Methotrexate-Loaded Extracellular Vesicles Functionalized with

Mar 22, 2018 - Thus, successful modification of drug delivery and novel therapeutic strategies are needed to overcome this obstacle. Extracellular ...
0 downloads 9 Views 4MB Size
Subscriber access provided by - Access paid by the | UCSB Libraries

Biological and Medical Applications of Materials and Interfaces

Methotrexate-loaded extracellular vesicles functionalized with therapeutic and targeted peptides for the treatment of glioblastoma multiforme Zhilan Ye, Tao Zhang, Wenshan He, Honglin Jin, Cuiwei Liu, Zhe Yang, and Jinghua Ren ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b18135 • Publication Date (Web): 22 Mar 2018 Downloaded from http://pubs.acs.org on March 22, 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 24 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

Page 1 / 24

Methotrexate-loaded Extracellular Vesicles Functionalized with Therapeutic and Targeted Peptides for the Treatment of Glioblastoma Multiforme Zhilan Ye1,2,#, Tao Zhang1, #, Wenshan He3, Honglin Jin1,Cuiwei Liu1, Zhe Yang4, Jinghua Ren1, *

1

Cancer Center, Union Hospital, Tongji Medical College of Huazhong University of Science

and Technology, Wuhan, China 430022

2

Research Center for Tissue Engineering and Regenerative Medicine, Union Hospital, Tongji

Medical College of Huazhong University of Science and Technology, Wuhan, China 430022

3

Department of Breast and Thyroid Surgery, Union Hospital, Tongji Medical College of

Huazhong University of Science and Technology, Wuhan, China 430022

4

Hubei Collaborative Innovation Center for Advanced Organic Chemical Materials; Ministry

of Education Key Laboratory for the Green Preparation and Application of Functional Materials, Hubei University, Wuhan, China 430022

#These authors contribute equally to this work *Correspondence to: Jinghua Ren, [email protected]

ACS Paragon Plus Environment

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

Page 2 of 24

Page 2 / 24

ABSTRACT Despite promising in-vitro evidence for effective glioblastoma treatment, most drugs are hindered from entering the central nervous system due to the presence of the blood-brain barrier (BBB). Thus, successful modification of drug delivery and novel therapeutic strategies are needed to overcome this obstacle. Extracellular vesicles (EVs), cell-derived membrane-encapsulated structures with diameters ranging from 50-1000 nm, have been explored as drug delivery system to deliver their cargo to the brain tissue. Moreover, tumor targeting and selective drug delivery has been facilitated by engineering their parent cells to secrete modified EVs. However, the method suffers from many shortcomings including poor repeatability, complex and time-consuming operations. In this context, we present an easy-to-adapt and highly versatile methodology to modify EVs with an engineered peptide capable of recognition and eradication of glioma. Based on molecular recognition between phospholipids on EVs lipid bilayer membranes and ApoA-I mimetic peptide, we have developed methotrexate (MTX) -loaded EVs functionalized with therapeutic (KLA) and targeted (LDL) peptide. In vitro experiments demonstrated EVs decorated with LDL or KLA-LDL could obviously ameliorate their uptake by human primary glioma cell line U87 and permeation into 3D glioma spheroids in contrast to blank EVs, and consequently the treatment outcome of the payload is improved. Both ex vivo and in vivo imaging experiments revealed that peptide LDL could obviously promote EVs extravasation across BBB and distribution in glioma site. Furthermore, compared with the mice administrated with MTX and MTX@EVs, MTX@EVs-KLA-LDL-treated mice showed the longest median survival period. In conclusion, functionalizing with peptide onto EVs surfaces may provide a substantial advancement in the application of EVs for selective target binding as well as therapeutic effects for brain tumor treatment.

KEYWORDS: extracellular vesicles, surface modification, drug delivery, 4F-LDL peptide, apoptotic peptide, cancer therapy. INTRODUCTION Glioblastoma multiforme (GBM) is the most common intracranial tumor with extremely poor prognosis. Despite advanced treatment, the median survival for patients with glioblastoma is ACS Paragon Plus Environment

Page 3 of 24 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

Page 3 / 24

about 15 months.1-2 Diffuse infiltrating growth of GBM makes complete surgical resection almost impossible, and an inevitable relapse always follows.3 The standardized treatment for GBM is adjuvant chemotherapy and radiotherapy after surgery.4 Although many existing pharmaceuticals have demonstrated their effectiveness in the management of GBM in vitro, therapeutic benefit of chemotherapy is often less effective owing to poor penetration through the blood-brain barrier (BBB) and weak infiltration into tumors.5-6 To achieve required concentration of drugs in tumor, high-dose administration is necessary, which inevitably leads to general toxicity and thus restricting therapeutic application in the clinic. Taken together, there is an urgent need to develop multi-functional nanocarrier, which should have the following characteristics: delivery drug, pass through the BBB and target tumor cells.7 Within the past decade, extracellular vesicles have emerged as prospective drug carriers for GBM therapy. EVs are vesicles with a heterogeneous diameter of 50-1000 nm that released by most cell types.8-9 By outward budding and shedding from the plasma membrane, EVs could be secreted to the extracellular environment and isolated from cell supernatants and bodily fluids.10 These membrane-derived vesicles represent important mediators of intercellular communication by transporting their contents to recipient cells.9-10 In view of their inherent property for transferring encapsulated cargoes, the exploitation of EV-based therapy is, therefore, of great interest. Over the past decade, EVs have opened a completely new paradigm for delivery of a wide range of therapeutic agents, varying from drug11, gene12 to peptide-protein13. Unarguably, as natural endogenous nanoparticles, EVs have the advantages of immune-tolerance and stability in circulation systems. Importantly, it has been elegantly demonstrated that cell-derived EVs allow the delivery of chemotherapeutic agents across BBB, which has always been quite a huge challenge in drug delivery.14 However, clinical application of EVs for central nervous system tumors is largely compromised by its insufficient targeting ability. Peptides are excellent candidates for targeting molecules because of fainter immunogenicity, small-scale structures, simple process and low fabrication costs.15 In order to direct EVs to glioma cell populations, it is feasible to modify the surface of EVs, which could help EVs to specifically identify tumor cells by receptor-ligand binding. The first

