Dual-Modified Novel Biomimetic Nanocarriers Improve Targeting and

Subscriber access provided by University of South Dakota

Biological and Medical Applications of Materials and Interfaces

Dual-Modified Novel Biomimetic Nanocarriers Improve Targeting and Therapeutic Efficacy in Glioma Shiyao Fu, Meng Liang, Yuli Wang, Lin Cui, Chunhong Gao, Xiaoyang Chu, Qianqian Liu, Ye Feng, Wei Gong, Meiyan Yang, Zhiping Li, Chunrong Yang, Xiang Yang Xie, Yang Yang, and Chunsheng Gao ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b18664 • Publication Date (Web): 24 Dec 2018 Downloaded from http://pubs.acs.org on December 24, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Dual-Modified Novel Biomimetic Nanocarriers Improve Targeting and Therapeutic Efficacy in Glioma Shiyao Fu

1, 3, #,

Meng Liang 1, #, Yuli Wang 1, #, Lin Cui

1, 3,

Chunhong Gao 1, Xiaoyang Chu 4,

Qianqian Liu 1, Ye Feng 1, Wei Gong 1, Meiyan Yang 1, Zhiping Li 1, Chunrong Yang 3, Xiangyang Xie 2, *, Yang Yang 1, *, Chunsheng Gao 1, * 1 State

Key Laboratory of Toxicology and Medical Countermeasures, Beijing Institute of Pharmacology and

Toxicology, Beijing 100850, China 2

General Hospital of Central Theater of the PLA, Wuhan 430070, China

3

Jiamusi University, Jiamusi 154002, China

4

307 Hospital of the PLA, Beijing 100071, China

# These

authors contributed equally to this work.

*Corresponding

Author.

Xiangyang

Xie

[email protected];

[email protected]; Chunsheng Gao [email protected].

1

ACS Paragon Plus Environment

Yang

Yang

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT: Glioma is a fatal disease with limited treatment options and very short survival. Although chemotherapy is one of the most important strategies in glioma treatment, it remains extremely clinically challenging largely due to the blood-brain barrier (BBB) and the blood-brain tumor barrier (BBTB). Thus, the development of nanoparticles with both BBB and BBTB penetrability, as well as glioma-targeting feature, is extremely important for the therapy of glioma. New findings in nanomedicine are promoting the development of novel biomaterials. Herein, we designed a red blood cell membrane-coated solid lipid nanoparticle (RBCSLN)-based nanocarrier dual-modified with T7 and NGR peptide (T7/NGR-RBCSLNs) to accomplish these objectives. As a new kind of biomimetic nanovessels, RBCSLNs preserve the complex biological functions of natural cell membranes while possessing physicochemical properties that are needed for efficient drug delivery. T7 is a ligand of transferrin receptors (TfR) with seven peptides that is able to circumvent the BBB and target to glioma. NGR is a peptide ligand of CD13 that is overexpressed during angiogenesis, representing an excellent glioma-homing property. After encapsulating vinca alkaloid vincristine (VCR) as the model drug, T7/NGR-RBCSLNs exhibited the most favourable antiglioma effects in vitro and in vivo by combining the dual-targeting delivery effect. The results demonstrate that dual-modified biomimetic nanoparticles provide a potential method to improve drug delivery to the brain and hence increasing glioma therapy efficacy. KEYWORDS: dual-targeting; biomimetic nanocarriers; blood-brain barrier; blood-brain tumor barrier; glioma;

2


Page 2 of 46

Page 3 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. INTRODUCTION

Glioma is considered to be one of the deadliest human cancers with a median survival time of 12-15 months.1 Chemotherapy is the common method for the subsequent treatment after surgical resection of glioma. However, the clinical efficacies of drug treatment in glioma remain very unsatisfying. The efficient drug delivery to glioma is largely restrained by the poor physicochemical properties, low-targeting-level nature of therapeutic molecules, and the existences of both the blood-brain barrier (BBB) and blood-brain tumor barrier (BBTB), which render many therapeutic agents ineffective for glioma intervention.2, 3 Therefore, there is an urgent, unmet medical need for development of deliverable therapeutic agents across physiological and pathological barriers for the treatment of glioma. Nanotechnology represents a hopeful approach for the delivery of drugs in glioma.4 An ideal nanocarrier for the treatment of glioma should be able to sufficiently penetrate the physiological and pathological barriers, while being sufficiently specific to reach to the target cells with suitable physiochemical features, biocompatibility, and efficient scale-up of production. In the last few decades, various types of nanovessels have been fabricated using a variety of materials. However, problems with the synthetic materials still exist and even cause toxicology issues.5,

6

Although endogenous materials (e.g., exosomes, lipoproteins,

and apoferritin) are considered “self” by the immune system, the production and purification characteristics of endogenous material-based nanocarriers are ineffective for scale-up of the manufacturing process with robust and reproducible procedures. 7-10

3



To address this dilemma, many attempts have focused on utilizing red blood cell membrane-coated nanoparticles (RBCNPs). Coating the intact red blood cell (RBC) membrane on the surface of nanoparticles endows RBCNPs with both the physicochemical properties of synthetic materials and the complicated biological functions of endogenous materials. RBCNPs possess a greater circulation half-life and display low immunogenicity compared with to synthetic materials-based nanocarriers. In addition, RBCs are the most readily available and abundant cells in body. This source is easy for scaling up and economic preparation.11 Recently, RBCNPs have received extensive attention for their unique advantages in applications of drug delivery.12 However, due to the insufficiency of targeting specificity, it may be hard for RBCNPs to cumulate in the glioma site based on nanosize alone. Receptor-mediated endocytosis is a crucial mechanism for the targeting drug transport to the brain. Thence, some efforts have been devoted to the development of active targeted RBCNPs that are modified with specific ligands. There are two approaches to realize the modification process of active targeted RBCNPs. The first one is direct modification. Due to the surfaces of RBC membrane carrying active groups such as amino, thiol, and carboxylic groups, which can be used to bind ligand onto the membrane through covalent linkage, thus modification of RBCNPs with a peptide onto their surfaces can be achieved by direct chemical reactions between RBCNPs and peptides.13 However, use of organic solvents and chemical reagents in the modification process may impair the integrity of the RBC membrane, as well as interfere the performance and biocompatibility in vivo, even influencing the reproducibility of preparation process. The second method is indirect modification by insertion of pro-modified lipids or proteins onto the biological membranes.

