Cellular uptake, antitumor response and tumor penetration of cisplatin-loaded milk protein nanoparticles

Biomaterials 34 (2013) 1372e1382

Contents lists available at SciVerse ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Cellular uptake, antitumor response and tumor penetration of cisplatin-loaded milk protein nanoparticles Xu Zhen, Xin Wang, Chen Xie, Wei Wu, Xiqun Jiang* Laboratory of Mesoscopic Chemistry and Department of Polymer Science & Engineering, College of Chemistry & Chemical Engineering, Nanjing University, Nanjing 210093, PR China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 4 September 2012 Accepted 25 October 2012 Available online 16 November 2012

The casein nanoparticles cross-linked by transglutaminase were prepared, and cisplatin (CDDP), as a model antitumor drug, was loaded into the casein nanoparticles. These nanoparticles were characterized by dynamic light scattering (DLS), transmission electron microscopy (TEM), and zeta potential. The uptake and penetration of nanoparticles in 2- and 3-dimensional SH-SY5Y cells were examined at 37
C and 4
C. The in vivo biodistribution of the nanoparticles was investigated using near-infrared fluorescence (NIRF) imaging and ion-coupled plasma mass spectrometry (ICP-MS). The antitumor effect of CDDP-loaded nanoparticles was evaluated on hepatic H22 tumor-bearing mice model via intravenous administration. It is found that the obtained nanoparticles showed spherical shape with the size of 257 nm, and drug loading content of 10%. These CDDP-loaded casein nanoparticles have the extraordinary capabilities to penetrate cell membrane barriers, target tumor and inhibit tumor growth. The tumor growth inhibition of CDDP-loaded nanoparticles is 1.5-fold higher than that of free CDDP. Further, the penetration examination of the CDDP-loaded casein nanoparticles in the tumor tissue demonstrated that the nanoparticles had the ability to penetrate the tumor and deliver CDDP to the tumor more deeply and affect the cells far from the vasculature. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: Casein Protein nanoparticles Drug delivery Multicellular tumor spheroids Tumor penetration

1. Introduction Recently, the nanoparticles based on the proteins such as albumin, gelatin and silk protein have been widely used as the drug delivery systems [1e3] due to their good biocompatibility, biodegradation into natural products, lack of toxicity, and nonantigenicity. Moreover, protein-based nanoparticles have active targeting ability except passive targeting based on the enhanced permeation and retention effect (EPR effect). For example, albumin nanoparticles can utilize albumin receptor (gp60)-mediated transcytosis through microvessel endothelial cells in angiogenic tumor vasculature and target the albumin-binding protein SPARC (Secreted Protein, Acidic and Rich in Cysteine), which is overexpressed in a majority of tumors [4]. Although two albumin-based particulate formulations have been approved for clinical use [5,6], the utilization of native proteins in terms of particulate drug delivery systems is still limited. In particular, the inherent bioactivities of protein-based nanoparticles have not been fully explored yet. For example, some proteins including homeodomain transcription factors [7], the herpes simplex virus-1 protein [8] and the

* Corresponding author. Tel./fax: þ86 25 83317761. E-mail address: [email protected] (X. Jiang). 0142-9612/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biomaterials.2012.10.061

HIV-1 transactivator Tat protein [9,10] are capable of penetrating cellular membrane, leading to increase drug retention in tumors and overcome drug resistance by increased intracellular drug concentrations in tumor. However, up to now, the membrane penetration of therapeutic cargos is mainly dependent on a series of short peptides called cell-penetrating peptides (CPPs), not cellpenetrating proteins and nanoparticles themselves. The strategy is often based on the conjugation of CPPs with cargos, thereby holding the risk to alter the biological activity of the cargoes. Moreover, it is generally accepted that CPPs mostly use energydependent endocytosis pathways across the membrane bilayer [11,12], resulting in endosomal sequestration and decreased bioavailability in the cytoplasm or nucleus [13]. In addition, as small molecular peptides, these CPPs are non-selective between tumor and normal tissues. Thus, the development of the protein-based nanoparticles drug delivery systems in which the particles themselves have drug-carrying and non-endocytotic cell-penetrating abilities is highly desirable. Casein is a main ingredient of milk proteins, which is comprised of as1-, as2-, b-, and k-casein (molar ratio 4:1:4:1, respectively) [14]. The caseins are proline-rich proteins and have distinct hydrophobic and hydrophilic domains. As a natural food product, casein is inexpensive, readily available, non-toxic and biodegradable [15]. In recently years, some casein-based oral drug delivery systems have

X. Zhen et al. / Biomaterials 34 (2013) 1372e1382

been developed [16,17]. In our previous work [18], we prepared casein hollow spheres by molecular self-assembly in aqueous solution, and found that these casein hollow spheres have extraordinary capability to penetrate cell barriers. Unlike most of CPPs which cross cellular membranes in an endocytosis fashion, the casein hollow spheres can penetrate cell membrane in a nonendocytosis-dominant uptake mechanism, irrespective of cancer cell types and temperatures [19]. However, due to their relatively larger hydrodynamic diameter (about 379 nm at pH ¼ 3.2 and 500 nm at pH ¼ 7.4), these casein hollow spheres were not considered to be an ideal injectable drug delivery system. In order to fully explore and utilize the bioactivity of casein, at present work, we further design and optimize the preparation strategy of casein particles. In particular, we use a natural crosslinker, transglutaminase (TGase) to replace glutaraldehyde which commonly used. These modifications significantly decrease the size of casein particles and improve the particle stability in various pH media. Further, we investigate the penetration of the casein nanoparticles in two-dimensional monolayer cells and threedimensional multicellular tumor spheroids (MCTS), and the distribution in tumor-bearing mice based on near-infrared fluorescence (NIRF) imaging. When these casein nanoparticles are used as the carriers to delivering cisplatin (CDDP) in cancer treatment, the tumor suppression is evaluated. Finally, the tumor penetration of CDDP-loaded casein nanoparticles in H22 tumor-bearing mouse is assessed by immunohistochemical analysis.

1373

2.4. Characterization of the nanoparticles The mean hydrodynamic diameter and size distribution of nanoparticles were evaluated by dynamic light scattering (DLS) using a Brookheaven BI9000AT system (Brookheaven Instruments Corporation, USA). Zeta potentials of the samples were measured by Zetaplus (Brookheaven Instruments Corporation, USA). All DLS measurements were done with a laser wavelength of 660.0 nm at 25 ◦C, and each batch was analyzed in triplicate. The obtained dried samples were mixed with KBr powder and pressed to a plate for FT-IR measurement. FT-IR spectra were recorded on a vacuum FT-IR Spectrometer (Bruker VERTEX80V, Germany). The morphology of casein nanoparticles was observed by transmission electron microscopy (TEM, JEOLTEM-100, Japan). One drop of the nanoparticle suspension was placed on a 200-mesh nitrocellulose-covered copper grid. The grid was allowed to dry at room temperature without staining, and was examined with the TEM. 2.5. Drug loading content and encapsulation efficiency The Pt content in the casein nanoparticles was measured by ion-coupled plasma mass spectrometry (ICP-MS, PerkineElmer Corporation, USA). Briefly, the CDDPloaded casein nanoparticles were decomposed in hot nitric acid. After being evaporated to dryness, they were dissolved in 2 N hydrochloric acid solution. Then, the Pt concentration in the solution was measured by ICP-MS. The following equations were used to evaluate the drug loading content and encapsulation efficiency.