ACS Paragon Plus Environment

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

Page 4 of 24

Page 4 / 24

demonstrations of receptor-mediated extracellular vesicle delivery to CNS were achieved by Alvarez-Erviti L et al. using genetic manipulation of dendritic cells as donor.16 In this system, pre-engineered donor cells enabled the expression of neuron-specific RVG peptide on surfaces of vesicles. These RVG-targeted vesicles can deliver therapeutic siRNA to the brain and silence specific gene. Thus, peptide modified vesicles have particular superiorities for tissue-specific drug delivery. Nevertheless, the method suffers from some drawbacks such as poor repeatability, time-consuming and laborious operations and low output. In the present work, we report a simple and versatile approach to functionalize drug-loaded EVs with the therapeutic peptide, KLA, and the targeted peptide, LDL, for selective binding to low-density lipoprotein receptor (LDLR) overexpressed on the BBB and GBM cell lines.17-18 The strategy is based on surface property of EVs that these particles are phospholipid-rich.19 The ApoA-I mimetic peptide (L-4F) facilitates the association of EVs and therapeutic/targeted peptide, which represents an alternative to modify the EVs by binding to the phospholipid vesicles. The functionalized EVs prompt the process of membrane receptor-mediated internalization both in vitro and in vivo, which improved the transport of proapoptotic peptide KLA and MTX to U87 glioma. Thus, functionalizing with peptide onto EVs surfaces may provide a unique opportunity to deliver anticancer drugs and therapeutic peptide selectively to brain tumor.

MATERIALS AND METHODS

Materials. The 4F-LDL peptide Ac-DWFKAFYDKVAEKFKEAF-GG-RLTRKRGLKLA and 4F-KLA-LDL peptide Ac-DWFKAFYDKVAEKFKEAF-GG-d (KLAKLAKKLAKLAK) -GG-RLTRKRGLKLA were synthesized by AB Biochem Co., Ltd (Shanghai, China) at a purity of greater than 95% as determined by HPLC. MTX was obtained from Sigma (USA). Annexin V-FITC/ PI apoptosis kit was supplied by Beyotime® Biotechnolo-gy Co.Ltd (Nantong, China). 1, 1-dioctadecyl-3,3,3,3-tetra-methylindotricarbocyanine iodide (DiR) was ordered from AAT Bioquest®, Inc (USA). PKH26 was from Sigma (USA). CFSE was from Abcam (Britain). Dulbecco's modified eagle's medium (high glucose) (DMEM), fetal bovine serum (FBS), trypsine-EDTA (0.25%), penicillin streptomycin and agarose were purchased

ACS Paragon Plus Environment

Page 5 of 24 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

Page 5 / 24

from Gibco (Invitrogen, USA). The mouse fibroblast cell line L929 and human primary GBM cell U87 were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). The cells were cultured at 37 °C, 5% CO2 in DMEM containing 10% FBS, 100 U/ml penicillin and 100 µg/ml streptomycin. BALB/c nude mice (male, 4-5 weeks old, 18-22 g) were purchased from HFK BIOSCIENCE Co, LTD (BeijingÈChina) and maintained under standard housing conditions. All animals involved in this study were operated in accordance with protocols evaluated and approved by the ethics committee of Tongji Medical College. Preparation of EVs. EVs were prepared as described.11 Briefly, L929 cells were exposed to ultraviolet irradiation (UBV, 300 Jm-2) for 1 h, and then 50 µg/ml chemotherapeutic drug MTX was added to cell supernatant for the formation of MTX-packaged EVs. After 16 h, the cell culture supernatant was collected and centrifuged for 10 min at 600 g to get rid of cells, and the collected supernatant was centrifuged for 2 min at 14,000 g to remove debris. At last, the supernatant was centrifuged for 60 min at 14,000 g to harvest MTX-encapsulating EVs (MTX@EVs). The purified EVs conjugated with LDL peptide (EVs-LDL) or KLA-LDL peptide (EVs-KLA-LDL) by simple co-incubation. Briefly, 20 µM LDL peptide or 20 µM KLA-LDL peptide was added into EVs-suspension firstly; secondly, the mixture was lightly agitated at room temperature (RT) for 3h,20 and stored at 4 °C for 24h; Last, the products were centrifuged at 14,000 g for 60 min to remove unconjugated peptide, and then were eluted with PBS by centrifugation for three times. Physicochemical Characterization of Evs. The particle size and zeta potential of EVs in PBS or PBS containing 20% serum were identified by a dynamic light scattering detector (Zetasizer Nano ZS, Malvern, UK). The morphology of EVs was investigated by a transmission electron microscope (TEM) (H-7000FA, Hitachi, Japan) following negative staining with 2% phosphotungstic acid. The Efficiency of Peptide Conjugation. The supernatant containing free peptide LDL or KLA-LDL was collected and then analyzed using High-performance liquid chromatography (HPLC) with 5-90% CH3CN/H2O containing 0.1% TFA over 30 min, C18 column, 190-300 nm. The peptide conjugation efficiency was calculated using the formula:

×100%

ACS Paragon Plus Environment

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

Page 6 of 24

Page 6 / 24

Encapsulation Efficiency (EE) and Drug Loading Capacity (DLC). The amount of MTX in MTX@EVs was measured by HPLC.11 Briefly, MTX@EVs were treated by lysis buffer, proteinase K, phenylmethylsulfonyl fluoride and DNase I according to previous description.11 The experimental conditions of HPLC were as follows: Diamond C18 column (150 mm×4.6 mm, pore size 5 µM), the mobile phase: CH3CN:H2O (1:1, v/v), flow rate 1.0 ml/min, and detection wavelength 304 nm. The EE% and DLC% were calculated as indicated below: !"#

$

!"#

%

×100%; &'

!"#

$

%

( )* +) , )*-

. /001

Cell Uptake of Evs. For EVs uptake experiments, fibroblast-derived EVs were labeled with PKH26 according to the manufacturer’s protocol of PKH26 Fluorescent Cell Linker Kit. In brief, 25 µg of EVs were re-suspended in 1 ml of diluentÈand then 2 µL of PKH26 was added into the diluent with incubation at room temperature for 5 min. After the reaction, 2 ml 0.5% BSA/PBS was put in the liquid to neutralize the surplus dye. The PKH26 stained EVs were washed twice with PBS at 14,000 g for 1 hour, and then collected for the experiments of cell uptake. U87 cells were seeded onto confocal dish at 2×104/well, and allowed to grow for 24 h. Thereafter, the cells were incubated with PKH26-labeled EVs, EVs-LDL or EVs-KLA-LDL in DMEM containing 5% FBS for 12h, and U87 cells should be pretreated with or without peptide LDL (2 µM) for 4h in the EVs-LDL group. In the end, the cells were stained by CFSE (2.5 µM) at room temperature for 20 min, and washed three times with PBS before observed with confocal microscopy (Olympus BX41F, Japan). Detection of Apoptosis. U87 cells were seeded at the density of 6×104 cells/well in 6-well plates. Following overnight incubation, cells were exposed to MTX, KLA-LDL and different drug-packaging EVs formulations in DMEM containing 5% FBS for 48 h. MTX and KLA-LDL concentration were 0.5 µg/ml and 0.4 µM respectively. For qualitative analysis, U87 cells were stained with DAPI as described previously. After staining, the cells were examined and photographed using a fluorescence microscope. For quantitative analysis, U87 cells were seeded at a density of 105 cells per well into 6-well plates. After incubation for 24 h, fresh media (as a control), MTX, MTX@EVs, KLA-LDL, EVs-KLA-LDL and MTX@EVs-KLA-LDL in DMEM containing 5% FBS were added into the plates with the final MTX at a dose of 0.5 µg/ml in the groups of MTX, MTX@EVs and MTX@EVs-KLA-LDL, and KLA-LDL at a dose of 0.4 µM in the groups of KLA-LDL, ACS Paragon Plus Environment

Page 7 of 24 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

Page 7 / 24

EVs-KLA-LDL and MTX@EVs-KLA-LDL. 48 hours after the treatments, cells were isolated and stained with annexin V-FITC and propidium iodide, and analyzed using flow cytometry (Canto II, BD Company, USA). EVs Penetration into Glioma Spheroid. The 3D glioma spheroids of U87 cells were developed as previously described.21 In brief, 2000 cells in 500 µL of medium was added to each well of a 48-well culture plate which was pre-coated with 200 µL of 2% low-melting-temperature agarose in free DMEM. Subsequently, the culture plates were allowed to agitate gently and cultured at 37 °C with 5% CO2 for 7 days. The homogeneous and tight glioma spheroids were chosen to incubate with PKH26-labeled EVs, EVs-LDL or EVs-KLA-LDL for 12 h with the final LDL and KLA-LDL at same concentration 1.6 µM. Finally, they were washed with PBS to remove unconjugated peptide and fixed in 4% paraformaldehyde before detection by confocal microscopy. Inhibitory Effect on Glioma Spheroid Growth. To evaluate inhibitory effect on tumor growth, the uniform spheroids were treated with 500 µL of 5% FBS DMEM containing MTX, KLA-LDL, MTX@EVs, EVs-KLA-LDL or MTX@EVs-KLA-LDL, with the final MTX and KLA-LDL concentration were 2 µg/ml and 1.6 µM respectively. And the glioma spheroids incubated with drug-free DMEM served as control. For the next 8 days, the major (dmax) and minor (dmin) diameters of glioma spheroid were measured every two days. The computing method of glioma spheroids volume was the following formula as V = 1/2×dmax×dmin2, whose change could reflect the growth inhibition of each group. In Vivo Evaluation of Brain Targeting of EVs-KLA-LDL. To study the brain distribution profiles of EVs in the BALB/c mice, the blank EVs and EVs-KLA-LDL were labeled with near infrared dye DiR.22 Briefly, the EVs and EVs-KLA-LDL were incubated with 1 µM DiR for 20 minutes, then were washed with 10 ml PBS. Each mouse received DiR-stained EVs or EVs-KLA-LDL through tail vein injection. At 1 h, 12 h, 24 h, 48 h post injection, the mice were sacrificed and the brains were collected to visualize under the IVIS spectrum imaging system (Bruker, GBR). Biodistribution of EVs-KLA-LDL in mice with glioma. The orthotopic model of glioma-bearing mice was established according to previous reports.23 Briefly, under the mouse adaptor stereotactic fixation device, U87 cells (1×105 in 5 µL PBS) were implanted

