Transarterial Infusion of iRGD-Modified ZrO2 Nanoparticles with Lipiodol Improves the Tissue Distribution of Doxorubicin and Its Antitumor Efficacy

LABORATORY INVESTIGATION

Transarterial Infusion of iRGD-modified ZrO2 Nanoparticles with Lipiodol Improves the Tissue Distribution of Doxorubicin and Its Antitumor Efficacy Yang Xie, MD, Xun Qi, MD, PhD, Ke Xu, MD, PhD, Xianwei Meng, PhD, Xiaowei Chen, MD, Fan Wang, MD, and Hongshan Zhong, MD, PhD

ABSTRACT Purpose: To evaluate the effect of transarterial infusion of iRGD-modified and doxorubicin-loaded zirconia-composite nanoparticles (R-DZCNs) with lipiodol in the improvement of the distribution of doxorubicin (DOX) in liver tumors and its antitumor efficacy. Materials and Methods: The effect of R-DZCNs was evaluated in vitro by tumor cellular uptake and cytotoxicity assays. For the in vivo study, DOX distribution and antitumor efficiency were assessed. In the DOX distribution study, VX2 tumor-bearing rabbits received transarterial infusion of lipiodol with DOX, doxorubicin-loaded zirconia-composite nanoparticles (DZCNs), or R-DZCNs, respectively. DOX distribution was assessed by immunofluorescence. In the antitumor study, tumor-bearing rabbits received transarterial infusions of lipiodol with DOX, DZCNs, R-DZCNs, or saline respectively. Tumor volume was measured using magnetic resonance imaging, and the expression of apoptosis-related factors (caspase-3, Bax, Bcl-2) was analyzed by immunohistochemistry and Western blotting. Results: R-DZCNs increased cellular uptake and caused stronger cytotoxicity. Compared with the DOX þ lipiodol or DZCNs þ lipiodol group, the R-DZCNs þ lipiodol group showed more DOX fluorescence spots (2,449.15 ± 444.14 vs. 3,464.73 ± 632.75 or 5,062.25 ± 585.62, respectively; P < .001) and longer penetration distance (117.58 ± 19.36 vs 52.64 ± 8.53 or 83.37 ± 13.76 μm, respectively; P < .001). In the antitumor study, the R-DZCNs þ lipiodol group showed smaller tumor volumes than the DOX þ lipiodol or DZCNs þ lipiodol group (1,223.87 ± 223.58 vs. 3,695.26 ± 666.25 or 2281.06 ± 457.21 mm3, respectively; P ¼ .005).The greatest extent of tumor cell apoptosis was observed in R-DZCNs þ lipiodol group immunohistochemistry and Western blotting results. Conclusions: Transarterial infusion of R-DZCNs with lipiodol improved the distribution of DOX and enhanced its antitumor efficacy.

ABBREVIATIONS DOX ¼ doxorubicin, DZCNs ¼ doxorubicin-loaded zirconia composite nanoparticles, HCC ¼ hepatocellular carcinoma, IHC ¼ immunohistochemistry, R-DZCNs ¼ iRGD peptide-modified and doxorubicin-loaded zirconia composite nanoparticles

Transarterial chemoembolization is recommended as the first-line treatment option for intermediate stage hepatocellular carcinoma (HCC). However, several previous clinical studies demonstrated the inadequate antitumor efficacy of chemotherapeutic treatment delivered by transarterial chemoembolization (1–3). The restricted distribution of a

chemotherapeutic drug within the tumor may account for the drug’s limited antitumor efficacy in transarterial chemoembolization (4,5). Previously, the transarterial infusion of drug-loaded nanoparticles with lipiodol in liver tumors showed effective treatment (6–9). To further improve the antitumor

From the Department of Radiology (Y.X., X.Q., K.X., X.C., F.W., H.Z.) and Key Laboratory of Diagnostic Imaging and Interventional Radiology of Liaoning Province (X.Q., H.Z.), First Affiliated Hospital of China Medical University Shenyang, 11,0001, Liaoning, China; and Laboratory of Controllable Preparation and Application of Nanomaterials (X.M.), Laboratory of Cryogenics, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing, China. Received November 8, 2018; final revision received April 2, 2019; accepted April 12, 2019. Address correspondence to H.Z.; E-mail: [email protected]

Appendix A and Figures E1–E2 can be found by accessing the online version of this article on www.jvir.org and clicking on the Supplemental Material tab.

None of the authors have identified a conflict of interest.

© SIR, 2019 J Vasc Interv Radiol 2019; ▪:1–10 https://doi.org/10.1016/j.jvir.2019.04.014

2 ▪ iRGD-modified ZrO2 Nanoparticles Improves Distribution of Doxorubicin

Xie et al ▪ JVIR

Figure 1. Overview of experimental design. IF ¼ immunofluorescence.

efficacy, elevating the active targeting effect of nanoparticles by surface modification may be a feasible strategy. The iRGD peptide contains a disulfide-linked arginine-glycineaspartic (RGD) tripeptide sequence and a C-terminal endbinding sequence (CendR) (10). The RGD tripeptide sequence can selectively recognize and bind to integrin avb3, which is overexpressed in the endothelium of tumor vessels or tumor cells (11). Sequence CendR can bind to Nrp-1, which is overexpressed in the tumor vessels, promoting vascular and tissue permeability (12). Pioneering studies have indicated that iRGD can improve the uptake of nanoparticles and their therapeutic drugs in liver tumors (13,14). Zirconia composite (ZrO2) nanoparticles, which can be easily synthesized, have been applied in the studies of therapy in liver tumors and have exhibited good biocompatibility (15). The iRGD-modified ZrO2 nanoparticle solution has been demonstrated to be a promising drug delivery system in the treatment of liver tumor (16). Based on these previous studies, the doxorubicin (DOX)loaded and iRGD-modified ZrO2 nanoparticles (R-DZCNs) were prepared. It was hypothesized that a transarterial infusion of R-DZCNs with lipiodol could improve the distribution of DOX in liver tumors and exhibit better antitumor efficacy than transarterial chemoembolization. The targeting effect of iRGD-modified ZrO2 nanoparticles was evaluated in vitro by using HepG2 cells (Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Academy of Sciences, Shanghai, China). Additional in vivo animal experiments were conducted to evaluate the antitumor effect of transarterial infusion of R-DZCNs with lipiodol in rabbit VX2 liver tumor models.

