Nanoparticles Penetrate into the Multicellular Spheroid-on-Chip: Effect

Nov 2, 2017 - Nanoparticles (NPs) are widely studied as tumor targeted vehicles. The penetration of NPs into the tumor is considered as a major barrie...
2 downloads 0 Views 5MB Size
Download PDF
Article pubs.acs.org/molecularpharmaceutics

Nanoparticles Penetrate into the Multicellular Spheroid-on-Chip: Effect of Surface Charge, Protein Corona, and Exterior Flow Ke Huang, Rena Boerhan, Changming Liu, and Guoqiang Jiang* Key Lab of Industrial Biocatalysis, Ministry of Education, Department of Chemical Engineering, Tsinghua University, Beijing, 100084, China

Downloaded via KAOHSIUNG MEDICAL UNIV on October 10, 2018 at 15:54:59 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Nanoparticles (NPs) are widely studied as tumor targeted vehicles. The penetration of NPs into the tumor is considered as a major barrier for delivery of NPs into tumor cell and a big challenge to translate NPs from lab to the clinic. The objective of this study is to know how the surface charge of NPs, the protein corona surrounding the NPs, and the fluid flow around the tumor surface affect the penetration and accumulation of NPs into the tumor, through in vitro penetration study based on a spheroid-on-chip system. Surface decorated polystyrene (PS) NPs (100 nm) carrying positive and negative surface charge were loaded to the multicellular spheroids under static and flow conditions, in the presence or absence of serum proteins. NP penetration was investigated by confocal laser microscopy scanning followed with quantitative image analysis. The results reveal that negatively charged NPs are attached more on the spheroid surface and easier to penetrate into the spheroids. Protein corona, which is formed surrounding the NPs in the presence of serum protein, changes the surface properties of the NPs, weakens the NP−cell affinity, and, therefore, results in lower NP concentration on the spheroid surface but might facilitate deeper penetration. The exterior fluid flow enhances the interstitial flow into the spheroid, which benefits the penetration but also strips the NPs (especially the NPs with protein corona) on the spheroid surface, which decreases the penetration flux significantly. The maximal penetration was obtained by applying negatively charged NPs without protein corona under the flow condition. We hope the present study will help to understand the spatiotemporal performance of drug delivery NPs and inform the rational design of NPs with highly defined drug accumulation localized at a target site. KEYWORDS: nanoparticle, surface charge, protein corona, 3D cell culture, microfluidics, tumor

1. INTRODUCTION Nanoparticles (NPs) as the ideal drug carriers for tumor therapy have been researched for the past few decades. Despite the clinical promise that NPs possess, the translation of NPs from lab to the clinic has been relatively slow owing to the poor control over the pharmacokinetics and accumulation of NPs in the tumor tissues in vivo.1−3 After injection, the NPs could extravasate from the tumor blood vessels due to the enhanced permeability and retention (EPR) effects, and the extravasated NPs need to penetrate into and distribute across the interstitium of the tumor; finally they are internalized by inner cells of the tumor.4,5 Due to the high tumor cell density and the high interstitial fluid pressure, NP penetration is seriously handicapped and NOs often cannot cross more than one or two cell layers.6,7 The penetration barrier is considered as a major challenge for the targeted delivery of NPs and one of the main reasons causing the huge gaps between animal experiments and clinical applications.1,8 Overcoming the penetration barrier requires the better understanding of the mass transfer mechanism of the NPs entering into the tumor and factors which impact on the penetration, distribution, and accumulation of NPs in the tumor. Penetration of NPs into the tumor is highly dependent © 2017 American Chemical Society

on the physicochemical characteristics of NPs and biological environment of the tumor.7,9−13 Particle size and surface properties are important features that may affect the penetration. As far as we know, there are a few reported studies on the former,14−17 but rare studies about the effects of surface properties. Investigation of the effects of NP surface properties on the penetration is a way to access the mechanism and to enhance the penetration. The surface charge of NPs is an elementary surface property, while the NPs will absorb serum proteins on their surface forming the so-called “protein corona” surrounding the NPs in a real biological environment, which may alter the NP surface properties and the fate of NPs in the physiological environment.18−21 Our previous study demonstrated that surface charge and protein corona codetermined the NP−cell interaction, which might further impact penetration.22 Studies on how the NP surface properties affect the NP penetration were handicapped by the absence of the proper in Received: Revised: Accepted: Published: 4618

