Shape effects of electrospun fiber rods on the tissue distribution and antitumor efficacy

    Shape effects of electrospun fiber rods on the tissue distribution and antitumor efficacy Hong Zhang, Yuan Liu, Maohua Chen, Xiaoming Luo, Xiaohong Li PII: DOI: Reference:

S0168-3659(16)30274-7 doi: 10.1016/j.jconrel.2016.05.011 COREL 8254

To appear in:

Journal of Controlled Release

Received date: Revised date: Accepted date:

3 March 2016 30 April 2016 5 May 2016

Please cite this article as: Hong Zhang, Yuan Liu, Maohua Chen, Xiaoming Luo, Xiaohong Li, Shape effects of electrospun fiber rods on the tissue distribution and antitumor efficacy, Journal of Controlled Release (2016), doi: 10.1016/j.jconrel.2016.05.011

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Shape effects of electrospun fiber rods on the tissue distribution and antitumor

RI P

Hong Zhang, Yuan Liu, Maohua Chen, Xiaoming Luo, Xiaohong Li*

T

efficacy

SC

Key Laboratory of Advanced Technologies of Materials, Ministry of Education, School of Materials Science and Engineering, Southwest Jiaotong University, Chengdu 610031, P.R. China

NU

* Corresponding author. School of Materials Science and Engineering, Southwest Jiaotong University,

E-mail Address: [email protected]

PT ED

Abstract:

MA

Chengdu 610031, P.R. China. Phone: +8628-87634068; fax: +8628-87634649.

The significant impact of drug-loaded nanocarriers on cancer chemotherapy lies in the ability to specifically target to tumors with alleviated systemic toxicities. In the current study, a versatile and scalable method has been developed to construct fiber rods from electrospun fibers by ultrasonication

CE

using encapsulated NaCl nanoparticles as void-precursors. The shape effects of doxorubicin

AC

(DOX)-loaded fiber rods with an average diameter of around 500 nm and different lengths are determined on the blood circulation, tumor accumulation and cellular uptake. Compared with microspheres, fiber rods indicate an up to 4-fold higher accumulation in tumors and an up to 3-fold longer terminal half-life of plasma DOX levels after intravenous injection. Fiber rods with shorter lengths show a significantly higher in vitro cytotoxicity to tumor cells, a higher DOX accumulation and cell necrosis in tumors, and a significantly lower metastasis in lungs. Among fiber rods with different lengths, fiber rods with an average length of 2 m induce significantly higher inhibition on tumor cell proliferation and induction of cell apoptosis, as wells as no detectable metastatic nodules in lung sections. Therefore, the shape effects of electrospun fiber rods hold great potential for enhancing systemic circulation and directing biodistribution to improve therapeutic outcomes.

Key words: electrospun fiber rod; shape effect; cellular uptake; blood circulation; tumor accumulation 1

ACCEPTED MANUSCRIPT 1. Introduction Chemotherapy represents the most widely used method for cancer treatment, employed alone or in combination with surgical removal or radiotherapy. While significant advances have been made in the

T

conventional systemic chemotherapy through directly killing tumor cells, it still suffers from many side

RI P

effects such as immunosuppression, nausea and hair loss. This is due to the non-specific distribution of chemotherapeutic agents in healthy tissues, and only a small percentage of administered drugs

SC

accumulate in tumors [1]. Invariably, the side effects impose a dose reduction, treatment delay, or discontinuance of therapy, thus substantial improvements need to be made in the target therapy to reach

NU

the next level of clinical relevance. In addition to the development of novel anticancer drugs, various drug delivery systems such as liposomes, nanoparticles and micelles, have been developed to improve

minimizing the systemic toxicity [2].

MA

the site specificity and bioavailability of drugs, thereby enhancing the therapeutic efficacy and

PT ED

To achieve a targeting capability of drugs and their formulations, there are several challenges needed to be addressed, including sufficient residence in the circulation before arriving at tumor sites, efficient travel within tumors and uptake into tumor cells, as well as a prompt release at the intended site for realization of therapeutic functions. One of the strategies is the attachment of targeting ligands with

CE

drugs or at the surface of nanocarriers, which bind to appropriate receptors overexpressed by tumor

AC

cells [3]. Alternatively, the passive targeting strategy takes advantages of the distinctive pathophysiological features of tumors and allows a preferential accumulation in tumor areas with leaky vasculatures, commonly referred to the enhanced permeation and retention (EPR) effect [4]. Though the first example of targeted liposomes was described in 1980s, these targeting technologies have not made a significant clinical impact on human health. Loomis et al. indicated that the in vivo tumor accumulation of drug and survival time of animals were not different between folate receptor-targeting and non-targeting liposomal doxorubicin (DOX), despite promising results from in vitro uptake studies demonstrating that the inclusion of folate increased the cellular uptake [5]. Some targeted drug delivery systems have already shown clinical efficacy, but there are contradicting data in the active targeting strategies regarding the added benefits for the inclusion of targeting molecules. Most likely, the recognition of targeting ligands by the reticuloendothelial system (RES) accelerates the clearance process, which offsets the benefit of active drug targeting to tumors [6]. Thus, an efficient passive 2

ACCEPTED MANUSCRIPT targeting is also beneficial to the nanocarriers based on ligand-mediated active targeting, but it is known that over 95% of administered doses are accumulated in organs other than tumors, particularly in liver, spleen, and lungs [7]. The passive targeting is largely mediated by the physicochemical

T

properties of nanocarriers, and attempts have been made to engineer drug carriers exhibiting a

RI P

prolonged half-life in the bloodstream and a lower uptake by RES and non-target tissues [8]. The shapes of nanocarriers have shown significant effects on the circulation half-life in blood and the

SC

distribution in tumor tissues. Geng et al. tested the effects of nanoparticle shapes on the circulation time following a systemic injection of 3.5 m-long filomicelles and spherical nanoparticles of 200 nm. The

NU

filomicelles circulated for up to one week in rodents, whereas the spherical nanoparticles were cleared within 2 days [9]. Thompson et al. indicated that rod-shaped microparticles (1 or 2 m equivalent

MA

spherical diameters) with high aspect ratios (length-to-diameter) displayed significantly improved margination compared to microspheres. But nanorods (500 nm equivalent spherical diameters), even

PT ED

with a high aspect ratio, did not exhibit enhanced margination compared to that of equivalent nanospheres under the high shear rate and disturbed blood flow [10]. Decuzzi et al. investigated the biodistribution of silica microspheres with diameters ranging from 700 nm to 3 m and non-spherical hemispheres, cylinders and discoids after intravenous injection into tumor-bearing mice. It was

CE

indicated that larger particles of non-spherical shapes may provide concentrations at intermediate sites

AC

that afford the deployment of novel modes of anticancer therapy, although smaller particles may well provide larger initial concentrations at tumor sites via different mechanisms of extravasation [11]. Therefore, an intensive consideration is needed in the evaluation of the contribution of particle sizes and shapes towards blood circulation, tissue distribution and cellular internalization [12]. With recent advances in the “bottom-up” and “top-down” approaches, the development of well-defined polymeric nanostructures of different shapes has become possible. The bottom-up approach relies on a spontaneous self-assembly of amphiphilic block copolymers to form elongated micelles as a result of the thermodynamic incompatibility between different blocks. This chemistry-centered approach is relatively simple and cost-efficient, particularly for nanoparticles of sub-100 nm range, but is limited to polymers with specific components and structures [13]. In the “top-down” strategy, the preparation of non-spherical polymeric nanoparticles mainly includes particle replication in nonwetting template (PRINT) and membrane stretching methods. The PRINT method is 3

ACCEPTED MANUSCRIPT a high-resolution molding technology which allows for the formation of nanoparticles in the micro- and nano-scale size ranges with an aspect ratio as high as 60 [14]. Non-spherical shape is also fabricated by stretching spherical particles in one or two dimensions after embedment into poly(vinyl alcohol) (PVA)

T

films, and rods or elliptical disks can be obtained with aspect ratios ranging from 2 to 15 [15]. The

RI P

top-down approaches indicate advantages in the fabrication of nanostructures with a wide range of shapes from sub-micron to micron at will, but shows limitations in the scale-up production and

SC

protection of entrapped drugs.