MATERIALS AND METHODS In Vitro Cell Cytotoxicity Assay and Cellular Uptake Assay The synthesis of R-DZCNs is described in Appendix A (available online on the article’s Supplemental Material page at www.jvir.org). Figure 1 (available online on the article’s Supplemental Material page at www.jvir.org) presents an overview of the experimental design. HepG2 cells were cultured in Dulbecco’s modified Eagle Medium with 10% fetal bovine serum at 37 C and 5% CO2. Cells were seeded in 96-well plates at a density of 3,000 cells/well. After adherence, cells were incubated with DOX-loaded zirconia composite nanoparticles (DZCNs) and R-DZCNs, each containing DOX at various concentrations (10, 5, 2.5, 1.25, 0.625, and 0.3125 μg/mL) for 24 hours. In the control group, normal culture medium was used as a replacement. Each well was treated with 10 μL of CCK8 reagent (Solarbio, Beijing, China) for 2 hours at 37 C. The absorbance was then measured at 450 nm, using a microplate reader (Multiskan FC, Thermo Fisher Scientific, Waltham, Massachusetts). The viability of the tumor cells was calculated as the ratio associated with the absorbency values of DZCNs or R-DZCNs group relative to that of the control group. Cellular uptake assay. HepG2 cells were seeded in 24well plates at a density of 1.2  104 cells/well and incubated overnight. The culture medium was then replaced with culture medium containing either DZCNs or R-DZCNs with 2 μg/mL of DOX and incubated for 4 hours (17). Then, cell nuclei were dyed with 40 ,6-diamidino-2-phenylindole (DAPI)

Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019

3

Figure 2. (a) Hepatic arteriography was performed to observe the location of and blood supply to the tumor before infusion (white arrow). (b) Radiography was used to evaluate the deposition of lipiodol after infusion (red arrow).

for 5 minutes. Excitation levels of red fluorescence (DOX) and blue fluorescence (cell nuclei) were observed by confocal laser scanning microscopy (model FV1,000-IX81, Olympus, Tokyo, Japan).

Animal Subjects Animal studies were performed in strict accordance with the Guidelines for the Care and Use of Laboratory Animals (eighth edition, publication no: 978-0-309-15400-0). The use of adult male New Zealand rabbits was approved by the Institutional Animal Care and Use Committee of the authors’ institution.

Preparation of VX2 Liver Tumor Model All rabbits used in the experiments weighed between 2.5 and 3.0 kg, with a median weight of 2.8 kg. Rabbits were anesthetized using intramuscular injection of pentobarbital sodium (3 mg/kg). Then, an 18-gauge puncture needle was used to percutaneously puncture the left or right lobe of the liver under ultrasonographic guidance. A piece of VX2 tumor tissue (approximately 1 mm in size) was implanted in the liver. Two weeks after implantation (18), tumor growth in the liver was confirmed by magnetic resonance (MR) imaging (General Electric, Boston, Massachusetts).

Transarterial Infusion of DOX, DZCNs, or R-DZCNs with Lipiodol The transarterial infusion of R-DZCNs combined with lipiodol (Jiangsu Hengrui Medicine, Lianyungang, China) is outlined in Figure E1 (available online on the article’s Supplemental Material page at www.jvir.org). Tumorbearing rabbits were anesthetized as described previously. The right femoral artery was exposed by open surgery and punctured with an 18-gauge needle, followed by insertion of a 5F vascular sheath (Terumo, Tokyo, Japan). Under the guidance of digital subtraction angiography (Artis zee

celling-mounted system; Siemens, Erlangen, Germany), a 2.2F microcatheter (Terumo, Tokyo, Japan) was selectively advanced into the proper hepatic artery, and the catheter tip was positioned at the terminal aspect of the proper hepatic artery. Hepatic arteriography was performed to observe the location and blood supply of the tumor before infusion (Fig 2a). Subsequently, the emulsion of lipiodol and R-DZCNs was slowly infused, and the deposition of lipiodol was observed by radiography (Fig 2b).