August 22, 2017 October 23, 2017 November 2, 2017 November 2, 2017 DOI: 10.1021/acs.molpharmaceut.7b00726 Mol. Pharmaceutics 2017, 14, 4618−4627

Article

Molecular Pharmaceutics

Figure 1. Microfluidic chip and spheroid-on-chip. (a) Structure of chip. The chip was constructed with three layers: the upper PDMS layer, the middle glass layer with a thickness of 0.17 mm, and the bottom PVC layer with a square-shaped window in the middle. (b) Four chambers of the chip. In each chamber, there are 5 semicircular weirs, each of which has 2 apertures. (c) The semicircular weir and the schematic and microscope photo of spheroid trapping. The scale bar is 50 μm.

vitro 3D cellular models. The evaluation, including cell culture and animal models followed by clinical studies, allows only a limited reductionist research and is stuck in a time- and moneyconsuming dilemma.23,24 Therefore, there is a demand to develop in vitro models that are sufficiently complex to realistically mimic aspects of the in vivo environment, yet convenient enough to enable an accurate prediction of NPs’ clinical performance. Among all the attempts aimed at developing alternative and proper in vitro models, the combination of 3D cell culture and microfluidics reveals many advantages and prospects. 3D cell culture can provide spatial cell-to-cell interactions and oxygen, nutrient, and metabolite gradients, whose cell morphology and signaling are more similar to physiological conditions than traditional 2D culture with the ability of real-time analysis and observation.25−28 Microfluidics manipulates small amounts of fluids in microscale fluidic channels, providing the significant advantages that it can incorporate fluid flow and mechanical forces to cells, which bring the cell-based assays a step closer to mimicking the in vivo microenvironment,29−31 for fluidic flows are a prevalent part of all in vivo tissues and affect cell interaction with NPs.32,33 Several in vitro setups which combine the 3D cell culture with microfluidics have been used to investigate NP accumulation at tumor under physiological flow conditions.15,34 3D cell culture with microfluidics provides a tool for investigating the effects of NP physicochemical characteristics on penetration and an insight into the mass transfer mechanism of the penetration. The objective of this study is to investigate the effect of surface charge of NPs, the protein corona, and the exterior flow on the penetration and accumulation of NPs in vitro, to elaborate the mass transfer mechanism of NPs penetrating in the tumor. We developed a spheroid-on-chip system by incorporating 3D multicellular spheroids into a specially designed microfluidic chip. NPs with the same size and matrix but carrying groups of different charge on their surface were loaded to the tumor-on-chip system. We investigated the effect of the surface charge on the NP penetration and accumulation

in the multicellular spheroids under the static condition and the flow condition with varied flow rate. Meanwhile, the penetration study was carried out in the presence and absence of serum proteins to determine how the protein corona surrounding the NPs affects the penetration. Together considering the effects of surface charge, flow condition, and protein corona, we proposed a mass transfer mechanism of NP penetration in the tumor and potential way for improving the penetration.

2. EXPERIMENTAL SECTION 2.1. Materials. Styrene (St, Tianjin Chemical Reagent Factory, China), dimethylaminoethyl methacrylate (DMAEMA), and methacrylic acid (MAA) (Beijing Eastern Acrylic Chemical Technology Co., Ltd., China) were distilled under reduced pressure and stored in a refrigerator. Coumarin-6 was purchased from J&K (J&K Scientific Ltd., China). The other chemicals are analytically pure. Deionized water was used throughout the whole experiments. 2.2. Synthesis of Surface Charged NPs. The P(St-coDMAEMA) NPs (PSD NPs, cationic, amine-modified) and P(St-co-MAA) NPs (PSM NPs, anionic, carboxylate-modified) with about 100 nm diameter were prepared by emulsion polymerization process as previously described.22 Briefly, the DMAEMA and MAA copolymerized with styrene in a nanoscaled emulsion droplet at a continuous agitation of 300 rpm at 80 °C for over 8 h, respectively. After the polymerization, the NPs were harvested by dialysis and lyophilized (FD-1A-50, Boyikang Instruments Co., Ltd.). The size and size distribution, as well as the zeta potential of NPs, were determined by dynamic light scattering (DLS) and zeta potential measurement (Nano ZS, Malvern Instruments, Malvern, U.K.) in PBS (10 mM, pH 7.4). 2.3. 3D Cell Culture and Multicellular Spheroid Characterization. HepG2 cells (China infrastructure of cell line resources) were cultured to form the multicellular spheroids. Beforehand, HepG2 cells were cultured in DMEM 4619

