Rod-Shaped Drug Particles for Cancer Therapy: The Importance of Particle Size and Participation of Caveolae Pathway

Authors

  • Wei He,

    1. Department of Pharmaceutics, School of Pharmacy, China Pharmaceutical University, Nanjing, P. R. China
    2. Key Laboratory of Druggability of Biopharmaceutics, China Pharmaceutical University, Nanjing, P. R. China
    Search for more papers by this author
  • Xiaofei Xin,

    1. Department of Pharmaceutics, School of Pharmacy, China Pharmaceutical University, Nanjing, P. R. China
    2. Key Laboratory of Druggability of Biopharmaceutics, China Pharmaceutical University, Nanjing, P. R. China
    Search for more papers by this author
  • Yongji Li,

    1. Department of Pharmaceutics, School of Pharmacy, China Pharmaceutical University, Nanjing, P. R. China
    2. Key Laboratory of Druggability of Biopharmaceutics, China Pharmaceutical University, Nanjing, P. R. China
    Search for more papers by this author
  • Xiaopeng Han,

    1. Department of Pharmaceutics, School of Pharmacy, China Pharmaceutical University, Nanjing, P. R. China
    2. Key Laboratory of Druggability of Biopharmaceutics, China Pharmaceutical University, Nanjing, P. R. China
    Search for more papers by this author
  • Chao Qin,

    Corresponding author
    1. Department of Pharmaceutics, School of Pharmacy, China Pharmaceutical University, Nanjing, P. R. China
    2. Key Laboratory of Druggability of Biopharmaceutics, China Pharmaceutical University, Nanjing, P. R. China
    Search for more papers by this author
  • Lifang Yin

    Corresponding author
    1. Department of Pharmaceutics, School of Pharmacy, China Pharmaceutical University, Nanjing, P. R. China
    2. Key Laboratory of Druggability of Biopharmaceutics, China Pharmaceutical University, Nanjing, P. R. China
    Search for more papers by this author

Abstract

It is disclosed how the sizes of rod-shaped paclitaxel-nanosuspensions (PTX-Ns) that are less than 500 nm affect their in vitro and in vivo performances. A size reduction from 500 to 160 nm enhances the cellular uptake and subsequent cytotoxicity, due to the participation of caveolae-mediated endocytosis; moreover, the ability of the PTX-Ns to penetrate tumors is well correlated with the extent to which the caveolae pathway participates in cellular uptake, as their ability to target caveolae is markedly promoted as their size decreased to 160 nm. Also, the size reduction markedly alters the in vivo performance and tumor targeting. It is disclosed that via enhanced tumor penetration and retention but not simply increased tumor accumulation, size reduction of PTX-Ns results in significant improvement in antitumor activities. Overall, this study highlights the importance of the size of the PTX-Ns and the participation of caveolae-mediated endocytosis in controlling their biological functions and will assist in the design and optimization of new nanosuspension formulations for disease therapy.

1 Introduction

Nanomedicines including liposomes, nanosuspensions of poorly water-soluble drugs, nanoemulsions, polymeric micelles, and other inorganic nanoparticles are advantageous for cancer diagnosis and therapy over traditional medicines because they can alter pharmacodynamics and pharmacokinetics of the drug, thereby allowing more drug to accumulate in the tumor site via the enhanced permeability and retention (EPR) effect and obtaining enhanced antitumor activities with reduced side effects.[1] To date, more than 20 nanomedicines, such as Doxil, DaunoXome, and Abraxane, have been approved by the Food and Drug Administration (FDA) for clinical use.[2] Nevertheless, the various physicochemical properties of nanomedicines, such as size, shape, surface charge and surface chemistry, influence their biological functions.[3] Of these parameters, the optimization of the particle size is the first consideration in the design of nanomedicines because the size would significantly affect both the in vitro and in vivo performance of the nanomedicine,[4] including cellular uptake, trafficking and internalization,[5] cytotoxicity,[6] the interaction between the nanomedicine and human blood proteins,[7] pharmacokinetics, biodistribution,[8] in vivo tumor targeting, and tumor penetration.[9] On the other side, these nanomedicines enter cells via a number of different endocytic patterns, such as macropinocytosis, clathrin-mediated internalization, caveolae-pathway, cholesterol-dependent manner, etc.[10] In particular, caveolae-mediated endocytosis has been considerably concerned due to bypassing the acidic endosomes and lysosomes; and thus, this provides an avenue for delivering a greater amount of active drugs with more efficient and immediate access to intracellular targets, thereby rendering it more preferable for subcellular drug delivery.[11]

Nanosuspensions of insoluble drugs are colloidal dispersions of nanoscale pure drug particles with a diameter of 100–1000 nm that are coated with a minimal amount of stabilizers and have been developed as the most successful nanomedicines, as more than ten nanosuspension products have been launched in market within the last 10 years.[12] Unlike conventional nanoparticles such as polymeric micelles, liposomes, nanoemulsions, polymersomes, and so on, nanosuspension formulations have an exceptionally high drug-loading capacity, consisting of particles with 100% drug and possessing 100% drug loading.[13] On the other hand, the drug particles in nanosuspension formulations always have irregular morphologies, rather than uniform spherical shapes. Therefore, in contrast with other nanoparticles, the physicochemical properties of nanosuspensions, particularly size that is a more directly and efficiently tunable factor for therapeutic enhancement,[13b,14] would generate more profound effects on their biological functions. However, for injectable nanosuspensions of drugs, very few systemic studies into the role of the size of the drug particles in the nanosuspensions on both the in vitro and in vivo functions have been performed to date. The influence of 184 and 818 nm riccardin D nanosuspensions with a spherical morphology on pharmacokinetics was recently reported,[15] demonstrating the former exhibited a pharmacokinetic property similar to solution and while the latter showed a high uptake in reticuloendothelial system (RES). Recently, Zhang and coauthors presented a report about the influence of the aspect ratio (AR) of 10-hydroxycamptothecin nanosuspensions (HCPT-Ns) on cellular internalization and their in vitro and in vivo anticancer efficiencies,[16] which shorter HCPT-Ns with AR = 1.3 had a much faster cellular uptake rate, less captured by the liver and a much higher anticancer efficacy than longer HCPT-Ns. Indeed, this report might provide guidelines for the design of nanosuspensions. Unfortunately, the previous report only mainly discovered the biological functions of nanosuspensions with particle sizes larger than 500 nm, which greatly compromised the benefits of their work because, to some extent, only nanosuspensions with particle sizes less than 500 nm, particularly those less than 200 nm, are most commonly used to treat disease. Especially, few reports are present to elaborate the relationship between particle size of nanoparticles and caveolar endocytosis. Therefore, it has become urgent to provide detailed information about how the particle sizes of nanosuspensions smaller than 500 nm affect their biological functions.

