High Drug Encapsulation Capacity in Cyclodextrin-Based Nanoparticles: Characterization and In Vivo Antitumor Efficacy in Tumor-Bearing Mice

Abstract

Insufficient drug encapsulation, typically at about 10 wt %, remains a major challenge in the design of nanotechnology-based drug delivery systems. In this study, we created cyclodextrin-based nanoparticles (CDNP_PEGs), synthesized by the hyperbranched polymerization of cyclodextrins using a diepoxy poly­(ethylene glycol) linker, which exhibited a high drug encapsulation capacity of up to 54 wt % using the xanthone derivative α-mangostin (MGS) as a model compound. Evaluation in a tumor-bearing mouse model indicated that a single intravenous administration via the tail vein induced a significantly greater antitumor effect compared with free MGS. These results demonstrate the potential of CDNP_PEGs as promising high-capacity carriers for the delivery of anticancer drugs.

Introduction

The development of nanotechnology-based medicines has garnered interest because of their ability to precisely manipulate pharmaceutical properties. Although small-molecule anticancer drugs have excellent cell permeability and are highly effective at killing cells, there is a trade-off because they also interact with healthy cells by free diffusion and induce systemic toxicity due to their small-molecule nature. Thus, the development of pharmaceuticals using nanotechnology was proposed. Various liposome medicines have been developed, including Doxil (a nanoparticle composed of a lipid bilayer encapsulating doxorubicin), which was approved by the FDA in 2007 (Table S1). The drug encapsulation rate of most marketed liposomal drugs is less than 10 wt % (Table S1). Nanosystems that encapsulate larger amounts of drug will improve delivery at more effective concentrations in target cells with minimal doses required, which may reduce side effects. To improve drug encapsulation efficiency, various strategies have been examined, including the selection of lipid types, optimization of lipid ratios, preparation methods, − and the structural modification of lipids.

We developed cyclodextrin-based nanoparticles (CDNPs), which were polymerized by chemically cross-linking cyclodextrins with an epichlorohydrin linker (ECH; CDNP_ECH). They are currently being evaluated as a drug delivery systems (DDS) platform. − Cyclodextrins (CD) are ring-shaped compounds containing glucose, and there are α, β, and γ-types depending on the number of sugars. The size-dependent lumen of each CD encapsulates a variety of intractable compounds and exhibits host–guest effects, and is used as a drug delivery system based on CD polymerization − and a food additive. , Recently, various CD derivatives have been developed and are widely used in water purification, , energy transfer, , and as food additives. ,

Previously, we found that CDNP_ECH can encapsulate and solubilize various poorly water-soluble, small-molecule compounds [doxorubicin, curcumin I, dihydromyricetin, resveratrol, and α-mangostin (MGS)] and serve as a delivery carrier for small-molecule drugs. − MGS (Figure A, with a xanthone derivative skeleton), a natural compound extracted from the peel of the mangosteen fruit, exhibits anticancer activity and induces apoptosis; − however, its use is been limited because of its poor solubility in water. MGS-encapsulated CDNP_ECH (MGS/CDNP_ECH) was found to have potential for MGS delivery. − However, the CDNP_ECH has a small particle diameter of 10 nm, and it is difficult to control the particle size to a size favorable for tumor accumulation on enhanced permeability and retention (EPR) effects. The use of nanoparticles for delivery prolongs the circulation of small-molecule anticancer agents, resulting in enhanced accumulation in tumors, which is known as the enhanced EPR effect. − Tumor associated endothelial cells are highly permeable to nutrients, and nanoparticles with a size of 10–150 nm can accumulate in tumor tissue via transcytosis and permeation through the cavity wall. The control of nanoparticle size, surface charge, and shape has been evaluated to improve drug efficacy. Moreover, the development of lipid-based nanoparticles, , polymeric micelles, , and inorganic materials , have been examined to optimize drug accumulation.

1.

(A) Scheme of CDNP synthesis and inclusion of MGS. (B) Mechanism of action of drug-encapsulated CDNPs to reach tumors via EPR effect.