Drug loading content % ¼ Encapsulation % ¼

Weight of the drug in nanoparticles  100% Weight of the feeding drug

Weight of the drug in nanoparticles  100% Weight of the feeding drug

2.6. In vitro CDDP release from the nanoparticles 2. Materials and methods 2.1. Materials Casein from bovine milk which is dephosphorylation and made of as1-, as2-, b-, and k-casein was purchased from SigmaeAldrich. Acrylic acid (AA, Nanjing Chemical Reagent Co., Ltd.) was distilled under reduced pressure in nitrogen atmosphere. Propionic acid was purchased from Shanghai Lingfeng Chemical Reagent Co., Ltd. Potassium peroxydisulfate (K2S2O8, Nanjing Chemical Reagent Co., Ltd.) was recrystallized from deionized water before use. Rhodamine B isothiocyanate (RBITC) was purchased from SunShineBio Co., Ltd (Nanjing, China). Cisplatin (CDDP) was provided by shandongboyuan Co., Ltd. (Jinan, China). Rat anti-mouse CD31 (BD PharmingenÔ), Alex-488 streptavidin, donkey anti-rabbit Alex-488, donkey anti-rat Alex-594 were obtained from Invitrogen. Cleaved caspase-3 antibody was purchased from Cell Signaling Technology Company. All other reagents were of analytical grade and used without further purification. Murine hepatic H22 cell line was obtained from Shanghai Institute of Cell Biology (Shanghai, China). Male ICR mice (6e8 weeks old and weighing 18e22 g) were purchased from Animal Center of Drum-Tower Hospital (Nanjing, China). 2.2. Preparation of casein nanoparticles Casein (80 mg) was dispersed in 40 mL of acrylic acid (133 mL) and propionic acid (35 mL) aqueous solution under magnetic stirring and the temperature was raised to
90 C. When the solution became clear, a predetermined amount of K2S2O8 was added into the reaction system under a nitrogen atmosphere to initiate the polymerization of AA monomers. As the opalescence appeared in the reaction system, which was a signature of the formation of casein-poly(acrylic acid) (casein-PAA) nanoparticles, the reaction was allowed to proceed for another 10 min at 90 ◦C. The suspension was filtered with filter paper to remove any aggregation. The resultant suspension was dialyzed against water (pH ¼ 3.0) for 24 h using a dialysis bag with a cutoff molecular weight of 14 kDa to remove residual monomers and other small molecules. To improve the stability of casein-PAA nanoparticles, the pH of the suspension was adjusted to around 5.5, and then a predetermined amount of TGase was added to the suspension to cross-link casein moiety of casein-PAA nanoparticles. The crosslinking reaction was allowed for 72 h at 37
C under magnetic stirring. Finally, the cross-linked product was again dialyzed against distilled water for 24 h to remove residual TGase, PAA in the particles and the other small molecules. 2.3. Preparation of CDDP-loaded casein nanoparticles CDDP was dissolved in the suspension of casein nanoparticles (1 mg/mL), which was allowed to shake at 37
C for 48 h to obtain CDDP-loaded casein nanoparticles. And then, this suspension was treated with the method developed by Kataoka’s group to remove free CDDP [20]. Briefly, the unbound CDDP was removed by dialysis against distilled water using a dialysis bag with a molecular weight cutoff 14 kDa for 48 h.

The release of CDDP from casein nanoparticles in phosphate buffered saline (PBS, 0.01 M phosphate buffer, pH 7.4, 0.15 M NaCl) at 37
C was evaluated by the dialysis method as reported previously [18]. Briefly, the CDDP-loaded nanoparticle solution of known platinum drug concentration was placed inside a dialysis bag (MWCO, 14 kDa). The dialysis bag was placed in PBS buffer and gently shaken at 37
C in a water bath at 80 rpm. The released Pt outside of the dialysis bag was sampled at a predetermined time interval, and measured by ion-coupled plasma mass spectrometry (ICP-MS, HewlettePackard 4500). 2.7. Stability of CDDP-loaded nanoparticles The stability of CDDP-loaded nanoparticles was observed by measuring the particle size as a function of time. The samples were dissolved in PBS with pH of 7.4 at 37
C and evaluated by DLS with a Brookheaven BI9000AT system. At various time points, the hydrodynamic diameters and the scattering light intensity were measured by DLS at 25
C. 2.8. In vitro cytotoxicity and cellular uptake Cytotoxicity of CDDP-loaded casein nanoparticles against human derived neuroblastoma cell line (SH-SY5Y cells) was assessed by MTT assay. SH-SY5Y cells were cultured in DMEM (Dulbecco’s modified Eagle essential medium), supplemented with 10% (v/v) inactivated FBS (fetal bovine serum), and antibiotics (10 U mL1 penicillin and 10 mgmL1 streptomycin). Cells were cultured in a humidified 5% CO2 incubator at 37
C. SH-SY5Y cells were seeded on 96-well plates with a density around 5000 cells/well and allowed to adhere for 24 h prior to the assay. The cells were co-incubated with a series of doses of free CDDP, empty casein nanoparticles and CDDP-loaded nanoparticles at 37
C for 24 h. Then, 50 mL of 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) indicator dye (5 mg/mL in PBS, pH ¼ 7.4) was added to each well, and the cells were incubated for another 2 h at 37
C in the dark. The medium was withdrawn and 200 mL acidified isopropanol (0.33 v/v HCl in isopropanol) was added in each well and agitated thoroughly to dissolve the formazan crystals. The solution was transferred to 96well plates and immediately monitored on a microplate reader (Bio-Rad, Hercules, CA, USA.). Absorption was measured at a wavelength of 490 nm and 620 nm as a reference wavelength. The values obtained were expressed as a percentage of the control cells to which no drugs were added. To trace the cellular uptake of nanoparticles, the casein nanoparticles were labeled with rhodamine B isothiocyanate (RBITC). The RBITC-labeled nanoparticles were prepared as follows: 1 mL of anhydrous DMSO containing 1 mg RBITC was added into 4 mL casein nanoparticles solution, and the mixture was stirred for 24 h at room temperature in the dark. Then, the RBITC-labeled casein nanoparticles were separated by centrifugation and unreacted RBITC was removed. Finally, the obtained RBITC-labeled nanoparticles were dispersed in aqueous solution for in vitro cellular uptake.