ACS Paragon Plus Environment

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

Page 8 of 24

Page 8 / 24

into the right striatum (1.8 mm lateral and 4 mm of depth) of the BALB/c nude mice. In order to examine whether EVs-KLA-LDL specifically bind to tumors in vivo, the EVs and EVs-KLA-LDL were labeled with DiR and then intravenously injected into glioma-bearing mice via the tail vein. The in vivo fluorescence imaging of the biodistribution of EVs or EVs-KLA-LDL were captured under the IVIS spectrum imaging system at 2, 6, 12 and 24 h post EVs injection. Twenty-four hours injection, brain, heart, liver, spleen, lung and kidney were harvested to analyze the distribution of EVs and EVs-KLA-LDL in the main organs by fluorescence imaging. Anti-glioma Effect and Toxicity Evaluation. U87luci-bearing glioma xenograft was developed as above. In order to evaluate the anti-glioma effect of different treatmentÈmice were

intravenously

injected

with

saline

(control),

MTX,

MTX@EVs,

and

MTX@EVs-KLA-LDL at days 7, 10, 13, 16 through tail vein. The dose of MTX was 5 mg/kg. To assess the tumor growth, the non-invasive bioluminescence imaging (BLI) were performed on days 6 and 27Èand tumor bioluminescence was quantified by measuring pseudocolor intensity. On days 27 post tumor injection, the main organs were harvested and stained with hematoxylin and eosin and the sera of the mice collected by centrifugation were used for blood chemistry tests. The survival time of mice in the four groups were recorded.

Statistical Analysis. All data were presented as mean ± standard deviation. The statistical differences were determined by student's T-test or one-way ANOVA. All analyses were carried out using SPSS 15.0. Statistical significance was set at *p< 0.05, and extreme significance was set at **p< 0.01.

RESULTS AND DISCUSSION

Manufacture and Characterization of EVs. As naturally-equipped nanocarriers, EVs have exhibited unique advantages as drug carrier, which can be utilized better through surface modification or cellular engineering. However, the surface modification that based on the intrinsic virtue of EVs has not been extensively explored. Intriguingly, EVs derived from cells are enriched with phospholipids on membrane surface, which represents a high level of stability and creates the opportunity for peptides conjugation.24 To take advantage of

ACS Paragon Plus Environment

Page 9 of 24 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

Page 9 / 24

this character, here we functionalized EVs with a synthetic multifunctional peptide which contains apoA-I mimetic peptide 4F (DWFKAFYDKVAEKFKEAF), ApoB LDLR binding domain (RLTRKRGLKLA) and therapeutic KLA (KLAKLAKKLAKLAK) for GBM treatment. First, cell-derived drug-encapsulating EVs were prepared. For the purpose of minimizing tumorigenic potential, mouse L929 fibroblastic cells were selected as source for EVs product in this study. As previously reported, encapsulating MTX into EVs was achieved by treatment of donor tumor cells with drug incubation followed by irradiation. Irradiation with ultraviolet light was applied to induce the donor cells apoptosis, and prompt EVs release. The release of EVs was triggered by irradiation and the average EVs yield was up to 40 µg from 100 ml (5 ×107 cells) of culture supernatant. Then, the complexation of L-4F with phospholipid enables the conjugation of functional peptide KLA-LDL to the purified EVs by simple incubation. TEM photographs illustrated that blank-EVs and all constructed EVs were spherical in shape and bilayer membrane structure (Fig. 1A). The particle size of these EVs observed by TEM was heterogeneous, with mean size of 200-300 nm, which was consistent with the range previously reported for typical EVs.25 The analysis of dynamic light scattering detector demonstrated that the size of EVs and MTX@EVs were 303.0 ± 12.9 nm and 325.8 ± 15.6 nm, and exhibited 307.6 ± 16.4 nm and 318.3 ± 15.5 nm after KLA-LDL conjugation respectively (Fig. 1B), which revealed that encapsulation of MTX or conjunction the peptide did not significantly influence the particle size (P > 0.05). The zeta potential of various EVs groups were about -10 mV (Fig. 1C), indicating that MTX and the synthetic peptide did not change the negative surfaces. The negative-charged EVs are beneficial for reducing the undesirable RES clearance. Moreover, the particle sizes and diameter distributions of MTX@EVs-KLA-LDL were stable without aggregation in PBS containing 20% serum (Figure S1). These results showed that MTX@EVs-KLA-LDL have good blood plasma stability. These results are not surprising, because there was no apparent change in the surface properties of the cell-derived EVs, whose stability in the blood is ideal, after drug loading and surface modification. The conjugation efficiency of LDL and KLA-LDL were approximately 1.98% and 1.77%, respectively (Figure S2-3). The amount of MTX encapsulated in EVs was analyzed by HPLC, which confirmed that the EE and the DLC values of MTX for EVs were 1.4 ± 0.2% and 5.1 ± 0.5%, respectively.