In this study, we made further improvements using chemical cross-linkers and found that CDNP with poly­(ethylene glycol) diglycidyl ether (PEG; CDNP_PEG) is a more hydrophilic cross-linker than ECH that that controls particle size (Figure A). Furthermore, MGS encapsulation surpassed that of previously reported CDNP_ECH and conventional liposome formulations. We characterized these CDNP_PEGs for drug encapsulation and evaluated their antitumor activity on tumor-bearing mice (Figure B).

Experimental Section

Materials

β-Cyclodextrin (CD), Poly­(ethylene glycol) diglycidyl ether (PEG, n = 8–9), Epichlorohydrin (ECH), Triton X-100, and α-Mangostin (MGS) were purchased from Tokyo Chemical Industry (Tokyo, Japan). d-(+)-Glucose was obtained from FUJIFILM Wako Pure Chemical Corporation (Tokyo, Japan), and dialysis membranes (<3.5 kDa) were purchased from Repligen (Massachusetts).

Synthesis

0.5 g of CD was dissolved in a 33 wt % aqueous NaOH solution, followed by the addition of 0.4 mL of 2.6 mM Triton X-100 and 1.5, 2.0, and 3.0 g for PEG_3, PEG_4, and PEG_6, respectably. The mixture was stirred at 400 rpm at room temperature (25 °C) for 18 h. After stirring, the product was purified to acetone precipitation and dialysis to obtain CDNP_PEG. The synthesis of CDNP_ECH was carried out following a previously reported method.

Characterization

The characterization of CDNP was conducted using dynamic light scattering (DLS), gel permeation chromatography coupled with multiangle light scattering and refractive index (GPC-MALS_RI), the phenol-sulfuric acid method, cryogenic transmission electron microscopy (Cryo-TEM), and small-angle X-ray scattering (SAXS).

DLS

The particle size of the sample solution was measured using a dynamic light scattering (DLS) instrument (Beckman-Coulter DelsaMax, California) at 25 °C with a scattering angle of 171°. The obtained autocorrelation function was analyzed using the CONTIN method to determine the hydrodynamic radius (R _h). The polydispersity index (PDI) was determined by cumulant analysis, and data are presented as the mean ± standard deviation (n = 3).

GPC-MALS_RI

Gel permeation chromatography (GPC)- multiangle light scattering (MALS) measurements were conducted using Shodex columns (SB802.5 and SB806, Shodex, Tokyo, Japan) at 40 °C, with 10 mM NaCl aqueous solution as the eluent. The CDNP solution (1 mg/mL) in 10 mM NaCl was filtered through a 0.22 μm PTFE membrane before injection into the column. The eluent from the column sequentially passed through a MALS detector (DAWN HELEOS II, wavelength: λ = 658 nm, Wyatt Technology, California) and a RI detector (RI range: 1.2–1.8 RIU, wavelength: λ = 488–690 nm, Wyatt Technology). The Rayleigh ratio (R _θ) was calculated using the following equation.

$$\frac{K_{\mathrm{c}}}{\mathrm{\Delta }R(\theta ,c)}=\frac{1}{M_{\mathrm{w}}}[1+\frac{R_{\mathrm{g},z}^2}{3}{q}^2+\mathrm{O}(q^4)]+2A_{2}c+\mathrm{O}(c^2)$$ 1

K is the optical constant, M _w is the weight-average molecular weight, and R _g,z is the z-average radius of gyration. The magnitude of the scattering vector q is determined using the following equation, where the scattering angle is 2θ.

$$q=\frac{4\pi }{\lambda }\sin \theta$$ 2

Phenol-Sulfuric Acid Method

The weight percentage of cyclodextrin (CD) (wt %) and number of CD (N _CD) in CDNP were determined using the phenol-sulfuric acid method by measuring the total glucose concentration in solution and converting it based on the seven glucose units per CD molecule.

For this analysis, 25 mg of dried CDNP was refluxed in 15 mL of 0.5 M sulfuric acid at 100 °C for 8 h, during which the CD components were hydrolyzed and decomposed into furfural derivatives. The resulting solution was then reacted with 5 wt % phenol and concentrated sulfuric acid. The final reaction product exhibited an orange color, and its absorbance was measured at a wavelength of 488 nm using a UV–vis spectrophotometer (Jasco, V-630, Tokyo, Japan). The glucose content in CDNP was quantified using a calibration curve prepared with d-(+)-glucose.