Paclitaxel (PTX), one of the most commonly used active compounds for cancer therapy in the clinic, acts by inhibiting microtubule dynamic instability to result in cellular division and apoptosis.[17] However, due to its low water solubility of less than 0.24 mg L−1 (0.3 × 10−6 m),[18] adverse side effects and poor tumor penetration, its use in clinic is significantly limited. To improve its water solubility and facilitate delivery, solubilizers, such as Cremophor EL and dehydrated ethanol, must be formulated into its marketed formulation. On the other hand, the addition of the solubilizers decreases the therapeutic outcome of PTX, and, therefore, only benefits 20%–40% of patients, due to intense side effects, like allergy, hypersensitivity, anaphylactic reactions, etc.[19] Thus, the development of a Cremophor EL-free PTX formulation is highly desirable. In a previous report, rod-shaped, beta-lactoglobulin (β-LG)-coated PTX-nanosuspensions (PTX-Ns) were reported.[20] Interestingly, the β-LG-coated PTX-Ns could be directly delivered to the cytosol, without retention within the endosomal–lysosomal system, which significantly improved the pharmacokinetics of PTX, as demonstrated by its markedly prolonged circulation time and increased area under the plasma concentration–time curve (AUC) compared with free PTX.

Considering the profound effect of the size of the nanomedicine on its biological functions, here, we further disclose how the size (160, 300, and 500 nm) of β-LG-coated PTX-Ns and the participation of caveolae-mediated endocytosis affect their cellular uptake, internalization, cytotoxicity, pharmacokinetics, biodistribution and in vivo efficiency. To the best of our knowledge, this is the first study to comprehensively elucidate how size influences the in vitro and in vivo performance of nanosuspensions of drugs, in particular caveolar endocytosis. This study would assist in the design and optimization of new nanosuspension formulations for disease therapy; moreover, the PTX-Ns are rigid particles; therefore, these results also provide guidelines to understand how the biological functions of other inorganic particles are impacted by their diameter.

2 Results

2.1 Preparation and Characterization of the 160, 300, and 500 nm PTX-Ns

β-LG could serve as an amphiphilic biomaterial and thus be capable of stabilizing nanosuspensions by its adsorption onto the drug particles via the interaction between the hydrophobic sites of protein and hydrophobic surface of drug particles.[20, 21] Hence, PTX-Ns were prepared by coating β-LG on the surface of drug particles using an antisolvent-precipitation method.[22] Their specific sizes were achieved by changing the duration of the ultrasonic treatment, with times for the 160, 300, and 500 nm PTX-Ns of 10, 5, and 2 min, respectively. The PTX/β-LG contents for 160, 300, and 500 nm PTX-Ns were 97.66%/2.4%, 93.47%/6.4%, and 92.03%/7.9%, respectively. Specially, the three PTX-Ns had similar size in width diameter, around 30 nm, and therefore, the fixed aspect ratio for 160, 300, and 500 nm PTX-Ns was around 5, 10, and 17, respectively. The polydispersity index (PI) values for the 160, 300, and 500 nm PTX-Ns were 0.09, 0.12, and 0.20, respectively, thereby demonstrating a narrow size distribution (Figure S1A, Supporting Information). Moreover, their zeta-potentials were −35.60 ± 2.25, −36.12 ± 1.39, and −38.97 ± 1.53 mV, respectively, thus indicating that the PTX-Ns were stable enough to inhibit aggregation, irrespective of their diameters. The negative surface charge on PTX-Ns was ascribed to the protein coating of β-LG possessing negative charge under pH conditions above its isoelectric point (pI = 5).[23] Rod-shaped particles with average diameters (lengths) of ≈160, 300, and 500 nm were observed in the transmission electron microscope (TEM) images (Figure S1B, Supporting Information), thereby confirming the dynamic light scattering (DLS) results. The particles in nanosuspensions with similar shape to that of raw drug crystals always have irregular shapes including rod-like, ellipsoid, Hauser's cube, triangle, and layer structure, instead of spherical morphology. Therefore, the rod-like morphology of PTX-Ns resulted from raw PTX particles with rod-like structure.[20]

The physical stability of PTX-Ns was studied by determining the particle size changes by DLS. Upon incubating in water, phosphate buffer saline (PBS, pH 7.4), 10% fetal bovine serum (FBS) of at 37 °C for 5 d, there were no significant changes regarding of particle size and PI (Figure S2, Supporting Information), thereby confirming the stability of PTX-Ns during use.

To further evaluate the stability of β-LG coating on PTX-Ns, tetramethyl-rhodamine (RITC)-PTX-Ns were collected by ultracentrifugation and subjected to fluorescence analysis after resuspending with different medium. As depicted in Figure S3 (Supporting Information), the fluorescence intensity from different sized RITC-PTX-Ns incubated in water 10% FBS, PBS, or roswell park memorial institute (RPMI)-1640 was kept constant during a period of 48 h. These results thus indicated that the fluorescent dye (RITC)-labeled β-LG would not dissociate from PTX-Ns during use.

The in vitro drug release profiles of PTX-Ns are displayed in Figure S4 (Supporting Information). More than 80% of PTX was released from the three PTX-Ns within 72 h; however, the size reduction did not accelerate the drug release, therefore demonstrating that the three PTX-Ns were able to release PTX in physiological conditions while the size change from 500 to 160 nm had neglected effect on the drug release.

2.2 Cellular Uptake and Trafficking

The uptake of the 160, 300, and 500 nm fluorescein isothiocyanate (FITC) labeled PTX-Ns (FITC-PTX-Ns) by the A549 cells at 37 °C is displayed in Figure 1A. There was a tendency toward increased cellular uptake as the concentration of FITC rose in all three PTX-Ns, thereby demonstrating that the uptake was strongly concentration dependent. Importantly, a size reduction of PTX-Ns markedly increased the cellular uptake; and the uptake of 160 nm PTX-Ns far outweighed that of 300 or 500 nm PTX-Ns, despite the difference in FITC concentrations. These results indicated that the cellular uptake of PTX-Ns were largely dependent upon their sizes.