4


Page 4 of 46

Page 5 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In this approach, RBCNPs are successfully modified with a positively charged targeting ligand.14 However, the ligands with positive charges are likely to be absorbed by the negatively charged RBC membrane, compromising their targeting ability. The surface electrical properties of RBC membranes may limit the modification of positive ligands onto the RBC membranes. In addition, most RBCNPs use PLGA nanoparticles as a core component. Although PLGA is approved by FDA, its low solubility and acidic by-products in degradation may hinder its utilization in brain diseases.15, 16 Moreover, the organic solvents used in the preparation of PLGA nanoparticles is another shortcoming that may lead to toxicity issues.17 Overall, engineering all of the desired properties of the ideal nanocarrier into a single nanocarrier particle is a tremendous challenge. By lipid-insertion, which modifies the surface of RBC membranes with negatively charged peptides, it may be possible to avoid the positive charge problems and achieve effective glioma-targeted drug delivery. Transferrin receptors (TfRs) are expressed on the surfaces of BBB and glioma cells.18 Therefore, the related ligands could be used to target to the cells of BBB and glioma. T7 peptide, which possesses negative charges, has a high affinity for TfR (Kd~10 nM). CD13 is highly expressed on tumor cells and involves in tumor growth and metastasis.19 Thus, CD13 represents an appropriate target for tumor targeting. NGR, a negatively charged peptide, was reported to selectively bind to CD13.20 Recently, it was reported that T7- or NGR- modified nanocarriers inhibit glioma growth.21,

22

Among

various nanoparticles, solid lipid nanoparticles (SLNs) based on natural lipids whose degradation products may not influence the extracellular/intracellular environment. In addition, production of SLNs, which needs no organic solvent, makes it one of the best

5



Page 6 of 46

choice for targeting brain diseases.23 Here, the first exploration of functional RBCSLNs as a novel drug delivery system targeted to glioma was reported. The dual-targeting RBCSLNs are modified with T7 and NGR (T7/NGR-RBCSLNs) by the lipid-insertion method, one of which penetrates the BBB and targets glioma cells, and the other penetrates the BBTB and targets glioma cells. After encapsulating vinca alkaloid vincristine (VCR) in the dual-modified T7/NGR-RBCSLNs, their anti-glioma effect was detected in vitro and estimated in vivo on intracranial C6 tumor-bearing animal models. These findings provide valuable animal experiment data to verify a peptide mediated nanotherapy of glioma, a malignant and hard to cure disease.

2. EXPERIMENTAL SECTION

2.1. Materials. NGR peptide (CYGGRGNG) with a cysteine on the N-terminal (Cys-NGR) and T7 peptide (CHAIYPRH) with a cysteine on the N-terminal (Cys-T7) were provided by Cybertron

medical

technology

Co.

(Beijing,

2-distearoyl-sn-glycero-3-phosphoethanolamine-N-maleimide

China).

(polyethylene

1, glycol)

(DSPE-PEG2000-Mal) were obtained from Xiàn ruixi Biological Technology Co., Ltd (Xiàn, China). Glycerol monostearate and Poloxamer-188 were provided by Fenglijingqiu Commerce and Trade Co., Ltd. (Beijing, China). 10-hydroxycamptothecin (VCR) was purchased from Yuxin Pharmaceutical Co, Ltd (Sichuan, China). The other chemicals were all reagent grade and purchased from MilliporeSigma, except for special stated. HUVECs (human umbilical vein endothelial cells), mouse brain endothelial bEnd.3 cells, and glioma C6 cells were purchased from the Cell Resource Centre of PUMCH (Beijing,

6


Page 7 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


China). Male and female (each in half) ICR mice (22-24 g) were provided by Vital River Laboratories (Beijing, China). All animal experiments were complied with the code of ethics in research, training and testing of drugs issued by the Animal Care and Use Ethics Committee in Beijing Institute of Pharmacology and Toxicology. 2.2. Synthesis and Characterization of Conjugates. DSPE-PEG2000-Mal were respectively conjugated to the cysteine residues of T7 and NGR to synthesize DSPE-PEG2000-T7 and DSPE-PEG2000-NGR. In brief, T7 and NGR peptides were reacted with DSPE-PEG2000-Mal (1.5 : 1 molar ratio) in chloroform included triethylamine (5 eq.) at 18-26 °C for 24 h with gentler mixing. The reactants were dialyzed for 48 h in a dialysis membrane (molecular weight cut-off of 3.5 kDa) with deionized water to eliminate the organic solvent and un-conjugated reactants. The solutions in the dialysis bag were concentrated and dried via a rotary evaporation process and obtained powders were kept at -20 °C. The products of the conjugation reactions were analyzed by a matrix-assisted laser desorption/ ionization time of flight mass spectrometry (MALDI-TOF MS). 2.3. RBC Membrane Derivation. RBC membrane was collected according to a previously reported by Dodge et al. 24 Briefly, whole blood was collected from ICR mice with heparin solution and then centrifuged to separate the serum and buffy coat. RBC membrane was collected through a hypotonic treatment. 2.4. Preparation of SLNs. Preparation of SLNs was performed by the solvent injection method with some modification.25 In brief, 56 mg of glycerol monostearate was dissolved in