Subsequently, the glucose concentration was converted to the CD content, and the number of CD molecules per CDNP particle (N _CD)­was calculated. Based on the obtained NCD and the molecular weight (M _w) of CDNP, the weight percentage (wt %) of CD in CDNP was determined.

$$\mathrm{CD}(\mathrm{mg}/\mathrm{ml})=\frac{\mathrm{Abs}}{\mathrm{the}\mathrm{slope}\mathrm{of}\mathrm{the}\mathrm{calibration}\mathrm{curve}}$$ 3

$$N_{\mathrm{C}\mathrm{D}}=\frac{\mathrm{C}\mathrm{D}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{c}\mathrm{.}(\mathrm{M})}{\mathrm{r}\mathrm{e}\mathrm{f}\mathrm{l}\mathrm{u}\mathrm{x}\mathrm{e}\mathrm{d}\mathrm{C}\mathrm{D}\mathrm{N}\mathrm{P}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{c}\mathrm{.}(\mathrm{M})}$$ 4

$$\mathrm{CD}(\mathrm{wt}\%)=\frac{N_{\mathrm{CD}}\times M_{\mathrm{w}}\mathrm{CD}}{M_{\mathrm{w}}\mathrm{CDNP}}$$ 5

Cryo-TEM

Cryogenic transmission electron microscopy (Cryo-TEM) was performed using a JEM-2100Plus electron microscope (JEOL, Tokyo, Japan) operated at an accelerating voltage of 200 kV. A 3 μL aliquot of CDNP or MGS/CDNP (each at a CDNP concentration of 10 mg/mL) was deposited onto a carbon grid (Quantifoil, Jena, Germany) and rapidly vitrified in liquid ethane using a Leica EM GP2 plunge freezer (Leica Microsystems, Wetzlar, Germany). The prepared samples were transferred to the electron microscope using a Gatan cryo-transfer stage (Gatan Inc., California).

SAXS

Small-angle X-ray scattering (SAXS) measurements were conducted at the BL-40B2 beamline of the SPring-8 synchrotron facility in Hyogo, Japan. A digital detector (Pilatus-3S 2M) was positioned 4 m from the sample, and the incident beam wavelength (λ) was set to 0.1 nm. The scattering vector q was defined as $q=\frac{4\pi }{\lambda }\sin \theta$ , where 2θ represents the scattering angle. The CDNP concentration was 10 mg/mL, and measurements were performed with an exposure time of 180 s. The obtained data were analyzed using a spherical model. This model is described by the following equation

$$I(q)=I_{\mathrm{a}\mathrm{v}}(0)P_{\theta }(q,R_{\mathrm{a}\mathrm{v}},\sigma )+I_{\mathrm{O}\mathrm{Z}}(q,\xi )$$ 6

$$P_{\mathrm{\Theta }}(q,R_{\mathrm{a}\mathrm{v}},\sigma )=\frac{{\int }_{-\infty }^{\infty }g(R_{\mathrm{a}\mathrm{v}},\sigma ){[\mathrm{\Delta }\rho V]}^2{\mathit{\Phi }}^2(q,R+r)\mathrm{d}r}{{\int }_{-\infty }^{\infty }g(R_{\mathrm{a}\mathrm{v}},\sigma ){[\mathrm{\Delta }\rho V]}^2\mathrm{d}r}$$ 7

$$V=\frac{4}{3}\pi r^3$$ 8

R _av represents the average radius, and ΔρV denotes the total excess scattering length of the particles. The spherical particle form factor Φ(q, r) and the Gaussian distribution function g(R _av, σ), which describes the two particle size distributions, are given as follows

$$\mathit{\Phi }(q,r)=3\frac{\sin (\mathit{qr})-\mathit{qr}\cos (\mathit{qr})}{{(\mathit{qr})}^3}$$ 9

$$g(R_{\mathrm{av}},\sigma )=\frac{f_1}{\sqrt{2\pi {\sigma }_1^2}}\exp (-\frac{{(R_{\mathrm{av},1}-r)}^2}{2{\sigma }_1^2})+\frac{f_2}{\sqrt{2\pi {\sigma }_2^2}}\exp (-\frac{{(R_{\mathrm{av},2}-r)}^2}{2{\sigma }_1^2})$$ 10

In this context, f represents the volume fraction of each particle population, satisfying the condition f ₁ + f ₂ = 1.