In Vivo DOX Distribution Study Twenty-four rabbits with liver tumors were divided into 3 groups of 8 rabbits (5). The equivalent DOX dose for each group was 3 mg/kg. All rabbits were infused with 0.25 mL of lipiodol. Subsequently, DOX (1 mL), DZCNs (1 mL), or R-DZCNs (1 mL) was infused in the DOX þ lipiodol group, the DZCNs þ lipiodol group, or the R-DZCNs þ lipiodol group, respectively. Four rabbits in each group were humanely sacrificed by the intravenous administration of sodium pentobarbital (10 mg/kg of body weight) 10 minutes after the infusion, and the other 4 rabbits were sacrificed at 4 hours after infusion (5). The tumors were harvested and embedded in tissue-freezing medium at 80 C and then cut into 5-μm sections. The sections were fixed with paraformaldehyde and blocked in 5% bovine serum albumin (BSA) for 30 minutes. The sections were incubated with CD31 antibody (1:20 dilution; Dako; Palo Alto, California) overnight at 4 C. Sections were incubated with fluorescein isothiocyanate-conjugated secondary antibody for 2 hours, and nuclei were stained with DAPI. Finally, DOX fluorescence and microvessels in the tumors were observed using confocal laser scanning microscopy. The images were evaluated using Image Pro-Plus software (Media Cybernetics, Warrendale, Pennsylvania). The distribution of DOX was evaluated by calculating the sum of DOX fluorescence spots and the DOX penetration distance. The microvessel density, the sum of DOX fluorescence spots, and the

4 ▪ iRGD-modified ZrO2 Nanoparticles Improves Distribution of Doxorubicin

Xie et al ▪ JVIR

Figure 3. (a) In vitro cytotoxicity assay revealed the viability levels of HepG2 cells treated with R-DZCNs were significantly lower than those treated with DZCNs at all loading concentrations of DOX after 24 hours of incubation (P < .05). (b) Fluorescence microscopy images of HepG2 cells after 4 hours incubation with DZCNs or R-DZCNs show stronger red fluorescence of DOX as well as merged fluorescence with DAPI in cells treated with R-DZCNs than in cells treated with DZCNs. Bar, 50 μm. (c) Quantitation of the intracellular red fluorescence of DOX in cells treated with R-DZCNs was greater than that of cells treated with DZCNs group (P < .05).

penetration distance were evaluated using the methods described previously (5,19).

In Vivo Antitumor Assay Animal Groups. Twenty tumor-bearing rabbits were randomized into 4 groups (n ¼ 5) (8). In the control group, rabbits received transarterial infusion of normal saline. The equivalent DOX dose for the remaining 3 groups was 1 mg per rabbit (18). All rabbits were infused with 0.25 mL of lipiodol. In addition, DOX (1 mL), DZCNs (1 mL), and R-DZCNs (1 mL) were infused in the DOX þ lipiodol, the DZCNs þ lipiodol, and the RDZCNs þ lipiodol groups, respectively. MR imaging was performed to monitor changes in tumor size. Rabbits in each group were humanely sacrificed immediately after the final MR images were obtained. Subsequently, liver tumors were harvested, and half of the tumor tissues were fixed in a formalin solution for immunohistochemistry (IHC), whereas the remaining tumor tissues were preserved at 80 C for Western blotting. Magnetic Resonance Imaging. MR imaging was performed to monitor the changes in tumor volume before transarterial infusion and at 7 and 14 days after transarterial infusion (8). A 3.0T clinical MR imaging scanner (General Electric) was used with a knee coil to improve the signal-tonoise ratio and spatial resolution. Axial T2-weighted turbo spin-echo (TR, 5.7; TE, 1.6; 3-mm layer thickness; number of excitation 1) images were acquired. Tumor volume was calculated according to the formula [V ¼ (a  b2)/2] (where a is the maximum diameter of the tumor, and b is the minimum diameter of tumor) (20). To minimize measurement error, tumor volume was assessed by 2 radiologists who were blinded to group allocation. The mean value of the 2 volume measurements was adopted. Immunohistochemistry. The expression levels of apoptotic proteins including caspase-3 and Bax and the antiapoptotic protein Bcl-2, were studied by IHC. IHC was

performed using 10% formalin-fixed and paraffin-embedded tissues. Tissues were cut into 5-μm sections, and sections were deparaffinized in dimethylbenzene and successively rehydrated in graded ethanol. Antigen retrieval was carried out by boiling in ethylenediaminetetraacetic acid. A 3% solution of hydrogen peroxide was used to quench endogenous peroxidases. Sections were blocked with 5% BSA and then incubated with caspase-3 antibody (1:500 dilution; Abcam, Cambridge, UK), Bax antibody (1:1,000 dilution; Proteintech, Chicago, Illinois), or Bcl-2 antibody (1:1,000 dilution; Proteintech) at 4 C overnight. Afterward, sections were rewarmed and incubated with horseradish peroxidase-conjugated secondary antibodies for 40 minutes at room temperature. Finally, sections were stained with 3,30-diaminobenzidine tetrahydrochloride, and nuclei were counterstained with hematoxylin. Images were obtained using inverted digital imaging light microscopy (Nikon, Tokyo Japan). The relative expression levels of caspase-3, Bax, and Bcl-2 were then evaluated in a semiquantitative manner related to the percentage of stained cells by using Image J software (US National Institutes of Health, Bethesda, Maryland). Western blotting. The expression levels of apoptosisrelated factors were analyzed by Western blotting with bactin used as an inner reference protein. Total protein was extracted from tumor tissue and homogenized in radioimmunoprecipitation assay lysis buffer. After the protein concentrations were measured, total protein aliquots of 40 μg were separated by sodium dodecyl sulfate-polyacrylamide electrophoresis. The separated proteins were transferred onto a polyvinylidene-fluoride membrane. The membranes were locked with 5% BSA and incubated with caspase-3 antibody (1:500 dilution; Abcam), Bax antibody (1:1,000; Proteintech), Bcl-2 antibody (1:1,000 dilution; Proteintech), or b-actin antibody (1:3,000 dilution; Abcam) overnight at 4 C. The membranes were washed with Tris-buffered saline tween and incubated with peroxidase-conjugated secondary antibody. After that, membranes were developed using

Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019

5

Figure 4. (a) Composite immunofluorescence images of DOX and microvessels show the differences among DOX penetration depth in DOX þ lipiodol, in DZCNs) þ lipiodol, and in R-DZCNs) þ lipiodol groups. Bar, 100 μm. (b) The sum of DOX fluorescence spots did not differ among the DOX þ lipiodol, DZCNs þ lipiodol, or R-DZCNs þ lipiodol group (P > .05 ) at 10 minutes after infusion but showed significant differences at 4 hours after infusion (P < .05). (c) Ten minutes after infusion, the DOX penetration distances did not differ among the 3 groups, whereas at 4 hours after infusion, the DOX penetration distance of the R-DZNCs þ lipiodol group was deeper than those of the DOX þ lipiodol and DZCNs þ lipiodol groups (P < .05).