DOI: 10.1021/acs.molpharmaceut.7b00726 Mol. Pharmaceutics 2017, 14, 4618−4627

Article

Molecular Pharmaceutics

Figure 2. Scheme of NP penetration experiments. (a) The PSD NPs (cationic, amine-modified polystyrene NPs) and PSM NPs (anionic, carboxylate-modified). (b) The scheme of NP penetration under static condition. (c) The scheme of NP penetration under flow condition. (d) The serum proteins are absorbed on the surface of NPs forming protein corona. (e) The scheme of NP penetration in the absence and the presence of serum proteins, under static condition and flow condition.

PDMS layer contains 4 culture chambers with height of 150 μm and channels connecting the chambers with the entrance; the middle layer is a 0.17 mm thickness glass layer for observing with the confocal laser scanning microscopy. The bottom layer is the support layer made of a PVC plate with a square-shaped window in the middle, which is designed to ensure the microscopy observation. In each chamber, there are five semicircular weirs (interior diameter of 300 μm) to trap the spheroids. Each weir has 2 apertures (50 μm width) to allow perfusion flow. The channel connected with the chamber is 150 μm high and 500 μm wide. The spheroids were loaded into the chip with the fluid flow driven by a syringe pump (LSP02-1B, Longer Pump, China). Prior to loading spheroids, the microfluidic chip was flushed with 75% ethanol and washed with deionized water in a sterile laminar flow hood. Then the spheroids were gently loaded into the microfluidic chip under a microscope to ensure that the spheroids were trapped within the weirs. After the spheroids were trapped, the device was moved into an incubator (BCJ160-S, SHANGHAI BOXUN, China). 2.5. NP Penetration Experiment. NP penetration was investigated on the spheroid-on-chip under a flow condition (Figure 2c). After spheroids were loaded and trapped, the feed fluid was replaced with the culture medium containing coumarin-6-loaded NPs (100 μg/mL). For evaluating the effect of protein corona on the penetration, culture media with and without fetal bovine serum were used, respectively. The chip was incubated at 37 °C for 2 h and was observed under a CLSM, after the incubation was terminated by replacing the culture medium with ice-cold PBS. To evaluate the effect of exterior flow, experiments were performed under different inlet flow rates. After penetration experiment, cellular uptake of NPs in the spheroid was evaluated by the fluorescence intensity of the internalized coumarin-6 detected with a flow cytometer (Supporting Information). In comparison, NP penetration was also studied under static condition. In this case, the formed multicellular spheroids were collected in the EP tubes and the culture medium was replaced

(Corning, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA), 100 U/mL penicillin, and 100 U/mL streptomycin at 37 °C in 5% CO2. The multicellular spheroids were prepared by a modified hanging drop approach.35 First, a culture dish consisting of a lid with porous PS surface and bottom plate with nonadsorbent poly(2-hydroxyethyl methacrylate) (pHEMA) surface was prepared. The porous PS surface of the lid was produced using a phase-separation method.36 A solution of 50 mg/mL PS in the tetrahydrofuran was mixed with the ethanol with a ratio of 2:1 (v/v). Then the mixed solution was sprayed onto the upper surface evenly followed by drying at ambient temperature. The pHEMA surface was coated on the bottom plate with a solution of 5 mg/mL pHEMA in 95% ethanol followed by drying at ambient temperature.37 Prior to cell seeding, both surfaces were sterilized by UV radiation for 30 min. For multicellular spheroid formation, HepG2 cell suspensions with different densities were prepared. Droplets of 15 μL of cell suspensions were placed over the porous PS surface of the lid. The bottom plate was filled with PBS in order to create a saturated environment and avoid the evaporation of the droplet. Then the lid was inverted 180° and incubated for 72 h at 37 °C in a humidified 5% CO2 atmosphere. After 72 h of incubation, the formed spheroids in the droplet were pipetted and collected into the bottom plate for later experiments. After the spheroids were collected, the cell nucleus was stained with fluorescent Hoechst 33342 (Yeasen) for 10 min to observe the spheroids under confocal laser scanning microscopy (CLSM, CarlZeiss LSM710 META, Carl Zeiss, Germany). The live/dead assay was also assessed. The spheroids incubated with culture medium were washed by centrifugation (100g for 5 min) and resuspension in PBS containing calcein-AM and propidium iodide and then observed under CLSM. 2.4. Microfluidic Chip Design and Spheroid-on-Chip. The microfluidic chip was fabricated by a well-developed soft lithography replica molding process with PDMS. The chip consisted of three layers (shown in Figure 1a). The upper 4620