Additionally, the Ornstein–Zernike equation, which describes the density fluctuations within the particles, is given by the following equation:

$$I_{\mathrm{OZ}}(q,\xi )=\frac{I_{\mathrm{OZ}}(0)}{1+q^2{\xi }^2}$$ 11

Drug Encapsulation and Drug Release

Drug Encapsulation

A schematic representation of the drug- encapsulation procedure is shown in Figure A. A 100 mM MGS solution dissolved in DMSO was mixed with a 10 mg/mL CDNP solution. The mixture was incubated at room temperature for 2 h and then centrifuged twice at 12,000 rpm for 5 min to remove free MGS, after which the supernatant was collected. The collected solution was then dialyzed using a 3.5 kDa dialysis membrane to remove DMSO. After dialysis, the solution was filtered through a 0.22 μm membrane filter and used for further experiments. The concentration of MGS in the complex was determined by measuring the UV absorbance at 323 nm using a UV–vis spectrophotometer (Jasco, V-630). The drug-encapsulating ratio was calculated as the ratio of the MGS concentration determined by UV measurement (C _MGS) to the CDNP concentration (C _CDNP) by weight.

$$\mathrm{drug}\mathrm{encapsulation}\mathrm{ratio}(\mathrm{wt}\%)=\frac{C_{\mathrm{MGS}}}{{C_{\mathrm{MGS}}+C}_{\mathrm{CDNP}}}\times 100$$ 12

2.

MGS Encapsulation in CDNP_PEGs. (A) The procedure for MGS encapsulation in CDNP­(MGS/CDNP) and analysis MGS encapsulation ratio in CDNP. (B) Concentration of MGS encapsulated in CDNP_PEG measured by UV spectroscopy. Data are presented as mean ± SD from three independent experiments (n = 3). (C) MGS encapsulation ratio in CDNP_PEGs. Data are presented as mean ± SD from three independent experiments (n = 3). (D) Number ratio of N _MGS/N _CD in a nanoparticle. Data are presented from a single experiment (n = 1).

Drug Release

A schematic representation of the drug release procedure is shown in Figure A. MGS-encapsulated CDNPs (MGS/CDNP) were prepared by incubating CDNPs (10 mg/mL) with MGS (10 mM) for 2 h, followed by centrifugation to remove free MGS. MGS/CDNP was enclosed within a dialysis membrane and dialyzed in a 150 mM NaCl solution at 37 °C. During the release experiment, 20 μL of the sample was collected at predetermined time intervals (0, 1, 2, 4, 6, 18, 24, and 48 h) and replaced with an equal volume of fresh NaCl solution.

4.

(A) Procedure for the drug release experiment. Comparison of drug release (B) concentration (mM) and (C) cumulative release (%) from CDNPs. Results represent the mean ± SD (n = 3). Differences between MGS/CDNP_PEG_6 and MGS/CDNP_ECH_4 or MGS/CDNP_ECH_3 were determined with an unpaired two-tailed Student’s t test. *p < 0.05 and **p < 0.01. (D) SAXS measurement of structural changes accompanying drug release after maximum MGS encapsulation. Data are presented from a single experiment (n = 1). (E) The mechanism of structural changes anticipated with drug release.

The concentration of MGS retained in CDNP was quantified by measuring the UV absorbance at 323 nm using a UV–vis spectrophotometer. The release profile was plotted as a semilogarithmic graph against time. The plotted data were fitted using the following equation ,

$$C(t)=C_1\exp \left(\frac{t}{{\tau }_1}\right)+C_2\exp \left(\frac{t}{{\tau }_2}\right)$$ 13

Here, τ₁ and τ₂ represent the characteristic release times of the drug, while C ₁ and C ₂ denote the weight fractions of the drug in each release phase.

Antitumor Effect

The experiments were performed in accordance with the institutional guidelines for the care and use of laboratory animals at Hiroshima University and were approved by the university’s animal ethics committee.