Super Signal West Femto Maximum Sensitivity Substrate (Thermo Fisher Scientific). Protein bands relating to caspase-3, Bax, Bcl-2, and b-actin were analyzed using Image J software. The expression ratio was calculated using the formula [caspase-3 (or Bax, Bcl-2)/b-actin].

among multiple groups. A P value <.05 was considered statistically significant.

Statistical Analysis

RESULTS In Vitro Cytotoxicity and Cellular Uptake Assay

Data processing and analysis were performed using SPSS version 23.0 software (IBM, Armonk, New York). All values are expressed as mean ± SD. A t-test was used to compare the differences between the 2 groups. One-way analysis of variance (ANOVA) followed by a least significant difference t-test was used to determine differences

The characteristics of R-DZCNs are shown in Figure E2 (available online on the article’s Supplemental Material page at www.jvir.org). As shown in Figure 3a, the viability of HepG2 cells in the R-DZCNs group was lower than that in the DZCNs group at each DOX concentration (27.48% ± 4.7 vs 39.68% ± 3.8,

6 ▪ iRGD-modified ZrO2 Nanoparticles Improves Distribution of Doxorubicin

Xie et al ▪ JVIR

Figure 5. MR imaging shows tumor sizes in each group were similar before infusion. (a) At 7 and 14 days after infusion, tumor sizes in the control group were larger than those of in the experimental groups (white arrow). Bar, 10 mm. (b) A comparison of tumor growth curves shows the tumor volume of the R-DZCNs þ lipiodol group was smaller than those of the DOX þ lipiodol and DZCNs) þ lipiodol groups (P < .05).

respectively at 10 μg/mL; P ¼ .025; 32.77% ± 3.36 vs 44.89% ± 4.25, respectively at 5 μg/mL; P ¼ .018; 40.71% ± 4.01 vs 52.86% ± 5.61, respectively at 2.5 μg/ mL; P ¼ .038; 45.45% ± 5.06 vs 64.66 ± 4.22, respectively at 1.25 μg/mL; P ¼ .07; 50.93% ± 5.43 vs 70.81% ± 6.46, respectively at 0.625 μg/mL; P ¼ .015; and 60.36% ± 5.23 vs 80.7% ± 8.55, respectively at 0.3125 μg/mL; P ¼ .031). In the cellular uptake assay, the purple signal in the merged images indicated the colocation of DOX (Fig 3, red fluorescence) and cell nuclei (Fig 3, blue fluorescence). The red fluorescence intensity in the R-DZCNs group was higher than that in the DZCNs group (118.53 ± 16.85 vs 159.52 ± 23.76, respectively; P ¼ .014) (Fig 3c).

significantly greater than that in the DOX þ lipiodol group (2,449.15 ± 444.14; P ¼ .015) and less than that in the R-DZCNs þ lipiodol group (5,062.25 ± 585.62; P ¼ .008). At 10 minutes after infusion, the DOX penetration distances in the DOX þ lipiodol, DZCNs þ lipiodol, and R-DZCNs þ lipiodol groups were 33.75 ± 7.3 μm, 27.36 ± 8.13 μm, and 34.78 ± 10.04 μm, respectively (P ¼ .275). At 4 hours after infusion, the DOX penetration distance in the R-DZCNsþ lipiodol group (117.58 ± 19.36 μm) was significantly greater than that in the DOX þ lipiodol (52.64 ± 8.53 μm) and DZCNs þ lipiodol (52.64 ± 8.53 μm; P < .001) groups (Fig 4c).

DOX Distribution In Vivo

In Vivo Antitumor Assay

In vivo the distribution of DOX fluorescence (Fig 4 red) in tumor was not homogeneous (Fig 4a). The red fluorescence was mainly perivascular. In contrast, little detectable red fluorescence was detected in the areas that were farther away from the vasculature (Fig 4a). Microvessel density among the DOX þ lipiodol, DZCNs þ lipiodol, and RDZCNs þ lipiodol groups showed no significant differences at 10 minutes (34.05 ± 5.06, 30.69 ± 6.7, and 31.75 ± 3.91 respectively; P ¼ .675) or at 4 hours after infusion (30.81 ± 5.68, 28.81 ± 4.19, and 31.75 ± 3.98, respectively; P ¼ .581). At 10 minutes after infusion, the sum of fluorescence spots in the DOX þ lipiodol, DZCNs þ lipiodol, and R-DZCNs þ lipiodol groups were similar (1417.60 ± 353.82, 1407.11 ± 520.58, and 1585.59 ± 484.1, respectively; P ¼ .364) (Fig 4b). At 4 hours after infusion, the sum of DOX fluorescence spots in the DZCNs þ lipiodol group (3,464.73 ± 632.75) was