DOI: 10.1021/acs.molpharmaceut.7b00726 Mol. Pharmaceutics 2017, 14, 4618−4627

Article

Molecular Pharmaceutics

Figure 3. Structure of spheroids. (a) The nucleus staining. (b) The live/dead cell distribution of spheroids. The scale bar is 50 μm.

Table 1. Diameter and the Zeta Potential of PSD and PSM NPs diameter/nm zeta potential/mV

PSD NPs

PSM NPs

PSD NPs + corona

PSM NPs + corona

96.7 ± 0.8 9.2 ± 0.7

91.9 ± 0.6 −34.9 ± 1.5

159.4 ± 0.4 −11.5 ± 0.2

202.2 ± 0.6 −12.3 ± 0.4

Figure 4. Confocal images of spheroids after 2 h NP penetration. (a) PSD NPs without corona under static condition. (b) PSD NPs with corona under static condition. (c) PSM NPs without corona under static condition. (d) PSM NPs with corona under static condition. (e) PSD NPs without corona under flow condition. (f) PSD NPs with corona under flow condition. (g) PSM NPs without corona under flow condition. (h) PSM NPs with corona under flow condition. The scale bar is 50 μm.

ImageJ and Matlab.38 With drawing a contour around the spheroid in ImageJ followed by 8 lines drawn from the center to the periphery of the spheroid in Matlab, the relative fluorescence intensity, as a function of distance from spheroid periphery, was quantified by averaging the intensities at the 8 lines (Figure S1). The NP accumulation in the spheroid and the mid penetration depth (d50) were calculated to denote the penetration amount and the spatial distribution of the NPs, respectively. NP accumulation in the spheroids is quantified as the area under fluorescence intensity−penetration depth curve

with fresh culture medium (also with and without fetal bovine serum, respectively) containing coumarin-6-loaded NPs (100 μg/mL). The spheroids were incubated at 37 °C for 2 h, treated, and observed as above. For CLSM observation (Figure S1), the spheroids were scanned from the surface to 150 μm depth (step of 10 μm) along the radial direction using 10× or 20× objectives. All experiments were performed for three spheroids at least. 2.6. Quantitative Image Analysis. The CLSM images (focused on the section with 30 μm depth) were analyzed by 4621

DOI: 10.1021/acs.molpharmaceut.7b00726 Mol. Pharmaceutics 2017, 14, 4618−4627

Article

Molecular Pharmaceutics

Figure 5. (a) The accumulation, (b) the mid penetration depth, and (c) the cellular uptake of NPs in the spheroids. The cellular uptake of NPs is presented by the relative geometric mean fluorescence intensity (GMFI), which was calculated from the flow cytometry spectra in Figure S3. (PSD +St: PSD NPs without corona under static condition. PSD+P+St: PSD NPs with corona under static condition. PSM+St: PSM NPs without corona under static condition. PSM+P+St: PSM NPs with corona under static condition. PSD+Fl: PSD NPs without corona under flow condition. PSD+P +Fl: PSD NPs with corona under flow condition. PSM+Fl: PSM NPs without corona under flow condition. PSM+P+Fl: PSM NPs with corona under flow condition. The mid penetration depth for PSD and PSM NPs with corona under flow condition is absent since it is hard to calculate the mid penetration depth accurately due to the very low penetration. The uptake data for PSD NPs under flow condition and for PSM NPs with corona under flow condition is absent since the fluorescent cells are too few for determination.)