Figure 1.

A) Fluorescence intensity of various concentrations of the different sized FITC-PTX-Ns that were incubated with A549 cells for 4 h at 37 °C (n = 5). B) CLSM and C) TEM observations of the different sized FITC-PTX-Ns particles after incubated with A549 cells for 2 h at 37 °C at a fixed FITC concentration of 100 µg mL−1. Green fluorescence indicates the location of PTX-Ns in cytoplasma.

To study the intracellular localization of the 160, 300, and 500 nm FITC-PTX-Ns after their internalization, the location of the PTX-Ns in the A549 cells was examined using confocal laser scanning microscope (CLSM). As displayed in Figure 1B, green fluorescence (PTX-Ns) from the 160 and 300 nm PTX-Ns is observed in the perinuclear region, and the former exhibits significantly stronger fluorescence, demonstrating that the two PTX-Ns could enter the cells and were primarily distributed in the cytosol. In contrast, the 500 nm PTX-Ns predominately resided on the cell surface.

TEM observation was performed to further confirm the cellular localization of the PTX-Ns. As depicted in Figure 1C, the 160 nm PTX-Ns were localized throughout the cytoplasm, only a few 300 nm PTX-Ns particles were observed in the cytoplasm, while the 500 nm PTX-Ns were only distributed on or around the cell membrane, without entering the cell.

On the other hand, the yellow fluorescence from the colocalization of the green fluorescence (PTX-Ns) and red fluorescence (endosomes/lysosomes) was displayed from 160 and 300 nm PTX-Ns, while the 500 nm PTX-Ns displayed little yellow fluorescence (Figure S5, Supporting Information). To confirm these results, a pH-resistant fluorophore, tetramethyl-rhodamine (RITC), was tagged with the three PTX-Ns and the location of RITC-labeled PTX-Ns (RITC-PTX-Ns, red) in lysosomes (green) was observed by CLSM. Despite the difference in cell lines, red fluorescence (PTX-Ns) around nucleus and yellow spots obtained by overlaying red fluorescence (PTX-Ns) and green fluorescence (lysosomes) dramatically decreased for increase in size of PTX-Ns (Figures S5 and S6, Supporting Information).

2.3 Internalization Pathways and Caveolar Endocytosis

The cellular internalization of the FITC-labeled 160, 300, and 500 nm PTX-Ns was determined in A549 and Caco-2 cells, with a fixed FITC concentration of 400 µg mL−1. After being exposed to inhibitors, the cells were incubated with the 160, 300, and 500 nm FITC-PTX-Ns for 2 h at 37 °C. The internalization of the 160 nm PTX-Ns in A549 cells was only markedly suppressed by nystatin (p < 0.01), thus indicating the involvement of caveolae-mediated internalization (Figure 2A). The cellular uptake of the 300 nm PTX-Ns was inhibited by nystatin and NaN3 + deoxyglucose (NaN3 + DG, p < 0.05), thereby demonstrating a requirement for an energy-dependent caveolae-internalization process. However, the internalization of the 500 nm PTX-Ns was not restrained by any endocytic inhibitor. To verify the participation of caveolae internalization, we examined the internalization pathways in another endothelial cell line, Caco-2 cells, whose surface are occupied by abundant caveolae, up to 30%.[24] Compared with A549 cells, Caco-2 cells took up the three PTX-Ns via multiple pathways (Figure 2B). This phenomenon could be explained that different cell lines would take up the nanoparticles through different pathways, owing to the difference in receptor expression on the cell surface.[25] However, the uptake of 160 nm PTX-Ns was most blocked by nystatin, reducing by ≈75%, and the uptake was also inhibited by another caveolae inhibitor, methyl-β-cyclodextrin (M-CD).[26] The internalization of 300 nm PTX-Ns was also suppressed by nystatin, with around 40% reduction in uptake. Overall, the three PTX-Ns entered the cells via different internalization pathways, but the 160 and 300 nm PTX-Ns exhibited a similar uptake pattern, caveolae-mediated endocytosis.

Figure 2.

Effect of inhibitors on the internalization of the different sized FITC-PTX-Ns in A) A549 or B) Caco-2 cells. The cells were incubated with FITC-PTX-Ns at 37 °C for 4 h at a FITC concentration of 400 µg mL−1 in the presence of various inhibitors. *p < 0.05 and **p < 0.01 versus the control (n = 5). Cpz, Cyto-D, nystatin, M-CD, and NaN3 with DG (NaN3 + DG) inhibited the clathrin-mediated endocytosis, macropinocytosis, caveolae-internalization process, cholesterol dependence, and energy-dependent mechanism, respectively. C,D) Caveolae-mediated endocytosis of the different sized RITC-PTX-Ns. The cells were cultured with RITC-PTX-Ns (5 µg mL−1 RITC) for 4 h at 37 °C. Cave-1 (green) or CTB (green) was marked by Alexa Fluor 488. The colocalization of RITC-PTX-Ns (red) with (C) Cave-1 or (D) CTB was observed by CLSM. Yellow spots (regions, arrows) indicate the colocalization of PTX-Ns with Cave-1 or CTB.

To study the colocalization of the three PTX-Ns with caveolae, caveolin-1 (Cave-1), a specific protein for formation of caveolae,[27] was marked by Alexa Fluor 488. Yellow spots from 160 and 300 nm PTX-Ns were displayed (Figure 2C), thereby demonstrating the colocalization of caveolae with PTX-Ns and the uptake in a mechanism of caveolar endocytosis; and the fluorescence from 160 nm was markedly stronger than 300 nm, revealing the cellular entry of 160 nm was more dependent on the caveolae-internalization process. Besides Cave-1, caveolar trafficking is also controlled by actin cytoskeleton and cholera toxin subunit B (CTB),[26b,28] so we stained the actin and CTB to further investigate the caveolar endocytosis. Again, marked colocalization of 160 or 300 nm PTX-Ns with CTB or actin was visualized (Figure 2D and Figure S8 (Supporting Information), and the yellow fluorescence from 160 nm was more profound, thus confirming the uptake via caveolar pathway.