MR Imaging. Prior to transarterial infusion, the tumor volumes showed no differences in each group (358.32 ± 28.33 mm3, 368.89 ± 38.85 mm3, 416.1 ± 44.29 mm3, and 398.97 ± 56.91 mm3, respectively; P ¼ .526) (Fig 5a). At 7 days after infusion, the tumor volume of the R-DZCNs þ lipiodol group (758.70 ± 213.58 mm3) was smaller than that of the control group (3349.58 ± 313.15 mm3; P < .001) and DOX þ lipiodol group (1578.01 ± 299.43 mm3; P ¼ .002) but was similar to that of the DZCNs þ lipiodol group (1151.10 ± 277.65 mm3; P ¼ .162). At 14 days after infusion, the tumor volume of the R-DZCNs þ lipiodol group was significantly smaller than that of the control group, the DOX þ lipiodol group and the DZCNs þ lipiodol group (1,223.87 ± 223.58 mm3 vs 8,590.69 ± 564.42 mm3 or 3,695.26 ± 666.25 mm3 or 2281.06 ± 457.21 mm3, respectively; P ¼ .005) (Fig 5b).

Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019

7

Figure 6. (a) IHC assays show much higher positivity in tumor cells stained with caspase-3 and Bax and fewer Bcl-2-stained cells in experimental groups than those in the control group, in a rabbit VX2 liver tumor model. Scale bar, 50 μm. (b) The percentages of cells stained positive for Caspase-3 in the DOX þ lipiodol, DZCNs þ lipiodol, and R-DZCNs þ lipiodol groups were 2.2-fold, 3.4-fold, and 5.7fold higher, respectively, than that of the control group. (c) The percentages of cells stained positive for Bax in those groups were 2.0fold, 3.8-fold, and 5.8-fold higher, respectively, than those in the control group (P < .05). (d) In contrast, the percentages of cells stained positive for Bcl-2 in the 3 experimental groups were decreased by 0.7-fold, 0.5-fold, and 0.2-fold, respectively, of the control group (P < .05).

Immunohistochemistry. IHC showed that greater numbers of cells stained positive for caspase-3 and Bax, as well as fewer cells stained positive for Bcl-2, were observed in the experimental groups compared with those in the control group (Fig 6a). The percentages of cells stained positive for caspase-3 in the DOX þ lipiodol group, the DZCNs þ lipiodol group, and the R-DZCNs þ lipiodol group were, respectively, 2.2-fold, 3.4-fold, and 5.7-fold higher than those in the control group (P < .001) (Fig 6b). The percentages of cells stained positive for Bax

in these groups were, respectively, 2.0-fold, 3.8-fold, and 5.8-fold higher than those in the control group (P ¼ .018) (Fig 6c). In contrast, the percentages of cells stained positive for Bcl-2 were, respectively, reduced to 0.7-fold, 0.5-fold, and 0.2-fold compared with those in the control group (P ¼ .01) (Fig 6d). Western Blotting. Western blotting assay showed increased caspase-3 and Bax proteins, as well as decreased Bcl-2 proteins in experimental groups compared with

8 ▪ iRGD-modified ZrO2 Nanoparticles Improves Distribution of Doxorubicin

Xie et al ▪ JVIR

Figure 7. (a) The Western blotting assay shows increased caspase-3 and Bax proteins and decreased Bcl-2 protein in the tumors of the 3 treatment groups compared with those in the control group in a rabbit VX2 liver tumor model. (b) Caspase-3 proteins in the DOX þ lipiodol, DZCNs þ lipiodol, and R-DZCNs þ lipiodol groups were 1.7-fold, 2.6 -fold, and 4.3-fold higher, respectively, compared to those in the control group. (c) In contrast, Bax protein levels were 1.9-fold, 2.5-fold, and 3.2-fold higher, respectively (P < .05). (d) However, Bcl2 proteins were 0.8-fold, 0.6-fold, and 0.3-fold, respectively, compared to those in the control group (P < .05).

those in the control group (Fig 7a). Caspase-3 protein bands in the DOX þ lipiodol group, the DZCNs þ lipiodol group, and the R-DZCNs þ lipiodol group were 1.7-fold, 2.6 -fold, 4.3-fold, respectively, higher than those in the control group (P < .001) (Fig 7b). Similarly, values for Bax were, respectively, 1.9-fold, 2.5-fold, and 3.2-fold higher than those in the control group (P ¼ .008) (Fig 7c). In contrast, Bcl-2 protein expression decreased to 0.8-fold, 0.6-fold, and 0.3-fold, respectively, of the control group (P ¼ .012) (Fig 7d).

DISCUSSION In the present study, both the in vitro cytotoxicity and the cellular uptake assays in HepG2 cells demonstrated the targeting effect of iRGD-modified DZCNs. The DOX fluorescence intensity in cell nuclei in the R-DZCNs group

was significantly higher than that in the DZCNs group. As DOX mediates its antitumor effect in cell nuclei (21), the stronger cytotoxicity in the HepG2 cells in the R-DZCNs group could be due to the higher cellular uptake of DOX in HepG2 cell nuclei. In the in vivo study, the rabbit VX2 liver tumor model was prepared. VX2 tumor is a hypervascular tumor which receives blood supply predominantly from the hepatic artery, similar to that in human HCC. The diameter of the rabbit hepatic artery is sufficient to allow for hepatic artery catheterization (22). Because of these characteristics, the rabbit VX2 liver tumor model was used in the present study. Greater amounts of DOX fluorescence spots and farther penetration distance were observed in the R-DZCNs þ lipiodol group at 4 hours after infusion, demonstrating improved DOX distribution in the tumor. Previous studies revealed that RGD can facilitate the accumulation of

Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019

nanoparticles through intravenous infusion (23) in rabbit VX2 liver tumors. In another study, the accumulation of RGD targeting nanoparticles through transarterial route but without lipiodol was similar to that of the nontargeted nanoparticles at 1 hour after infusion but higher at 24 hours (24). In the present study, it is likely that iRGD-modified nanoparticles can improve the DOX distribution through transarterial infusion with lipiodol compared with that of free DOX and DZCNs. Additionally, the binding of the CendR sequence to Nrp-1 promoted the permeability of tumor vessels and triggered the extravasation of nanoparticles, which also contributed to a better distribution of DOX in tumors. In the present study, the effect of lipiodol on tumors cannot be discounted. Apart from the effect of inducing ischemic necrosis of the tumor, lipiodol can also facilitate the intratumoral delivery of nanoparticles. First, embolization allows more nanoparticles to accumulate in tumor tissues (8). Second, a reduced interstitial fluid pressure that occurred after embolization contributed to the improved delivery of nanoparticles (25). Thus, infusing nanoparticle with lipiodol can facilitate the drug’s accumulation and distribution in tumors. In the drug distribution study, a lower dose of DOX, 3 mg/kg, was used than in previous studies (8 mg/kg) (5,25). Two factors were considered for the use of the lower concentration in the present study. First, using a relatively small dose of DOX, the targeting effect of R-DZCNs can be better demonstrated, as a DOX concentration that is too high can overestimate the distribution of free DOX in the tumor, according to a previous study (5). Second, a lower dose of DOX would be better for the safety and welfare of subjects. The selection of the 2 time points (10 minutes and 4 hours) in which to observe DOX distribution was based on previous studies which demonstrated the first detection of DOX in tumor tissue at 10 minutes after infusion with lipiodol (5) and the complete elution of DOX from lipiodol in less than 4 hours (26). Ten minutes after infusion, the sum of DOX fluorescence spots and penetration distances in each group showed no significant differences. It would therefore seem that a 10-minute observation window was insufficient for R-DZCNs to be eluted from lipiodol and perform its targeting abilities. The improved DOX distribution in the R-DZCNs þ lipiodol group seemed likely to provide better inhibition of tumor growth associated with tumor cells apoptosis. This was demonstrated by the present in vivo antitumor study. MR imaging was performed to evaluate the tumor growth by monitoring the volume changes. Caspase-3 and Bax are 2 factors associated with tumor cell apoptosis. In contrast, Bcl-2 is considered an anti-apoptotic factor. In the R-DZCNs þ lipiodol group, the stronger inhibition of tumor growth indicated by smaller tumor volume as well as the greater extent of tumor cell apoptosis demonstrated that R-DZCNs can lead to an increased antitumor efficacy. There are several limitations in the present study. First, although VX2 tumor has many features in common with HCC, it should be recognized that, as an implanted tumor,

9

VX2 tumor cannot totally mimic the microenvironment in HCC. Second, a previous study demonstrated that the targeting effect of RGD in HepG2 cells remained unsatisfactory (27), meaning a more efficient targeting peptide for HCC needs to be explored in the future. Third, DOX distribution was assessed at 10 minutes and at 4 hours after infusion. However, the distribution of DOX in tumor is a dynamic process that occurs in a consecutive manner, so that more time points of measurements may help in better evaluating the change in DOX distribution in the tumor. In conclusion, the transarterial infusion of R-DZCNs with lipiodol exhibited a targeting effect to liver tumor tissues. This highlights the potential of R-DZCNs with lipiodol to improve the distribution of antitumor drugs within tumor and enhance their antitumor efficacy.

ACKNOWLEDGMENTS Research was supported by the National Key R&D Program (No. 2018YFC0115500) and National Natural Science Foundation of China (No. 81630053). This work was supported by the Key Laboratory of Diagnostic Imaging and Interventional Radiology of Liaoning Province. The authors also thank the Key Laboratory of Immunodermatology for performing confocal laser scanning microscopy and the Ministry of Health and Ministry of Education.

REFERENCES 1. Brown KT, Do RK, Gonen M, et al. Randomized trial of hepatic artery embolization for hepatocellular carcinoma using doxorubicin-eluting microspheres compared with embolization with microspheres alone. J Clin Oncol 2016; 34:2046. 2. Meyer T, Kirkwood A, Roughton M, et al. A randomised phase II/III trial of 3-weekly cisplatin-based sequential transarterial chemoembolisation vs embolisation alone for hepatocellular carcinoma. Br J Cancer 2013; 108: 1252–1259. 3. Facciorusso A, Bellanti F, Villani R, et al. Transarterial chemoembolization vs bland embolization in hepatocellular carcinoma: a metaanalysis of randomized trials. United European Gastroenterol J 2017; 5: 511–518. 4. Primeau AJ, Rendon A, Hedley D, Lilge L, Tannock IF. The distribution of the anticancer drug doxorubicin in relation to blood vessels in solid tumors. Clin Cancer Res 2005; 11:8782–8788. 5. Liang B, Xiong F, Wu H, et al. Effect of transcatheter intraarterial therapies on the distribution of doxorubicin in liver cancer in a rabbit model. PLoS One 2013; 8:e76388. 6. Nishiofuku H, Tanaka T, Fukuoka Y, et al. Intraarterial therapy using micellar nanoparticles incorporating SN-38 in a rabbit liver tumor model. J Vasc Interv Radiol 2017; 28:457–464. 7. Refaat T, West D, El Achy S, et al. Distribution of iron oxide core-titanium dioxide shell nanoparticles in VX2 tumor bearing rabbits introduced by two different delivery modalities. Nanomaterials (Basel) 2016; 6. 8. Jeon MJ, Gordon AC, Larson AC, Chung JW, Kim YI, Kim DH. Transcatheter intra-arterial infusion of doxorubicin loaded porous magnetic nano-clusters with iodinated oil for the treatment of liver cancer. Biomaterials 2016; 88:25–33. 9. Ma TC, Shao HB, Xu Y, Xu K. Three treatment methods via the hepatic artery for hepatocellular carcinoma: a retrospective study. Asian Pac J Cancer Prev 2013; 14:2491–2494. 10. Sugahara KN, Teesalu T, Karmali PP, et al. Tissue-penetrating delivery of compounds and nanoparticles into tumors. Cancer Cell 2009; 16: 510–520. at V. RGD-based strategies to target alpha(v) beta 11. Danhier F, Le Breton A, Pre integrin in cancer therapy and diagnosis. Mol Pharm 2012; 9:2961–2973.