7



1.0 mL of ethanol to form the oil phase, then the oil phase was mixed with 200μL of VCR (5 mg) or Cy5.5 (hydrophobic probe, 100 μL) solutions (containing 1% poloxamer-188) and ultrasonic sonicated at 60 °C, next the emulsion was injected through an injection needle into the stirred (1000 rpm) aqueous solutions (10 mL, containing 1% poloxamer-188) and stirred for 30 min under 40 °C. After that, the multiple emulsion was poured into a 10 mL of ice-cold water (~ 4 °C) and stirred continuously (1000 rpm) under 4 °C for 30 min. Then the suspension was centrifugated (10 000 rpm for 15 min). In the last, the sublayers of the centrifugation were filtrated through a 0.22 μm membrane to obtain the concentrated product of VCR-loaded N-SLNs or Cy5.5-labelled N-SLNs. To control the particle size of the final products, the final dispersion was extruded 3 times via polycarbonate membrane filters (100 nm) by an Avanti Polar Lipids mini extruder (Alabaster, AL). 2.5. Preparation of RBCSLNs and Peptide-Modified RBCSLNs. RBC membranes were sonicated for 6 minutes using a bath sonicator (KQ3200, Kunshan, China) with 37 kHz frequency of and 100 W power (working model: work 1 second and stop 1 second). Next, the prepared RBC membranes were filtrated through the 400, 200, and 100 nm polycarbonate porous membranes (each size membrane repeated 3 times) in the above mini extruder. To prepare RBCSLNs, 3 mL of VCR-loaded N-SLNs or Cy5.5-labelled N-SLNs at 10 mg.mL-1 was mixed with 1 mL of RBC membrane vesicles and subsequently filtrated through a 100 nm membrane for 5 times to prepare VCR-loaded RBCSLNs or Cy5.5-labelled RBCSLNs. To prepare T7-modified RBCSLNs (T7-RBCSLNs), NGR-modified RBCSLNs (NGR-RBCSLNs), and T7/NGR-co-modified RBCSLNs (T7/NGR-RBCSLNs), the lipid-insertion technique was used. 8


Page 8 of 46

Page 9 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For T7-RBCSLNs preparation, DSPE-PEG2000-T7 (0.5%, 1%, 2%, 3%, 4%, and 6%, molar ratio) was incubated with RBCSLNs in PBS (pH 7.4) at 37°C with stirring for approximately 4 h. NGR-RBCSLNs were prepared by the similar programs, except replacing the DSPE-PEG2000-T7 with

DSPE-PEG2000-NGR.

For

T7/NGR-RBCSLNs

preparation,

the

content

of

DSPE-PEG2000-NGR and DSPE-PEG2000-T7 was 2% and 4%, respectively. 2.6. Connection Efficiency of the Biomimetic Nanocarriers. The primary amine group of the DSPE-PEG2000-peptide was conjugated the with carboxyl group of the fluorescence probe, forming the target product, a DSPE-PEG2000-peptide-fluorescence probe. DSPE-PEG2000-NGR was labeled by the 5-(6)-carboxytfluorescein diacetate (CFDA), in the following process. In brief, CFDA (5 mg) was added into dimethyl sulfoxide (1 mL), then N, N'-dicyclohexyl carbodiimide (2.4 mg) and of NHS (1.3 mg) were added to the mixture. Stirred the mixture for 24 h under 18-26 °C and removed the undissolved substances via centrifuged at 4000 rpm for 15 min. Next, obtained the supernatants and dissolved with 2 mg of DSPE-PEG2000-NGR and 1 μL of triethylamine at 18-26 °C and stirred for 24 h without light. After that, the solution was dialyzed (3.5 kDa) against deionized water in the dark for 24 h to eliminate the non-target products. The goal substances (DSPE-PEG2000-NGR-CFDA) were freeze-dried to obtain loose solid powders. DSPE-PEG2000-T7 was labeled by a fluorescence probe of 5-Carboxy-X-rhodamine (5-ROX), under a similar program as described for DSPE-PEG2000-NGR-CFDA. The products contained dialyzed solutions (DSPE-PEG2000-T7-5-ROX) were also freeze-dried to obtain loose solid powders. The fluorescence probe-labelled dual-modified RBCSLNs (T7-5-ROX/NGR-CFDA-RBCSLNs) were prepared following the same procedures as the preparation of T7/NGR-RBCSLNs, except 9



DSPE-PEG2000-T7

and

DSPE-PEG2000-NGR

were

partially

Page 10 of 46

substituted

with

DSPE-PEG2000-T7-5-ROX and DSPE-PEG2000-NGR-CFDA, respectively. A certain amount of DSPE-PEG2000-T7-5-ROX, DSPE-PEG2000-NGR-CFDA, or RBCSLNs were dispersed in a phosphate buffer solution (PBS) at the pH of 7.4, respectively. Spectrum scanning at the range from 200 to 600 nm was performed by a ultraviolet-visible spectrophotometer and the maximum absorption wavelengths of DSPE-PEG2000-T7-CFDA and DSPE-PEG2000-T7-5-ROX were determined. The standard solutions of the two fluorescence probes were prepared (0.5 to 3.50 μg·mL−1), and the absorbance

at

maximum absorption wavelength (493 nm for CFDA, 578 nm for 5-ROX) was determined by the same spectrophotometer above. These data were collected and calculated to fit the linear regression. T7-5-ROX/NGR-CFDA-RBCSLN suspensions were diluted 10 times by PBS (pH 7.4), and its absorbency (ATotal) at the maximum absorption wavelength was assayed, respectively. Next, the samples of T7-5-ROX/NGR-CFDA-RBCSLNs were transferred to the upper chambers of ultrafiltration centrifuge tubes (MWCO 300 kDa) and centrifuged for 15 min at 8000 r·min-1, followed by the collection of sublayer liquids. The collected samples were diluted 4 times by PBS (pH 7.4), and their absorbency (AFree) at the maximum absorption wavelength was also assayed. The following equation was performed to calculate the connection efficiency of T7-5-ROX/NGR-CFDA-RBCSLNs: Connection efficiency = (10ATotal - 4AFree)/10ATotal ×100% 2.7. Ligand Density Screening. To explore the influence of T7 and NGR ligands