2.4 Transport of PTX-Ns across Caco-2 Cell Monolayer

Caveolar pathway would enable the nanoparticles to enter the cells without entrapment in lysosomes and therefore facilitate the nanoparticles to achieve transcytosis. Therefore, the transcellular transport of different sized PTX-Ns across Caco-2 cell monolayer was studied. The Caco-2 cell monolayer was observed by CLSM in x-y-z scanning mode after incubation with PTX-Ns for 2 h. Upon incubation with 160 nm RITC-PTX-Ns, strong red fluorescence signals with weak green fluorescence (actin) in basolateral side was observed in the merged image (Figure 3A); in contrast, tiny red fluorescence for 300 nm RITC-PTX-Ns and no red fluorescence for 500 nm RITC-PTX-Ns were displayed in the merged images. To further investigate the transport, CLSM observation for the XY plane of Caco-2 cell monolayer was performed along the Z-axis from apical side of cells to the bottom in basal side and the images were taken at depth distance of 5 µm (Figure S9, Supporting Information). Colocalization of 160 or 300 nm PTX-Ns with actin was displayed, while the red fluorescence signals from 160 nm PTX-Ns was markedly stronger than that of 300 nm PTX-Ns. Finally, the size distribution of transported PTX-Ns was examined by DLS. The size-distribution patterns of 160 and 300 nm PTX-Ns in basolateral media were similar to that in apical media (Figure 3B); while the intensity signal of 500 nm PTX-Ns could not be detected in basolateral media. These results suggested that the 160 and 300 nm PTX-Ns could transport across the Caco-2 cell monolayer and obtain transcytosis, with the former being more profound.

Figure 3.

Transcytosis of RITC-PTX-Ns (red) across Caco-2 cell monolayer. A) The XZ vertical CLSM images of Caco-2 cell monolayer after cultured with RITC-PTX-Ns at 37 °C for 2 h at 100 µg mL−1 RITC. Actin (green) was stained by FITC-phalloidine. White arrows (yellow or orange fluorescence) indicated the transcellular RITC-PTX-Ns across Caco-2 cell monolayer. The scale bar is 10 µm. B) Size distributions (intensity) of different sized PTX-Ns in apical and basolateral medium after incubation with Transwell filter grown Caco-2 cell monolayer at 37 °C for 2 h.

2.5 Cytotoxicity and Transwell Migration

The cytotoxicity of the formulations was evaluated in H22 cells. The three PTX-Ns exhibited dose-dependent cytotoxicity when the PTX concentration was greater than 0.5 µg mL−1 (Figure 4A). A trend of decreased cell viability was observed when the size of the PTX-Ns was reduced from 500 to 160 nm.

Figure 4.

A) Cell viability after incubation with PTX-Ns with 160, 300, and 500 nm in H22 cells for 24 h at 37 °C (n = 5). *p < 0.05 and **p < 0.01. B) H22 cell apoptosis induced by Taxol and the different sized PTX-Ns. The cells were incubated with 5 µg mL−1 of the PTX formulations for 72 h at 37 °C. An Annexin V-FITC/PI kit was used to stain the early and late apoptotic cells and the percentage of apoptotic cells was determined by flow cytometery (FCM).*p < 0.05, **p < 0.01, and ***p < 0.001. C,D) In vitro migration inhibition by Taxol and the different sized PTX-Ns. The cells were treated with PTX formulations at a PTX concentration of 10 µg mL−1 at 37 °C for 2 h. The control was untreated cells. (C) Observation of cancer cell migration. Migrated cells stained with 0.1% crystal violet were imaged by optical microscope. Blue spots indicate the migrated cells. The scale bar is 100 µm. (D) Quantitative determination of cell migration. **p < 0.01 and ***p < 0.001.

The total percentages of apoptotic H22 cells were 32.2%, 17.5%, 6.8%, and 17.9% for the 160, 300, and 500 nm PTX-Ns, and Taxol (Figure 4B), respectively, with percentages of early/late apoptotic cells of 11.6%/20.6%, 4.3%/13.2%, 2.3%/6.5% and 5.3%/13.6% for these PTX formulations (Figure S10, Supporting Information), respectively. The tendency toward increased apoptosis was present as the size of the PTX-Ns decreased.

To study the effect of size of the PTX-Ns on the inhibition of cancer cell migration, a transwell migration assay and a highly metastatic breast cell line, 4T1 cells, were utilized in our study, with untreated cells as control. Blue areas and cell migration ratio from Taxol were smaller than that of control (Figure 4C,D), thus this result indicated that Taxol (free drug) could suppress cancer cell migration. The blue areas and migration ratio from the different sized PTX-Ns were in order: 160 nm << 300 nm < 500 nm. In contrast with Taxol, the 160 and 300 nm PTX-Ns significantly suppressed the migration, while migration ratios from Taxol and 500 nm PTX-Ns were similar. These results demonstrated that enhanced inhibition of cell migration was obtained via size reduction of PTX-Ns.

2.6 Penetration in Tumor Spheroids

We used a multicellular spheroid model to further investigate the influence of the size of the PTX-Ns on tumor penetration. Interestingly, the distribution of the fluorescence intensity within the spheroids was markedly increased as the size of the PTX-Ns decreased from 500 to 160 nm (Figure 5A), despite the differences at each time point. This result was accompanied by almost homogeneous fluorescence from the surface to center in the tumors treated with the 160 nm PTX-Ns. To further quantitatively determine penetration ability of the three PTX-Ns, the penetration ratio was obtained by calculating the ratio of the penetration depth (normal to the spheroid surface) and spheroid radius. The penetration ratios for the 160, 300, and 500 nm PTX-Ns were ≈100%, 20%, and 13% (Figure 5B), respectively, indicating that the smaller sized particles exhibited a greater ability to penetrate the tumor.

Figure 5.

A) The penetration of the different sized PTX-Ns into the A549 tumor spheroids. The tumor spheroids were treated with FITC-PTX-Ns at 50 ng mL−1 FITC for 1, 2, and 4 h, respectively, at 37 °C. The control was free FITC. The spheroids were incubated with FITC-PTX-Ns and the location of the PTX-Ns (green fluorescence) in the spheroids was observed at specific time intervals using CLSM. The scale bar is 250 µm. B) The penetration ratio of PTX-Ns with different particle sizes in the A549 tumor spheroids. The penetration ratio was obtained by calculating the ratio of the penetration depth (normal to the spheroid surface) and spheroid radius. The tumor spheroids were treated with FITC-PTX-Ns with a concentration of 50 ng mL−1 FITC. *p < 0.05 and **p < 0.001.