Electrospinning is a versatile technique for nanofiber fabrication and possesses such advantages as

NU

simplicity, cost-effectiveness, flexibility and potential to scale up. Electrospinning provides an opportunity for direct encapsulation of drugs into a broad range of polymers with a high loading

MA

efficiency [16]. Electrospun fibers have been implanted intratumorally or adjacent to tumor tissues for those unresectable or inoperable solid tumors, or at the resection margins after surgical removal of solid

PT ED

tumors [17]. But the physical implants require direct accessibility through surgical procedures, having a large invasiveness. In this context, fiber rods were constructed from electrospun fibers under ultrasonication, and the lengths were modulated by the amount of NaCl nanoparticles encapsulated into fibers. In order to determine the cellular uptake and tissue distributions of fiber rods, coumarin 6 was

CE

loaded into poly(styrene-maleic anhydride) copolymers (PSMA) to obtain coumarin-containing fiber

AC

rods (RDCOU). PSMA fiber rods were selected because they are not biodegradable, enabling evaluation of the temporal biodistribution without resorption as a confounding variable. The cellular uptake of fiber rods with different lengths was evaluated on tumor cells and macrophages, and the distribution in tumors and other tissues was determined after intravenous injection. Poly(ethylene glycol)-polylactide (PELA) was used to construct DOX-loaded fiber rods (RDDOX) of different lengths. In comparison with microspheres, the cytotoxicity, pharmacokinetics, antitumor efficacy and antimetastasis capabilities of RDDOX were evaluated on tumor-bearing mice.

2. Materials and methods 2.1. Materials PELA (Mw = 42.3 kDa, Mw/Mn = 1.23) containing 10 wt% of PEG was prepared by bulk ring-opening polymerization using stannous chloride as catalyst [18]. PSMA (Mw = 170 kDa, Mw/Mn 4

ACCEPTED MANUSCRIPT = 2.5) containing 14.8 mol% of maleic anhydride was obtained from Shanghai Zhaocheng Science and Technology Development Co., Ltd.

(Shanghai, China). Coumarin 6, 4’,6-diamidino-2-phenylindole

(DAPI), trypsin and dialysis bags (1 kDa cutoff) were procured from Sigma-Aldrich Inc. (St. Louis,

T

MO). Sodium bis(2-ethylhexyl) sulfosuccinate (AOT, 96%) was purchased from Aladdin Industrial Co.,

RI P

Ltd. (Shanghai, China), and DOX hydrochloride was from Dalian Mellon Biological Technology Co., Ltd. (Dalian, China). Rabbit antimouse antibodies of caspase-3 and Ki-67, goat antirabbit

SC

IgG–horseradish peroxidase (HRP) and 3,3-diaminobenzidine (DAB) developer were obtained from Biosynthesis Biotechnology Co., Ltd. (Beijing, China). All other chemicals and solvents were of

NU

reagent grade or better, and received from Changzheng Regents Co. (Chengdu, China), unless

2.2. Preparation of drug-loaded fiber rods

MA

otherwise indicated.

PT ED

Electrospun fiber rods were constructed by scission of electrospun fibers containing NaCl nanoparticles by ultrasonication. NaCl nanoparticles with an average size of around 500 nm were prepared using reverse microemulsions as described previously with some modifications [19]. Briefly, 10 mL of CaCl2 solution in formamide (4.4%, w/v) was added dropwise into 10 mL of AOT solution in

CE

n-heptane (16%, w/v) to form microemulsions. After kept stirring for 1 h at room temperature, 15 mL

AC

of acetone was added for emulsion breaking, followed by ultrasonication for 5 min to disperse the formed nanoparticles. The suspensions were centrifuged to collect nanoparticles, followed by acetone washing and vacuum dried. Drug-loaded electrospun fibers with inoculated NaCl nanoparticles were prepared as described previously with some modifications [20]. Briefly, 0.5 g of polymers (PSMA or PELA) were dissolved in dry dimethyl formamide, and different amount of NaCl nanoparticles from 25 to 125 mg were added into polymer solutions. The formed suspensions were placed into a 1-mL syringe and pushed by a microinject pump (Zhejiang University Medical Instrument Co., Hangzhou, China) at a flow rate of 0.4 mL/h. The electrospinning was performed under 20 kV/10 cm using a high-voltage power supply (Tianjing High Voltage Power Supply Co., Tianjing, China). Fibers were collected on an aluminum foil wrapped on a grounded rotating mandrel. After removal of solvent residues under vacuum, the fibrous mats were put in a glass vial containing 10 mL of distilled water, which was cooled by a water-ice slurry to maintain the processing temperatures below 25 °C. Ultrasonication was carried 5

ACCEPTED MANUSCRIPT out using a Vibracell 500 W sonicator (Techcomp Limited Co., Beijing, China) with a probe diameter of 13 mm, working at 20 kHz. The total run time was 3 min under an amplitude of 60% with a 2 s ON and 2 s OFF.

T

DOX hydrochloride was mixed with excess triethylamine in dimethyl sulfoxide overnight to obtain a

RI P

DOX base, which was added into above PELA solutions to prepare RDDOX. For comparison, DOX-loaded PELA microspheres (MSDOX) and coumarin-loaded PSMA microspheres (MSCOU) with a

SC

diameter close to that of fibers were prepared by the solvent evaporation process as described previously with some modifications [18]. Briefly, polymer (PSMA or PELA) and drugs (DOX or

NU

coumarin 6) were dissolved in methylene chloride. The solutions were added dropwise into distilled water containing 5% of PVA, followed by emulsification under a high-speed homogenizer. The solution

MA

was stirred at room temperature for 24 h to evaporate methylene chloride. The resulting microspheres

PT ED

were collected by centrifugation, washed with distilled water and vacuum dried.

2.3. Characterization of drug-loaded fiber rods The morphologies of electrospun fibers, fiber rods and microspheres were observed by SEM (FEI Quanta 200, The Netherlands) equipped with a field-emission gun (20 kV) and a Robinson detector

CE

after 2 min of gold coating to minimize the charging effect. From SEM images, at least 100 fiber

AC

fragments were randomly selected and their diameters and lengths were measured to generate average values by using the tool of Photoshop 10.0 edition. The microsphere size was measured by a Nano-ZS laser particle analyzer (Zetasizer Nano ZS90, Malvern Co. UK). The presence of DOX or coumarin 6 in fiber rods or microspheres was examined by a fluorescence microscope (Olympus IX51-FL, Japan). The loading contents of DOX or coumarin 6 in fiber rods or microspheres were determined after extraction with phosphate buffer saline (PBS) as described previously [21]. Briefly, 5 mg of fiber rods or microspheres were dissolved in 1 mL of methylene chloride, followed by the addition of 10 mL PBS (pH 7.4). The mixture was vortexed thoroughly and centrifuged to collect the supernatant. The supernatant was then detected by a fluorospectrophotometer (Hitachi F-7000, Japan) under the excitation/emission wavelengths of 475/589 nm for DOX and 467/497 nm for coumarin 6. The DOX or coumarin 6 contents were obtained using a standard curve from known concentrations of DOX or coumarin 6 solutions. The extraction efficiency was calibrated by adding a certain amount of DOX or 6

ACCEPTED MANUSCRIPT coumarin 6 into PELA or PSMA solutions along with the same concentrations as above and extracted using the same process. The drug loading amount indicated the amount (in mg) of drug encapsulated per 100 mg of fiber rods or microspheres. The loading efficiency indicated the amount of drug

RI P

T

encapsulated compared with the total amount used for the preparation of fiber rods or microspheres.

2.4. In vitro drug release from fiber rods

SC

In vitro release of DOX and coumarin 6 from fiber rods with different lengths or microspheres was determined in dialysis bags after incubation in PBS for up to 4 weeks. Briefly, fibrous rods or

NU

microspheres were sealed in dialysis bags and immersed into 30 mL of PBS, which was kept in a thermostated shaker at 37 °C. To ensure a sink condition, the total amounts of DOX and coumarin 6

MA

from fiber rods or microspheres in the release media were 0.2 mg/mL and 1.5 μg/mL, respectively. At predetermined time intervals, 1.0 mL of released solution was withdrawn and replaced with an equal

PT ED

amount of fresh PBS. The amount of DOX or coumarin 6 released from fiber rods or microspheres was determined by a fluorospectrophotometer as described above.

2.5. In vitro cellular uptake of fiber rods

CE

The cellular uptake of fiber rods with different lengths was observed by confocal laser scanning

AC

microscopy (CLSM) after incubation of RDCOU with mouse mammary tumor cells 4T1 and mouse macrophages RAW 246.7 as described previously [22]. Briefly, 4T1 and RAW 246.7 cells were from American Type Culture Collection (Rockville, MD) and maintained in RPMI 1640 media and DMEM high glucose media (Invitrogen, Grand Island, NY), respectively, supplemented with 10% fetal bovine serum (FBS, Gibco BRL, Grand Island, NY). Fiber rods and microspheres were sterilized by electron-beam irradiation using a linear accelerator (Precise™, Elekta, Crawley, UK) with a total dose of 80 cGy. 4T1 or RAW 246.7 cells were seeded onto glass coverslips in 24-well tissue culture plates (TCP) at a density of 1×104 cells/well and allowed to be attached overnight before treatment. The cells were then treated with RDCOU or MSCOU-containing media at 60 μg/mL for 24 h, followed by PBS wash and fixation with 4% paraformaldehyde for 30 min at room temperature. After DAPI staining and PBS rinsing, cells were observed under CLSM (Leica TCS SP2, Germany). The uptake of RDDOX was also detected on 4T1 cells following the same way and compared with that of MSDOX. 7

ACCEPTED MANUSCRIPT The amount of cellular uptake of fiber rods with different lengths was determined after cell lysis as described previously [23]. Briefly, 4T1 and RAW 246.7 cells were prepared as a confluent layer in 48-well TCP, followed by incubation with RDCOU or MSCOU-containing media at 60 g /mL for 6 h.

