, to examine whether Ad-HA/M-β-CyD induces apoptosis, we performed

Annexin V-FITC/PI assay in HCT116 cells after incubation with Ad-HA/M-β-CyD for

2 h (Fig. 8b). As a result, Ad-HA/M-β-CyD induced the highest apoptotic activity, and the ratio of double (Annexin V/PI) positive cells in HCT 116 cells was higher (35%) than that of M-β-CyD (23%). Moreover, the addition of HA reduced the Annexin V/PI positive cells in HCT116 cells. These results suggest that Ad-HA/M-β-CyD induces apoptosis in HCT116 cells.

Cytochrome c translocates from the mitochondria to the cytoplasm in apoptotic cells.42) Therefore, we measured the cytochrome c level after treatment with Ad-

HA/M-β-CyD for 2 h in HCT116 cells (Fig. 8c). The level was increased by the treatment with Ad-HA/M-β-CyD, and was decreased by the addition of HA, suggesting that Ad-HA/M-β-CyD induces the cytotoxic activity thorough apoptosis.

As described above, FA-M-β-CyD induces mitophagy in KB cells by acting into the cells.17) Interestingly, Ad-HA/M-β-CyD induces apoptosis, although M-β-CyD enters the HCT116 cells after the treatment with Ad-HA/M-β-CyD. In the former case, FA-

M-β-CyD upregulated mitochondrial transmembrane potential in KB cells via mitochondrial folate transporter on the mitochondria, resulting in mitophagy.17)

Hereafter, we should clarify the detailed mechanism for apoptosis induced by Ad-

HA/M-β-CyD.

Conclusion

Chemical and Pharmaceutical Bulletin Advance Publication

In this study, we demonstrated the in vitro antitumor activity of a supramolecular complex

of M-β-CyD with Ad-HA in CD44-positive HCT116 cells through apoptosis (Fig. 9).

Recently, a number of HA-based drug carriers for antitumor drugs have been developed,43-

46) and they improved antitumor activity of the drugs in vivo. However, antitumor drugs

often cause adverse effects, but M-β-CyD is expected to provide slight side effects because

it has been widely used as pharmaceutical excipients, compared to common antitumor

drugs. Therefore, Ad-HA/M-β-CyD may be useful as a safe tumor-selective

supramolecular antitumor agent.

Conflict of Interest

The authors declare no conflict of interest.