ACS Paragon Plus Environment

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

Page 10 of 24

Page 10 / 24

Figure 1. Characterization of EVs. (A) Representative TEM characterization of EVs. The scale bar corresponds to 100 nm in the images. (B) Size distribution and (C) zeta potential of EVs, MTX@EVs, EVs-KLA-LDL, and MTX@EVs-KLA-LDL.

Cell Uptake of Evs. The low density lipoprotein (LDL) receptor has been confirmed upregulated on both the BBB and glioma cells. There is a relatively low expression in normal brain tissues, which therefore makes LDLR an attractive therapeutic target for the therapy of glioma.17-18 Accordingly, a synthetic peptide containing the LDL receptor (LDLR) binding domain of apoB 100 was introduced onto the surface of EVs for targeting both the BBB and glioma. To study the internalization of fibroblast-derived EVs by human glioma cells, CFSE-labeled U87 cells were administered with PKH26-labeled EVs for 12 h and cellular uptake of fibroblast-derived EVs was assessed by confocal microscopy. As shown in Fig. 2, PKH26 red fluorescence intensity of U87 cells treated with PKH26-labeled EVs-LDL or EVs-KLA-LDL was obviously stronger than those treated with EVs, suggesting that the uptake of EVs by LDLR over-expressed U87 cells was improved by the peptide modification on EVs surface. However, if U87 cells were pretreated with peptide LDL in the EVs-LDL group (pre-LDL), red fluorescence intensity of U87 cells is weaker than that of not being pretreated with, which further demonstrate the targeting of EVs-LDL. More intriguingly, EVs-KLA-LDL treatment results in rounded morphological changes of the cells. Following treatment with EVs-KLA-LDL, the cells were then stained with viability dye CFSE. From Fig. 2A, a number of cells that uptake EVs-KLA-LDL showed less viability dye CFSE (green) than cells that uptake EVs-LDL, which confirmed that the therapeutic peptide KLA produced cytotoxicity. These results clearly indicated the potential of membrane-targeting modification for enhancing the cellular delivery of EVs, and also demonstrated antitumor capabilities of coupling of KLA to a selective targeting domain onto EVs.

ACS Paragon Plus Environment

Page 11 of 24 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

Page 11 / 24

Figure 2. Uptake of EVs by U87 cells. (A) Fluorescence images for PKH26-labeled EVs, EVs-LDL (pre-LDL), EVs-LDL and EVs-KLA-LDL in CFSE-stained U87 cells and (B) the corresponding quantification of red fluorescence intensity of PKH26. Red: EVs labeled by PKH26. Green: U87 cells stained by CFSE. The scale bar corresponds to 50 µm in the images.

Detection of Apoptosis. The morphological changes of the nucleus can monitor the apoptosis of

the

cells.

Specifically,

cells

undergoing

apoptosis

can

be

identified

by

chromatin gathered toward the center and the formation of nuclear fragment. As shown in Fig. 3A, the cells treated with free MTX (0.5 µg/ml) or KLA-LDL (0.4 µM) peptide showed moderate apoptosis. MTX-loaded EVs and KLA-LDL-modified EVs treatment groups possessed obviously more apoptotic nuclei. The U87 cells treated with MTX-loaded EVs functionalized with therapeutic and targeted peptides showed convoluted nuclei with obvious cavitation and fragmentation. For quantitative analysis, apoptosis was determined with FITC-Annexin V/PI assay using flow cytometry. Compared with free MTX, MTX@EVs induced apoptosis more efficiently and the apoptosis percentage increased from 8.73% to 24.17% (Fig. 3B). Consistent with morphological changes, EVs-KLA-LDL induced significant apoptosis compared with the same concentration of KLA-LDL. In the EVs-KLA-LDL group, 31.28 % of U87 cells were observed to be apoptotic, whereas it was ACS Paragon Plus Environment

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

Page 12 of 24

Page 12 / 24

23.01% of KLA-LDL treated U87 cells. Taken together, it could be concluded that both drugs and therapeutic peptides within EVs are more efficient than their free form. The increase in the cytotoxicity might be due to the encapsulation of drugs into EVs, which could shield drugs

from

being

pumped

out

by

the

efflux

transporters.

Moreover,

the

MTX@EVs-KLA-LDL group, whose apoptosis percentage was 49.32%, engendered the strongest cytotoxicity against U87 cells among all MTX formulations group, arguing that targeted modification of EVs can increase the cellular uptake of drug-encapsulating EVs, then cause a more remarkable effect of promoting tumor cells apoptosis.