T

Cells were wash with PBS and lysed with 0.5% Triton X-100 for 2 h at 4 C. The cell lysate was

RI P

collected and the amount of coumarin 6 was detected by a fluorospectrophotometer as described above. The cellular uptake was normalized to the total proteins of the cell lysate, which was determined by

SC

BCA protein assay kit (Pierce, Rockford, IL). The uptake efficiency of RDDOX was also detected on

NU

4T1 cells following the same way and compared with that of MSDOX.

2.6. In vitro cytotoxicity of fiber rods

MA

The cytotoxicity of fiber rods was determined by Cell Counting Kit-8 reagent (CCK-8, Dojindo Laboratories, Kumamoto, Japan) after incubation with RDDOX of different lengths and MSDOX as

PT ED

described previously [24]. Briefly, 4T1 cells were seeded in 96-well TCP at 5000 cells/well and allowed to be attached overnight before treatment. The DOX stock solution was diluted in RPMI 1640 media with a series of concentration, and RDDOX or MSDOX releasing equivalent amount of DOX during 72 h (from in vitro release data) were applied. Fiber rods and microspheres without drug

CE

inoculation were also tested. After incubation for 72 h and removal of RDDOX or MSDOX-containing

AC

media, cells were treated with 200 L of RPMI 1640 containing 20 L CCK-8 in each well, and incubated for 4 h according to the reagent instruction. The absorbance at 450 nm was measured for each well by a microplate spectrophotometer (Elx-800, Bio-Tek Instrument Inc., Winooski, VT). The cell apoptosis was quantified by an Annexin V-FITC apoptosis detection kit (Beijing 4A Biotech Co., Beijing, China) after treatment with RDDOX of different lengths. Briefly, 4T1 cells were grown into 6-well TCP at density of 2 × 105 cells/well and treated as above with a final DOX concentration of 2.0 g/mL. After exposure for 72 h, cells were gently rinsed with ice-cold PBS, harvested with 0.25% trypsin, and suspended in 250 L of binding buffer. The cells were incubated for 15 min with 5 μL of Annexin V-FITC, and then incubated with 10 μL of propidium iodide for 5 min before analysis with flow cytometry (BD Accuri C6, BD Biosciences, CA).

2.7. Pharmacokinetics analysis of fiber rods after intravenous injection 8

ACCEPTED MANUSCRIPT Pharmacokinetic profiles of RDDOX with different lengths were determined after intravenous injection as described previously [25], and compared with that of MSDOX. Male Sprague Dawley rats weighing 180220 g were from Sichuan Dashuo Biotech Inc. (Chengdu, China), and the animal

T

protocols were approved by the University Animal Care and Use Committee. Briefly, RDDOX and

RI P

MSDOX were administrated to the rats by intravenous injection through caudal veins at a dose of 2 mg DOX/kg body weight, using free DOX injection as the control. At 0.5, 1, 3, 6, 9, 12, 24, 48, 72, 96, 120,

SC

144, and 168 h after administration, blood samples of around 500 L were collected in heparinized tubes via amputation of tails. The blood samples were immediately centrifuged to recover plasma. The

NU

DOX concentrations were determined as described above, and the DOX contents were obtained using a standard curve from known concentrations of DOX in blank plasma. The pharmacokinetic parameters

MA

were calculated using PK solver [26], including total area under the plasma concentration-time curve from time zero to infinity (AUC0-), terminal half-life (T1/2β), time-averaged total body clearance (CL),

PT ED

maximum plasma concentration (Cmax) and mean residence time (MRT).

2.8. Tissue distribution of fiber rods and drugs in tumor-bearing mice The tissue distribution were determined on tumor-bearing mice after intravenous injection of fiber

CE

rods with different lengths and compared with that of microspheres [27]. Briefly, female BALB/c mice

AC

weighing 1822 g were obtained from Sichuan Dashuo Biotech Inc. (Chengdu, China). 4T1 cells were expanded in RPMI 1640 media containing 10% FBS, and the collected cells were suspended in PBS at 1  107 cells/mL. Xenograft 4T1 tumors were established by injection of 100 L cell suspensions into the mammary fat pad of each mouse as described previously [21]. Tumors were allowed to grow for 10 days to a volume of about 100 mm3, and treated by tail vain injection of RDCOU or MSCOU at a dose of 25 mg/kg. At 1, 24 and 168 h after injection, three mice of each group were sacrificed to collect heart, liver, spleen, lungs, kidneys, tumor and blood. Tissues were cut into small pieces, washed with ice-cold saline to remove bloodstains, and weighed. PBS was added by a 3-fold volume to the weight of a tissue and the mixture was homogenized, followed by precipitation of proteins and extraction of coumarin 6 as described previously [28]. The organic layer was collected by centrifugation for the determination of coumarin 6 content by a fluorospectrophotometer as above. The percentage of injected dose (ID%) indicated the ratio of the actual amount of RDCOU or MSCOU in a tissue to the total amount of injected 9

ACCEPTED MANUSCRIPT drug, and the percent dose rate (ID%/g) was used to represent the fiber rod or microsphere accumulation in per gram of a tissue [29]. The tissue distribution of RDDOX with different lengths was also determined on tumor-bearing mice as above. RDDOX and MSDOX were injected into mice at a dose

T

of 2 mg DOX/kg. The DOX content in the tissues was detected at 1 and 24 h after administration as

RI P

described above.

SC

2.9. In vivo toxicity of fiber rod treatment

The in vivo toxicity of RDDOX treatment was determined on tumor-bearing mice with respect to the

NU

hematological detection and histological analysis of normal tissues. Briefly, tumor-bearing mice were injected with RDDOX of different lengths or MSDOX via tail veins. At 24 h after the treatment, blood was

MA

collected from each mouse. The important hematological markers such as red blood cells (RBC), white blood cells (WBC), hemoglobin (HGB) and platelet (PLT), and the key blood chemistry parameters

PT ED

such as aspartate transaminase (AST), creatine kinase (CK), lactate dehydrogenase (LDH) and blood urea nitrogen (BUN) levels were assayed as described previously [30]. The mice were killed to retrieve lung and heart, followed by fixation in 4% neutral buffered formaldehyde and paraffin embedment.

AC

previously [21].

CE

Tissue sections were stained with hematoxylin and eosin (HE) for pathological study as described

2.10. Antitumor efficacy of fiber rods The 4T1 tumor-bearing mice were treated with RDDOX of different lengths, MSDOX and free DOX at an equivalent DOX dose of 2 mg/kg via a tail vein injection, using saline treatment as the control. The tumor size, body weight and survival rates of animals were monitored. The length and width of a tumor were measured with a vernier caliper, and the tumor volume was calculated as described previously [31]. The number of live animals at each time point was plotted in Kaplan Meier survival curves, and the median survival time was obtained for comparison of treatment efficacy. On day 21 after treatment, animals were sacrificed, and tumors were rapidly excised and fixed in 4% neutral buffered formaldehyde. The tissues were processed routinely for HE staining and observed under a light microscope (Nikon Eclipse E400, Japan). To investigate the proliferation and apoptosis of tumor cells, immunohistochemical (IHC) assessment of Ki-67 and caspase-3 expressions was 10

ACCEPTED MANUSCRIPT conducted on tumor sections as describe previously [21]. To quantify the protein expressions, Ki-67 or caspase-3-positive cells were counted in five randomly selected areas from IHC staining images, and

T

compared with the total number of cells in these areas.

RI P

2.11. Antimetastasis efficacy of fiber rods

The metastasis of 4T1 cells to lungs was evaluated from the formation of metastatic nodules and

SC

histological analysis of lung tissues. Briefly, lungs were retrieved on day 21 after injection, washed with PBS and fixed in Bouin’s solutions. Surface metastatic nodules in lungs were determined under

NU

dissecting microscopy as described previously [32]. Meanwhile, lungs were fixed in 4% neutral buffered formaldehyde and were processed routinely for HE staining as above for histological

MA

evaluation.