Chemical and Pharmaceutical Bulletin Advance Publication

Refrences

1) Lam J. K. W., Chow M. Y. T., Zhang Y., Leung S. W. S., Mol. Ther. Nucleic Acids, 4, e252 (2015). 2) Sliwkowski M. X., Mellman I., Science, 341, 1192-1198 (2013). 3) Bertrand N., Wu J., Xu X., Kamaly N., Farokhzad O. C., Adv. Drug Deliv. Rev., 66, 2-25 (2014). 4) Maeda H., Nakamura H., Fang J., Adv. Drug Deliv. Rev., 65, 71-79 (2013). 5) Xu X., Ho W., Zhang X., Bertrand N., Farokhzad O., Trends Mol. Med., 21, 223-232 (2015). 6) Vite C. H., Bagel J. H., Swain G. P., Prociuk M., Sikora T. U., Stein V. M., O’Donnell P., Ruane T., Ward S., Crooks A., Li S., Mauldin E., Stellar S., De Meulder M., Kao M. L., Ory D. S., Davidson C., Vanier M. T., Walkley S. U., Sci. Transl. Med., 7, 276ra226-276ra226 (2015). 7) Jono H., Anno T., Motoyama K., Misumi Y., Tasaki M., Oshima T., Mori Y., Mizuguchi M., Ueda M., Shono M., Obayashi K., Arima H., Ando Y., Biochem. J., 437, 35-42 (2011). 8) Yang D.-S., Stavrides P., Kumar A., Jiang Y., Mohan P. S., Ohno M., Dobrenis K., Davidson C. D., Saito M., Pawlik M., Hum. Mol. Genet., 26, 843-859 (2017). 9) Zimmer S., Grebe A., Bakke S. S., Bode N., Halvorsen B., Ulas T., Skjelland M., De Nardo D., Labzin L. I., Kerksiek A., Sci. Tra. Med., 8, 333ra350-333ra350 (2016). 10) Motoyama K., Arima H., Nishimoto Y., Miyake K., Hirayama F., Uekama K., FEBS Lett., 579, 1707-1714 (2005). 11) Arima H., Motoyama K., Matsukawa A., Nishimoto Y., Hirayama F., Uekama K., Biochem. Pharmacol., 70, 1506-1517 (2005). 12) Grosse P. Y., Bressolle F., Pinguet F., Br. J. Cancer, 78, 1165-1169 (1998). 13) Onodera R., Motoyama K., Okamatsu A., Higashi T., Kariya R., Okada S., Arima H., Int. J. Pharm., 452, 116-123 (2013). 14) Yokoo M., Kubota Y., Motoyama K., Higashi T., Taniyoshi M., Tokumaru H., Nishiyama R., Tabe Y., Mochinaga S., Sato A., Sueoka-Aragane N., Sueoka E., Arima H., Irie T., Kimura S., PLoS One, 10, e0141946 (2015). 15) Onodera R., Motoyama K., Arima H., J. Incl. Phenom. Macrocycl. Chem., 70, 321-326 (2011). 16) Onodera R., Motoyama K., Okamatsu A., Higashi T., Arima H., Sci. Rep., 3, 1104 (2013). 17) Kameyama K., Motoyama K., Tanaka N., Yamashita Y., Higashi T., Arima H., Int. J. Nanomed., 12, 3433-3446 (2017). 18) Suzuki M., Kobayashi H., Kanayama N., Nishida T., Takigawa M., Terao T., Biochim. Biophys. Acta, 1591, 37-44 (2002). 19) Miyake H., Hara I., Okamoto I., Gohji K., Yamanaka K., Arakawa S., Saya H., Kamidono S., J. Urol., 160, 1562-1566 (1998). 20) Hou X., Yang C., Zhang L., Hu T., Sun D., Cao H., Yang F., Guo G., Gong C., Zhang X., Tong A., Li R., Zheng Y., Biomaterials, 124, 195-210 (2017). 21) Yang Y., Zhang Y.-M., Li D., Sun H.-L., Fan H.-X., Liu Y., Bioconjug. Chem., 27, 2834- 2838 (2016). 22) Badwaik V., Liu L., Gunasekera D., Kulkarni A., Thompson D. H., Mol. Pharmaceut., 13, 1176-1184 (2016). 23) Yin H., Zhao F., Zhang D., Li J., Int. J. Pharm., 483, 169-179 (2015). 24) Ono N., Arima H., Hirayama F., Uekama K., Biol. Pharm. Bull., 24, 395-402 (2001). 25) Oh E. J., Park K., Kim K. S., Kim J., Yang J.-A., Kong J.-H., Lee M. Y., Hoffman A. S., Hahn S. K., J. Control. Release, 141, 2-12 (2010). 26) Stella V. J., Rao V. M., Zannou E. A., Zia V., Adv. Drug Deliv. Rev., 36, 3-16 (1999). 27) Kurkov S. V., Madden D. E., Carr D., Loftsson T., J. Pharm. Sci., 101, 4402-4408 (2012). 28) Leong N. J., Prankerd R. J., Shackleford D. M., McIntosh M. P., J. Pharm. Sci., 104, 1492- 1498 (2015). 29) Matsumura Y., Kataoka K., Cancer Sci., 100, 572-579 (2009).