2.7 Pharmacokinetics, In Vivo Imaging, and Biodistribution

The plasma concentration–time curves of PTX were calculated after mice were intravenously injected with Taxol or the different sized PTX-Ns and are displayed in Figure 6A. The injection of Taxol (free drug) produced low initial plasma levels, followed by a rapid decrease to undetectable levels within 0.5 h, with an elimination half-life (T1/2) of less than 15 min. In contrast, the plasma levels of the 160, 300, and 500 nm PTX-Ns were significantly higher than Taxol; importantly, the plasma levels of the PTX-Ns exhibited a strong size dependency, with the smaller sized particles exhibiting increased plasma levels. It was worth noting that, of the three PTX-Ns, the 160 nm PTX-Ns displayed the highest plasma levels with two plasma concentration peaks, thus indicating that they had a profound mitigating effect on phagocytosis and escaped the RES.[29]

Figure 6.

A) Drug plasma concentration–time curves of PTX after an intravenous injection of Taxol or the PTX-Ns with different particle sizes at a PTX dose of 10 mg kg−1, based on the animal's body weight (n = 4). B) In vivo images of the H22 tumor-bearing mice that were administered the different sized Cy7-PTX-Ns at 0.5, 1, 2, 4, and 8 h, at a fixed Cy7 dose of 0.5 mg kg−1 based on animal's body weight. The control was free dye (Cy7). C) Accumulation of Cy7-PTX-Ns with different particle sizes in the H22 tumor-bearing mice. The fluorescence intensity indicates the quantity of PTX-Ns in the tissues (n = 5). D) Accumulation of the Cy7-PTX-Ns in the tumor tissue following administration. The fluorescence intensity indicates the quantity of the PTX-Ns in the tissues (n = 5). *p < 0.05 and **p < 0.01 versus the 500 nm PTX-Ns. E) Typical TEM images used to observe the PTX-Ns in tumor cross-sections at 1 h after administration. Arrows indicate the PTX-Ns.

The pharmacokinetic parameters of PTX are displayed in Table 1. Compared with Taxol, the 160, 300, and 500 nm PTX-Ns increased their AUC values by ≈26-, 14-, and 0.1-fold, decreased their systemic clearance (CL) values by ≈35-, 12-, and 0.2-fold, and prolonged their T1/2 by 533-, 290-, and 204-fold, respectively. It appeared that the smaller sized PTX-Ns exhibited significant improvements in PTX pharmacokinetics, including increased AUC, extended circulation time, and decreased blood clearance.

Table 1. Pharmacokinetic parameters of PTX after an intravenous injection of Taxol or the 500, 300, and 160 nm PTX-Ns in rats at a PTX dose of 10 mg kg−1 (n = 4)
FormulationsCmax [mg L−1]Tmax [h]AUC0–∞ [mg h L−1]T1/2 [h]CL [L h−1]F [%]
  1. Tmax, time to maximum concentration.

Taxol4.59 ± 3.610.17 ± 0.004.94 ± 4.520.018 ± 0.0121.34 ± 0.83
500 nm8.32 ± 5.360.14 ± 0.0485.55 ± 3.453.68 ± 2.831.11 ± 0.51112.35
300 nm29.27 ± 11.140.17 ± 0.0071.19 ± 53.025.23 ± 4.110.109 ± 0.0871441.10
160 nm62.81 ± 17.570.21 ± 0.059133.53 ± 14.499.60 ± 3.530.038 ± 0.0042703.04

To examine accumulation in the tumor site, whole-body fluorescence images of the cyanine 7 monoacid (Cy7)-labeled PTX-Ns (Cy7-PTX-Ns) were captured at 0.5, 1, 2, 4, and 8 h following the injection of the Cy7-PTX-Ns into the H22 tumor-bearing mice. As displayed in Figure 6B, within 0.5–2 h after injection, intense fluorescence was observed in the main organs, particularly the liver and kidney; importantly, strong fluorescence in the tumor sites was also observed, irrespective of the particle size of PTX-Ns, thereby demonstrating the tumor-targeting ability of the three PTX-Ns. 8 h later, the fluorescence in these tissues is significantly reduced, and the three PTX-Ns were eliminated from the body. The highest levels of fluorescence were still observed in the liver and spleen. In terms of tumor accumulation, the fluorescence intensity from the three PTX-Ns was similar in the 0.5–2 h period; however, 2 h later, fluorescence tended to increase as the size decreased from 500 to 160 nm.

To further confirm the results described above, a quantitative biodistribution study was performed by collecting the major organs at 1, 2, and 8 h and obtaining the fluorescence intensity using ex vivo imaging. Despite the differences in particle size, the fluorescence intensity of the three PTX-Ns in the liver and kidney was higher than the other organs (Figure 6C). The lowest fluorescence intensity was observed for the 500 nm PTX-Ns in all of the healthy organs, and highest levels of fluorescence were observed for the 300 nm PTX-Ns in the spleen and lung at 2 h or at 8 h. These results thus suggested that the PTX-Ns preferentially accumulated in two organs (liver and kidney) and the largest concentration of the 300 nm PTX-Ns was distributed in the spleen and lung at 2 h or at 8 h postinjection, owing to that the liver, kidney, spleen, and lung are the organs of RES.[30] These results also indicated that it was easier for the spleen and lung to take up the 300 nm PTX-Ns. Again, the fluorescence intensity from the three PTX-Ns was similar at 1 or 2 h postinjection; however, at 8 h, there was a tendency for the fluorescence intensity to increase as the size of the PTX-Ns was decreased (Figure 6D), thereby indicating that the enhanced retention of the PTX-Ns at the tumor site was obtained via size reduction.

To study their delivery into the inside of tumor tissue, the PTX-Ns were observed in tumor cross-sections after 1 h administration by TEM. The distributions of the 160 and 300 nm PTX-Ns inside the tumor were visualized, with greater amounts of the former (Figure 6E). However, the 500 nm PTX-Ns were only located around the surface of the tumor tissue.