PT ED

2.12. Statistical analysis

The results are reported as mean ± standard deviation (SD). Whenever appropriate, comparisons among multiple groups were performed by analysis of variance (ANOVA), while a two-tailed Student’s t-test was used to discern the statistical difference between two groups. A probability value (p) of less

AC

CE

than 0.05 was considered to be statistically significant.

3. Results and discussion

3.1. Characterization of drug-loaded fiber rods In the current study, fiber rods were fabricated by ultrasonication of electrospun fibers with encapsulated NaCl nanoparticles. To date, few attempts have been made to produce fiber rods directly from electrospun fibers. Mechanic methods, such as mortar grinding, cryogenic milling and cutting were found to be effective for brittle fibers, resulting in a combination of chips, flakes and short fibers of around 50100 m in length [33]. Sawawi et al. indicated that an ultrasonication of poly(styrene) and poly(methyl methacrylate) produced fiber rods of around 10 m long, while it was not effective for ductile fibers of poly(L-lactide) or poly(acrylonitrile) [34]. Our strategy was the incorporation of NaCl nanoparticles as void-precursors in fibers, and the dissolution of NaCl nanoparticles led to the formation of voids in fibers, which were supposed to be preferably broken under sonication [35]. 11

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Fig. 1. (a) SEM and fluorescent images of RDDOX of different lengths and MSDOX. (b) The lengths of

PT ED

fiber rods decreased with the increase in the weight ratios of NaCl nanoparticles to matrix polymers (n = 3).

Fig. 1a shows SEM and fluorescent images of RDDOX and MSDOX. RDDOX was well dispersed in

CE

water without significant entanglement, and MSDOX had smooth surfaces with an average diameter of around 470 nm. As shown in Fig. 1b, the lengths of fiber rods decreased with the increase in the

AC

amount of NaCl nanoparticles in the electrospinning suspensions. The inoculations of 25, 50, 80 and 125 mg of NaCl nanoparticles resulted in the formation of RDDOX with 8.6 ± 2.0, 6.1 ± 1.4, 4.5 ± 1.3 and 2.5 ± 0.9 μm in lengths, which were name as RDDOX-8, RDDOX-6, RDDOX-4 and RDDOX-2, respectively. Fig. S1 (Supplementary Material) shows fluorescent images of RDCOU. The lengths of 8.1 ± 1.7, 6.1 ± 1.5, 4.1 ± 1.3, and 2.4 ± 0.7 were obtained for RDCOU-8, RDCOU-6, RDCOU-4 and RDCOU-2, respectively. The diameters of RDDOX and RDCOU, measured from SEM images, were about 0.49 ± 0.13 and 0.43 ± 0.16 μm, respectively. Efficient drug encapsulation was one of the advantages of electrospun fibers [36], and the scission of electrospun fibers led to a drug leakage from fibers. The same amount of DOX was used for RDDOX and MSDOX preparations, and the DOX loading efficiencies were 67.3%, 74.3%, 83.5%, 86.3%, and 71.1% for RDDOX-2, RDDOX-4, RDDOX-6, RDDOX-8, and MSDOX, respectively. In the current study, the DOX loading contents of RDDOX and MSDOX were set as 4.0 wt% through adjusting the amount of DOX used for fiber rod and microsphere preparations. 12

ACCEPTED MANUSCRIPT Similar loading efficiency was detected for RDCOU and MSCOU, and the coumarin loading content was

AC

CE

PT ED

MA

NU

SC

RI P

T

set as 0.1 wt%.

Fig. 2. (a) Percent release of DOX and (b) coumarin 6 from fiber rods of different lengths and microspherres after incubation in PBS at 37 C (n = 3).

3.2. In vitro release profiles of DOX and coumarin from fibrous rods In vitro release profiles of DOX were investigated on RDDOX with different lengths and MSDOX. The saturation solubility of DOX in PBS was about 20 µg/mL [37], and a sink condition was maintained during the drug release tests. As shown in Fig. 2a, fiber rods and microspheres indicated a burst release of DOX during 1 day of incubation, followed by a sustained release for over 4 weeks. RDDOX-2, RDDOX-4, RDDOX-6 and RDDOX-8 showed an initial release of around 29.9%, 24.3%, 21.7% and 17.5%, respectively, which were lower than that from MSDOX at around 33.8% during the 1st day. After incubation for 4 weeks, RDDOX-2, RDDOX-4, RDDOX-6 and RDDOX-8 showed an accumulated release of 13

ACCEPTED MANUSCRIPT around 63.4%, 50.7%, 47.3% and 38.0%, respectively, and there were significant differences among the groups (p < 0.05). The higher initial burst and accumulated release during 4 weeks from RDDOX with shorter lengths were due to the increased release from the cross sections. There was no significant

T

difference in the DOX release rates between MSDOX and RDDOX-2 (p > 0.05), which were significantly

RI P

higher than those of RDDOX-4, RDDOX-6 and RDDOX-8 (p < 0.05). As shown in Fig. 2b, the cumulative release of coumarin 6 from both fiber rods of different lengths and microspheres was less than 0.1%,

SC

which was similar to other studies [38]. It was indicated that most of the incorporated coumarin 6 remained in the fiber rods and microspheres during incubation and the fluorescence signals in the cell

MA

3.3. In vitro cellular uptake of fiber rods

NU

or tissue samples should represent the retention of fiber rods and microspheres.

The effects of particle shapes have been characterized at the cellular level, and the clearance by

PT ED

macrophages significantly affects the circulation in blood and distribution in tumor tissues. Morton et al. tested the cellular uptake of hyaluronic acid-coated poly(lactide-co-glycolide) (PLGA) particles, indicating nearly 10-fold higher uptake of 320 nm-long rod particles than that of 200 nm-long ones in breast cancer cells [39]. Contrary to the above observations for elongated nanostructures, several other

CE

studies have found that the spherical gold and polymer nanoparticles were internalized to a greater

AC

extent than their corresponding rod-shaped or cylindrical particles [40]. After evaluation on nanospheres or nanorods of different diameters and lengths, Zhou et al. indicated that the nanorods with a medium length of lower than 500 nm had a much longer blood circulation and faster cellular uptake than the nanospheres or long nanorods [41]. Inconsistent even controversial effects of particle size and shapes on the biological profiles have been observed in these studies. Therefore, the cellular uptake and intracellular accumulation of RDCOU with different lengths need to be clarified. The cellular uptake of fiber rods was observed by CLSM after incubation with 4T1 and RAW 264.7 cells. As shown in Fig. S2a,b (Supplementary Material), both the microspheres and fiber rods were taken up into the cells, and the coumarin fluorescence was observed mainly in the cytoplasm. The fluorescence intensity of coumarin in 4T1 cells was stronger after incubation with fiber rods than that of microspheres (Fig. S2a). The elongated particles enhanced the propensity to be internalized over their spherical counterparts in nonphagocytic cells [42]. In addition, fiber rods were taken up to a lesser 14

ACCEPTED MANUSCRIPT extent by RAW 264.7 macrophages compared to 4T1 cells, and there was no obvious difference for microspheres (Fig. S2a). The cellular uptake efficiency was determined from the amount of fiber rods and microspheres normalized to the protein contents of cells. As shown in Fig. 3a, MSCOU were taken

T

up by 4T1 cells at a significantly lower efficiency compared to RDCOU-2 and RDCOU-4 (p < 0.05).

RI P

MSCOU exhibited an over 2.5 folds higher uptake by RAW 264.7 macrophages than RDCOU-4, RDCOU-6, and RDCOU-8 (p < 0.05), and a slightly higher uptake than RDCOU-2 (p > 0.05). In addition, a lower

SC

uptake of RDCOU-6 and RDCOU-8 with lager lengths was observed in 4T1 and RAW 264.7 cells, due to the large energy requirement for endocytosis of particles with high aspect ratios [43]. Champion et al.

NU

indicated that worm-like particles with a high aspect ratio of around 20 exhibited negligible uptake by macrophages [44]. In addition, the cellular uptake of RDDOX indicated a similar profile of RDCOU by

MA

4T1 cells (Fig. S2b). As shown in Fig. 3b, free DOX exhibited the highest cellular uptake, as free DOX easily entered 4T1 cells by a passive diffusion. RDDOX-2 was taken up to a significantly higher extent

PT ED

by 4T1 cells compared to MSDOX and RDDOX with other lengths (p < 0.05). Thus, these findings showed that the shapes of fiber rods independently influenced the internalization by tumor cells and macrophages.