Figure 3. The morphological changes of U87 cells nuclei (A) and the flow cytometry results (B) of the apoptosis of U87 cells after treatments with MTX, MTX@EVs, KLA-LDL, EVs-KLA-LDL and MTX@EVs-KLA-LDL. U87 cells that treated with PBS served as control. The scale bar corresponds to 20 µm in the images.

EVs Penetration into Glioma Spheroid. It was well documented that three dimensional (3D) multicellular spheroids provide more physiologically relevant models compared to two dimensional (2D) cell culture, as they better represent multicellular composition, obstacle of drug transport, tissue hypoxia and limited internal nutrition in interior of tumor.26 And thus

ACS Paragon Plus Environment

Page 13 of 24 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

Page 13 / 24

tumor spheroid models are increasingly being used for drug discovery in oncology.27 To evaluate the capability of EVs penetration through tumor tissues, we cultured U87 glioma 3D spheroids. At 7 days post inoculation, the U87 spheroids turned to be dense and uniform. To assess the penetrating ability of PKH26 (red) -labeled EVs, U87 glioma spheroids were incubated with EVs, EVs-LDL or EVs-KLA-LDL for 12 h, and then observed under confocal microscopy (Fig. 4). For quantitative analysis, the depth of penetration into the glioma spheroids of EVs was around 70 µm, while it was around 117 µm and 117.5 µm for that of EVs-LDL and EVs-KLA-LDL, respectively. These results indicated that EVs modified with targeting peptide could increase the uptake by U87 cells as well as augment their permeation capacity into tumors.

Figure 4. The permeation of EVs into glioma spheroid after incubation for 12 h. PKH26-labeled EVs (A, B), EVs-LDL (C, D) and EVs-KLA-LDL (E, F). Image A, C and E: ACS Paragon Plus Environment

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

Page 14 of 24

Page 14 / 24

EVs penetration in X, Y, Z axes. Image B, D and F: multilayer scans of the penetrating ability of EVs starting from the top of glioma spheroid every 20 µm.

Inhibitory Effect on Glioma Spheroid Growth. The penetrating superiority of LDL peptide modified EVs would produce stronger growth inhibition on glioma spheroids. To verify this hypothesis, the glioma spheroids were treated with MTX, MTX@EVs, KLA-LDL, EVs-KLA-LDL and MTX@EVs-KLA-LDL with final MTX concentration 2 µg/ml and KLA-LDL concentration 1.6 µM. Then we measured the size of every glioma spheroid every other day under a microscope. As shown in (Fig. 5A), the volume of the glioma spheroid without any treatment continued to increase during 8 days, and the volume of it on days 8 became about 3.97 times of the primary spheroid. By comparison, the glioma spheroids of other groups grew slowly in different degrees. From Fig. 5B, the volumes of glioma spheroids on days 8 were 2.89, 1.95, 1.70, 1.50 and 0.99 times of the primary volume for MTX, KLA-LDL, MTX@EVs, EVs-KLA-LDL and MTX@EVs-KLA-LDL group, respectively. This study demonstrated that both drug within EVs and targeted modification of drug-packaging-EVs could kill more tumor cells compared to drug alone, as well as predicting the tendency of delivery and cytotoxicity of EVs-based drug delivery system in vivo experiments.

ACS Paragon Plus Environment

Page 15 of 24 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

Page 15 / 24

Figure 5. (A) Time-lapse images of the development of the spheroid at days 0, 2, 4, 6 and 8 post treatments with drug-free DMEMÈMTX, MTX@EVs, KLA-LDL, EVs-KLA-LDL and MTX@EVs-KLA-LDL, at MTX concentration 2 µg/ml and KLA-LDL concentration 1.6 µM. Scale bar represents 200 mm. (B) Quantification of results shown in (A). Glioma spheroids without any MTX and KLA-LDL treatments served as control. ap < 0.05, compared with

ACS Paragon Plus Environment

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

Page 16 of 24

Page 16 / 24

control; bp < 0.05, compared with MTX; cp < 0.05, compared with KLA-LDLÈdp < 0.05, compared with MTX@EVsÈep < 0.05, compared with EVs-KLA-LDL.

In Vivo Evaluation of Brain Targeting of EVs-KLA-LDL. It has been demonstrated that cell-derived EVs facilitate the delivery of chemotherapeutic agents across BBB, but tumor-targeting ability of EVs warrants further exploration. To overcome this issuekwe propose a method of targeting modification of EVs to promote the penetration across BBB. Ex vivo DiR-fluorescence of the brains were obtained after intravenous injections of DiR-labeled EVs or EVs-KLA-LDL. As shown in Fig. 6A, the DiR intensity for the EVs-KLA-LDL group was significantly stronger than that for the EVs group at 1 h, 12 h, 24 h and 48 h post-injection. The ratio of area under the curve (AUC0-t) of EVs-KLA-LDL to EVs was 1.13 from the semi-quantitative fluorescence intensity-time curve (Fig. 6B). In this work, we first confirm that EVs-KLA-LDL could cross the BBB and penetrate the brain more efficiently than blank-EVs, which might be attributed to the interaction between LDL peptide and LDLR over-expressed at the BBB.