2.8 In Vivo Antitumor Effect

The 160, 300, and 500 nm PTX-Ns were intravenously injected into the H22 tumor-bearing mice via the tail vein, with Taxol (free drug) and saline as the controls. A graph of the time versus tumor volume is depicted in Figure 7A. The tumor volume from the saline group increased by ≈40-fold, indicating that tumor growth was not inhibited. However, the tumor growth in the Taxol group was markedly inhibited 25 d postadministration, with the volume only increasing 27-fold compared with that on day 1, thereby confirming the antitumor effect of PTX. In contrast, the tumor growth in the three PTX Ns-injected groups was significantly inhibited, and a trend toward enhanced size-dependent inhibition of tumor growth was observed. In detail, 7 d later, the most rapid tumor growth was observed in the 500 nm PTX-Ns group, which displayed an ≈33-fold increase in tumor volume at day 25 compared with day 1, and faster tumor growth 17 d later compared with the Taxol group. The injection of the 300 nm PTX-Ns resulted in slower tumor growth after 7 d compared with the 500 nm PTX-Ns group, with a 23-fold increase in the tumor volume on day 25. The group treated with the 160 nm PTX-Ns exhibited a further reduction in tumor growth 13 d later compared with that of the 300 nm PTX-Ns group, leading to an only 15-fold increase in the tumor volume on day 25. In summary, the tumor volumes increased by 7-, 17-, and 27-fold from day 1 to day 25 for the 160, 300, and 500 nm PTX-Ns, respectively, compared with the saline group.

Figure 7.

Antitumor activity of Taxol and the PTX-Ns with different particle sizes in the H22 tumor-bearing mice (n = 9), in which saline was used as a control. The mice were administered PTX formulations every 2 d at a fixed PTX dose of 20 mg kg−1 based on animal's body weight. The alterations in the tumor volume were examined every 2 d to assess the particles' antitumor activity. A) Fold change in the tumor volumes in the H22 tumor-bearing mice. A comparison of the tumor volume between the groups was performed on the 30th day. *p < 0.05, **p < 0.01, and ***p < 0.001. B) TUNEL and C) Ki67 analysis of the tumors. The brown-stained cells represent the positive cells. The scale bar is 20 µm. D) Quantitative analysis of cellular apoptosis (n = 3). *p < 0.05, **p < 0.01, and ***p < 0.001. E) Quantitative analysis of cellular proliferation (n = 3). *p < 0.05, **p < 0.01, and ***p < 0.001.

The antitumor effect of these PTX formulations was further studied by analyzing cellular apoptosis and proliferation in paraffin sections of the tumor tissues using the terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) and Ki67 kits, respectively. As shown in Figure 7B–E, the numbers of apoptotic cells (brown) or the degree of cell proliferation (brown) in the Taxol group were greater or less than that of saline group, with percentages of apoptotic and proliferating cells of 12.1% and 5.5%, respectively, for the Taxol group, and 3.0% and 14.0%, respectively, for the saline group. Thus, these results demonstrated the antitumor activity of PTX. The percentages of apoptotic cells (brown) in the tumors from mice treated with the three PTX-Ns were ≈38%, 25%, and 17% for the 160, 300, and 500 nm PTX-Ns, respectively, and followed the order: 160 nm > 300 nm > 500 nm. These results were further confirmed by the examining cell proliferation. Again, the antitumor effect of PTX-Ns was significantly size dependent.

3 Discussion

It was demonstrated that a reduction in the size of the PTX-Ns from 500 to 160 nm, with 300 nm as the cut-off point, resulted in increased cellular internalization, and therefore, an increase in cancer cell apoptosis and enhanced inhibition of cancer cell migration. The cellular uptake of the 160 nm PTX-Ns was markedly increased compared with the 300 and 500 nm particles. This observation was ascribed to the fact that the smaller PTX-Ns have greater surface areas relative to their total mass compared with the larger PTX-Ns, which increases their opportunity to interact with the cell membrane.[31] The reduction in size from 500 to 160 nm led to an increased localization of the PTX-Ns in the endosomal–lysosomal systems and a decrease in the residence of the PTX-Ns on cell surface, thereby improving the cellular internalization. It was also explained that rod-like nanoparticles with small aspect ratio preferred to enter the cells via a rocket pattern with the long axis oriented perpendicular to the membrane and consequently obtained enhanced internalization; while rod-shaped particles with large aspect ratio were inclined to enter the cells through a submarine mode with the long axis of the particles oriented parallel to the membrane, thereby discounting the internalization efficiency.[32] The discrepancy in the internalization efficiency of the different sized PTX-Ns thus generated completely different cytotoxicity results, with increased cell apoptosis and improved suppression of cancer cell migration as the size decreased. The percentage of apoptotic cells from the group treated with the 160 nm PTX-Ns was approximately fivefold and twofold greater than the cells treated with the 500 and 300 nm particles, respectively (Figure 4B), and the cell migration ratio from 160 nm PTX-Ns was around 3.5- and 3-fold less than that of 500 and 300 nm particles (Figure 4D). Larger PTX-Ns, especially 500 nm PTX-Ns, tended to located on the cell membrane and did not enter the cells (Figure S5, Supporting Information), therefore the cytotoxicity of the 300 nm particles was compromised and was markedly lower than the 160 nm particles. However, cytotoxicity of the 300 nm PTX-Ns was still similar to that of Taxol (free drug). In general, it was reported that enhancing the cellular uptake of PTX nanocrystals to obtain intracellular drug release could promote its therapeutic effects against cancer cells;[33] drug particles with a smaller size would promote faster drug release upon entering the cells, hence enhancing their cytotoxicity.[34] In our previous report,[20] we discovered that ≈25% of the H22 cells underwent apoptosis following treatment with the 200 nm PTX-Ns, which was 1.5- and 3.7-fold higher than the 300 and 500 nm particles, respectively. Therefore, these results indicated that the size of the PTX-Ns particles played a critical role in their internalization efficiency and resulting in vitro cytotoxicity, with 300 nm as a turning point. At sizes higher than this point, such as 500 nm, the cytotoxicity is much poorer than the free drug, while at sizes lower than the point, such as 160 nm or 200 nm, a significant increase in the number of apoptotic cells was obtained.