CE

3.4. In vitro cellular toxicity and apoptosis of fiber rods

AC

The cell viability was tested on 4T1 cells against free DOX, RDDOX of different lengths and MSDOX with an equivalent amount of DOX released. Fig. S3a (Supplementary Material) shows the cytotoxicity of fiber rods and microspheres without drug inoculation. Over 90% of cell viability was remained after incubation for 72 h at different concentrations ranging from 50 to 200 μg/mL, showing no significant cytotoxicity for all the fiber rods and microspheres. Fig. 3c summarizes the cell viability after incubation with RDDOX and MSDOX for 72 h, showing a higher cytotoxicity of RDDOX-2 than other fiber rods and MSDOX. The IC50 values of RDDOX-2, RDDOX-4, RDDOX-6, RDDOX-8 and MSDOX against 4T1 cells were around 0.19, 0.27, 0.37, 1.31 and 0.39 μg/mL, respectively. As control, the IC50 value of free DOX was 0.49 μg/mL. The cytotoxicity of RDDOX increased with the decrease in the rod lengths, which might be attributed to the better cellular internalization of RDDOX with shorter lengths (Fig. 3b). Compared with that of free DOX, the IC50 decreased by around 2.6, 1.8, 1.3, and 1.3 folds after DOX was encapsulated into RDDOX-2, RDDOX-4, RDDOX-6, and MSDOX, respectively, which might be 15

ACCEPTED MANUSCRIPT

AC

CE

PT ED

MA

NU

SC

RI P

T

attributed to the sustained release of DOX from fiber rods and microspheres (Fig. 2a).

Fig. 3. (a) The uptake amount of RDCOU with different lengths and MSCOU by 4T1 and RAW 264.7 macrophages, normalized to the total proteins of the cell lysate (n = 5). (b) The uptake amount of DOX in 4T1 cells after incubaiton with RDDOX of different lengths and MSDOX, normalized to the total proteins of the cell lysate (n = 5; *: p < 0.05). (c) The viability of 4T1 cells after treatment with RDDOX with different lengths and MSDOX, compared with free DOX (n = 5). 16

ACCEPTED MANUSCRIPT The effects of various DOX formulations on the cell apoptosis were investigated after dual staining with Annexin V-FITC and propidium iodide. Fig. S3b summarizes the staining results after treatment on 4T1 cells for 72 h with RDDOX of different lengths and MSDOX. The percentage of total apoptotic

T

cells including early and late apoptotic cells was averaged at around 69.0%, 61.2%, 55.5%, 45.3% and

RI P

60.3% after treatment with RDDOX-2, RDDOX-4, RDDOX-6, RDDOX-8 and MSDOX, respectively. Compared with that of free DOX at around 49.8%, RDDOX-2, RDDOX-4 and MSDOX accelerated cell

AC

CE

PT ED

MA

NU

SC

apoptosis, which was consistent with above cytotoxicity results.

Fig. 4. (a) Plasma DOX levels of rats after intravenous administration of free DOX, RDDOX with different lengths and MSDOX (n = 5). (b) The distribution of RDCOU with different lengths and MSCOU in heart, liver, spleen, lung, kidney, tumors and blood at 1, 24 and 168 h after intravenous administration 17

ACCEPTED MANUSCRIPT to tumor-bearing mice (n = 4). (c) The distribution of DOX in heart, liver, spleen, lung, kidney, tumors and blood at 1 and 24 h after intravenous administration of RDDOX with different lengths and MSDOX,

RI P

T

compared with free DOX (n = 4).

3.5. In vivo pharmacokinetics of fiber rods

The pharmacokinetics of DOX-loaded fiber rods with different lengths and microspheres was

SC

investigated in Sprague Dawley rats and compared with that of microspheres. Fig. 4a shows the plasma

NU

concentration-time profiles of free DOX, DOX-loaded fiber rods and microspheres after intravenous injection. Biphasic curves were indicated for all the groups with a rapid distribution phase, followed by

MA

a slow elimination phase. There was a more significant decrease in the DOX levels after injection of free DOX than those of fiber rods and microspheres. After 6 h of administration of free DOX, the plasma DOX level reached 15.1  3.7 ng/mL, which was significantly lower than those of RDDOX-2,

PT ED

RDDOX-4, RDDOX-6, RDDOX-8, and MSDOX (p < 0.05), at 70.7  11.0, 60.1  9.8, 66.6  13.0, 54.2  9.8, and 61.9  8.4 ng/mL, respectively. The DOX content in plasma was undetectable after injection for 72 h, but the plasma concentrations of DOX were around 40 ng/mL for RDDOX with different lengths and

CE

MSDOX. In addition, there was more significant decrease in the plasma DOX concentrations after MSDOX injection than those of RDDOX. After injection of MSDOX for 7 days, the plasma DOX level was

AC

6.9  3.4 ng/mL, which was significantly lower than 16.9  4.3, 22.4  9.8, 30.6  2.7 and 31.7  6.9 ng/mL for RDDOX-2, RDDOX-4, RDDOX-6, and RDDOX-8, respectively (p < 0.05). Table 1 summarizes the pharmacokinetic parameters after administration of free DOX, RDDOX with different lengths and MSDOX. Free DOX was quickly eliminated from the circulating system after intravenous administration, showing T1/2β of 13.7 h, CL of 70.5 ng/mL/h, MRT of 14.6 h, and AUC0– of 545 ng/mL*h. MSDOX indicated a significantly longer systemic circulation time than that of free DOX (p < 0.05), at T1/2β of 88.2 h, AUC0– of 4910 ng /mL*h and MRT of 123.7 h. In addition, RDDOX showed significantly higher T1/2, AUC0–, and MRT, and significantly lower CL values than MSDOX. As shown in Table 1, while AUC0– showed no apparent difference among RDDOX-2, RDDOX-4, RDDOX-6, and RDDOX-8 at around 6500 ng /mL*h, RDDOX-8 showed T1/2β at 265.2 h and MRT at 378.2 h, which were significantly higher than those of fiber rods with shorter lengths (p < 0.05). Thus, the pharmacokinetics tests demonstrated that fiber rods had slower plasma elimination rates than free DOX 18

ACCEPTED MANUSCRIPT and MSDOX. This might be due to the less significant macrophage phagocytosis of fiber rods (Fig. 3), which could reduce the RES clearance and prolong the blood circulation [45].

CL (ng/mL/h)

Cmax (ng/mL)

70.5 ± 6.6

97.9 ± 0.4

NU

lengths and MSDOX to rats (n = 5).

RI P

T

Table 1. Pharmacokinetic parameters after intravenous administration of free DOX, RDDOX of different

123.7 ± 18.9

40.6 ± 5.5

123.7 ± 5.1

156.3 ± 20.1

39.6 ± 3.3

174.3 ± 7.0

181.6 ± 28.8

36.4 ± 3.9

169.4 ± 16.2

6886 ± 772

266.2 ± 31.5

20.7 ± 2.0

163.1 ± 15.2

6296 ± 109

378.2 ± 51.6

19.2 ± 2.9

147.0 ± 3.0

AUC0– (ng/mL*h)

MRT (h)

DOX

13.7 ± 4.7

545 ± 91

14.6 ± 3.8

MSDOX

88.2 ± 11.4

4910 ± 594

RDDOX-2

110.4 ± 14.9

6690 ± 619

RDDOX-4

128.6 ± 14.3

6432 ± 804

RDDOX-6

187.2 ± 26.7

RDDOX-8

265.2 ± 31.6

PT ED

MA

SC

T1/2β (h)

CE

3.6. In vivo tissue distribution of fiber rods

To gain a deep insight into the in vivo biodistribution behaviors of fiber rods, RDCOU with different

AC

lengths and MSCOU were intravenously injected into tumor-bearing mice, and tissues were collected for the determination of fluorescence intensities. As shown in Fig. 4b, MSCOU indicated a wide distribution after 1 h of injection, accumulating mainly in liver, spleen and kidney. Fiber rods were preferentially accumulated in lungs, and significantly higher lung retentions were detected for fiber rods with longer lengths (p < 0.05). After 24 h of injection, the accumulation of microspheres in liver, kidney, and spleen decreased, indicating a systemic clearance of microspheres as time went by. However, the amount of fiber rods in lungs increased, and the difference among fiber rods with different lengths became more evident compared with those at 1 h after injection. In addition, the accumulation of fiber rods in tumor tissues indicated a 2-fold increase after 24 h, while a slight decrease in the tumor retention was detected for microspheres. After 168 h of injection, there were still some fiber rods circulating in the blood, while few microspheres were detected. Fiber rods indicated a considerably higher tumor accumulation 19

ACCEPTED MANUSCRIPT than microspheres (p < 0.05), and RDCOU-2 and RDCOU-4 showed 4-fold higher accumulation than MSCOU. It is indicated that the nanoparticle shape plays a role in drifting from the center to vessel wall and penetration into the tumor tissues. In contrary to a streamline parallel to a vessel wall of spherical

T

particles under flow conditions, non-spherical particles exhibit more complex rotational and translation

RI P

trajectories, drifting from one side of the vessel to the other during flow, which increases the lateral drift of fiber rods towards the blood vessel walls in the microcirculation [13]. Thus, the enhanced tumor

SC

accumulation of fiber rods was correlated with a reduction in macrophage clearance (Fig. 3) and a better permeation or diffusion into tumor tissues.