1374

X. Zhen et al. / Biomaterials 34 (2013) 1372e1382

The SH-SY5Y cells were seeded at a density of 1  106 cells/well in a 6-well plate containing a cover glass and allowed to adhere for 24 h in a humidified atmosphere of 5% CO2 at 37
C to achieve a confluence of approximately 80%. After 24 h incubation, the medium was replaced by fresh temperature-equilibrated complete medium, and a determined amount of the RBITC-labeled casein nanoparticles were added into the plates. With 4 h further incubation, the cover glass containing adherent cells was taken out, washed three times with PBS to remove free RBITClabeled casein nanoparticles and fixed through inversely putting the cover glass onto the glass slide. Cells were observed by confocal laser scanning microscopy (CLSM, Zeiss LSM 710, Germany) at an excitation wavelength of 543 nm. 2.9. Penetration of the nanoparticles in multicellular tumor spheroids (MCTS) In order to produce multicellular tumor spheroids, a layer of poly(2hydroxyethylmethacrylate) (PHEMA) film was coated on the bottom of tissue culture flasks (T25). 450 mg PHEMA was dissolved into 30 mL 95% ethanol solution and the mixture turned slowly for 24 h at 37
C. After PHEMA was completely dissolved, 4 mL mixture was pipeted into a tissue culture flask. The flask then was allowed to dry for 48 h at 37
C. To ensure sterile, PHEMA coated flask must be exposed to ultraviolet light for 1 h before use. The SH-SY5Y monolayer cells were trypsinized to ensure single-cell suspension and count the cell numbers using a hemocytometer. 5  105 SH-SY5Y cells in 5 mL of fresh DMEM medium were placed into PHEMA-coated flask. The cells were incubated at 37
C in humidified atmosphere with 5% CO2 and the culture medium was replaced every other day. Multicellular tumor spheroids (about 200e350 mm in diameter) formed spontaneously in 7e9 days. The uptake of nanoparticles by MCTS was observed by CLSM. SH-SY5Y multicellular tumor spheroids with diameters between 250 and 350 mm

were harvested after approximately 7e9 days of growth. For each experiment, about 20 spheroids were handpicked with a Pasteur pipette and transferred to a 5 mL eppendorf tube. Appropriate concentrations of RBITC-labeled nanoparticles were then added to the spheroids suspension and co-cultured at 37
C and 4
C, respectively, for a defined time period. The medium was then removed and spheroids were washed with PBS (pH ¼ 7.4) before observation with CLSM. 2.10. Real-time NIRF imaging NIR-797-isothiocyanate was used to label CDDP-loaded casein nanoparticles. Briefly, 2 mg of NIR-797-isothiocyanate was dissolved in 1 mL of DMSO. 200 mL of the dye solution was added to 4 mL of nanoparticle solution and shaken gently at 37
C for 12 h. Unconjugated NIR-797 was removed by ultrafiltration (MWCO, 100 kDa) for 24 h. All animal studies were performed in compliance with guidelines set by the Animal Care Committee at Drum-Tower Hospital. ICR mice were inoculated on the left flank with H22 tumor cells (4e6  106 cells per mouse). On day 7, when the tumor volume reached about 100 mm3, the NIR-797 labeled and CDDP-loaded nanoparticles were intravenously administrated into H22 tumor bearing mice. After i.v. administration, the time-dependent biodistribution in tumor bearing mice was imaged using a MaestroÔ EX fluorescence imaging system (Cambridge Research & Instrumentation, CRi, USA). The mice were anesthetized and placed on an animal plate heated to 37
C. The NIRF at 745 nm was collected and exposure time was set to 1600 ms. Scans were conducted at 1 h, 2 h, 4 h, 8 h, 24 h, 48 h, 72 h, 96 h, 120 h, and 144 h post administration. Thereafter, tumor bearing mice were sacrificed. Tumor, heart, liver, spleen, kidneys, stomach, intestine and lungs were harvested for isolated organ imaging to estimate the tissue distribution of nanoparticles.

Fig. 1. (A) FT-IR spectra of casein, PAA, casein-PAA nanoparticles and casein nanoparticles; (B) The hydrodynamic diameter distribution of casein nanoparticles and casein-PAA nanoparticles (inset); (C) The hydrodynamic diameter distribution of CDDP-loaded casein nanoparticles. (D) TEM image of CDDP-loaded casein nanoparticles.

X. Zhen et al. / Biomaterials 34 (2013) 1372e1382 2.11. Biodistribution examination in vivo ICR mice were inoculated on the left flank with H22 tumor cells (4-6  106 cells per mouse). The CDDP-loaded nanoparticles were administered intravenously at a dose of 3 mg/kg on a CDDP basis 7 days after the inoculation. The mice were sacrificed at 1, 2, 4, 8, 12, 24 and 48 h after administration (n ¼ 3 at each time point). Subsequently, the tumor, liver, spleen, kidney, lung, and heart were excised, and blood was collected. The blood and each of the organs were decomposed in hot perchloric acid and nitric acid. After being evaporated to dryness, resulting precipitates were dissolved in 0.5 M hydrochloric acid solution. Quantitative analysis of Pt was performed by ICP-MS. The data were normalized to the tissue weight. 2.12. In vivo anticancer efficacy H22 cells (4-6  106 cells per mouse) were inoculated subcutaneously to ICR mice at the left flank in the same way as described above. 7 days after inoculation (tumor volume reached about 100 mm3), the mice were randomly allocated to four groups for different formulations, and this day was designated as Day 1. Each group contains 10 mice. Tumor-bearing mice were injected via the tail vein with saline, empty casein nanoparticles, free CDDP (3 mg/kg), and CDDP-loaded casein nanoparticles (3 mg/kg on CDDP basis), respectively. The tumor volumes were measured in two dimensions every other day using a slide caliper for 17 days. The tumor volume was calculated as V ¼ a  b2/2 (a and b: the longest and shortest diameter of tumor, respectively). Moreover, the survival rates were monitored as an indicator of systemic toxicity. 2.13. Penetration in the tumor