10


Page 11 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


density on the cell uptake, fluorescence probe Cy5.5 labeled RBCSLNs (T7-RBCSLNs and NGR-RBCSLNs) with different peptide molar ratios (0.5%, 1%, 2%, 3%, 4% and 6%) were prepared. Next, C6 cells (5×105 cells in each well) were embedded in a 6-well plate and cultivated for 24 h. After that, different formulations were added into the cell wells and incubated (37 °C) for 2 h. Then, the wells were washed with PBS (~4 °C), trypsin solution and finally rinsed 3 times by PBS (~4 °C). With centrifugation and re-suspending in PBS (Cy5.5 of 150 ng.mL-1), about 105 cells were assayed at once by the flow cytometry (FCM). 2.8. Characterization of Biomimetic Nanocarriers. The particle diameters and related distributions of the prepared formulations were analyzed using a Nanophox dynamic light scattering (Sympatec GmbH, Germany). The T7/NGR-RBCSLNs loading VCR was photographed by a H-7650 transmission electron microscopy (TEM) (HITACHI, Japan) to observe its morphology. The structural information of T7/NGR-RBCSLNs loading VCR was characterized by a J-810 circular dichroism (CD) spectroscopy (JASCO, Japan). The stability of T7/NGR-RBCSLNs loading VCR in PBS with 10% fetal bovine serum (FBS) at 37 °C was measured via a Formulaction Turbiscan Lab® Expert (L'Union, France). The data of this tests were processed via the software comes from the instrument. The VCR encapsulation efficiency (EE) and drug loading capacity (DL) of prepared nanoparticles were calculated via the following equations: EE%=(Wtotal drug-Wfree drug)/Wtotal drug×100% DL%=(Wtotal drug-Wfree drug)/Wtotal drug and carriers×100% Where Wtotal drug, Wfree drug and Wtotal drug and carriers means the total drug in nanoparticles, the quantity of free drug in nanoparticle suspensions and the weight of total drug and carriers, 11



respectively. 2.9. In Vitro Release Profile. Dialysis bags were employed to determinate the in vitro VCR release profiles of various VCR-loaded RBCSLNs formulations. PBS (0.1 M, pH 7.4) was used as the release medium. Aound 1 mL of different RBCSLNs sample was transferred to a dialysis bag (12-14 kDa). Then, the sample loaded bags were immersed in 30 mL of release medium (37 °C) with continual stirring. At preseted time points, 800 μL of release medium was removed for analysis and the same volume of blank medium was added in. The VCR in the obtained samples was measured via an HPLC as reported.22 2.10. Cell Binding Test. To investigate the cell binding affinity of various RBCSLNs, various Cy5.5-labelled RBCSLNs (5 μM) were respectively kept (at 37 °C) with 3 kinds of cells (HUVECs, C6 and bEnd.3 cells) in culture plates for 2 h. Next, the plate wells were rinsed with PBS

(~4 °C) for 3 times. With centrifugation and re-suspending in PBS, the cell

suspensions were assayed qualitatively using a UltraVIEW Vox confocal laser scanning microscopy (CLMS) (PerkinElmer) and evaluated quantitatively via the FCM. 2.11. Cytotoxicity Assay. This test was estimated by the MTT method. The C6 cells were inoculated into the 96-well plates (~4500 cells per well and incubated with different RBCSLNs at various concentrations. Then, MTT solutions (20 μL, 5 mg.mL-1) were transported in the wells after 72 h. Incubated for 4 h, the cell viability (%) was calculated based on the collected UV data (490 nm) via a Model 680 plate reader. 2.12. Cell Transportation Assay. The establishment of the in vitro BBB model was based on the report.26 In brief, bEnd.3 cells (1.0×105 cells) were plated on the upper side of

12


Page 12 of 46

Page 13 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


R transwell insert. The transportation experiment on BBB was carried out as the Corning○

the transendothelial electrical resistance (TEER) exceed 200 Ω.cm2.27 The culture mediums in the upper transwell chambers were replaced with Cy5.5-labelled RBCSLNs (50 μM) formulations in culture medium. After an incubation of 4 h, solution collected from the bottom transwell chamber was assayed by the Cary Eclipse fluorescence spectrophotometer (Agilent, Australia). In order to construct the BBTB model in vitro, the upper inserts and bottom transwell of transwell were plated with HUVECs and C6 cells, respectively. The density ratio of HUVECs/C6 was 5:1.28 The culture mediums in the upper transwell chambers were replaced with Cy5.5-labelled RBCSLNs. After 4 h of incubation, solution collected from the lower chamber was assayed by a fluorescence spectrophotometer. 2.13. Inhibitory Effect on In Vitro BBB/Glioma Cells. The establishment of the BBB/glioma co-culture model was

on the basis of a report.29 Briefly, bEnd.3 cells (1.0×105

cells/insert) were plated on the upper side of the transwell insert. C6 cells (2000 cells/chamber) were plated on the bottom transwell chamber. After a incubation of 5 days, the model was constructed and could be utilized in the subsequent experiments. Free VCR or different VCR-loaded formulations (final concentration was 1000 ng.mL-1) were appended to the transwell insert of the co-culture model. After a 48 h incubation, surviving C6 cells (%) in the bottom transwell chamber were estimated using the assay of sulforhodamine-B staining.29 2.14. Brain Targeting Test in Zebrafish. Zebrafish were raised as the report describled.30 Dividing the fishes into several groups, the hearts of zebrafish were injected with free Cy5.5 solutions or various Cy5.5-labelled RBCSLNs suspensions (10 nL, 0.1 mg·mL-1) 13



Page 14 of 46

via the micro-sprinkler. Ten minutes later, the fish brain was observed and photographed by a CLSM. 2.15. In vivo Distribution Study. The animal model of glioma-bearing mice was constructed according to the report.31 The glioma-bearing mice in different groups were injected

by

Cy5.5-labelled

RBCSLNs,

Cy5.5-labelled

T7-RBCSLNs,

Cy5.5-labelled

NGR-RBCSLNs and Cy5.5-labelled T7/NGR-RBCSLNs through the tail venin. After 6 h, the in vivo imaging was carried out using the IVIS® Spectrum-CT (PerkinElmer, Hebron, KY). Bioluminescent and fluorescent data were processed by the Caliper Living Image® software (Alameda, CA). The basal immunofluorescence level in each animal was collected before the injection of formulations. Then, different Cy5.5-labelled RBCSLNs suspensions were injected into the mice via tail vein and related fluorescence data were collected. Six hours later, the anesthetization was conducted, and the heart-through perfusion were performed with normal saline and paraformaldehyde solutions (4%). The brain were removed and fixed in paraformaldehyde solutions (4%), dehydrated with sucrose solution for 24 h and embedded in OCT compound (Tissue-Tek, Sakura, Torrance, CA). The brains were sliced into 5 μm sections via frozen section, stained with DAPI and then subjected to the CLSM assay. 2.16. In Vivo Anti-Cancer Evaluation. Animals of the above model were assigned to 6 groups (n=10): normal saline, free VCR, RBCSLNs loading VCR, T7-RBCSLNs loading VCR, NGR-RBCSLNs loading VCR, and T7/NGR-RBCSLNs loading VCR. At the 8 days of the model establishment, the mouse was injected with various tested formulations (1 mg VCR per kg) 4 times every other day. At the 16 days, 4 animals were randomly grabbed from every group, treated with anesthetic, and scanned on brain cancer via a Siemens magnetic resonance 14