The antitumor efficacy of the MGS/CDNP complex was evaluated using a mouse model bearing CT26 tumors. Initially, 7-week-old BALB/c mice (average body weight: approximately 22 g) were subcutaneously inoculated with CT26WT cells (1.0 × 10⁶ cells/mouse, 100 μL), which were mixed with Matrigel at a 1:1 ratio. By day 4 postinoculation, the tumor volume reached approximately 100 mm³. Following this confirmation, the mice were randomly assigned into groups of six (n = 6), and each group received an intravenous injection (0.15 mL) of the designated sample solution at a dose equivalent to 10 mg/kg of MGS in CDNP.

MGS was prepared by dissolving it in a solution of 0.4% DMSO, 2% EtOH, and 2% Tween 80 in 150 mM NaCl, and administered at a dose of 10 mg/kg. MGS/CDNP solution was intravenously injected at a dose equivalent to 10 mg/kg of MGS (approximately 0.22 mg per mouse). Throughout the experimental period, tumor volume was continuously monitored. The V of tumor volume (mm³) was calculated using the following equation

$$V=\frac{L\times W^2}{2}$$ 14

Here, L (mm) represents the tumor’s longest diameter, while W (mm) denotes the shortest diameter.

Statistical analysis was performed using a two-tailed Student’s t test. Data are presented as mean ± standard deviation (s.d.), with error bars representing standard deviation (n = 6). Statistical significance was defined as p-value <0.05, with significance levels indicated as follows: *p < 0.05, **p < 0.01, and ***p < 0.001.

Results and Discussion

Synthesis and Characterization of CDNP_PEGs

Three types of cyclodextrin nanoparticles (CDNP_PEG) were synthesized at weight ratios of 3, 4, and 6 using poly­(ethylene glycol) diglycidyl ether (PEG) as a cross-linker of CD (Figure A, and Table ). Table shows the sample codes and weight ratios. For the CDNP_X_Y annotation, X indicates the cross-linker, and Y is the value of the weight ratio.

1. Sample Code and Particle Properties of CDNPs.

sample code	feeding weight ratio (PEG:CD)	D h (nm)	PDI	elution time on GPC (min)	M w (105 g/mol)	N CD	CD (wt %)	density ratio (V CD × N CD/V CDNP)
CDNP_PEG_3	3.0:1	15.7 ± 0.9	0.57 ± 0.11	18.0	0.39	7.8	23.1	0.52
CDNP_PEG_4	4.6:1	47.1 ± 2.4	0.21 ± 0.10	16.5	12	230	21.2	0.7
CDNP_PEG_6	6.2:1	132.2 ± 8.6	0.83 ± 0.21	15.0	32	1400	49.1	0.22
CDNP_ECH_4	4.3:1	10.8	N.A.	19.5	1.4	64.5	50.9	14.0

^aStatistical analysis and D _h and PDI values for CDNP_ECH were not available due to experimental limitations.

The hydrodynamic diameter (D _h) of CDNP_PEG was analyzed by DLS and revealed sizes of 16.0, 44.8, and 120.0 nm for CDNP_PEG_3, CDNP_PEG_4, and CDNP_PEG_6, respectively, indicating the formation of nanostructures through polymerization (Table ). At the PEG weight ratios above 7, gelation occurred and no nanoparticles were obtained. Each CDNP was characterized using size exclusion chromatography for differential refractive index and light scattering (Figure S1 and Table ). The order of the longest elution time was CDNP_PEG_3, CDNP_PEG_4, and CDNP_PEG_6, respectively, which is in order of increasing particle size (Table ). The molecular weights were determined using Berry plots, showing values of 0.39 × 10⁵, 12 × 10⁵, and 32 × 10⁵ g/mol for CDNP_PEG_3, CDNP_PEG_4, and CDNP_PEG_6, respectively (Table ). Particle size and molecular weight increased proportionally with the weight ratio of PEG.

The number of CDs in a single nanoparticle (N _CD) was evaluated by quantifying the total glucose content using the phenol-sulfuric acid method and converting it based on the seven glucose units in β-CD (Table and Figure A). The calculated N _CD values were 7.8, 230, and 1400 for CDNP_PEG_3, CDNP_PEG_4, and CDNP_PEG_6, respectively (Table ). The weight percentage of the CDs (CD wt %) for a single nanoparticle was calculated by dividing the CD weight by the total weight of the CDNP. The CD wt % values were 23.1, 21.2, and 49.1 wt % for CDNP_PEG_3, CDNP_PEG_4, and CDNP_PEG_6, respectively (Table ). The wt % for N _CD and CD increased with the weight ratio of PEG.