1375

H22 tumors of mice were harvested at 12 h and 24 h post-administration of biotin-conjugated casein nanoparticles. Tumors were fixed in 4% paraformaldehyde for 4 h at 4
C, incubated in 25% sucrose solution for 12 h and frozen in Optimal Cutting Temperature (O.C.T.) embedding medium. 6 mm sections were cut for histological analysis. For immunostaining, the sections were rehydrated in PBS containing 0.1% (v/v) Triton X-100 at pH ¼ 7.4 for 10 min followed by incubation with 3% BSA for 1 h at 37
C, then exposed to the appropriate primary monoclonal antibody (1:400, rat monoclonal anti-mouse CD31, BD Pharmingen, San Jose, California) for 1 h in a humidified chamber at 37
C. The slides were washed three times with PBS containing 0.05% (v/v) tween 20 at pH ¼ 7.4 for 5 min each and then counterstained with a Alexa 594 conjugated donkey anti-rat secondary antibody (1:1000, Molecular Probes, Eugene, OR) in a humidified chamber at 37
C in the dark for 30 min. Next, the slides were washed three times with PBS and counterstained with Alex 488 conjugated streptavidin (1:1000, Invitrogen, Camarillo, CA) in a humidified chamber at 37
C in the dark for 30 min. After then, the slides were washed three times with PBS and the nucleus were labeled with DAPI. The slides were mounted by one drop of FluoromountÔ Aqueous Mounting Medium (Invitrogen, Caelsbad, USA) and observed with a Zeiss LSM 710 confocal microscope. 2.14. Observation of apoptosis in tumor The location of the apoptotic cells in tumor was examined. H22 tumors were taken from the mouse groups received saline, empty casein nanoparticles, free CDDP (3 mg/kg) and CDDP-loaded nanoparticles (3 mg/kg CDDP eq.) at 3 day postadministration, respectively. The tumors were fixed in 4% paraformaldehyde for 4 h at 4 ◦C, incubated in 25% sucrose solution for 12 h and frozen in Optimal Cutting Temperature (O.C.T.) embedding medium. 6 mm sections were cut for histological analysis. For immunostaining, the sections were rehydrated in PBS containing 0.1%

To trace the location of the nanoparticles in the tumor tissue, the casein nanoparticles were conjugated with biotin. The biotin-conjugated nanoparticles were prepared based on the method reported [21]. Briefly, the nanoparticle solution was mixed biotin hydrazine at room temperature and stirred. 1 h later, a predetermined amount of NaCNBH3 was added to reduce the Schiff base generated, and the solution kept stirring for 24 h. Finally, the resulting nanoparticles were purified by dialyzed against distilled water extensively.

Fig. 2. (A) The stability of CDDP-loaded casein nanoparticles in PBS at 37
C (B) CDDP release profile of CDDP-loaded casein nanoparticles in pH ¼ 7.4 PBS saline at 37
C.

Fig. 3. (A) In vitro cytotoxicity of free CDDP and CDDP-loaded casein nanoparticles against SH-SY5Y cells after co-incubation for 24 h; (B) In vitro cytotoxicity of empty casein nanoparticles against SH-SY5Y cells after co-incubation for 24 h.

1376

X. Zhen et al. / Biomaterials 34 (2013) 1372e1382

(v/v) Triton X-100 in PBS at pH ¼ 7.4 for 10 min followed by incubation with 3% BSA for 1 h at 37
C, then exposed to the appropriate primary monoclonal antibody (1:400, rat monoclonal anti-mouse CD31, BD Pharmingen, San Jose, California) and primary polyclonal antibody against cleaved caspase-3 (Asp175, Cell Signaling Technology) for 1 h in a humidified chamber at 37
C, respectively. The slides were washed three times with PBS containing 0.05%(v/v) tween 20 in PBS at pH ¼ 7.4 for 5 min each and then counterstained with Alexa 594 conjugated donkey anti-rat secondary antibody (1:1000, Molecular Probes, Eugene, OR) and Alexa 488 conjugated donkey anti-rabbit secondary antibody (1:1000, Molecular Probes, Eugene, OR) in a humidified chamber at 37
C in the dark for 30 min, respectively. Next, the slides were washed three times with PBS and the nucleus were labeled with DAPI. The slides were mounted by one drop of FluoromountÔ Aqueous Mounting Medium (Invitrogen, Caelsbad, USA) and observed with a Zeiss LSM 710 confocal microscope. 2.15. Statistical analysis Quantitative data were expressed as mean  SD. Statistical comparisons were made by ANOVA analysis and Student’s t-test. A P value <0.05 was considered statistically significant.

3. Results and discussion 3.1. Preparation of casein nanoparticles In our previous work [18,19], the casein particles with hollow structure were prepared. It was found that these casein particles showed the cell-penetrating behaviors with energy-independent fashion. As driven by this inspiring result, we hypothesize the casein nanoparticles should be favorable for drug delivery. However, these casein particles had a mean hydrodynamic diameter of 379 nm at pH ¼ 3.2 and 500 nm at pH ¼ 7.4, which are not suitable for effectively targeting into solid tumors. Thus,

we encountered the problems of the size and kinetic stability of such casein particles. In order to improve the delivering ability of casein particles in vivo, we optimize the preparation strategy and switch the cross-linker of casein nanoparticles from glutaraldehyde to TGase to reduce the potential toxicity and enhance the stability of nanoparticles [22]. Initially, we prepared the caseinpoly(acrylic acid)(casein-PAA) nanoparticles by polymerizing acrylic acid in the presence of casein, and followed by selectively cross-linked casein component by catalyze the acyl transfer reaction between the g-carboxyl groups of glutaminyl residues and primary amines via TGase [23,24]. TGase are widely distributed in various organs, tissues, and body fluids [25]. TGasecatalyzed reactions are extremely specific for a particular glutamine residue in native protein substrates. The cross-linked protein products are highly resistant to environmental changes and proteolytic degradation [26]. Thus, we removed PAA component from casein-PAA nanoparticles by dialysis against water at pH 7.4 to obtain the pure casein nanoparticles due to the non-covalent interaction between casein and PAA. Finally, CDDP was loaded into the casein nanoparticles. The loading content of CDDP in the nanoparticles is 10% and the encapsulation efficiency is 72%. To demonstrate the removal of PAA component from the nanoparticles after the dialysis, FT-IR measurements of the casein nanoparticles before and after dialysis were carried out. As shown in Fig. 1A, the absorption peak corresponding to carbonyl stretching of PAA located at w1718 cm1 disappears completely after dialysis for 24 h at pH of 7.4, indicating that PAA component is removed from the nanoparticles and the pure casein nanoparticles are obtained.

Fig. 4. CLSM images of SH-SY5Y cells after 4 h incubation with casein nanoparticles labeled by Rhodamine B. (A) at 37
C and (B) at 4
C; (C) CLSM images of multicellular tumor spheroids incubation with casein nanoparticles labeled by Rhodamine B at 37
C and at 4
C; (D) Mean fluorescence intensity in multicellular tumor spheroids for RBITC-labeled casein nanoparticles at 37
C and 4
C. Scale bar ¼ 50 mm.

X. Zhen et al. / Biomaterials 34 (2013) 1372e1382

3.2. Stability and in vitro release of CDDP-loaded casein nanoparticles The size distribution of empty casein nanoparticles and CDDPloaded casein nanoparticles were measured by DLS and shown in Fig. 1B and C. It is found that the size of nanoparticles is about 265 nm and 280 nm at pH of 7.4 before and after removal of PAA, respectively. After CDDP loading, the size of casein nanoparticles decreases to 257 nm at pH of 7.4, with a remarkably narrow distribution. These sizes are much smaller than that (500 nm) of the previous nanoparticles cross-linked by glutaraldehyde [19]. TEM image also provide the direct evidence for the size and shape of the CDDP-loaded casein nanoparticles (Fig. 1D). TEM image clearly shows the nearly spherical shape of the CDDP-loaded casein nanoparticles with the size of about 250 nm in dry state. In addition, the zeta potential of the casein nanoparticles is positive (18.3 mV) at pH ¼ 3.0 and negative (20.0 mV) at pH ¼ 7.4.