NU

Fig. 4c summarizes the biodistribution of DOX in 4T1 tumor-bearing mice at 1 and 24 h after injection of free DOX, DOX-loaded fiber rods and microspheres. After 1 h of free DOX injection, a

MA

high retention in kidney was detected, indicating a rapid clearance from urea and discharge from the body. After 24 h of MSDOX injection, the DOX concentration in all the tissues indicated a decrease

PT ED

compared with those after 1 h of injection, and few DOX was detected in the blood circulation. However, the tumor accumulation of RDDOX-2 and RDDOX-4 indicated a slight increase after 24 h compared with those after 1 h of injection, and showed over 2.5 folds higher than that of MSDOX (p < 0.05). It indicated that fiber rods had a longer blood circulation, a higher accumulation in tumors, and a

AC

CE

higher retention capability in tumors than microspheres.

3.7. In vivo toxicity of fiber rods For acute toxicity test, RDDOX with different lengths and MSDOX were intravenously injected into tumor-bearing mice with an equivalent concentration of DOX at 2.0 mg/kg. The body weight monitoring, the hematological analysis and histological observation of lung and heart tissues were performed to evaluate the toxicity of various DOX formulations. Fig. S4a (Supplementary Material) summarized the body weight changes. Free DOX showed significantly higher weight loss than others, suggesting that the encapsulation of DOX in fiber rods and microspheres could decrease the systematic toxicity of DOX. It was noted that RDDOX-8 suppressed the growth of body weight, indicated that fiber rods with long lengths might cause some additional toxicity to mice. Blood routine testing was performed at 24 h after treatment, and the hematological results were summarized in Fig. S4b. There were slight decreases in WBC and PLT counts after treatment with free DOX. The hematological 20

ACCEPTED MANUSCRIPT markers of RBC, HGB, PLT and WBC indicated no significant differences after treatment with RDDOX and MSDOX in comparison with saline treatment (p > 0.05). It suggested no syndrome, such as hemolytic anemia, acute infection or bone marrow dysfunction was caused by these DOX formulations.

T

It is known that biochemical parameters such as AST, BUN, LDH and CK levels, are associated with

RI P

the physiological status of liver, kidney and heart [30]. As shown in Fig. S4b, the four important parameters in mice treated with RDDOX and MSDOX were similar to that of the control group, indicating

SC

that these DOX formulations did not cause any irreversible damages to liver, kidney and heart during the treatment period. However, LDH and CK values in mice treated with free DOX were 2 folds higher

NU

than those of the control group, demonstrating that the free DOX treatment led to serious damages to heart [46]. Additionally, HE staining of heart and lung tissues was used to assess the histological

MA

toxicities. The histopathological observations indicated normal lung sections after treatment with free DOX, DOX-loaded microspheres and fiber rods. As presented in Fig. S4c, serious cardiotoxicity was

PT ED

observed after free DOX injection, showing apparent hyperemia and myocardial fiber breakage with an infiltration of acute inflammatory cells. Except RDDOX-8 treatment, no obvious histopathological abnormalities such as degenerations or lesions, were observed after treatment with RDDOX-2, RDDOX-4,

CE

RDDOX-6 and MSDOX.

AC

3.8. In vivo antitumor efficacy of fiber rods The antitumor efficacy of DOX-loaded fiber rods was evaluated with respect to the inhibition of tumor growth, animal survival, histological and IHC analysis of tumors retrieved. Fig. 5a summarizes the changes in tumor volumes after 30 days of treatment with RDDOX of different lengths and MSDOX in comparison with free DOX and saline treatment as the control. In the saline treatment group, mice experienced a rapid tumor growth to about 1200 mm3 after 30 day, exhibiting significant differences compared with DOX treatment (p < 0.05). The inoculation of DOX into microspheres led to a better tumor growth inhibition, at average tumor volumes of 710 and 900 mm3 after treatment with MSDOX and free DOX, respectively (p < 0.05). This might be due to the enhanced blood circulation, improved accumulation in tumor tissue and controlled DOX release in the tumor mass. Among the DOX-loaded fiber rods, RDDOX-2 and RDDOX-4 showed a better antitumor efficacy than MSDOX, and the RDDOX-2 treatment showed a significantly lower tumor volume at around 570 mm3 than other groups (p < 0.05). 21

PT ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Fig. 5. (a) Tumor growth and (b) survive rates of 4T1 tumor-bearing mice after intravenous

CE

administration of free DOX, RDDOX with different lengths and MSDOX, using saline injection as the

AC

control (n = 6). (c) Typical IHC staining images of Ki-67 and capase-3 of tumors retrieved on day 21 after intravenous administration of free DOX, RDDOX with different lengths and MSDOX. Fig. 5b shows the animal survival rate in a Kaplan Meier plotting, indicating that all the treatments could prolong the life of tumor-bearing mice compared with the control. The median survival time of animals after MSDOX treatment was 37 days, which was longer than that of free DOX treatment at 32 days. Mice treated with RDDOX-2 rods, RDDOX-4, RDDOX-6, and RDDOX-8 led to median survival periods of 42, 40.5, 35.5, and 32 days, respectively. The most significant extension of animal survival time was observed after RDDOX-2 treatment, and there was still 20% of mice survival at day 62 after treatment. Fig. S5 (Supplementary Material) shows HE staining images of tumors retrieved after treatment for 21 days with RDDOX of different lengths and MSDOX compared with saline and free DOX treatment. There were still a large amount of living cells in tumors after saline treatment, and the tumor cells 22

ACCEPTED MANUSCRIPT showed obvious nucleolus cleavage and high extent of malignance. Necrotic areas could be obviously seen in tumor tissues after DOX treatment, and the necrotic regions after RDDOX-2 and RDDOX-4 treatment were larger than those after MSDOX and free DOX treatment, along with apparent vacuolus

T

degeneration of tumor cells.

RI P

The cell proliferation and apoptosis in the tumors retrieved were assessed by IHC staining of Ki-67 and caspase-3. Fig. 5c shows typical images of IHC staining of Ki-67 on tumor sections, indicating

SC

significantly higher cell proliferation in free DOX group than those after fiber rod and microsphere treatment. The positive cells were counted from five different areas for each sample, and Fig. S6a

NU

(Supplementary Material) summarizes the results after treatment. The treatment with MSDOX led to 42.6  7.3% of proliferative cells in tumors, which was significantly lower than that of free DOX treatment

MA

at 72.8  6.1%. In addition, RDDOX-2 showed significantly fewer Ki-67-positive cells (20.8  3.6%) than MSDOX and RDDOX with longer lengths (p < 0.05). Fig. 5c shows caspase 3-stainning images and

PT ED

Fig. S6b summarizes the counting results of positively stained cells. The treatment with RDDOX-2 showed significantly higher caspase 3-positive cells of around 83.2% than those of MSDOX, RDDOX-4, RDDOX-6, and RDDOX-8 (p < 0.05), at around 67.2%, 69.6%, 32.6%, and 18.8%, respectively.

CE

3.9. In vivo antimetastasis effect of fiber rods

AC

Metastasis usually occurs at a later stage of cancer and remains the main cause of cancer related deaths. 4T1 cells, as a metastatic breast cancer cell lines, primarily metastasize to lung through a hematogenous route [47]. The antimetastasis effect was determined form the surface metastatic nodules and histological staining of lung tissues retrieved after treatment for 21 days with fiber rods and microspheres, in comparison with free DOX and saline treatment. Fig. S7 (Supplementary Material) shows the lungs after fixation in Bouin’s solution, indicating the appearance of metastatic nodules on the lung surface from tumor-bearing mice. The lung surface was occupied by metastasized colonies in mice treated with saline, and the number of metastatic nodules was significantly reduced after treatment with free DOX and MSDOX. After treatment with RDDOX of different lengths, fewer metastatic nodules were observed in the lungs for RDDOX with shorter lengths. As shown in Fig. 6a, the numbers of metastatic nodules on the lung surface from tumor-bearing mice after RDDOX-2 and RDDOX-4 treatment were less than 5, which was significantly lower than those after treatment with MSDOX, 23

ACCEPTED MANUSCRIPT RDDOX-6, and RDDOX-8 (p < 0.05), at 13.2  2.8, 13.5  3.4, and 19.3  4.1, respectively. Histological examination of tumor metastasis was processed by HE staining of lung tissues retrieved. As shown in Fig. 6b, saline-treated group exhibited the most significant tumor metastasis in lungs compared to free

T

DOX, RDDOX-8 and RDDOX-6, while the treatment with MSDOX and RDDOX-4 resulted in fewer and

RI P

smaller metastatic nodules in lungs although tumor metastasis was still obvious. In particular,

AC

CE

PT ED

MA

NU

SC

RDDOX-2-treated group showed no clearly detectable foci of tumor metastasis in lung sections.