Figure 6. Ex vivo DiR-fluorescence of the brains after intravenous injections of DiR-labeled EVs or EVs-KLA-LDL. (A) Ex vivo DiR-fluorescence of the brains at various time points (1h, 12h, 24h, 48h) following intravenous administration of DiR-labeled EVs (upper row) and EVs-KLA-LDL (lower row). (B) The corresponding semi-quantitative analysis of the fluorescent intensity.

ACS Paragon Plus Environment

Page 17 of 24 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

Page 17 / 24

Biodistribution of EVs-KLA-LDL in mice with glioma. The targeting efficiency of DiR-labeled EVs- KLA-LDL on glioma-bearing mice was investigated with the vivo imaging as shown in Fig. 7A. DiR-labeled EVs-KLA-LDL and EVs rapidly distributed all over the body after injection. More importantly, an obvious stronger fluorescence of DiR was observed in the brain of EVs-KLA-LDL group compared with EVs group at the serial time up to 24 h. The signal in the brain did not disappear by 24 h, which was consist with the ex vivo imaging of EVs-KLA-LDL (Fig. 6A). To further investigate the systemic location of these two formation EVs, we removed the major organs 24 h after injection. The ex vivo imaging of the tumor-bearing brains exhibited that the mean fluorescence intensity of the EVs-KLA-LDL were higher than that of EVs in glioma region (Fig. 7C), which revealed that EVs-KLA-LDL distributed more than EVs in glioma related tissues. However, the biodistribution of EVs-KLA-LDL in other organs was almost the same as EVs including heart, liver, spleens, lung and kidney (Fig. 7B). From these data, we could infer that the constructed targeting peptide facilitate the enrichment of EVs in the glioma in vivo.

Figure 7. In vivo imaging of the biodistribution of EVs and EVs-KLA-LDL after injection. (A) The fluorescent image of the experimental animals at the different time points as indicated in vivo. (B) Ex vivo imaging of other major organs including liver, spleen, kidney, heart and lung at 24 h. (C) Ex vivo imaging of the brains harvested at 24 h. (D) Semi-quantitative analysis of different organs fluorescent intensity at 24 h after injection.

ACS Paragon Plus Environment

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

Page 18 of 24

Page 18 / 24

Anti-glioma Effect and Toxicity Evaluation. Stable expression of the enzyme luciferase in U87luci, allowing non-invasive and cost-effective intracranial BLI, was used to monitor intracranial growth and survival in many research fields.28 Accordingly, we use the U87luci-bearing glioma model to perform our experiments. Tumor growth was monitored using in vivo quantitative bioluminescence (Fig. 8A). The results in Fig. 8B showed individual bioluminescence images of the brain tumors from the four experimental groups (Saline, MTX, MTX@EVs and MTX@EVs-KLA-LDL), which were taken at 7 and 27 days following the intracranial implantation of U87luci cells. Normalized to days 7, BLI signals from the animals with saline treatment and MTX solution treatment were 5.5-fold and 4.7-fold increase on days 27, respectively. Meanwhile, the animals administrated with MTX@EVs displayed slower tumor growth, which is about 1.8-fold increase (Fig.8C). This indicated that EVs facilitate the delivery of MTX across BBB, and inducing the promotion of glioma therapy. Further, the fluorescence intensity of MTX@EVs-KLA-LDL group was significantly decrease, which was mainly contributed by that MTX-loaded EVs were functionalized with KLA-LDL peptide, and then revealed the best therapeutic efficacy. In addition, the survival of U87 glioblastoma-bearing mice administered with different treatments was studied. Life-span extension treated with a multi-formulation of MTX on days 7, 10, 13 and 16 after glioma implantation was shown in Fig. 8D. Contrasted with control group, the survival time of the mice with MTX@EVs-KLA-LDL treatment was significantly longer (P < 0.05). However, free MTX treatment did not contribute to prolong the survival time (P > 0.05). Using the log-rank test, the mid-survival period for control group, MTX, MTX@EVs and MTX@EVs-KLA-LDL group were 28 days, 30 days, 41 days, and 48 days, respectively. To further analysis the toxic side-effects of main tissues including heart, liver, spleen, lung, and kidney, hematoxylin and eosin (HE) staining was carried out. As shown in Fig. 9A, histological assessment did not reveal any apparent histopathological abnormalities or lesions in all groups. Moreover, we collected the blood sera of the mice receiving different treatments for blood chemistry tests. Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) are commonly used as indicator of liver function, and blood urea nitrogen (BUN) and creatinine (CRE) are closely related to the kidney function of mice. The results showed that AST and CREA reflecting organ function impairment of the MTX-treated mice were higher than the control group. However MTX@EVs or MTX@EVs-KLA-LDL did not cause abnormal functions of these organs (Fig. 9B). Taken together, it suggested that

ACS Paragon Plus Environment

Page 19 of 24 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

Page 19 / 24

EVs, a safe and effective vehicle for drug deliver, did not lead to toxicity to the major organs in mice.