1377

The kinetic stability of CDDP-loaded nanoparticles was monitored by DLS. The time-dependent changes in the light scattering intensity and the average hydrodynamic diameter of the nanoparticles in PBS with pH ¼ 7.4 at 37
C are shown in Fig. 2A. It can be seen that the size of the CDDP-loaded nanoparticles is hardly changed even after 7-day store. Also, only a slightly decrease in the relative light scattering intensity is observed with time. These results suggest that the CDDP-loaded casein nanoparticles are kinetically stable at least in one week period and appropriate for further in vivo systemic drug delivery. The release profile of CDDP from CDDP-loaded casein nanoparticles in 0.15 M NaCl PBS with pH of 7.4 at 37
C is shown in Fig. 2B. It can be seen that the release of CDDP is in a sustained manner from the casein nanoparticles. Actually, 5 days incubation with PBS causes 50% release of the total loaded CDDP. In addition, no significantly initial burst in release curve is observed. This release pattern implies that CDDP may bind to the casein nanoparticles via the interaction between platinum of CDDP and

Fig. 5. (A) The NIRF images of H22 tumor-bearing mice following i.v. injection of NIR-797 labeled and CDDP-loaded casein nanoparticles; (B) The fluorescence intensity of the tumor region with time post-injection; (C) ex vivo fluorescence intensity images of the tumors and major organs at 144 h post injection.

1378

X. Zhen et al. / Biomaterials 34 (2013) 1372e1382

carboxylic group of nanoparticles, as indicative of particle size decrease after CDDP loading. 3.3. In vitro cytotoxicity and cellular uptake To assess the antitumor effect of the CDDP-loaded casein nanoparticles, in vitro cytotoxicity against SH-SY5Y cells was initially assayed. The inhibition ratio was examined by MTT assay after the cells were incubated with a series of doses of free CDDP and CDDP-loaded nanoparticles for 24 h. The dose-dependent cytotoxicities of free CDDP and CDDP-loaded nanoparticles are observed (Fig. 3A). Compare to free CDDP, the CDDP-loaded nanoparticles show slight lower toxicity in a range of experimental concentrations due to slow but sustained CDDP release from CDDP-loaded nanoparticles. It is also found that the empty casein nanoparticles are non-toxic even at the highest concentration of 200 mg/mL (Fig. 3B). For the anti-tumor agent, the internalization of the drug into cells is the prerequisite for its pharmacological effect. To examine the cellular uptake of the casein nanoparticles, SH-SY5Y cells were incubated with RBITC-labeled casein nanoparticles at 37
C, and then observed by CLSM. As shown in Fig. 4A, the internalization of RBITC-labeled casein nanoparticles in cells can be clearly observed. The cells exhibit red color in the cytoplasm region with a diffused distribution and blue fluorescence of DAPI from the nucleus, suggesting that casein nanoparticles are mainly distributed uniformly in the entire cell cytoplasm and not in the nucleus. To further investigate the effect of temperature or energy on cellular uptake of casein nanoparticles, SH-SY5Y cells were incubated with the casein nanoparticles at 4
C where the endocytosis is inhibited largely.

From Fig. 4B, it can be seen that the casein nanoparticles are still effectively taken up by the cells at 4
C with the intracellular distribution same as at 37
C, that is, localization in the entire cytoplasm but not nucleus. This result suggests that cellular uptake of a large fraction of casein nanoparticles may follow a temperature- or energy-independent mechanism. Cellular uptake by two-dimensional monolayer SH-SY5Y cells may not accurately reflect the delivery efficiency because the nanoparticles not only have to be taken up by cells but also should be transported to the deeper tissue regions. Particularly for solid tumors, the penetrating ability of nanoparticles is crucial since the interstitial fluid pressures (IFP) are high and vessels are few in tumor regions. To evaluate the transmembrane and penetration of casein nanoparticles in vitro, the three-dimensional multicellular tumor spheroids (MCTS) are cultured to mimic the solid tumors [27,28]. The casein nanoparticles were incubated with the MCTS for 2 h, 4 h, 8 h, 12 h and 24 h at 37
C and 4
C. Similar to the observed interactions between the casein nanoparticles and the cells in monolayer cultures, the casein nanoparticles (red color) are strongly internalized in the MCTS whether at 37
C or at 4
C (Fig. 4C). The spheroids exhibit a diffused distribution of red color in the perpheral region and the red regions diffuse deeper into the spheroids as the time elapsed (Fig. 4C). After 24 h of incubation, the casein nanoparticles localize 40e50 mm deep within the MCTS at 37
C, as shown by CLSM. This result suggests that the casein nanoparticles are able to not only penetrate cell membrane barriers but also transport to the MCTS in the deeper regions. Encouragingly, it is also noted that the casein nanoparticles are shown to be internalized within the spheroids at 4
C which is agreement with the result of monolayer cultures. To our

Fig. 6. Time profiles of Pt concentration in plasma (A) and Pt accumulation in tumor (B) after treatments with CDDP-loaded casein nanoparticles; (C) Biodistribution profiles of Pt accumulation in different organs of H22 tumor-bearing mice after treatments with CDDP-loaded casein nanoparticles. Data are expressed as means  SD (n ¼ 3).