Caveolae-mediated endocytosis played a critical role in the cellular uptake, transcytosis and tumor penetration of the PTX-Ns. After blocking the caveolae-mediated pathway, the cellular uptake of the 160, 300, and 500 nm particles by the cells was reduced by more than 75%, greater than 30% and 0%, respectively. Our previous report indicated that the cellular uptake in the H22 cell line was reduced by 40% when the caveolae-mediated pathway was blocked.[20] Indeed, the colocalization of 160 and 300 nm PTX-Ns with Cave-1, CTB or actin further confirmed the participation of caveolar endocytosis. On the other hand, caveolae are cup-shaped with a neck size of less than 60 nm,[35] therefore, it is difficult that particles greater than this size obtain cellular entry via caveolar pathway.[24b] However, a previous report indicated that caveolae could well accommodate up to 100 nm particles, probably owing to that the caveolar size and shape were dynamic and relied on signaling activated by the interplay between particles and caveolae.[27b] Herein, we found that the β-LG-coated PTX-Ns with a diameter of even up to 300 nm could enter the cells via caveolae, even though the reported neck size of caveolae was less than 60 nm. This result may be related with the protein coating onto the particles and the rod shape of PTX-Ns. Correspondingly, the penetration ratios (%) for the tumor spheroids were ≈100%, 100%,[20] 20%, and 13% for the 160, 200, 300, and 500 nm PTX-Ns, respectively; moreover, in the separate tumor tissues, the location capability of the 160, 300, and 500 nm PTX-Ns inside the tumor was as follows: 160 nm >> 300 nm >> 500 nm. Accordingly, the tumor penetration ability of PTX-Ns is directly related to the extent to which the caveolae pathway participates in cellular uptake. It has been shown that caveolae can traffic their cargo across cells and penetrate across the endothelial cell barrier, thereby delivering the cargo inside the tissue.[36] In the transport study, we truly provided evidence that the 160 and 300 nm PTX-Ns, especially the former, obtained transcytosis for the participation of caveolar endocytosis. Here, a new strategy for targeting caveolae was discovered and achieved by reducing the size of the PTX-Ns to 300 nm, without ligand modifications such as anti-amyloid precursor protein (APP) antibodies.[36a] In addition, the ability to target caveolae was significantly promoted as the size decreased to 160 nm. This discovery can also be adapted to other nanomaterials by altering the cellular internalization mechanism through the size control, thus affecting their cellular uptake and cytotoxicity.

It was demonstrated that smaller PTX-Ns could dramatically increase the systemic exposure and exceptionally prolong the circulation time. In contrast with Taxol (free drug), the three PTX-Ns, in particular 160 and 300 nm particles, showed higher mean maximum plasma concentrations, thereby indicating that the PTX-Ns were not degraded, were maintained in the circulation system and, therefore, could act similar to other nanoparticles to treat the disease.[37] Moreover, the blood clearance of the PTX-Ns was dramatically slower than Taxol, with a trend that as their sizes decreased from 500 to 160 nm, the clearance was significantly delayed. The prolonged circulation was due to the liver and kidney accumulation of the PTX-Ns (Figure 6C) because the accumulation could make the liver function as a depot that released the drug into the blood over time.[13a,38] The integrity and extended circulation of the PTX-Ns would have potential for the efficient use of the EPR. Remarkably, the pharmacokinetics of the PTX-Ns was improved as the size decreased, which were demonstrated as an increased AUC, prolonged T1/2 and decreased blood clearance. In general, larger particles are prone to accumulate in the RES tissues, such as the liver and spleen, where many phagocytic cells exist, therefore accelerating blood clearance.[39] Second, the smaller the particles, the less the absorption of the opsonins.[7, 40] Importantly, of the three PTX-Ns, the 160 nm PTX-Ns displayed a much longer T1/2 and higher AUC than the other two formulations. An additional comparison with the previously reported 200 nm PTX-Ns revealed that the pharmacokinetic profiles of the 160 and 200 nm PTX-Ns were similar,[20] with a second plasma concentration peak appearing within 2–4 h from both of them. Thus, a size of ≈200 nm was the critical cut-off point for the PTX-Ns for in vivo performance, and at sizes equaling to or below this limit, the in vivo behavior of the PTX-Ns was exceptionally positive for cancer therapy, due to its significantly enhanced pharmacokinetics. However, sizes above this point, such as 300 nm, would lead to worse therapeutic outcome. This assertion was confirmed by studying their in vivo antitumor effects. Moreover, the appearance of the second peak in the plasma concentrations, which is the major contributor to the prolonged circulation and increased systemic exposure for the 160 and 200 nm PTX-Ns, is an important indicator of the significantly improved pharmacokinetics. The 160 nm PTX-Ns exhibited significantly delayed blood clearance and higher AUC values than the 300 or 500 nm particles, and only the 160 nm PTX-Ns exhibited the second peak concentration at ≈4 h (Figure 6A). The presence of a second peak plasma concentration for the 160 nm PTX-Ns implies that PTX-Ns with a diameter ≤200 nm can markedly reduce phagocytosis and promote escape from the RES. This might be ascribed to the involvement of caveolae pathway in internalizing the ≤200 nm PTX-Ns because this pathway could allow the nanoparticles to enter the cells, achieve transcytosis, and ultimately return to the circulation.[36]

Despite the differences in their particle sizes, the PTX-Ns preferentially accumulated in the liver and kidney. Apparently, all of the PTX-Ns tended to accumulate in the liver and kidney as well as the spleen and lung, suggesting that the PTX-Ns would be markedly sequestered by the RES after intravenous injection because they behave as conventional nanoparticles. However, the extent of RES uptake of the PTX-Ns differed according to their size. As described above, opsonin binding to the particles decreases as the particle size decreases, and PTX-Ns with a diameter ≤200 nm mentioned above would exhibit improved in vivo performance. Thus, the trend that the uptake of the 160 nm PTX-Ns was the least in the kidney was consistent with the observation that the 160 nm PTX-Ns generated the highest plasma concentration; indeed, the 160 nm particles exhibited the least accumulation in the heart, further confirming the benefits of the 160 nm particles. It might also explained by the fact that it is difficult for the spleen and liver to filter particles less than 200 nm,[41] thus allowing the 160 nm PTX-Ns to escape from these organs more efficiently. In particular, more 300 nm PTX-Ns particles were sequestered by the RES (lung, liver, and spleen) compared with the 500 nm particles, but the 300 nm PTX-Ns displayed higher plasma levels with more enhanced in vivo performance than the 500 nm particles. These findings might be related to the specific aspect ratio of the 300 nm PTX-Ns, which produces better affinity for the RES organs;[16] however, the participation of caveolae-mediated pathway in internalizing the 300 nm PTX-Ns would benefit their transcytosis, thereby allowing them to re-enter the blood circulation.[36a]