μm) was designed less than the diameter of the spheroids, which ensured that the spheroids can be captured tightly and fixed in the weirs against the flow. 3.3. NP Penetration in Spheroid-on-Chip. The penetration of NPs with different size has been reported in several studies, but the effects of NP surface charge and their absorbing proteins (protein corona) on their penetration, especially under the dynamic flow condition, were rarely considered. In our previous study, we prepared the polystyrene NPs decorated with different surface charged groups but the same size and found that the surface charge and protein corona have significant effects on the behavior of NPs upon interacting with cells.22 Here, using these two types of NPs (their physicochemical characteristics are outlined in Table 1), we investigated the penetration and accumulation of NPs in the spheroid-on-chip to explore the effects of surface charge, protein corona, and exterior flow (Figure 2). In the presence of serum proteins, the NPs absorbed serum proteins on their surface, resulting in the change of zeta potential and particle size, so zeta potential and particle size were determined after equilibrium adsorption, and the data is also listed in Table 1. 3.3.1. NP Penetration under Static Condition. After 2 h of penetration, NP distribution in the multicellular spheroid was observed by confocal microscopy. The spheroids were scanned from the surface to 150 μm depth in the spheroids. The images focused on the section of 30 μm depth were collected (Figure 4) for quantitative analysis, since the fluorescence intensity attenuated seriously in the deeper sections. The fluorescence intensity, as a function of distance from spheroid periphery, is shown in Figure 4. The NP accumulation in the spheroids and the mid penetration depth of NPs are shown in Figure 5. Under the static condition, we observed that both PSD NPs and PSM NPs (shown by green fluorescence) appeared as a fluorescent rim surrounding the spheroids (Figure 4a,b,e,f), which indicated that the major NPs were attached on the spheroid surface or penetrated in the first cell layer. However, the accumulation of PSM NPs (with negatively charged group on their surface), whether in the presence or absence of serum proteins, was significantly larger than that of PSD NPs (Figure 5a). The presence of serum proteins, which were absorbed on the NPs to form the protein corona, changed the surface charge

(a.u.). The mid penetration depth is defined as depth along the radius in the spheroid where the NP concentration declined by half.

3. RESULTS 3.1. 3D Cell Culture and Multicellular Spheroid Characterization. The multicellular spheroids with different size were obtained by modified hanging drop approach after 72 h incubation. The nucleus staining images (Figure 3a, blue fluorescence) show that the HepG2 cells form a multilayer-like structure, comprising a scattered core surrounded by a compact rim.27 The interstitial space forms a tortuous network in the spheroid, which is the main pathway for the NPs entering into the spheroid. The diameter of the spheroids is increased with the cell numbers. There are approximately 1000 cells in a spheroid with the diameter of 200 μm. By staining with calceinAM (Figure 3b, corresponding to green cells, denoting the viable cells) and propidium iodide (corresponding to red cells, denoting the dead cells) after 72 h of incubation, the spheroids (200−300 μm) display a necrotic core and a viable rim, which is attributed to the limited flow of oxygen and nutrients into the core. 3.2. Spheroid-on-Chip and Characterization. The three-layer PDMS microfluidic chip consists of four culture chambers, in which there are five semicircular weirs for capturing multicellular spheroids. The fluid tracer by eosin solution (Figure 1b) shows that fluid fills the chambers and is homogeneously distributed in the chambers. The simulation (Figure S2) also demonstrates the homogeneous flow in the chambers, which is necessary for precise control of the pressure and shear stress around the spheroid and fluid velocity in the channel, ensuring parallel experiments. With an inlet flow rate of 50 μL/min, the flow velocity in the region surrounding the spheroid is approximately 500 μm/s, which is analogous to the blood flow rate in capillaries of humans.39 Therefore, we chose the inlet flow rate of 50 μL/min for most experiments. The PDMS chips were connected with a syringe pump to create a constant flow, and multicellular spheroids with diameter of approximately 200 μm were loaded into the PDMS chip by the flow and can be intercepted in the semicircular weirs (Figure 1c). The height of the chamber (150 4622

DOI: 10.1021/acs.molpharmaceut.7b00726 Mol. Pharmaceutics 2017, 14, 4618−4627

Article

Molecular Pharmaceutics

Figure 6. Relative fluorescence intensity distribution in the spheroids after PSM NPs were loaded with different flow rates: (a, b, c) without protein corona under flow rate of 0, 50, 100 μL/min, respectively; (d, e, f) with protein corona under flow rate of 0, 25, 50 μL/min, respectively.