Page 15 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


imaging (MRI) (Munich, Germany) to obtain the tumor diameters. Tumor inhibitions were evaluated by such equation: Rv=(Vdrug/ Vsaline) × 100%. The Vdrug represents tumor volume after treated by drug, and Vsaline represents tumor volume after treated by physiological saline. The left animals of all groups (n=6) were applied in the survival test. The time between the tumour transplantation day and the animal sacrifice day was used to calculate the survival time. Then, the Kaplan Meier survival curve of each group was plotted. During the experiment, the body weights of mice were monitored every day. 2.17. Histology of Tumors. After the achievement of tumor inhibition data, the animals were killed and their brain tissues were collected. Tumor neutral-buffered formalin(10%)-fixed and paraffin-embedded tissues for sliced at a thickness of 0.5 μm, the sections were treated by Hematoxylin and Eosin (HE) staining. Then the sections were observed and photographed by an Olympus microscopy (Olympus Company, Japan). 2.18. Safety Evaluation. To reckon the safety of T7/NGR-RBCSLNs loading VCR, healthy mice were treated as the description of "2.16 In Vivo Anti-Cancer Evaluation". The mice in the control group were administrated with PBS. At the day 16, animals were anesthetized and sacrificed, tissues of heart, liver, brain, lung and kidney were harvested and stained by HE. At the same time, haemogram assay was also performed to estimate the safety profiles of VCR-loaded T7/NGR-RBCSLNs. 2.19. Statistical Analysis. Quantitative data are presented as the means ± standard deviation (SD). The difference between any two groups was calculated by ANOVA. The P value less than 0.05 (P0.05), but the cell uptake of T7-RBCSLNs was remarkably increased as the DSPE-PEG2000-T7 molar proportion ranged from 1% to 4%. When continually increased the proportion to 6%, there were no remarkable improvements in cell uptake than that of 4% ratio (P> 0.05). It could be owed to the receptor saturation phenomenon of TfRs. Due to the limited number of receptors and endocytosis recycling, receptor-mediated internalization is a saturated cell pathway, this constraints the cellular uptake quantity of T7-RBCSLNs. In the same way, when increased molar ratio of DSPE-PEG2000-NGR from 1% to 2%, there was a significant 17



increase in cell uptake of NGR-RBCSLNs (Figure S2 B). However, when the DSPE-PEG2000-NGR proportion larger than 2%, C6 cells exhibited no remarkable enhancement in the uptake of Cy5.5-labelled NGR-RBCSLNs (P>0.05). Though the NGR peptide can increase cell uptake, the single receptor-mediating ligands on RBCSLNs will restrict the cell uptake amount of RBCSLNs due to receptor saturation phenomenon. Therefore, a molar proportion of 4% for DSPE-PEG2000-T7 and 2% for DSPE-PEG2000-NGR were chosen for subsequent studies. 3.4. Characterization of Dual-Modified Biomimetic Nanocarriers. The physicochemical properties of the four diverse RBCSLNs are presented in Table 1. The VCR encapsulation efficiency (EE) of the four biomimetic nanoparticles were larger than 50%, and the surface modification with T7 and/or NGR did not influence the final EE. The drug loading yield (DL) of the four formulations were around 2%. As VCR is a hydrophilic drug, which is hard to entrapped into hydrophobic matrix as SLN or PLGA, thus the EE and DL of VCR loaded SLNs here were relatively low. For an ideal nanovessels, particle size is a prerequisite and a key determinant of the fate of nanoparticles, not only in vivo but also in vitro. After the drug encapsulation evaluation, the particle size of prepared RBCSLNs was measured using a dynamic light scattering analyser. As Table 1 demonstrated, mean particle size of the VCR-loaded T7/NGR-RBCSLNs was 123.67 ± 0.65 nm with a polydispersity index of 0.057 ± 0.015, which indicated a narrow size distribution (Figure 2 A). According to the nanoparticles size requirements,32 The nanosize of about 120 nm is suitable for the particles in blood to across into the tissue, reach the surfaces of cells, and alleviate intracellular transportation. As exhibited in Figure 2 B, VCR-loaded T7/NGR-RBCSLNs were uniform and round, with a characteristic core-shell structure. In addition, TEM imaging results were 18


Page 18 of 46

Page 19 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


consistent with that revealed by the dynamic light scattering analyser. To improve targeting efficiency, the natural materials were usually modified with peptides on surface. However, the use of solvents, reagents, and other materials in the modifying processes may alter the targeting characteristics of the related carriers, affecting the their physiological disposition features. Herein, a CD spectrum was applied to evaluate the secondary structure alters of RBC membrane proteins when they were modified with ligands. As displayed in Figure 2 C, results from CD spectrum indicated that the secondary structure of RBC membrane proteins had almost unchanged after the surface modification. It meant that the T7- and NGR-modified RBCSLNs might not alter RBCSLNs’ activities. More works are in need to study such issue in the future. The stability of VCR-laoded T7/NGR-RBCSLNs in physiological conditions is a precondition for their utilization in vivo. Hence PBS contained FBS (10%, V/V) was used to simulate the conditions in vivo. The stability of T7/NGR-RBCSLNs against PBS (37°C) was estimated by the Turbiscan Lab® Expert. Based on the criterions of this instrument33, the transmission showed in Figure 2 D was less than 0.5%, which indicated there were no notable aggregations or sedimentations occurring in the release medium for T7/NGR-RBCSLNs over 60 h, which may be attributed to the slightly negative charge and steric stabilization of RBC membranes. The in vitro drug release of VCR-loaded T7/NGR-RBCSLNs was also carried out to characterize VCR release properties of T7/NGR-RBCSLNs. Compared to the rapid release of free VCR, VCR-loaded T7/NGR-RBCSLNs (Figure 2 E) demonstrated sustained release behaviours, and cumulative VCR release from T7/NGR-RBCSLNs was less than 50% in 48 h, with no initial release burst detected. The