Fig. 6. The number of metastatic nodules on lung surface retrieved on day 21 after intravenous administration of free DOX, RDDOX with different lengths and MSDOX, using saline injection as the control (n = 3; *: p < 0.05). (b) HE staining images of lung sections retrieved on day 21 after intravenous administration of free DOX, RDDOX with different lengths and MSDOX, using saline injection as the control. ‘T’ represents metastatic tumors. As discussed above, the shape of fiber rods affected the cellular uptake, tissue distribution and blood 24

ACCEPTED MANUSCRIPT circulation process, as well as the antitumor and antimetastasis abilities. Compared with microspheres, fiber rods indicated a significantly lower internalization by macrophages, a higher uptake by 4T1 cells (Fig. 3), a higher AUC0– and a lower CL (Table 1), as well as a higher tumor accumulation and

T

retention capability (Fig. 4), leading to a higher antitumor efficacy. In addition, the longer circulation of

RI P

fiber rods in blood, preferential accumulation in lungs and higher plasma DOX levels (Fig. 4) inhibited the tumor cell migration and growth at the metastasis sites, indicating a lower metastasis after fiber rod

SC

treatment than that of microspheres. Among fiber rods with different lengths, RDDOX-2 was taken up maximally by 4T1 cells, inducing the highest toxicity and most significant tumor cell apoptosis (Fig. 3).

NU

Compared with RDDOX-6 and RDDOX-8, RDDOX-2 and RDDOX-4 showed higher tumor accumulations of DOX (Fig. 4), inducing more apparent necrosis and fewer metastatic nodules on the lung surface (Fig.

MA

6). Additionally, compared with RDDOX-4, the RDDOX-2 treatment demonstrated considerably lower tumor growth and tumor cell proliferation, and higher animal survival rate and tumor cell apoptosis

PT ED

(Fig. 5), as well as no clearly detectable foci of tumor metastasis in lung sections (Fig. 6).

4. Conclusion

A versatile and scalable method has been developed to construct fiber rods from electrospun fibers

CE

by ultrasonication using encapsulated NaCl nanoparticles as void-precursors. The inoculation of

AC

different amount of NaCl nanoparticles results in fiber rods with lengths from 2 to 8 m. Compared with microspheres, fiber rods indicates a significantly lower internalization by macrophages, a higher uptake by 4T1 cells, a longer circulation in blood and a higher tumor accumulation. Fiber rods with shorter lengths show a significantly higher in vitro cytotoxicity to tumor cells and a higher tumor accumulation of DOX. Among the fiber rods investigated, the RDDOX-2 treatment leads to the most significant inhibition on the tumor growth and metastasis. It suggests that a comprehensive understanding of the shape effects of electrospun fiber rods on the biological processes is expected as a guideline for the design of novel drug carriers to improve therapeutic outcomes.

Supplementary Material available: The fluorescent images of RDCOU, CLSM images of 4T1 cells after uptake of RDCOU and RDDOX, flow cytometry analysis of cells, in vivo toxicity, HE staining images of tumors and typical images of Bouin-fixed lungs after RDDOX treatment on tumor-bearing 25

ACCEPTED MANUSCRIPT mice are included.

Acknowledgements

T

This work was supported by National Natural Science Foundation of China (31470922 and

RI P

21274117), Specialized Research Fund for the Doctoral Program of Higher Education

SC

(20120184110004), and Fundamental Research Funds for the Central Universities (2062015YXZT06).

References

D.J. Crommelin, A.T. Florence, Towards more effective advanced drug delivery systems, Int. J.

NU

[1]

Pharm. 454 (2013) 496–511.

C.M. Dawidczyk, C. Kim, J.H. Park, L.M. Russell, K.H. Lee, M.G. Pomper, P.C. Searson,

MA

[2]

State-of-the-art in design rules for drug delivery platforms: Lessons learned from FDA-approved

[3]

PT ED

nanomedicines, J. Controlled Release 187 (2014)133–144. Y. Zhong, F. Meng, C. Deng, Z. Zhong, Ligand-directed active tumor-targeting polymeric nanoparticles for cancer chemotherapy, Biomacromolecules 15 (2014) 1955−1969. [4]

F. Danhier, O. Feron, V. Préat, To exploit the tumor microenvironment: Passive and active tumor

K. Loomis, K. McNeeley, R.V. Bellamkonda, Nanoparticles with targeting, triggered release, and

AC

[5]

CE

targeting of nanocarriers for anti-cancer drug delivery, J. Controlled Release 148 (2010) 135–146.

imaging functionality for cancer applications, Soft Matter. 7 (2011) 839–856. [6]

K.F. Pirollo, E.H. Chang, Does a targeting ligand influence nanoparticle tumor localization or uptake? Trends biotechnol. 26 (2008) 552–558.

[7]

Y.H. Bae, K. Park, Targeted drug delivery to tumors: Myths, reality and possibility, J. Controlled Release 153 (2011) 198–205.

[8]

O.C. Farokhzad, R. Langer, Impact of nanotechnology on drug delivery, ACS Nano 3 (2009) 16–20.

[9]

Y. Geng, P. Dalhaimer, S. Cai, R. Tsai, M. Tewari, T. Minko, D.E. Discher, Shape effects of filaments versus spherical particles in flow and drug delivery, Nat. Nanotechnol. 2 (2007) 249–255.

[10] A.J. Thompson, E.M. Mastria, O. Eniola-Adefeso, The margination propensity of ellipsoidal 26

ACCEPTED MANUSCRIPT micro/nanoparticles to the endothelium in human blood flow, Biomaterials 34 (2013) 5863–5871. [11] P. Decuzzi, B. Godin, T. Tanaka, S.Y. Lee, C. Chiappini, X. Liu, M. Ferrari, Size and shape effects in the biodistribution of intravascularly injected particles, J. Controlled Release 141 (2010)

T

320–327.

RI P

[12] S. Venkataraman, J.L. Hedrick, Z.Y. Ong, C. Yang, P.L.R. Ee, P.T. Hammond, Y.Y. Yang, The effects of polymeric nanostructure shape on drug delivery, Adv. Drug Delivery Rev. 63 (2011)

SC

1228–1246.

[13] J.M. Williford, J.L. Santos, R. Shyam, H.Q. Mao, Shape control in engineering of polymeric

NU

nanoparticles for therapeutic delivery, Biomater Sci. 3 (2015) 894–907. [14] J. Xu, D.H. Wong, J.D. Byrne, K. Chen, C. Bowerman, J.M. DeSimone, Future of the particle

MA

replication in nonwetting templates (PRINT) technology, Angew. Chemi. Int. Ed. 52 (2013) 6580–6589.

PT ED

[15] N. Doshi, S. Mitragotri, Macrophages recognize size and shape of their targets, PLoS One. 5 (2010) e10051.

[16] A. Greiner, J.H. Wendorff, Electrospinning: A fascinating method for the preparation of ultrathin fibers, Angew. Chemi. Int. Ed. 46 (2007) 5670–5703.

CE

[17] X..Y. Xie, .Y. Xie, H. Wang, C.Y. Liu, S.L. Yan, X.H. Li, Antitumor activities of emulsion

AC

electrospun fibers with core loading of hydroxycamptothecin via intratumoral implantation, Int. J. Pharm. 425 (2012) 19–28. [18] X. Li, Y. Zhang, R. Yan, W. Jia, M. Yuan, X. Deng, Z. Huang, Influence of process parameters on the protein stability encapsulated in

polyDL-lactide-poly(ethylene glycol) microspheres, J.