With respective to tumor distribution, size reduction of PTX-Ns did not alter the tumor accumulation but resulted in prolonged retention in the tumor site. Considerable amounts of the three PTX-Ns were deposited in the tumor site, and their accumulation within the 0.5–2 h period was similar (Figure 6B,D), despite the differences in their particle sizes. These results indicated that the tumor accumulation of the elongated PTX-Ns ≤500 nm took advantage of the EPR effect and was oriented through vascular pores via a rocket mode that can allow rod-shaped nanoparticles to transverse across the blood vessels due to their reduced steric hindrance and viscous drag near the vessel pore walls,[31] thus compromising the effect of the different sizes. However, the 160 nm PTX-Ns had a 100% penetration ratio and were located inside the tumor; in contrast, the penetration ratios for the 300 and 500 nm PTX-Ns were less than 20% (Figures 5A and 6E). Only a very small amount of the 300 nm PTX-Ns was delivered inside the tumor and the 500 nm PTX-Ns were only located on the surface of tumor tissue. This phenomenon was ascribed to that the 160 nm PTX-Ns had much better transcytosis effect than 300 or 500 nm PTX-Ns. Moreover, at 8 h postadministration, the tumor retention of the 160 nm PTX-Ns was also significantly longer than that of the 300 or 500 nm particles. For spherical nanoparticles such as silica particles or gold particles, the tumor retention was not well correlated with the deep penetration due to their fast clearance from tumors.[42] In fact, the tumor retention of drug is definitely important for cancer therapy. In general, small molecular compounds readily obtain tumor accumulation and locate in the tumor inside; however, they are easily removed from tumor tissues into circulation system or surrounding tissues, thereby having limited antitumor effect.[43] Herein, we found that prolonged tumor retention of rod-like nanoparticles was achieved by the improved tumor penetration which was obtained via size reduction. This is explained that the rod-shaped nanoparticles delivered into the tumor inside are difficult to flow because of the steric hindrance produced by the rod-morphology and ineffective lymphatic drainage surrounding the tumor.[38a,44] Additionally, in our previous report,[20] we found that the 200 nm PTX-Ns also exhibited a 100% penetration ratio for multicellular spheroids. Therefore, for tumor retention of PTX-Ns, a size of around 200 nm was a vital cut-off point.

We had demonstrated that the PTX-Ns with a size not greater than 200 nm could significantly inhibit the tumor growth. The accumulation of the three PTX-Ns in the tumor was very similar within the initial postinjection period (0.5–2 h). However, the three PTX-Ns exhibited significant differences in their antitumor effects, and the smaller particles exhibited enhanced antitumor activity. As well known, the tumor microenvironment, including the extracellular matrix, low transcapillary pressure gradient, elevated interstitial fluid pressure, and fiber network of the connective tissue, hinders the ability of the drug nanocarriers to penetrate into tumor, therefore resulting in the deposition of nanocarriers and drug release in the perivascular region of the tumor site. Accordingly, improved tumor penetration of the nanocarrier and prolonged retention in tumor inside are two critical strategies for tumor therapy.[45] Therefore, the reduction in the size of the PTX-Ns that would lead to the improved antitumor effect was mainly ascribed to the increase in their tumor penetration and resulting prolonged tumor retention. Moreover, we previously found that the 200 nm PTX-Ns displayed an only ninefold increase in tumor volume on day 25 compared with day 1, which was similar to 160 nm (sevenfold increase in tumor volume).[20] Consequently, 200 nm is an important cut-off size, namely, for the PTX-Ns >200 nm, such as the 300 nm particles, the antitumor effect would be decreased, whereas PTX-Ns ≤200 nm would significantly inhibit tumor growth. These results provide scientists with an opportunity to deeply understand how size affects the in vivo behavior of pure drug nanoparticles and thus indicates the factors that should be focused on when designing nanosuspensions for disease treatment.

4 Conclusions

The present study indicates that the improved delivery of PTX-Ns to cancer cells within solid tumors is strongly size dependent. A decrease in the size of the PTX-Ns from 500 to 160 nm significantly enhanced cellular internalization and thus promoted cancer cell apoptosis due to the participation of caveolae-mediated endocytosis. Importantly, we found that the ability of the PTX-Ns to deliver tumor inside was well correlated with the extent to which the caveolae pathway participated in cellular uptake, and thus a new strategy for targeting caveolae was discovered, as the ability to target caveolae was markedly promoted as the size was decreased to 160 nm. To our knowledge, it is the first report showing that the particles' ability to target caveolae is obtained by decreasing the size of the nanoparticles. The reduction in the size of the PTX-Ns could markedly increase the systemic exposure, exceptionally prolong the circulation time, improve the tumor accumulation, and therefore enhance their antitumor effects. Taken together, this study highlights the importance of the size of the protein-coated PTX-Ns and the participation of caveolae-mediated endocytosis in controlling their biological functions in terms of cellular uptake, cytotoxicity, tumor delivery, pharmacokinetics, biodistribution, and antitumor activities and would assist in the design and optimization of new nanosuspension formulations for disease therapy.

5 Experimental Section

Please refer to the Supporting Information for full Experimental Section. The animals received care in compliance with the Principles of Laboratory Animal Care and the Guide for the Care and Use of Laboratory Animals. Animal experiments followed a protocol approved by the China Pharmaceutical University Institutional Animal Care and Use Committee.

Acknowledgements

W.H. and X.F.X. contributed equally to this work. This study was supported by the National Natural Science Foundation of China (Nos. 81402869, 81473152, and 81503011), the Natural Science Foundation of Jiangsu Province (No. BK20140671), the Fundamental Research Funds for the Central Universities, and the Fostering Plan of University Scientific and Technological Innovation Team and Key Members of the Outstanding Young Teacher of Jiangsu Qing Lan Project (2014 and 2016). The authors also thank the Cellular and Molecular Biology Center of the China Pharmaceutical University for assistance with confocal microscopy. The authors had declared that no competing interest existed.

Conflict of Interest

The authors declare no conflict of interest.