X. Zhen et al. / Biomaterials 34 (2013) 1372e1382

knowledge, no study has previously reported the internalization of nanoparticles within the spheroids at 4
C. The semiquantitative analysis of mean fluorescence intensity of RBITClabeled casein nanoparticles in MCTS is shown in Fig. 4D. It is demonstrated that the mean intensity increases as the time elapsed whether at 37
C or at 4 ◦C. The similar trend indicates that the casein nanoparticles internalize in MCTS at 37
C or at 4 ◦C in the same mechanism. Compared to the mean fluorescence intensity at 37
C, the lower mean fluorescence intensity at 4
C results from the rigidity of 3-dimensional SH-SY5Y cells at 4
C, rather than different mechanism of cellular uptake of casein nanoparticles at 37
C and at 4
C. This result verifies again that the cellular uptake of casein nanoparticles may follow a temperature- or energy-independent mechanism. The special mechanism to internalize into the cells and the ability to diffuse deeper into the spheroids for the casein nanoparticles seem to hold a promise in drug delivery in vivo. 3.4. Real-time NIRF imaging To investigate the fate of CDDP-loaded casein nanoparticles in a living system, the non-invasive near-infrared fluorescence (NIRF) imaging technique was used to visualize the tissue distribution of the nanoparticles in vivo. The CDDP-loaded casein nanoparticles were labeled with a NIRF dye, NIR-797, and then injected into subcutaneous hepatic H22 tumor-bearing mice via the tail vein. Fig. 5A depicts the in vivo fluorescence distribution of NIR-797 labeled and CDDP-loaded nanoparticles. The different fluorescence intensities are represented by different colors as shown in color histogram. Within the first 8 h post-injection, a significant fluorescence signal is observed on the abdomen of mouse. For example, at 1 h and 2 h p.i. whole intestinal tract and liver is illuminated, indicating that some of casein nanoparticles are rapidly eliminated from the circulation and cleared by hepatobiliary excretion process. Nevertheless, at 8 h p.i. the fluorescence signal appears in the tumor. Moreover, the fluorescence intensity at the tumor site increases as the time elapsed, peaks at 48 h and then sustains for at least 144 h. Compared to the tumor, the fluorescence intensities of the other organs decrease rapidly, and are barely observable at 48 h p.i., except for the liver. Finally, the NIRF signal of the liver also disappears entirely at 96 h p.i. and only the signal in the tumor region is left. Similar result was observed in the gelatin-based nanoparticles [29]. These results imply that some of CDDP-loaded casein nanoparticles can escape from the recognition by the phagocytic cells and the reticuloendothelial system to target the tumor passively. Quantitative data of time-dependent fluorescence intensity for tumor were acquired by detecting total signals at tumor site with the method developed by Liu’s group [30]. As shown in Fig. 5B, fluorescence intensity at tumor site increases significantly in the initial 48 h postadministration, that is, a 12-fold higher than that at 2 h (15.1 versus 1.3). Further, only a slight decrease in fluorescence intensity for tumor from 48 h to 144 h suggests that the nanoparticles can not be sharply cleared from tumor tissue. The mouse was also sacrificed at 144 h post-injection, and ex vivo fluorescence intensity images were obtained for the tumor tissue as well as major organs such as heart, liver, spleen, lungs, kidneys, stomach, and intestine. As shown in Fig. 5C, a strong NIRF signal is observed in tumor tissue while other tissues show negligible NIRF signal. The ex vivo fluorescent images of the excised organs further confirm the higher accumulation of the CDDP-loaded casein nanoparticles in the tumors. These results indicate that CDDP-loaded casein nanoparticles could target and successfully accumulate in the tumor tissues, due to the long blood circulation time and prominent EPR effect.

1379

3.5. Biodistribution in vivo In order to quantitatively estimate the biodistribution and tumor targeting of CDDP-loaded casein nanoparticles, the ICR mice bearing subcutaneous H22 tumors were sacrificed at multiple time points after administering with CDDP-loaded casein nanoparticles. The platinum contents in tissues, organs and blood were determined by ICP-MS. The data are expressed as percentage of dose/g of wet tissue at each test point. As shown in Fig. 6A at 12 h post-administration, the Pt concentration in blood is 0.68% dose/mL of blood. Compared to our previous work [31], this value is 2-fold higher than that of free CDDP (0.32%) at same time-point. The Pt concentration in the tumor at 12, 24 and 48 h post-injection reaches 4.1%, 4.2%, 4.0% dose/g, respectively, which is much higher than that in the blood. The constant of Pt concentration in the tumor from 24 h to 48 h post-injection suggests that the nanoparticles can retain in the tumor area for a long time. Compared to our previous work [31], the Pt concentration in the tumor for CDDP-loaded casein nanoparticles is 2.3-fold, 4.1-fold and 4.9-fold higher than that of free CDDP formulation at 12 h, 24 h and 48 h, respectively. Additionally, the maximum accumulation of CDDP-loaded nanoparticles in tumor tissue occurs at about 12e48 h p.i., which is about 4.0-fold higher than that at 1 h p.i. This result is in good accordance with

Fig. 7. (A) Tumor volume of H22 tumor-bearing mice that received different treatments as indicated. The same CDDP dose (3 mg/kg) was administered on day 1 for free CDDP and CDDP-loaded casein nanoparticles groups. Data are presented as mean  SD (n ¼ 10). * represents P < 0.05 since the 11th day and ** represents P < 0.01 since the 7th day; (B) Survival rates of tumor-bearing mice treated with different treatments as indicated. Saline and empty casein nanoparticles are used as control groups.

1380

X. Zhen et al. / Biomaterials 34 (2013) 1372e1382

the observation in in vivo NIRF imaging, demonstrating that the delivery of CDDP through casein nanoparticles prolongs the retention of CDDP as well as enhance its accumulation in the tumor site. From the data in Fig. 6, the nanoparticle formulation shows cumulative accumulation in each organ, especially in the liver and spleen of the mice. The Pt level in the organs continues to increase up to at least 12 h after injection. The obvious accumulation of the nanoparticles in liver and spleen suggests that the partial nanoparticles are intercepted by the phagocytes in the reticuloendothelial system (RES). It is reported that a large fraction of free CDDP appears to be excreted through the glomerular filtration as the highest Pt level is detected in the kidney at 3 min after injection. Therefore, such a rapid accumulation of free CDDP in the kidney may be associated with its renal toxicity [32]. In contrast, it is worth noting that the accumulation of CDDP-loaded nanoparticles in kidney at 8, 12, 24, and 48 h is only 1.93%, 2.96%, 2.13%, and 1.07% dose/g, respectively. The slightly accumulation in the kidney means the low renal toxicity of the nanoparticles. Hence, it is reasonable to say that the CDDP-loaded nanoparticles are promising for high tumor accumulation efficacy and minimum side effects for cancer therapy. 3.6. In vivo anticancer efficacy We next investigate the in vivo antitumor efficacy of CDDPloaded casein nanoparticles compared with free CDDP in the mice bearing subcutaneous inoculated H22 tumors. The treatments were done by injecting free CDDP at dose of 3 mg/kg and CDDPloaded casein nanoparticles at the same dose on a CDDP basis. In two control groups, the mice were i.v. administered saline and empty casein nanoparticles. Fig. 7A shows the growth curves of H22 tumors in mice after intravenous administration of saline, empty casein nanoparticles,

free CDDP and CDDP-loaded casein nanoparticles. Neither saline nor casein nanoparticle treatment has any measurable effect on tumor inhibition. The tumor volumes in both groups increase rapidly. The mean tumor size of mice in saline and empty casein nanoparticles groups is 6150  1102 mm3 and 6380  1189 mm3, respectively, on the 17th Day. On the other hand, although the groups of CDDP-loaded casein nanoparticles and free CDDP show similar antitumor activity within 7 day, but after 7 day, CDDPloaded casein nanoparticles exhibit the better therapeutic efficacy than free CDDP. Compared to a mean tumor volume of 2807  1133 mm3 and TGI of 50.2% for free CDDP treatment group, the mean tumor size of mice in CDDP-loaded casein nanoparticles group is only 1586  618 mm3 and tumor growth inhibition (TGI) reaches 74.2% on the 17th Day. In addition, the animals receiving free CDDP exhibited decreasing activity level and lack luster hair during the experimental process. These symptoms were mostly improved in the treatment of CDDP-loaded casein nanoparticles. Fig. 7B shows the survival rates of tumor-bearing mice in each treatment group. All of the mice treated with saline and empty casein nanoparticles died in 41 and 43 days, respectively. In comparison, the CDDP-loaded casein nanoparticles treatment group displays distinctly therapeutic effect and the mice show the longest survival time. Compared to only one animal survival on the 49th day for free CDDP group, the group receiving the CDDP-loaded casein nanoparticle has half of the mice survival on the 49th day. The median survival time for the groups receiving saline, empty casein nanoparticles, free CDDP and CDDP-loaded casein nanoparticles is 33, 35, 39 and 49 days, respectively, suggesting that the CDDP-loaded casein nanoparticles improve the survival rate of H22 tumor-bearing mice. This result indicates again the superior antitumor effect and the reduced side effects of nanoparticle formulation compared with free CDDP.