Controlled Release 68 (2000) 41–52. [19] L. Zhang, H.M. Cheng, J.M. Ma, Preparation of NaCl nanopartides in nonaqueous reverse microemulsion, Acta Phys. Chim. Sin. 18 (2002) 79–81. [20] F. Chen, H. Wan, T. Xia, X. Guo, H. Wang, Y. Liu, X. Li, Promoted regeneration of mature blood vessels by electrospun fibers with loaded multiple pDNA-calcium phosphate nanoparticles, Eur. J. Pharm. Biopharm. 85 (2013) 699–710. [21] X. Luo, H. Zhang, M. Chen, J. Wei, Y. Zhang, X. Li, Antimetastasis and antitumor efficacy promoted by sequential release of vascular disrupting and chemotherapeutic agents from 27

ACCEPTED MANUSCRIPT electrospun fibers, Int. J. Pharm. 475 (2014) 438–449. [22] C. Liu, J. Yuan, X. Luo, M. Chen, Z. Chen, Y. Zhao, X. Li, Folate decorated and

intracellular drug delivery, Mol. Pharm. 11 (2014) 4258−4269.

T

reduction-sensitive micelles assembled from amphiphilic polymer–camptothecin conjugates for

RI P

[23] A. Wakasugi, M. Asakawa, M. Kogiso, T. Shimizu, M. Sato, Y. Maitani, Organic nanotubes for drug loading and cellular delivery, Int. J. Pharm. 413 (2011) 271–278.

camptothecin

and

folate

conjugations

SC

[24] Y. Zhang, M. Chen, X. Luo, H. Zhang, C. Liu, H. Li, X. Li, Tuning multiple arms for on

star-shaped

copolymers

to

enhance

NU

glutathione-mediated intracellular drug delivery, Polym. Chem. 6 (2015) 2192– 2203. [25] K. Alhareth, C. Vauthier, F. Bourasset, C. Gueutin, G. Ponchel, F. Moussa, Conformation of

MA

surface-decorating dextran chains affects the pharmacokinetics and biodistribution of doxorubicin-loaded nanoparticles, Eur. J. Pharm. Biopharm. 81 (2012) 453–457.

PT ED

[26] Y. Zhang, M. Huo, J. Zhou, S. Xie, PKSolver: An add-in program for pharmacokinetic and pharmacodynamic data analysis in Microsoft Excel, Comput. Meth. Prog. Bio. 99 (2010) 306–314.

[27] Y.M. Kang, G.H. Kim, J.I. Kim, B.N. Lee, S.M. Yoon, J.H. Kim, M.S. Kim, In vivo efficacy of

CE

an intratumorally injected in situ-forming doxorubicin/poly(ethylene glycol)-b-polycaprolactone

AC

diblock copolymer, Biomaterials 32 (2011) 4556–4564. [28] S.Q. Liu, N. Wiradharma, S.J. Gao, Y.W. Tong, Y.Y. Yang, Bio-functional micelles self-assembled from a folate-conjugated block copolymer for targeted intracellular delivery of anticancer drugs, Biomaterials 28 (2007) 1423–1433. [29] Y. Hamada, K. Gonda, M. Takeda, A. Sato, M. Watanabe, T. Yambe, S. Satomi, N. Ohuchi, In vivo imaging of the molecular distribution of the VEGF receptor during angiogenesis in a mouse model of ischemia, Blood 118 (2011) e93–100. [30] X. Han, Z. Li, J. Sun, C. Luo, L. Li, Y. Liu, Y. Du, S. Qiu, X. Ai, C. Wu, Stealth CD44-targeted hyaluronic acid supramolecular nanoassemblies for doxorubicin delivery: Probing the effect of uncovalent pegylation degree on cellular uptake and blood long circulation, J. Controlled Release 197 (2015) 29–40. [31] R. Liu, J.B. Wolinsky, J. Walpole, E. Southard, L.R. Chirieac, M.W. Grinstaff, Y.L. Colson, 28

ACCEPTED MANUSCRIPT Prevention of local tumor recurrence following surgery using low-dose chemotherapeutic polymer films, Ann. Surg. Oncol. 17 (2010) 1203–1213. [32] W. Song, Z. Tang, D. Zhang, Y. Zhang, H. Yu, M. Li, S. Lv, H. Sun, M. Deng, X. Chen,

RI P

docetaxel and cisplatin, Biomaterials 35 (2014) 3005–3014.

T

Anti-tumor efficacy of c(RGDfK)-decorated polypeptide-based micelles co-loaded with

[33] O. Kriha, M. Becker, M. Lehmann, D. Kriha, J. Krieglstein, M. Yosef, S. Schlecht, R.B.

controlled

movement

of

short

electrospun

SC

Wehrspohn, J.H. Wendorff, A. Greiner, Connection of hippocampal neurons by magnetically polymer

fibers—A

route

to

magnetic

NU

micromanipulators, Adv. Mater. 19 (2007) 2483–2485.

[34] M. Sawawi, T.Y. Wang, D.R. Nisbet, G.P. Simon, Scission of electrospun polymer fibres by

MA

ultrasonication, Polymer 54 (2013) 4237–4252.

[35] K. Friedemann, T. Corrales, M. Kappl, K. Landfester, D. Crespy, Facile and large-scale

PT ED

fabrication of anisometric particles from fibers synthesized by colloid-electrospinning, Small 8 (2012) 144–153.

[36] Y. Yang, X. Li, L. Cheng, S. He, J. Zou, F. Chen, Z. Zhang, Core-sheath structured fibers with pDNA polyplex loadings for the optimal release profile and transfection efficiency as potential

CE

tissue engineering scaffolds, Acta Biomater. 7 (2011) 2533–2543.

AC

[37] S.P. Jiang, S.N. He, Y.L. Li, D.L. Feng, X.Y. Lu, Y.Z. Du, H.Y. Yu, F.Q. Hu, H. Yuan, Preparation and characteristics of lipid nanoemulsion formulations loaded with doxorubicin, Int. J. Nanomed. 8 (2013) 3141–3150.

[38] W. Lu, Y. Zhang, Y.Z. Tan, K.L. Hu, X.G. Jiang, S.K. Fu, Cationic albumin-conjugated pegylated nanoparticles as novel drug carrier for brain delivery, J. Controlled Release 107 (2005) 428–448. [39] S.W. Morton, K.P. Herlihy, K.E. Shopsowitz, Z.J. Deng, K.S. Chu, C.J. Bowerman, J.M. DeSimone, P.T. Hammond, Scalable manufacture of built-to-order nanomedicine: Spray-assisted layer-by-layer functionalization of PRINT nanoparticles, Adv. Mater. 25 (2013) 4707–4713. [40] K. Zhang, H. Fang, Z. Chen, J.S.A. Taylor, K.L. Wooley, Shape effects of nanoparticles conjugated with cell-penetrating peptides (HIV Tat PTD) on CHO cell uptake, Bioconjugate chem. 19 (2008) 1880–1887. [41] Z. Zhou, X. Ma, E. Jin, J. Tang, M. Sui, Y. Shen, E.A. Van Kirk, W.J. Murdoch, M. Radosz, 29

ACCEPTED MANUSCRIPT Linear-dendritic drug conjugates forming long-circulating nanorods for cancer-drug delivery, Biomaterials 34 (2013) 5722–5735. [42] S.E. Gratton, P.A. Ropp, P.D. Pohlhaus, J.C. Luft, V.J. Madden, M.E. Napier, J.M. DeSimone,

T

The effect of particle design on cellular internalization pathways, Proc. Natl. Acad. Sci. U. S. A.

RI P

105 (2008) 11613–11618.

[43] J.D. Robertson, G. Yealland, M. Avila-Olias, L. Chierico, O. Bandmann, S.A. Renshaw, G.

SC

Battaglia, pH-sensitive tubular polymersomes: formation and applications in cellular, ACS Nano 8 (2014) 4650–4661.

NU

[44] J.A. Champion, S. Mitragotri, Shape induced inhibition of phagocytosis of polymer particles, Pharm. Res. 26 (2009) 244–249.

MA

[45] J. Xiao, X. Duan, Q. Yin, Z. Zhang, H. Yu, Y. Li, Nanodiamonds-mediated doxorubicin nuclear delivery to inhibit lung metastasis of breast cancer, Biomaterials 34 (2013) 9648–9656.

PT ED

[46] M.G. Ignatova, N.E. Manolova, R.A. Toshkova, I.B. Rashkov, E.G. Gardeva, L.S. Yossifova, M.T. Alexandrov, Electrospun nanofibrous mats containing quaternized chitosan and polylactide with in vitro antitumor activity against HeLa cells, Biomacromolecules 11 (2010) 1633–1645. [47] C. Qin, B. He, W. Dai, H. Zhang, X. Wang, J. Wang, X. Zhang, G. Wang, L. Yin, Q. Zhang,

CE

Inhibition of metastatic tumor growth and metastasis via targeting metastatic breast cancer by

AC

chlorotoxin-modified liposomes, Mol. Pharm. 11 (2014) 3233–3241.

30

ACCEPTED MANUSCRIPT

AC

CE

PT ED

MA

NU

SC

RI P

T

Graphical Abstract

31