Solubilization and Encapsulation Capacity of Drugs in CDNP

The solubilization and encapsulation capacity of MGS were evaluated for the three types of CDNP_PEGs. The MGS was injected into the dispersion of CDNP_PEG, and the MGS complexed with CDNP_PEG was purified by dialysis. The amount of trapped MGS in each CDNP_PEG was quantified by UV–vis measurements (Figure A). The results indicated that the concentrations of complexed MGS in the solution were 5.8, 10.2, and 6.9 mM for CDNP_PEG_3, CDNP_PEG_4, and CDNP_PEG_6, respectively. The CDNP_PEGs significantly improved the solubility of MGS in water (Figure B), whereas free MGS is poorly soluble in water (2.03 × 10^–4 mg/L, 4.95 × 10^–7 mM at 25 °C).

The MGS encapsulation ratios for CDNP_PEG were 29.7, 54.1, and 35.1 wt % for CDNP_PEG_3, CDNP_PEG_4, and CDNP_PEG_6, respectively, indicating successful internalization of MGS into CDNP_PEG (Figure C). In this study, we refer to MGS-encapsulated CDNPs as MGS/CDNPs. Compared with the previously reported MGS encapsulation rate for CDNP_ECH, which was up to 10.7 wt % (Figure C), the encapsulation rate for CDNP_PEG was nearly 5-fold higher (Figure C). The number ratio of CD to MGS in MGS/CDNP_PEGs, was calculated based on the encapsulation efficiency and N _CD. The number ratios of MGS per CD were 3.6, 7.1, and 2 for CDNP_PEG_3, CDNP_PEG_4, and CDNP_PEG_6, respectively (Figure D). Other drugs, including doxorubicin, donepezil, THSG, dihydromyricetin, galantamine, tacrine, resveratrol, and curcumin, were also evaluated and exhibited successful encapsulation in CDNP_PEG (Figure S2 and Table S2). Based on these results, systems using CDNP_PEG are powerful for preparing dispersions of hydrophobic compounds in aqueous media.

Characterization of MGS/CDNP_PEGs

The particle morphology of MGS/CDNP_PEG_4 was observed using cryo-TEM (Figure A,B). The results indicated that MGS/CDNP_PEG_4 exhibited a spherical morphology without anisotropy, with the observed nanoparticles measuring approximately 20 nm in size. Further structural analysis was done using small-angle X-ray scattering (SAXS) to evaluate the structure before and after MGS encapsulation (Figure C and Table S3). The results indicated that CDNP_PEGs exhibited a Gaussian chain structure. After MGS encapsulation in CDNP_PEGs, the structure was transformed into a spherical configuration. In contrast, a previous study of CDNP_ECH proposed a model in which the Gaussian chain structure was maintained before and after MGS encapsulation, with one MGS coordinated to a single CD (Figure D). The structural transformation of MGS/CDNP_PEG_4 was further confirmed through small-angle neutron scattering (SANS) analysis. The contrast between deuterated and light hydrogen regions can be detected in SANS, because the scattering length densities of hydrogen and deuterium are very different. In D₂O, CDNP_PEG_4 adopted a Gaussian chain structure, whereas MGS/CDNP_PEG_4 exhibited a spherical morphology (Figure S3A). Furthermore, SANS analysis using deuterated MGS (D-MGS) in H₂O was conducted to evaluate the distribution of MGS in the nanoparticles. The results indicated that D-MGS was spherically distributed within the particles (Figure S3B). This distribution does not indicate that MGS is distributed in the core, as it may not be fitted into a two-layer sphere, such as a core–shell. The results of the analysis in the one-layer sphere indicate that MGS was distributed throughout the nanoparticles. To examine the detailed structural changes, CDNP_PEG_4 was encapsulated with 7.1–100 mM concentrations of MGS. The MGS concentration-dependent structure was analyzed by SAXS (Figure D). The results indicated that spherical structures were induced when the MGS concentration was over 14.3 mM. The MGS encapsulation rate for CDNP_PEG_4 was 3.4–54 wt % (Figure E and Table ). The structural change was induced over 6.9 wt %, and the range of MGS per CD was 0.9–2.5 (Figure F). These results suggest that the structural transformation of CDNP_PEG is induced by the presence of multiple MGS molecules per CD, forming micelle-like subparts in the nanoparticles through hydrophobic interactions. Based on these results, the high encapsulation behavior of hydrophobic molecules in CDNP_PEG may be associated with the structural change from a Gaussian chain to a spherical structure. The semiflexible PEG linker (Persistence length: 0.16 nm) may provide greater flexibility within the nanoparticle and facilitate the formation of hydration layers compared with previously used epichlorohydrin linkers (Persistence length: 0.52 nm). , The density ratio of CDs in a single nanoparticle (V _CD × N _CD/V _CDNP) is 0.52, 0.7, and 0.22% for CDNP_PEG_3, CDNP_PEG_4, and CDNP_PEG_6, respectively (Table ). The density of CD within a single nanoparticle and the ratio of MGS to CD are in a proportional relationship (Figure G). For drug encapsulation by CDNPs, these results suggest that the density of CDs and the linker flexibility required to form stable hydrophobic domains, in which multiple MGS molecules interact with multiple CDs, induce structural changes once the maximum binding capacity of MGS to CDs is reached (Figure H). The hydrophilicity of the linker may play a role; however, further studies, including the coordination of water molecules, will be needed in the future.