Chemical and Pharmaceutical Bulletin Advance Publication

30) Maeda H., Wu J., Sawa T., Matsumura Y., Hori K., J. Control. Release, 65, 271-284 (2000). 31) He C., Hu Y., Yin L., Tang C., Yin C., Biomaterials, 31, 3657-3666 (2010). 32) Yamada Y., Hashida M., Harashima H., Biomaterials, 52, 189-198 (2015). 33) Nam H. Y., Kwon S. M., Chung H., Lee S.-Y., Kwon S.-H., Jeon H., Kim Y., Park J. H., Kim J., Her S., Oh Y.-K., Kwon I. C., Kim K., Jeong S. Y., J. Control. Release, 135, 259-267 (2009). 34) Vercauteren D., Vandenbroucke R. E., Jones A. T., Rejman J., Demeester J., De Smedt S. C., Sanders N. N., Braeckmans K., Mol. Ther., 18, 561-569 (2010). 35) Oommen O. P., Duehrkop C., Nilsson B., Hilborn J., Varghese O. P., ACS Appl. Mater. Interfaces, 8, 20614-20624 (2016). 36) Contreras-Ruiz L., de la Fuente M., Párraga J. E., López-García A., Fernández I., Seijo B., Sánchez A., Calonge M., Diebold Y., Mol. Vis., 17, 279-290 (2011). 37) Arima H., Kihara F., Hirayama F., Uekama K., Bioconjug. Chem., 12, 476-484 (2001). 38) Kihara F., Arima H., Tsutsumi T., Hirayama F., Uekama K., Bioconjug. Chem., 13, 1211- 1219 (2002). 39) Kihara F., Arima H., Tsutsumi T., Hirayama F., Uekama K., Bioconjug. Chem., 14, 342-350 (2003). 40) Anno T., Higashi T., Motoyama K., Hirayama F., Uekama K., Arima H., Int. J. Pharm., 426, 239-247 (2012). 41) Arima H., Motoyama K., Higashi T., Ther. Deliv., 8, 215-232 (2017). 42) Gogvadze V., Orrenius S., Zhivotovsky B., Biochim. Biophys. Acta, 1757, 639-647 (2006). 43) Tripodo G., Trapani A., Torre M. L., Giammona G., Trapani G., Mandracchia D., Eur. J. Pharm. Biopharm., 97, 400-416 (2015). 44) Han M., Huang-Fu M. Y., Guo W. W., Guo N., Chen J., Liu H., Xie Z., Lin M., Wei Q., Gao J., ACS Appl. Mater. Interfaces, in press (2017). 45) Vogus D. R., Evans M. A., Pusuluri A., Barajas A., Zhang M., Krishnan V., Nowak M., Menegatti S., Helgeson M. E., Squires T. M., Mitragotri S., J. Control. Release, in press (2017). 46) Chi Y., Yin X., Sun K., Feng S., Liu J., Chen D., Guo C., Wu Z., J. Control. Release, 261, 113-125 (2017).

Chemical and Pharmaceutical Bulletin Advance Publication

Figure legends

Fig. 1. Preparation pathway of Ad-HA and Ad-HA/M-β-CyD.

1 Fig. 2. H-NMR spectrum of Ad-HA in D2O.

Fig. 3. (a) Effect of M-β-CyD on chemical shift of Ad-HA in 1H-NMR spectrum. (b)

Benesi-Hildebrand plots for changes in 1H-NMR spectrum of Ad-HA by the addition of M-β-

CyD in D2O. The stability constant was calculated by Benesi-Hildebrand plot. Each point

represents the mean ± S.E. of 3 experiments.

Fig. 4. (a) Cytotoxic activity of Ad-HA/M-β-CyD in HCT116 cells (CD44 (+)) in the

absence and presence of HA. (b) Effect of the temperature on the cytotoxic activity of Ad-

HA/M-β-CyD in HCT116 cells. HCT116 cells were incubated with M-β-CyD (5 mM) and

Ad-HA for 2 h at 37ºC or 4ºC. The cytotoxic activity was assayed by the WST-1 method.

Each value represents the mean ± S.E. of 3-4 experiments. *p < 0.05, compared with control.

†p < 0.05, compared with M-β-CyD. ‡p < 0.05, compared with Ad-HA/M-β-CyD. §p < 0.05,

compared with Ad-HA/M-β-CyD at 37ºC.