Figure 7. Accumulation and mid penetration depth after PSM NPs were loaded with varying flow rates: (a, b) without protein corona; (c) with protein corona.

the fluorescence intensity showed, except for PSM NPs applied without serum proteins, that the surface concentration and accumulation of NPs were both significantly reduced under flow condition in comparison with those under the static condition. The most difference under flow condition lies in the penetration of PSM NPs. The accumulation of bare NPs rose a little, but the accumulation of NPs with protein corona dropped sharply to almost zero. The cellular uptake amount of PSM NPs in the spheroid changed correspondingly. These results reveal that exterior flow, protein corona, and surface charge have synthetic and cross-coupling effects on NP penetration, which will be discussed in the following section. 3.3.3. The Effects of Flow Rate. Since the above study revealed the important impacts of exterior fluid flow on penetration of PSM NPs, we further investigated the penetration of PSM NPs in the spheroid-on-chip under different flow rates. The fluorescence intensity distribution, the accumulation, and the mid penetration depth are shown in Figure 6 and Figure 7, respectively. For all cases, exterior fluid flow decreased the PSM NP concentration on the spheroid surface and higher flow rate resulted in lower surface concentration. This result indicates that the flow around the multicellular spheroid handicaps the NP attachment on the spheroid surface due to the shear force. For bare NPs and NPs with protein corona, exterior fluid flow

and NP size and impacted the penetration and accumulation of the two types of NPs. The NP concentration on the spheroid surface of both NPs was decreased in the presence of serum proteins (Figure 4b,f, the fluorescence intensity at depth = 0 denotes the NP concentration on the spheroid surface), indicating that the NP attachment on the spheroid surface was handicapped when NPs carried protein corona; while the mid penetration depth was increased, which means deeper penetration. However, the accumulation of the two types of NPs varied in the opposite direction: larger for PSM NPs vs smaller for PSD NPs. The cellular uptake amount of the NPs in the spheroid was generally dependent on the NP penetration (Figure 5c) but was not fully determined by it. For instance, against the big gap between the penetration of PSD and PSM NPs, their cellular uptake became closer. The intrinsic capacity of cell internalization of the NPs also affected the cellular uptake amount in the spheroid. 3.3.2. NP Penetration under Flow Condition. The NP penetration into the spheroid-on-chip was investigated with the inlet flow rate of 50 μL/min. The results after 2 h penetration are shown in Figure 4 and Figure 5. Under flow condition, we only observed an obvious fluorescent rim surrounding the spheroids when PSM NPs were loaded in the absence of serum proteins, while for PSD NPs, as well as PSM NPs with protein corona, the fluorescent rims faded. The quantitative analysis of 4623

DOI: 10.1021/acs.molpharmaceut.7b00726 Mol. Pharmaceutics 2017, 14, 4618−4627

Article

Molecular Pharmaceutics

Figure 8. Effect of surface charge, protein corona, and the exterior flow on the penetration of NPs in the multicellular spheroids. (a) NP penetration into spheroid through the intercellular space based on the convective mass transfer. (b) The effects of surface charge of the NPs on the interaction between the NPs and the cells and the penetration. (c) The supposed penetration process of NPs with protein corona. (d) The effects of exterior fluid flow.

Table 2. Effect of the Surface Charge, Protein Corona, and Exterior Flow on the Surface Concentration CS, Mass Transfer Coefficient k, and Penetration Flux NLa surface charge

a

exterior flow

protein corona

NPs

PSD

PSM

PSD

PSM

bare NPs

NPs + corona

zeta potential CS k NL

positive (−) (−) (−)

negative (+) (+) (+)

negative (−) (+) (−)

negative (−) (+) (+)

positive/negative (−) (+) (±)

negative (−) (/) (−)

(+) indicates increasing, (−) indicates decreasing, and (/) indicates unclear.