Figure 8. Therapeutic efficacy of glioma after intravenous administration of different MTX formulations. (A) Schematic representation of tumor growth in the orthotopic implantation model. Mice were treated with saline, MTX, MTX@EVs, and MTX@EVs-KLA-LDL on days 7, 10, 13, 16, and sacrificed on days 27 post tumor injection. (B) Representative BLI images of U87-luci-bearing mice with treatments of MTX, MTX@EVs and EVs-KLA-LDL on days 7 and 27. Control groups received saline only. (C) Normalized bioluminescence intensity of U87MG-luci glioma at days 27 against that of days 7 for different groups. ap < 0.05, compared with saline; bp < 0.05, compared with MTX; cp < 0.01, compared with MTX@EVs. (D) Survival curves of GBM-bearing mice treated with saline, MTX, MTX@EVs, and MTX@EVs-KLA-LDL at days 7, 10, 13, 16 through tail vein post U87luci glioma cells implantation. The dose of MTX was 5 mg/kg.

ACS Paragon Plus Environment

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

Page 20 of 24

Page 20 / 24

Figure 9. Toxicity evaluation of glioma-bearing nude mice after administration with different treatments including saline, MTX, MTX@EVs and MTX@EVs-KLA-LDL. (A) Histochemical analysis of heart, liver, spleen, lung, and kidney sections stained with hematoxylin and eosin. Scale bar represents 100 µm. (B) The blood chemistry tests of the mice receiving treatments of saline, MTX, MTX@EVs or MTX@EVs-KLA-LDL.

CONCLUSIONS Chemotherapy for glioblastoma is hampered by the low permeability across BBB and weak infiltration into tumor tissue. In this work, we have taken advantage of the intrinsic virtue of EVs that enriched with phospholipids on membrane surface and developed MTX-loaded EVs functionalized with a synthetic multifunctional peptide. By interacting between apoA-I mimetic peptide 4F and phospholipids, MTX-loaded EVs are functionalized with therapeutic and targeted peptides for glioblastoma treatment. The LDL and KLA functionalization ACS Paragon Plus Environment

Page 21 of 24 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

Page 21 / 24

promoted MTX-loaded EVs uptake by brain tumor cells, penetration into glioma spheroid, which subsequently caused stronger cytotoxicity and inhibitory effects of MTX on the growth of glioma spheroids. In vivo experiments revealed that the engineering EVs demonstrated an obvious increased BBB permeation, glioma accumulation, and accordingly enhanced therapeutic effects of MTX compared with those of other groups in vivo. Taken together, functionalizing chemotherapeutics-loaded EVs with peptide in a facile and controllable way facilitates them targeting BBB and brain tumor cells with highly efficient anticancer properties, which represents a substantial advancement in the application of EVs for brain tumor treatment.

ASSOCIATED CONTENT Supporting Information Size distribution of MTX@EVs-KLA-LDL in PBS containing 20% serum; the amount of MTX encapsulated into EVs was measured by HPLC; the amount of peptide LDL or KLA-LDL was assessed by HPLC.

AUTHOR INFORMATION Corresponding Authors E-mail: [email protected] Notes The authors declare no competing financial interests.

ACKNOWLEDGMENTS We gratefully acknowledge financial support from the National Natural Science Foundation of China (81372712) and the Fundamental Research Funds for the Central Universities (No. 2015YGYL020). References ACS Paragon Plus Environment

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

Page 22 of 24

Page 22 / 24 ! (*+

(

,- %

.((- % /

(

01

" #

,2

'#

$ %

&

'

' (

)

%3 ! ,

'

4% 5%

+

6 2 *77 8

&

) )

) < $ +5 (

# 8 , ! #

+

8

'#

$

$

9

: #* $ +

9 4 0

8

*

(

73 %* #

$ +

( 2 +

3

%%

-

9 9

0

) ?*&

;

9 8 %

# +7% $

'

6

*

, ) D 4* - D %;

+ 6

6

)&

* % ,$ %! 8 '* $

, A < F 4* @ D

, < #

( ,

7

9



%$'

& $6 (

) 2

7 * $ )77% $

+

;

= #

"

%

6 )%& *

8 " 7 $

? $

9 # +7 1 *5 $

A A

@ 8 '( ,

> =% & &

# < 01 $ +

! *,

$

% 2

+

"

4 4 , @*+

9 ! %& 3 '

#

) ! 8

$

# %%

6

ACS Paragon Plus Environment

)

,

)

(

6 # )
')

Page 23 of 24 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

Page 23 / 24 D

, 8 2*


'

*% & 2 @

% 3 D

F @* A

01

3

9 D =

%%

'

$ &



%$9 D

'

, G @* A

!* %6

,

1 %

?

, (

, D 9

73

' 8

%

%

" 5

6$

2

#

9 (*

$

, F

4% +

01 6

%

+

!

< >

8

&

(

&

- +

0

3

)

0 B ! & % 7+

'

# %%

A 2

?*,

, )'

> "%3 $

& ) B

!

, ( D

9 9 )$ &

"

B ! 3,

2 !

7

6

F 2

%$2 #

'

*% 6

,

-

6

9 D

, G 9

,

,(

73 ' 8

6

9 2

4 2 % +

,

$3

D

, A

! 1 *5 $

*%

=& $ +

D 9

@ 9 2 1+

53 2* ' $ 6,

A

& )#

2