Fig. 8. Penetration of CDDP-loaded casein nanoparticles in tumors. Immunofluorescence staining of casein nanoparticles penetrated in tumor 12 h and 24 h post-injection. Green nanoparticles and red blood vessels show in the images of H22 tumor sections. Scale bar ¼ 50 mm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

X. Zhen et al. / Biomaterials 34 (2013) 1372e1382

3.7. Tumor penetration of CDDP-loaded nanoparticles The ability of drug-loaded nanoparticles to efficiently propagate throughout the entire tumor and infect cells distant from the injection site is more significant to achieve effective chemotherapy responses in the cancer treatment [33,34]. Thus, considering extraordinary capability to penetrate cell barriers for the casein nanoparticles, it is crucial to directly observe the penetration of CDDP-loaded casein nanoparticles throughout the tumor. Fig. 8 presents the distribution of casein nanoparticles relative to blood vessels in H22 tumor tissue sections. The red fluorescence signal arises from Alex-594-conjugated secondary antibody complexed with primary rat anti-mouse CD31 antibody, which corresponds to the location of blood vessels. The green fluorescence signal arises from Alex-488-labeled streptavidin firmly conjugated with biotin anchored casein nanoparticles to visualize nanoparticles’ position. It can be seen from Fig. 8 that most casein nanoparticles seem to be located in or around the blood vessels at 12 h p.i., suggesting that a quantity of nanoparticles in blood can extravasate through the leaky vessels. Nevertheless, it is noteworthy that more nanoparticles extravasation is observed at 24 h p.i. Although some casein nanoparticles are still distributed in the vascular lumen, most are localized outside of the tumor vasculature, demonstrating that the CDDP-loaded casein nanoparticles have the ability to penetrate

1381

further away from the blood vessels and affect more viable cancer cells. Further, we examined the distribution of CDDP delivered by casein nanoparticles in the tumors based on the apoptosis of H22 tumor cells. The tumor samples were taken from the mice at 3 day post-administration, and the expression of cleaved caspase-3 was studied relative to the vasculature using immunohistochemical staining as described in experimental section. Fig. 9 presents the location of apoptotic cells relative to the vasculature. The red fluorescence signal corresponds to the location of blood vessels. The green fluorescence signal arises from Alex-488-conjugated secondary antibody complexed with primary cleaved caspase-3 (Asp 175) antibody, which represents the apoptotic cells in tumor. As control, the images from the tumors treated with saline and empty casein nanoparticles show that cleaved caspase-3 positive cells are barely detected, while only limited apoptotic cells are observed in the tumor treated by free CDDP. Distinctly, it can be seen that a larger fraction of apoptotic cells are distributed near the blood vessels, and the farthest apoptotic cells are about 70 mm away from the blood vessels. Nevertheless, the apoptotic cells are hardly observed in the region distant from the vessels. This result demonstrates again that the CDDP-loaded casein nanoparticles can enter into the tumor region through the vasculature and then CDDP is released from the nanoparticles to

Fig. 9. Penetration of CDDP-loaded casein nanoparticles in tumors. Cleaved caspase-3 (apoptosis assay, green), CD31 (blood vessels, red) and DAPI (nuclear, blue) co-staining images of H22 tumor sections from mice receiving saline (A), blank casein nanoparticles (B), free CDDP (3 mg/kg) (C) And CDDP-loaded casein nanoparticles (3 mg/kg CDDP eq.) (D), respectively. Scale bar ¼ 50 mm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

1382

X. Zhen et al. / Biomaterials 34 (2013) 1372e1382

permeate the tumor more deeply and affect the cells far from the vasculature. 4. Conclusion In this study, casein nanoparticles cross-linked by transglutaminase were prepared. These nanoparticles are well dispersed and stable in aqueous solution. When CDDP was loaded in the casein nanoparticles, the size of nanoparticles decreased to 257 nm, and drug loading content of CDDP in the casein nanoparticles reached 10%. In vitro cytotoxicity revealed that CDDP-loaded casein nanoparticles have similar cytotoxicity to free CDDP against SH-SY5Y cells. The casein nanoparticles could penetrate multicellular tumor spheroids whether at 37
C or not 4
C, confirming that the casein nanoparticles not only penetrate cell barriers but also transport to the spheroids in the deeper regions with an energy-independent mechanism. In vivo NIRF imaging and ICP-MS analyses demonstrated that CDDP-loaded casein nanoparticles had prominent tumor targeting ability and sufficient CDDP-loaded nanoparticles accumulation in the tumor. The CDDP-loaded casein nanoparticles also showed superior antitumor efficacy on hepatic H22 tumor-bearing mice than free CDDP. Further it was demonstrated that the nanoparticles had the ability to penetrate the tumor after their extravasation through the vasculature and distribute further away from the blood vessels as the time elapsed, and make CDDP permeate the tumor more deeply and affect the cells far from the vasculature. Acknowledgments This study was supported by National Natural Science Foundation of China (No. 51033002 and 51273090). References [1] Elzoghby AO, Samy WM, Elgindy NA. Albumin-based nanoparticles as potential controlled release drug delivery systems. J Control Release 2012; 157:168e82. [2] Ding D, Zhu ZS, Liu Q, Wang J, Hu Y, Jiang XQ, et al. Cisplatin-loaded gelatinpoly(acrylic acid) nanoparticles: synthesis, antitumor efficiency in vivo and penetration in tumors. Eur J Pharm Biopharm 2011;79:142e9. [3] Lammel AS, Hu X, Park S-H, Kaplan DL, Scheibel TR. Controlling silk fibroin particle features for drug delivery. Biomaterials 2010;31:4583e91. [4] Desai N, Trieu V, Yao Z, Louie L, Ci S, Yang A, et al. Increased antitumor activity, intratumor paclitaxel concentrations, and endothelial cell transport of cremophor-free, albumin-bound paclitaxel, ABI-007, compared with cremophor-based paclitaxel. Clin Cancer Res 2006;12:1317e24. [5] Sparreboom A, Scripture CD, Trieu V, Williams PJ, De T, Yang A, et al. Comparative preclinical and clinical pharmacokinetics of a cremophor-free, nanoparticle albumin-bound paclitaxel (ABI-007) and paclitaxel formulated in cremophor (Taxol). Clin Cancer Res 2005;11:4136e43. [6] Hawkins MJ, Soon-Shiong P, Desai N. Protein nanoparticles as drug carriers in clinical medicine. Adv Drug Deliv Rev 2008;60:876e85. [7] Kim YH, Choi CY, Lee S-J, Conti MA, Kim Y. Homeodomain-interacting protein kinases, a novel family of co-repressors for homeodomain transcription factors. J Biol Chem 1998;273:25875e9. [8] He B, Gross M, Roizman B. The g134.5 protein of herpes simplex virus 1 complexes with protein phosphatase 1a to dephosphorylate the a subunit of the eukaryotic translation initiation factor 2 and preclude the shutoff of protein synthesis by double-stranded RNA-activated protein kinase. PNAS 1997;94:843e8. [9] Albini A, Soldi R, Giunciuclio D, Girauo E, Benelli R, Primo L, et al. The angiogenesis induced by HIV-1 Tat protein is mediated by the Flk-1/KDR receptor on vascular endothelial cells. Nat Med 1996;2:1371e5.