Fig. 5. Cytotoxic activity of Ad-HA/M-β-CyD in (a) HCT116 cells (CD44 (+)) and (b)

NIH3T3 cells (CD44 (-)). HCT116 cells or NIH3T3 cells were incubated with M-β-CyD (0-

5 mM) and Ad-HA for 2 h at 37ºC. The cytotoxic activity was assayed by the WST-1 method. Each value represents the mean ± S.E. of 3-4 experiments. *p < 0.05, compared with

M-β-CyD. †p < 0.05, compared with control. ‡p < 0.05, compared with siControl.

Chemical and Pharmaceutical Bulletin Advance Publication

Fig. 6. (a) Cellular association of Ad-HA/5-DTAF-M-β-CyD with HCT116 cells (CD44 (+)).

Cells were incubated with Ad-HA/5-DTAF-M-β-CyD in the absence and presence of HA for

2 h. (b, c) Effect of the endocytosis inhibitors on the cellular association of (b) Ad-HA/5-

DTAF-M-β-CyD and (c) 5-DTAF-M-β-CyD with HCT116 cells. Cells were pretreated with the endocytosis inhibitors for 1 h, and then incubated with Ad-HA/5-DTAF-M-β-CyD or 5-

DTAF-M-β-CyD for 2 h. The fluorescence intensity was measured by a flow cytometer.

Each value represents the mean ± S.E. of 3 experiments. *p < 0.05, compared with control.

†p < 0.05, compared with 5-DTAF-M-β-CyD. ‡p < 0.05, compared with Ad-HA/5-DTAF-M-

β-CyD.

Fig. 7. (a) Intracellular distribution of Ad-HA/5-DTAF-M-β-CyD in HCT116 cells (CD44

(+)) and (b) colocalization of Ad-HA/5-DTAF-M-β-CyD with rhodamine 123 in HCT116 cells (CD44 (+)). Cells were incubated with Ad-HA/5-DTAF-M-β-CyD in the absence and presence of HA for 2 h. After the incubation with Hoechst 33342 and rhodamine 123, the cells were observed with a fluorescence microscope. The experiments were performed independently three times, and the representative data were shown.

Fig. 8. (a) Effect of Ad-HA/M-β-CyD on the efflux of cholesterol from HCT116 cells (CD44

(+)). Cells were incubated with Ad-HA/M-β-CyD for 1 h. The concentration of cholesterol in the culture medium was determined by Cholesterol E test Wako. (b, c) Apoptotic activity of Ad-HA/M-β-CyD in HCT116 cells (CD44 (+)) determined by (b) Annexin V-FITC/PI assay and (c) cytochrome c assay. HCT116 cells were pretreated with HA for 1 h, and then incubated with Ad-HA/M-β-CyD for 2 h. After the trypsinization, the cells were incubated with Annexin V-FITC and PI for 30 min. In the case of cytochrome c assay, the cells were

Chemical and Pharmaceutical Bulletin Advance Publication

incubated with anti-cytochrome c-FITC, and the fluorescence intensity was measured by a flow cytometer. Each value represents the mean ± S.E. of 3 experiments. *p < 0.05, compared with control. †p < 0.05, compared with M-β-CyD. ‡p < 0.05, compared with Ad-

HA/M-β-CyD.

Fig. 9. Proposed scheme for antitumor activity of Ad-HA/M-β-CyD in HCT116 cells (CD44

(+)).

Chemical and Pharmaceutical Bulletin Advance Publication

Fig. 1

OH OH O

O O III HA O HO HO O OH NH O n HA

1-Adamantane EDC, HoBt

methyl amine in DMSO/H2O

Ad NH OH O III O O O HO HO O OH NH O n Ad-HA

M-β-CyD

(R = H or CH3) in H2O

Ad-HA/M-β-CyD

Chemical and Pharmaceutical Bulletin Advance Publication Fig. 2

DS of Ad : 19 per 1 HA

CH3 of HA

CH2 of Ad

2.0 1.8 1.6 1.4 1.2 ppm

Chemical and Pharmaceutical Bulletin Advance Publication Fig. 3

(a) 1H-NMR spectrum (b) Benesi-Hilderband plot

y = 0.8084x + 8.447 Concn. of M-β-CyD R² = 0.9928 15 5 mM Hz

3 mM δ -CyD ∆ β 10 -

1 mM 0 ∆

0.5 mM 1/ 5 Ad-HA/M- 0.1 mM 4 4 -1 KC = 1.04 × 10 ±0.01 ×10 M Ad-HA 0 mM 0 1.8 1.7 1.6 1.5 0510 ppm Concn. of M-β-CyD (mM)