19



incomplete drug release during the 48 hours may be mainly due to the sustained drug release character of SLN. During the initial 10 h, the drugs on the surfaces of SLNs might diffuse from RBC membranes and made the initial release (40%); then the drug release rate slowed down and re-started the release at 52 h (45%); finally the cumulative drug release reached 80% at 72 h. In addition, similar release patterns were observed in the tested four formulations at every time point, implying that ligand modifications did not significantly affect the release behavior of VCR. Overall, the tested four preparations had similar physicochemical characteristics and release profiles, which almost eliminated the other factors (except for surface modification) that influence the in vivo effects of RBCSLNs. 3.5. Cellular Uptake. To investigate whether the cellular affinity of the RBCSLNs change after ligand insertion, different Cy5.5-labelled RBCSLNs were added into model cell contained wells. bEnd.3 cells, the major constituent of the BBB34, were chosen as the TfR-overexpressing cells employed to study the influences of T7. HUVEC cells which over-expressed CD13 were utilized as a tumor-angiogenesis model to reckon the neovasculature targeted power of NGR.35 NGR-RBCSLNs demonstrated insufficient recognition to bEnd.3 cells (Figure 3 A). Therefore, the cell uptake of NGR-RBCSLNs was not ideal according to the CLSM results. The intracellular fluorescence of NGR-RBCSLNs decreased to a similar level as RBCSLNs. In contrast, both T7-RBCSLNs and T7/NGR-RBCSLNs were significantly internalized into bEnd.3 cells compared to NGR-RBCSLNs, implying that the function of T7 on the RBCSLN surface enhances its brain targeting abilities. Cell-carrier binding process (Figure 3 B) was remarkably suppressed by the free T7 peptide (1 mg.mL-1), suggesting the special bind 20


Page 20 of 46

Page 21 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


between T7 modified carriers (T7-RBCSLNs and T7/NGR-RBCSLNs) and bEnd.3 cells via TfRs. Based on the above results, the T7 and NGR dual-modified RBCSLNs may across the BBB via the NGR motif to exert anti-glioma effects. To test this assumption, a CD13-positive HUVEC cell line was employed to assay cell uptake of the RBCSLNs modified with NGR peptides, including NGR-RBCSLNs and T7/NGR-RBCSLNs. As displayed in Figure 3 C, NGR modified RBCSLNs (NGR-RBCSLNs and T7/NGR-RBCSLNs) showed more potent intracellular fluorescence than that of T7-RBCSLNs, indicating that the NGRS on these biomimetic nanoparticles can greatly aid the cell uptake. As we thought, the binding process was notably hindered by surplus free NGR peptides (1 mg.mL-1), for the fluorescence intensity of NGR modified nanoparticles (NGR-RBCSLNs and T7/NGR-RBCSLNs) decreased to a similar level as RBCSLNs (Figure 3 D). These findings suggested that if the CD13 levels in HUVEC declined, functional NGR biomimetic nanoparticles (NGR-RBCSLNs and T7/NGR-RBCSLNs) would ineffectively bind to the target cell through the NGR ligand. Therefore, the cell uptake efficiency of NGR modified nanoparticles was unsatisfactory. C6 cells were employed to estimate the glioma-targeting efficiency of the nanoparticles. As displayed in Figure 3 E and F, both functional T7 biomimetic nanoparticles (T7-RBCSLNs and T7/NGR-RBCSLNs) and functional NGR biomimetic nanoparticles (NGR-RBCSLNs and T7/NGR-RBCSLNs) exhibited notable endocytosis in C6 cells compared to the RBCSLNs. Moreover, among these nanoparticles, T7/NGR-RBCSLNs demonstrated the greatest cell uptake. These results suggest that RBCSLNs whether modified with functional T7/NGR or not greatly influences the tumor homing capacity of RBCSLNs. Above all, the results of cell uptake strongly support our assumption that T7 and NGR are the key factors determining cell recognition and uptake

21



Page 22 of 46

of RBCSLNs in this experiment. 3.6. Cytotoxicity Assay. MTT test was performed to measure the cellular toxicity of different VCR contained nano-carriers against C6 cells. As displayed in Figure S3, the negligible toxicity of the blank RBCSLNs in C6 cells confirmed the safety of the biomimetic nanocarriers. Furthermore, no drug loaded RBCSLNs with modification showed no cytotoxicity over blank RBCSLNs, suggesting that modified RBCSLNs were as safe as RBCSLNs. With the increase of VCR concentrations, the anti-proliferative activities of free VCR significantly increased in C6 cells, offering the anti-glioma effect to the drug load carriers. The free VCR exhibited the strongest cellular toxicity (IC50 was 31.7 ng.mL-1) among the tested formulations, while the difference between free VCR and VCR-loaded RBCSLNs was relatively small at the low VCR concentration range (0.1, 1 and 10 ng.mL-1). Free VCR displayed a more powerful inhibitory action than the various types of VCR-loaded RBCSLNs at the high concentration range (100 and 1000 ng.mL-1). It can be speculated from the above results that the free VCR is internalized into cells through a passive diffusion process, under in vitro high concentration gradient. In contrast, VCR-loaded biomimetic nanoparticles underwent a gradual and slow drug release process (Figure 2 E) after coming into the cytoplasm. Hence, free VCR displayed the more potent cellular-inhibitory action on C6 cells than the other types of VCR-loaded biomimetic nanocarriers. Among these drug-loaded biomimetic

nanocarriers,

increased

cell

uptake

caused

an

expected

improved

anti-proliferation action. It indicates that T7/NGR-RBCSLNs loading VCR could remarkably increase cellular toxicity of VCR (IC50 79.24 ng.mL-1), in comparation with VCR-loaded T7-RBCSLNs (IC50 253.7 ng.mL-1) and NGR-RBCSLNs (IC50 248.1 ng.mL-1). The above results 22