[10] Liu Y, Jones M, Hingtgen CM, Bu G, Laribee N, Tanzi RE, et al. Uptake of HIV-1 Tat protein mediated by low-density lipoprotein receptor-related protein disrupts the neuronal metabolic balance of the receptor ligands. Nat Med 2000;6:1380e7. [11] Andaloussi SEL, Lehto T, Mager I, Rosenthal-Aizman K, Oprea II, Simonson OE, et al. Design of a peptide-based vector, PepFect6, for efficient delivery of siRNA in cell culture and systemically in vivo. Nucleic Acids Res 2011;39: 3972e87. [12] Lundin P, Johansson H, Guterstam P, Holm T, Hansen M, Langel U, et al. Distinct uptake routes of cell-penetrating peptide conjugates. Bioconjug Chem 2008;19:2535e42. [13] Margus H, Padari K, Pooga M. Cell-penetrating peptides as versatile vehicles for oligonucleotide delivery. Mol Ther 2012;20:525e33. [14] Elzoghby AO, El-Fotoh WSA, Elgindy NA. Casein-based formulations as promising controlled release drug delivery systems. J Control Release 2011; 153:206e16. [15] Livney YD. Milk proteins as vehicles for bioactives. Curr Opin Colloid Interface Sci 2010;15:73e83. [16] Bachar M, Mandelbaum A, Portnaya I, Perlstein H, Even-Chen S, Barenholz Y, et al. Development and characterization of a novel drug nanocarrier for oral delivery, based on self-assembled b-casein micelles. J Control Release 2012; 160:164e71. [17] Shapira A, Markman G, Assaraf YG, Livney YD. Beta-casein-based nanovehicles for oral delivery of chemotherapeutic drugs: drug-protein interactions and mitoxantrone loading capacity. Nanomedicine 2010;6:547e55. [18] Liu C, Yao W, Zhang L, Qian H, Wu W, Jiang X. Cell-penetrating hollow spheres based on milk protein. Chem Commun 2010;46:7566e8. [19] Liu C, Zhen X, Wang X, Wu W, Jiang X. Cellular entry fashion of hollow milk protein spheres. Soft Matter 2011;7:11526e34. [20] Nishiyama N, Yokoyama M, Aoyagi T, Okano T, Sakurai Y, Kataoka K. Preparation and characterization of self-assembled polymer-metal complex micelle from cis-dichlorodiammineplatinum(II) and poly(ethylene glycol)-poly(alpha, beta-aspartic acid) block copolymer in an aqueous medium. Langmuir 1999; 15:377e83. [21] Zhu ZS, Xie C, Liu Q, Zhen X, Zheng XC, Wu W, et al. The effect of hydrophilic chain length and iRGD on drug delivery from poly(epsilon-caprolactone)poly(N-vinylpyrrolidone) nanoparticles. Biomaterials 2011;32:9525e35. [22] Zeiger E, Gollapudi B, Spencer P. Genetic toxicity and carcinogenicity studies of glutaraldehyde - a review. Mutat Res 2005;589:136e51. [23] Teixeira LSM, Feijen J, van Blitterswijk CA, Dijkstra PJ, Karperien M. Enzymecatalyzed crosslinkable hydrogels: emerging strategies for tissue engineering. Biomaterials 2012;33:1281e90. [24] Motoki M, Aso H, Seguro K, Nio N. Alpha-S1-casein film prepared using transglutaminase. Agric Biol Chem 1987;51:993e6. [25] Kim S-Y, Grant P, Lee J-H, Pant HC, Steinert PM. Differential expression of multiple transglutaminases in human brain. J Biol Chem 1999;274: 30715e21. [26] Smiddy MA, Martin JEGH, Kelly AL, de Kruif CG, Huppertz T. Stability of casein micelles cross-linked by transglutaminase. J Dairy Sci 2006;89:1906e14. [27] Hamilton G. Multicellular spheroids as an in vitro tumor model. Cancer Lett 1998;131:29e34. [28] Kim T-H, Mount CW, Gombotz WR, Pun SH. The delivery of doxorubicin to 3-D multicellular spheroids and tumors in a murine xenograft model using tumorpenetrating triblock polymeric micelles. Biomaterials 2010;31:7386e97. [29] Ding D, Wang J, Zhu ZS, Li RT, Wu W, Liu BR, et al. Tumor accumulation, penetration, and antitumor response of cisplatin-loaded gelatin/Poly(acrylic acid) nanoparticles. ACS Appl Mater Interfaces 2012;4:1838e46. [30] Yang K, Zhang S, Zhang G, Sun X, Lee S-T, Liu Z. Graphene in mice: ultrahigh in vivo tumor uptake and efficient photothermal therapy. Nano Lett 2010;10: 3318e23. [31] Ding D, Zhu ZS, Li RT, Li XL, Wu W, Jiang XQ, et al. Nanospheres-incorporated implantable hydrogel as a trans-tissue drug delivery system. ACS Nano 2011; 5:2520e34. [32] Nishiyama N, Okazaki S, Cabral H, Miyamoto M, Kato Y, Sugiyama Y, et al. Novel cisplatin-incorporated polymeric micelles can eradicate solid tumors in mice. Cancer Res 2003;63:8977e83. [33] Netti PA, Berk DA, Swartz MA, Grodzinsky AJ, Jain RK. Role of extracellular matrix assembly in interstitial transport in solid tumors. Cancer Res 2000;60: 2497e503. [34] McKee TD, Grandi P, Mok W, Alexandrakis G, Insin N, Zimmer JP, et al. Degradation of fibrillar collagen in a human melanoma xenograft improves the efficacy of an oncolytic herpes simplex virus vector. Cancer Res 2006;66: 2509e13.