Chemical and Pharmaceutical Bulletin Advance Publication Fig. 4

(a) (b) 120 § 100 100 80 80 60 *‡ * 60 40 40 * 20 *† Cell viability (%) Cell viability Cell viability (%) Cell viability 20

0 0 37°C 4°C Ad-HA/M-β-CyD

Chemical and Pharmaceutical Bulletin Advance Publication (a) HCT116 cells (CD44 (+)) Fig. 5 100 ●: M-β-CyD ■: Ad-HA/M-β-CyD 80

60

40

Cell viability (%) Cell viability 20 * 0 02.55 Concn. of M-β-CyD (mM)

(b) NIH3T3 cells (CD44 (-))

100 ●: M-β-CyD ■: Ad-HA/M-β-CyD 80

60

40

Cell viability (%) Cell viability 20

0 02.55 Concn. of M-β-CyD (mM)

(c) CD44 knockdown HCT116 cells

100

80

60 †‡

40 †

20 Cell viability (%) Cell viability 0

Ad-HA/M-β-CyD

Chemical and Pharmaceutical Bulletin Advance Publication Fig. 6 (a) Cellular association

500 *† 400 *‡ 300 * 200

intensity (%) intensity 100

Mean fluorescence 0

(b) Endocytosis inhibitors

100

80 ‡ 60 ‡ 40

intensity (%) intensity 20 Mean fluorescence 0

Ad-HA/5-DTAF-M-β-CyD

(c) Endocytosis inhibitors 140 120 100 80 60

intensity (%) intensity 40 20 Mean fluorescence 0

5-DTAF-M-β-CyD

Chemical and Pharmaceutical Bulletin Advance Publication Fig. 7

(a) DIC Hoechst 33342 5-DTAF Merge

Control

5-DTAF-M-β-CyD

Ad-HA/ 5-DTAF-M-β-CyD

Ad-HA/ 5-DTAF-M-β-CyD with HA

Chemical and Pharmaceutical Bulletin Advance Publication (b) Phase contrast Hoechst 33342 5-DTAF Rhodamine 123 Merge

Control

5-DTAF- M-β-CyD

Ad-HA/ 5-DTAF-M-β-CyD

Ad-HA/ 5-DTAF-M-β-CyD with HA

Chemical and Pharmaceutical Bulletin Advance Publication Fig. 8

(a) Cholesterol efflux (b) Annexin V-FITC/PI assay Control Ad-HA/M-β-CyD

1.5 * PI PI 1 *†

0.5 Annexin V-FITC Annexin V-FITC

Ad-HA/M-β-CyD Ad-HA with HA

Efflux of cholesterol (µg/mL) 0 PI PI (c) Cytochrome C assay 40000 *† Annexin V-FITC Annexin V-FITC 30000 * M-β-CyD 20000 †‡

10000 PI

Mean fluorescence intensity Mean fluorescence 0

Annexin V-FITC

Chemical and Pharmaceutical Bulletin Advance Publication Fig. 9

Chemical and Pharmaceutical Bulletin Advance Publication

Table 1. Physicochemical Properties of Ad-HA/M-β-CyD

Sample Mean diameter (nm) Polydispersity index ζ-Potential (mV) Ad-HA 153.3 ± 7.3 0.38 ± 0.05 -3.8 ± 1.3 Ad-HA/M-β-CyD 140.3 ± 10.1 0.25 ± 0.00 -4.2 ± 1.0

The particle size, polydispersity index and ζ-potential were measured by a Zetasizer Nano.

The concentrations of Ad-HA and M-β-CyD were 4.2 nM and 102 nM, respectively. Each

value represents the mean ± S.E. of 3-4 experiments.

Chemical and Pharmaceutical Bulletin Advance Publication