Page 23 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


demonstrate the synergistic action of T7 and NGR on the dual-modified RBCSLNs, promotes their cellular inhibition against C6 line that is positive of TfR and CD13. It further supports the findings on cell uptake of biomimetic nanovessels shown in Figure 3. 3.7. Penetrating Efficiency Investigation. The BBB is considered the first obstacle restricting shift of drugs from the circulation to brain. To investigate the potential penetrating abilities of nanocarriers, a BBB model in vitro was constituted to assay the permeation abilities of T7/NGR-RBCSLNs by imitating in vivo conditions. Figure S4 A exhibited that, four hours incubation later, the ratio of penetrated Cy5.5-labelled T7/NGR-RBCSLNs (3.39 ± 0.34%) and T7-RBCSLNs (3.35 ± 0.28%) were remarkably larger than those of Cy5.5-labelled NGR-RBCSLNs (1.11 ± 0.22%) and RBCSLNs (1.09 ± 0.17%), indicating that T7 possess notable penetrating potential to the BBB model in vitro. In addition, the BBTB is a typical pathological obstruction for the transportation of nanoparticles. To simulate the BBTB better, a HUVEC/C6 cell co-culture model was constituted to evaluate the target specificity and transcellular ability of the prepared RBCSLNs. Figure S4 B displayed that, four hours incubation later, 4.43 ± 0.27% of Cy5.5-labelled T7/NGR-RBCSLNs and 4.45 ± 0.31% of Cy5.5-labelled NGR-RBCSLNs had penetrated the BBTB model, which were notably bigger than that of Cy5.5-labelled T7-RBCSLNs (1.19 ± 0.15%) and RBCSLNs (1.21 ± 0.09%). These results suggest that the functional NGR modified biomimetic nanoparticles significantly influences their BBTB penetrating efficiency (Figure S4 B). Moreover, NGR can easily penetrate the BBTB into the glioma while cannot efficiently across the BBB (Figure S4 A). Overall, the above results suggest that RBCSLNs modified with T7 and NGR possess the potential to penetrate both 23



BBB and BBTB in vitro. 3.8. Inhibitory Effect on In Vitro BBB/Glioma Cells. In order to mimic the in vivo glioma, a model consist of BBB/tumor cells was established. In such co-culture model, the tested various formulations were incubated in the upper transwell inserts, and the survival percentages of C6 cells in the bottom transwell chamber were finally evaluated via a sulforhodamine-B staining analysis. After incubated with free VCR and other VCR-contained formulations, the C6 cells survival percentage was 97.15 ± 12.88%, 91.64 ± 11.43%, 92.32 ± 10.22%, 47.57 ± 4.48%, and 31.47 ± 3.67%, respectively (Figure S5). These results indicate that VCR-loaded T7-RBCSLNs and T7/NGR-RBCSLNs exhibit a remarkable cell-proliferation inhibiting activities by delivery VCR across the BBB and then target to the C6 cells. In contrast, there was almost no free VCR had acrossed the in vitro BBB model. Though free VCR had a stronger C6 cell inhibition action than those VCR-contained RBCSLN formulations in the cytotoxicity test (Figure S3), T7 functionalized RBCSLNs (T7-RBCSLNs and T7/NGR-RBCSLNs) had a stronger C6 cell inhibition than free VCR, it could be due to higher targeting capability of T7 peptides. These findings are in accordance with the results of BBB model in vitro as displayed in Figure S4 A. In short, NGR ligand can improve the targeting ability of RBCSLNs to the glioma after penetrating the BBB. The dual modifications RBCSLNs possess a stronger anti-cancer effect than that of mono-modified RBCSLNs. Among all tested formulations, the T7/NGR-RBCSLNs loading VCR demonstrated the strongest inhibitory effects through facilitating the VCR crossing the BBB and subsequently entering into glioma cells. 3.9. Brain Targeting Test in Zebrafish. After in vitro assessment of penetration 24


Page 24 of 46

Page 25 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


efficiency (Figure S4), the BBB penetrating ability of the above Cy5.5-labelled RBCSLNs formulations were tested in vivo via zebrafish, which possess mammals resembling BBB structures. The Cy5.5 (red) signal distributing pattern in the brain and around the vascellum in the fish, indicates the BBB penetrating capability of the tested objects. Thus, it offers a fast, efficient, and economic model to screen the brain targeting nanoparticles. After the injection, free Cy5.5 with red fluorescence was observed only in the vascellums (Figure 4 A). This implied that free Cy5.5 was restricted in the vessels and was unable to across the BBB into the central nerve system. Additionally, zebrafish injected with Cy5.5-labelled RBCSLNs or Cy5.5-labelled NGR-RBCSLNs showed little fluorescence in brain. In contrast, Cy5.5-labelled T7 functionalized RBCSLNs (T7-RBCSLNs and T7/NGR-RBCSLNs) displayed a high-level cumulation in the zebrafish brain. The fluorescence distributions in the brain and extravascular tissues manifested the existed capability of T7 functionalized RBCSLNs to deliver drugs cross the BBB and come into the zebrafish brain. It can be concluded from the above results that the T7 functionalized RBCSLNs could greatly enhance the brain targeting delivery efficiency in vivo. The results obtained in zebrafish were in accordance with the in vitro results exhibited in Figure S4 A. Further investigations are in need to confirm the actual anti-glioma effects of T7/NGR-RBCSLNs in vivo. 3.10. In vivo Distribution Study. Drug therapy on glioma in clinic is unsatisfactory, mainly owing to the pathophysiological barriers as BBB and BBTB, and the low brain targeting nature of drugs. Special accumulation of nanocarriers in the cancer site would improve the anti-glioma effects of pharmaceutical molecules in brain. In order to evaluate the glioma targeted efficiency of T7/NGR-RBCSLNs in mice, bio-distribution of various 25



Page 26 of 46

Cy5.5-labelled RBCSLNs was photographed via gathering fluorescence signals from mice beard with C6 cells. As displayed in Figure 5, T7 functionalized RBCSLNs (T7-RBCSLNs and T7/NGR-RBCSLNs) were primarily distributed in the brain of mice (mean fluorescent intensities were 132 and 158 pmol·M-1·cm-1, respectively), while other RBCSLNs formulations possessed reduced brain targeting ability (the mean fluorescent intensities were lower than 35 pmol·M-1·cm-1). The trend found in this section was in accordance with the findings observed in Figure 4. To