10 ▪ iRGD-modified ZrO2 Nanoparticles Improves Distribution of Doxorubicin

12. Teesalu T, Sugahara KN, Kotamraju VR, Ruoslahti E. C-end rule peptides mediate neuropilin-1-dependent cell, vascular, and tissue penetration. Proc Natl Acad Sci U S A 2009; 106:16157–16162. 13. Schmithals C, Koberle V, Korkusuz H, et al. Improving drug penetrability with iRGD leverages the therapeutic response to sorafenib and doxorubicin in hepatocellular carcinoma. Cancer Res 2015; 75:3147–3154. 14. Fei W, Zhang Y, Han S, et al. RGD conjugated liposome-hollow silica hybrid nanovehicles for targeted and controlled delivery of arsenic trioxide against hepatic carcinoma. Int J Pharm 2017; 519:250–262. 15. Meng XW. What potential is there for the use of ZrO2 nanostructures for image-guided thermotherapy? Nanomedicine 2015; 10:3311–3313. 16. Chen X, Fu C, Wang Y, Wu Q, Meng X, Xu K. Mitochondria-targeting nanoparticles for enhanced microwave ablation of cancer. Nanoscale 2018; 10:15677–15685. 17. Fang Z, Sun Y, Xiao H, et al. Targeted osteosarcoma chemotherapy using RGD peptide-installed doxorubicin-loaded biodegradable polymeric micelle. Biomed Pharmacother 2017; 85:160–168. 18. In Joon Lee M, Cheol-Hee Ahn, Eui-Joon C, In Jae C, Jin Wook Chung M, Young Il K. Improved drug targeting to liver tumors after intra-arterial delivery using superparamagnetic iron oxide and iodized oil preclinical study in a rabbit model. Invest Radiol 2013; 48. 19. Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastas: correlation in invasive breast carcinoma. N Engl J Med 1991; 324:1–8. 20. Deng G, Zhao DL, Li GC, Yu H, Teng GJ. Combination therapy of transcatheter arterial chemoembolization and arterial administration of

21.

22.

23.

24.

25.

26.

27.

Xie et al ▪ JVIR

antiangiogenesis on VX2 liver tumor. Cardiovasc Intervent Radiol 2011; 34:824–832. Pommier Y, Leo E, Zhang HL, Marchand C. DNA topoisomerases and their poisoning by anticancer and antibacterial drugs. Chem Biol 2010; 17: 421–433. Burgener FA, Violante MR. Comparison of hepatic VX2-carcinomas after intra-arterial, intraportal and intraparenchymal tumor cell injection. An angiographic and computed tomographic study in the rabbit. Invest Radiol 1979; 14:410–414. Hu G, Lijowski M, Zhang H, et al. Imaging of Vx-2 rabbit tumors with alpha(nu)beta3-integrin-targeted 111In nanoparticles. Int J Cancer 2007; 120:1951–1957. Tian M, Lu W, Zhang R, et al. Tumor uptake of hollow gold nanospheres after intravenous and intra-arterial injection: PET/CT study in a rabbit VX2 liver cancer model. Mol Imaging Biol 2013; 15: 614–624. Liang B, Chen S, Li L, et al. Effect of transcatheter intra-arterial therapies on tumor interstitial fluid pressure and its relation to drug penetration in a rabbit liver tumor model. J Vasc Interv Radiol 2015; 26: 1879–1886. Lewis AL, Gonzalez MV, Lloyd AW, et al. DC bead: in vitro characterization of a drug-delivery device for transarterial chemoembolization. J Vasc Interv Radiol 2006; 17(2 Pt 1):335–342. Ma Y, Ai G, Zhang C, et al. Novel linear peptides with high affinity to alphavbeta3 integrin for precise tumor identification. Theranostics 2017; 7:1511–1523.

Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019

APPENDIX A. SCHEMATIC ILLUSTRATION OF THE TRRANSARTERIAL INFUSION OF R-DZCNs AND LIPIODOL Transarterial infusion of iRGD-modified ZrO2 nanoparticles with lipiodol improves the tissue distribution of doxorubicin and its antitumor efficacy.

MATERIALS AND METHODS Preparation of ZrO2 Nanoparticles Silica dioxide (SiO2) nanoparticles were used as template for the preparation of zirconia composite (ZrO2) nanoparticles. After dehydration with absolute ethanol, 1 mL SiO2 (238mg) was dispersed in a mixed solution containing 1.2 mL aqueous ammonia, 120 mL ethanol, and 40 mL acetonitrile. Zirconium (IV) (0.5 mL; Bailingwei, Beijing, China) was dispersed in 10 mL acetonitrile and 30 mL ethanol, and the mixture was then added to the SiO2 solution. After continuously stirring for 12 hours, the silica/ zirconia was then obtained. At 80 C, the silica/zirconia was treated with aqueous solution containing 100 mL deionized water and 4 mL sodium hydroxide (1 M) for 4 hours. After washing with deionized water and centrifugation for five times, the ZrO2 nanoparticles were obtained.