3.

(A) Cryo-TEM images of MGS encapsulated in CDNP_PEG_4. The scale bar is 200 nm. (B) The scale bar is 50 nm. SAXS measurement of CDNP with Encapsulated MGS. (C) Comparison of SAXS profiles with and without MGS encapsulation in CDNP_PEGs and CDNP_ECH with MGS encapsulation. (D) SAXS results with different feeding concentration of MGS. Change in drug encapsulation ratio in (E) and [MGS]/[CD] in (F) as a function of MGS concentration. (G) Relationship between CD density ratio and [MGS]/[CD] in particles. (H) The mechanism of structural changes anticipated with drug encapsulation. Data are presented from a single experiment (n = 1).

2. MGS-Conc. Dependent Encapsulation Properties of CDNP_PEG_4.

feeding MGS conc. (mM)	MGS encapsulation ratio in CDNP (wt %)	number ratio of N MGS:N CD	dissolved MGS conc. (mM)
100	54	7:1	10.2
71.4	36	4.7:1	8.8
35.7	19	2.5:1	4.6
14.3	6.9	0.9:1	1.7
7.1	3.4	0.4:1	0.8

Release Behavior of MGS from MGS/CDNP

The release behavior of MGS from MGS/CDNPs was examined (Figure A). The release rate half-life (t ₅₀) and half-life concentration (C ₅₀) were evaluated based on the amount of MGS remaining over time (Figure B,C and Table ). The t ₅₀ values were 57.3, 64.1, and 51.3 h, whereas the C ₅₀ values were 0.51, 0.53, and 0.59 mM for CDNP_PEG_3, CDNP_PEG_4, and CDNP_PEG_6, respectively (Table ). The elimination rates were C ₁ and C ₂ at t ₁ and t ₂. The initial phase revealed a period of rapid release, in which 23.4% of the encapsulated drug was released, followed by a slower release phase, in which 76.0% of the drug was released from CDNP_PEG_4 over 141 h. The initial rapid release was the result of free diffusion caused by the osmotic pressure gradient between MGS/CDNP_PEG_4 and the solvent, whereas the slower release phase was associated with the gradual release of MGS molecules stabilized within the CDNP_PEG_4 over time (Figure B,C).