further

estimate

the

cancer

targeted

ability

of

T7/NGR-RBCSLNs,

immunofluorescence analysis was performed after treatments with different types of Cy5.5-labelled RBCSLNs in glioma mice model. As exhibited in Figure 6, no glioma-site related fluorescence was found in the mice of Cy5.5-labelled RBCSLNs group. Cy5.5-labelled NGR-RBCSLNs displayed little cumulations in the glioma sites. For Cy5.5-labelled T7-RBCSLN mice, fluorescence was distributed throughout the brain, suggesting that T7-RBCSLNs traversed the BBB. Cy5.5-labelled T7/NGR-RBCSLNs demonstrated marginally higher fluorescence than Cy5.5-labelled T7-RBCSLNs in brain, while with special cumulation in the tumor area, suggesting the accurate cancer targeted features of T7/NGR-RBCSLNs. Above results are in accordance with the results of Figure 5, and further confirmed that the T7/NGR-RBCSLNs could cross the BBB and BBTB to realize the glioma targeting delivery. NGR-RBCSLNs had an inferior tumor targeting potential than that of T7 linked RBCSLNs (T7-RBCSLNs and T7/NGR-RBCSLNs), it might be due to insufficient in vivo BBB penetration by NGR-RBCSLNs, while the good cell uptake of NGR-RBCSLNs (Figure S4 B) might originate from the enhanced entries of RBCSLNs into glioma through NGR in BBTB model in vitro.

26


Page 27 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


These results once again demonstrate the superiorities of using dual-modified biomimetic nanocarriers in brain targeted drug delivery. 3.11. In Vivo Anti-Cancer Evaluation. Clinical therapeutic benefits of patients are primarily reflected by the life quality and survival time during and after the cancer treatment. After treated with the various tested formulations, the anticancer effect was evaluated by MRI via measuring the cancer volume. In accordance with the results of in vivo distribution (Figure 6), anti-golioma results demonstrated the remarkable cancer targeting efficacy of VCR-loaded T7/NGR-RBCSLNs in the animal model. As displayed in Figure 7 A, tumor diameters at the day 16 were notably decreased based on MRI in response to the therapy of VCR-contained T7/NGR-RBCSLNs in comparison with those of control groups. Relative tumor inhibition rates for the experiment groups at the day 16 (Figure 7 B) were 97.28 ± 11.72% (free VCR), 96.31 ± 8.49% (VCR-loaded RBCSLNs), 89.35 ± 9.58% (VCR-loaded NGR-RBCSLNs), 59.28± 5.48% (VCR-loaded T7-RBCSLNs), and 41.32 ± 6.48% (VCR-loaded T7/NGR-RBCSLNs), respectively. The above data imply that the anti-golioma effect of T7/NGR-RBCSLNs loading VCR is stronger than that of other preparations in the model animals. To further explore the in vivo antiglioma efficacy of VCR-loaded T7/NGR-RBCSLNs loading VCR, Kaplan-Meier survival curves (Figure 7 C) and relative body weights (Figure S6) of the model mice were plotted and monitored. For the duration of the experiment, the body weight of physiological saline and free VCR groups induced 15%, but the VCR-loaded biomimetic nanoparticle groups exhibited no notable decrease (Figure S6). As exhibited in Figure 7 C, the VCR-loaded T7/NGR-RBCSLNs group had a remarkably prolonged median 27



survival (36 days), which was 2, 1.9, 1.8, 1.7 and 1.2-fold of normal saline (18 days), free VCR (19 days), VCR-contained RBCSLNs (20 days), VCR-contained NGR-RBCSLNs (21 days) and VCR-contained T7-RBCSLNs (29 days), respectively. These results are consistent with the previous findings, suggesting the superiority of T7/NGR-RBCSLNs to the other nanoparticles investigated in the BBB/glioma cell model (Figure S5) and bio-distribution (Figure 6). This also demonstrated the synergistic effects of brain targeting from T7 and NGR. Though VCR-loaded RBCSLNs (without modification) prolonged the median survival time from 18 days to 20 days, no notable difference was found between the VCR-loaded RBCSLNs and normal saline groups, which could be attributed to the insufficient tumor-targeting ability of non-modified RBCSLNs. Due to the weak BBB penetrating efficiency, the median survival time of VCR-loaded NGR-RBCSLNs group was just 21 days, which was higher than the normal saline group. Owing to high brain-targeting efficacy, VCR- poor ed T7-RBCSLNs group displayed a significant prolonged median survival than other groups, which were 1.6 folds of normal saline group. Based on the reports,18, 36 single T7 peptide can guide the nanocarrier cross both BBB and BBTB, and finally enter into the cells of glioma. It suggests that the incorporating T7 onto the surface of RBCSLNs enhances the anti-glioma efficacy in vivo. Histological variations of glioma tissue in response to different treatments were analyzed via HE staining. As displayed in Figure 7 D, the arrangement of tumor cells lose polarities and the structure is disordered; the tumor cells are irregular and their sizes are inconsistent; pleomorphism of cells and a frequent hyperchromasia of the nucleus. Gliomas from the VCR-loaded T7/NGR-RBCSLNs group showed unnatural tissues and cells, displaying a hypocellular and necrotic zone (cell lysised and liquefied, the larger white regions in Figure

28


Page 28 of 46

Page 29 of 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


7 D), whereas tumors from other groups exhibited more hypercellular region and abnormal nucleus polymorphism. Above findings indicate that T7/NGR-RBCSLNs loading VCR exert a stronger antiglioma activity than other formulations. The tendency found in the histological assay was in accordance with the findings of the distribution in vivo (Figure 5 and 6) and the anticancer effect in vivo (Figure 7 A, B and C). The aim to construct a targeted drug carrier is to gain the optimum therapeutic efficiency with acceptable safety features in the clinical application. To estimate the toxicity of VCR-loaded T7/NGR-RBCSLNs, normal mice were injected with physiological saline (normal saline), free VCR, and T7/NGR-RBCSLNs loading VCR. Organs as heart, brain, liver, lung and kidney were removed and treated with HE staining for histological examination. The three tested formulations exhibited negligible organ damage (Figure S7 A), suggesting that none of the treatments cause systemic toxicity at the current VCR dosage. As exhibited in Figure S7 B and C, there was no notable reduction in red blood cell (RBC) or white blood cell (WBC) in mice with the treatment of free VCR or VCR-loaded T7/NGR-RBCSLNs. In Figure S7 D, results of mean corpuscular volume (MCV) revealed that MCV in free VCR group was notably smaller than that of the other two groups (P

Dual-Modified Novel Biomimetic Nanocarriers Improve Targeting and

Recommend Documents