Preparation of DZCNs and R-DZCNs To prepare doxorubicin-loaded zirconia composite nanoparticles (DZCNs ), 20 mg of DOX (Meilun Biotechnology, Dalian, China) and 40 mg of ZrO2 nanoparticles were dispersed into 15 mL of 33.33% ethanol-deionized water (v/v), poured into a 50 mL conical flask, then mixed by ultrasound. After vacuum pumping until the solvent was drained, the residual was collected. After centrifugation and washing with deionized water three times, the sedimented DZCNs and supernatants were collected respectively. The preparation of iRGD-modified and doxorubicin-loaded ZrO2 nanoparticles (R-DZCNs) is outlined in Figure E1. Firstly, 40 μL of trimethylchlorosilane was added to the mixture of 40 mg ZrO2 nanoparticles and 10 mL of 50% ethanoldeionized water (v/v) and stirred overnight. After washing five times with absolute ethanol, the amino-modified nanoparticles were obtained. Second, 20 mg DOX was loaded into amino-modified nanoparticles. Third, 4 mg of iRGD peptide was added to an aqueous solution containing 20 mg/mL 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC; Bailingwei, Beijing, China) and 18 mg/mL N-hydroxy succinimide (NHS; Bailingwei, Beijing, China). The pH value was adjusted to 5.5 and the mixture was shaken at room temperature for 30 minutes to activate the iRGD (Jier Biochem Co, Shanghai, China) peptide.

10.e1

After that, the amino-modified DZCNs and activated iRGD were added to 100 mL borate buffer (4 mol boric acid: 1 mol borax) and the pH value was adjusted to 8.3. The mixture was shaken for 24 hours at room temperature. Finally, the RDZCNs nanoparticles were collected after washing with deionized water and centrifugation three times. Characteristics, such as the structure and diameter of R-DZCNs were observed by transmission electron microscopy (TEM; JEOL, Tokyo, Japan). Zeta-potential and hydrodynamic particle sizes were measured with a Zeta-Sizer (Zetasizer Nano ZS90; Malvern Instruments, Malvern, U.K.). Drug encapsulation ratio and drug loading ratio were calculated by the following equations:  Drug-encapsulation efficiency ¼ (drug input  drug in supernatant)/drug input  100%  Drug-loading efficiency ¼ (drug input  drug in supernatant)/drug loading material  100%.

Cumulative Drug Release Test In Vitro R-DZCNs (1 mL, containing 10mg DOX) or the mixture of lipiodol (0.25 mL) and DOX aqueous solution (1 mL, containing 10 mg of DOX) were placed in dialysis bags (Solarbio, Beijing, China) and then immersed in 50 mL of phosphate buffered saline (PBS) respectively. At specific time intervals (1, 2, 3, 4, 5, 6, 8, 10, 12, and 24 hours), 1 mL of PBS was extracted and replaced with fresh PBS. The DOX concentration in PBS was measured using a microplate reader (Multiskan FC, Thermo Fisher, Waltham, MA,) at a wavelength of 450 nm and the cumulative release of DOX was calculated as DOX dosage in PBS at each time point.

RESULTS Characterization of R-DZCNs and DOX Loading and Release As shown in Figure E2a, the hollow structure of ZrO2 nanoparticles was observed using TEM. The mean diameter of R-DZCNs was 165.21 ± 8.2 nm. The hydrodynamic diameter of R-DZCNs was 230.1 nm with a concentrated size distribution (Fig E2b). The encapsulation efficiency of DOX was approximately 50%, while the drug loading efficiency was approximately 20%. The molar concentration of DOX released from R-DZCNs in vitro were (1.11, 2.45, 3.28, 4.10, 4.69, 5.13, 6.01, 6.37, 6.67, 7.07)  103 mmol at 1, 2, 3, 4, 5, 6, 8, 10, 12, and 24 hours respectively, while these values in lipiodol were (6.42, 9.58, 12.09, 13.61, 14.16, 14.72, 15.31, 15.51, 15.72, 15.76)  103 mmol at 1, 2, 3, 4, 5, 6, 8, 10, 12, and 24 hours, respectively (Fig E2c).

10.e2 ▪ iRGD-modified ZrO2 Nanoparticles Improves Distribution of Doxorubicin

Xie et al ▪ JVIR

Figure E1. R-DZCNs synthesis. (a) ZrO2 nanoparticles were amino-modified to facilitate the linking of iRGD peptide on the surface of nanoparticles. DOX was loaded into amino-modified nanoparticles. Finally, the iRGD peptide was activated and conjugated to the amino groups on the surface of ZrO2 nanoparticles. (b) Transarterial infusion of iRGD-modified and R-DZCNs with lipiodol.

Figure E2. The physiochemical characterization of R-DZCNs. (a) TEM shows a hollow structure of ZrO2 nanoparticles. The mean diameter was calculated at 165.21 ± 8.2 nm. Bar, 100 nm. (b) Dynamic light-scatter monitoring shows the size distribution of DZCNs and R-DZCNs was concentrated, and the mean hydrodynamic diameter was 230.1 nm. (c) The cumulative drug release curve shows a sustained DOX release profile from R-DZCNs, and the cumulative DOX release ratio was approximately 35.5% after 24 hours. TEM ¼ transmission electron microscopy.