3. Encapsulation Efficiency and Changes in MGS Concentration in Particles during Drug Release.

release time (h)	encapsulation ratio (wt %)	number ratio of N MGS:N CD	encapsulated MGS conc. (mM)
0	55	7.3	13.6
2	34	4.5	8.3
4	34	4.4	8.3
6	27	3.5	6.6
12	18	2.4	4.4

4. Characterization of Drug Release from CDNP_PEGs.

sample code	t 50(h)	C 50 (mM)	C 1 (%)	τ1 (h)	C 2 (%)	τ2 (h)
CDNP_PEG_3	57.3	0.51	34.5	3.3	63.9	219.1
CDNP_PEG_4	64.1	0.53	28.3	1.8	71.3	181.0
CDNP_PEG_6	51.3	0.59	23.4	2.8	76.0	122.5
CDNP_ECH_4	10.7	0.07	51.9	8.3	43.8	52.8

A SAXS analysis of the release behavior of MGS/CDNP_PEG_4 indicated that the structure transitioned to a Gaussian chain after 12 h (Figure D). This likely occurred because of a reduction of MGS within the particles during the release process, resulting from their inability to maintain a core spherical structure (Figure E). Furthermore, calculations for the MGS encapsulation efficiency at each time point revealed that structural changes occurred within the range of 18–27 wt % (Table ). A comparison of the MGS encapsulation efficiency required for spherical structure formation suggested that the structural transition of CDNP_PEG_4 occurred at an encapsulation threshold of approximately 18 wt %.

Antitumor Effect of MGS/CDNP in a Tumor-Bearing Mouse Model

The antitumor effect of MGS/CDNP_PEG_4 was determined in a mouse model implanted with CT26 cells (mouse colorectal cancer cells) (Figure A). The study included the following five groups: Control, CDNP_PEG_4, MGS, MGS/CDNP_ECH_4, and MGS/CDNP_PEG_4. Each sample was administered 4 days after tumor implantation, and tumor volume changes were monitored over time. The results indicated that the MGS/CDNP_PEG_4-treated group exhibited the most significant tumor growth suppression, confirming its potent antitumor effect (Figures B and S4A). From day 8 onward, the induction of a superior antitumor effect compared with MGS/CDNP_ECH_4 was observed. On day 23, the tumor volume ratio upon treatment, normalized to the initial volume at the start of administration, was 0.05, indicating a marked therapeutic effect on tumor growth (Figure C). No weight loss or decrease in the number of surviving mice was observed, suggesting that this nanotechnology is safe (Figures D and S4B). These results suggest that the observed effects are attributable to MGS delivery, given that CDNP alone did not inhibit cell proliferation in vitro (Figure S4C).

5.

(A) Overview of the schedule for antitumor effect experiments. (B) Changes in tumor volume after injection CT26WT cells to mice. Results represent the mean ± SD (n = 6). Differences between MGS/CDNP_PEG_4 and MGS/CDNP_ECH_4 were determined with an unpaired two-tailed Student’s t test. *p < 0.05 and **p < 0.01. (C) Tumor volume ratio at day 23, normalized to the initial volume at the start of administration. (D) Changes in mouse body weight after sample injection.

Conclusion

We synthesized and carried out a structural analysis of CD nanoparticles to enhance the use of poorly water-soluble drugs. CDNPs along with a PEG linker enhanced drug encapsulation efficiency. Structural changes of the nanoparticles were induced upon drug encapsulation. Moreover, CDNP_PEG, which showed a high encapsulation rate, induced antitumor effects with a single intravenous administration. This DDS carrier system will be useful for future applications requiring nanotechnology.

Glossary

Abbreviations

ANSTO

Australian Nuclear Science and Technology Organization

TOC

table of contents

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsbiomedchemau.5c00220.

• Additional experimental details, materials, and methods; supplementary physicochemical and structural characterization of cyclodextrin nanoparticles and drug-loaded nanoparticles, including DLS, GPC-MALS-RI, Cryo-TEM, and SAXS; supporting figures (Figures S1–S4) and supporting tables (Tables S1–S3) (PDF)

All authors have given approval to the final version of the manuscript. CRediT: Tomoki Kosugi conceptualization, data curation, writing - original draft; Reina Kobayashi data curation.

This work was supported by the JSPS the grant-in-aid for early career scientists (24K21111), JSPS grant-in-aid for scientific research A (23H00305), MEXT Promotion of Distinctive Joint Research Center Program (JPMXP 0621467946), JST A-STEP (JPMJTR23UC), the canon foundation (S23–075), the sumitomo foundation (2300369).

The authors declare no competing financial interest.