The serum proteins, which are adsorbed onto the surface of the NPs, forming a protein corona surrounding the NPs and altering the surface properties of the NPs (Figure 2d), make the penetration process more complicated. At the first step of the penetration, the interaction between the NPs and the superficial layer of cells will be highly dependent on the protein corona. Nevertheless, the proteins are mainly physically absorbed on the NPs and the adsorption is reversible.40,41 With the further penetration into the spheroid, the proteins absorbed on the NPs might dissociate and the protein corona would collapse due to the space resistance rising from the narrow and tortuous intercellular channel. Therefore, we suppose that the protein corona will be lost when NPs penetrate into the spheroid (Figure 8c), i.e., the protein corona will only dominate the interaction with the superficial layer of cells, while the interaction with inner cells is still governed by the NPs’ own surface. 4.2. Effects of Surface Charge, Protein Corona, and Exterior Flow on the Penetration. According to the convective mass transfer mechanism, the penetration flux is dependent on the NP concentration on the spheroid surface (CS) and the mass transfer resistance into the tumor. Assuming a quasi-stationary mass transfer and a zero NP concentration in the spheroid interior (Figure 4 shows that the NP concentration in the spheroids is very low), applying the convective mass transfer equation, penetration flux (NL) can be expressed as

has opposite effects on their accumulation. It turns out that the accumulation and the mid penetration depth are both increased with the flow rates for bare NPs, while accumulation is decreased with the flow rates for NPs with protein corona (the mid penetration depth of NPs with protein was not calculated due to the very low accumulation). These results manifest that, under fluid flow condition, the protein corona must play a critical role in the interaction between the NPs and the multicellular spheroid.

4. DISCUSSION 4.1. Process of NP Penetration into the Multicellular Spheroid. A supposed process of NP penetration in a multicellular spheroid is illustrated in Figure 8a. The NP penetration into tumor is considered to be based on convective mass transfer. In the absence of serum proteins, the detailed process for NP penetration might be as follows. First, the NPs in the medium culture gather around and attach on the spheroid surface via interaction with superficial layer of cells (e.g., electrostatic interaction and hydrophobic interaction, etc.), with corresponding NP concentration of CS. The attached NPs penetrate into the spheroid most through the intercellular pathway via convective mass transfer arising from the fluid diffusive flux (static condition) and interstitial flow (in the presence of flow around the spheroid), while part of the NPs binds to the cell surface and undergoes internalization further. 4624

DOI: 10.1021/acs.molpharmaceut.7b00726 Mol. Pharmaceutics 2017, 14, 4618−4627

Article

Molecular Pharmaceutics NL ∝ kCS

the outside of the spheroid (Figure 8b), which hinders penetration of NPs with positive charge (PSD NPs, smaller k) and accelerates the penetration of NPs with negative charge (PSM NPs, larger k). Therefore, the penetration of PSM NPs (except for the PSM NPs with protein under flow conditions; the reason will be discussed later) is higher than that of the PSD NPs. Moreover, the forming electric field in the spheroids turns out to be a major factor to prevent positively charged NP penetration, and it decreases the mass transfer coefficient k so much that the protein corona and flow reflect little effect on the positively charged NP accumulation in the spheroid. Another thing should be noted, that surface charge and protein corona might have different effects on penetration and cell internalization of NPs. The cell internalization capacity of PSD and PSM NPs has been evaluated in our previous study by suspension of HepG2 cells.22 The results showed that PSD NPs offered higher cellular uptake than PSM NPs and the NPs with protein corona offered higher cellular uptake than NPs without corona. The almost opposite effects on penetration and cell internalization resulted in the fact that higher cell internalization capacity generally met lower penetration of NPs. The reason might be that the internalization of the superficial cells hindered further penetration. The different effects of surface charge and protein corona on the cell internalization process also resulted in the gap between the cellular uptake and accumulation amount of the NPs in the spheroid (Figures 5a and 5c). The NP penetration into the tumor is an intricate process, and many other factors affecting the penetration might interact with the surface charge, protein corona, and exterior flow, exerting integrated effects on the NP penetration. For instance, the high interstitial pressure in solid tumors is considered as one of the main obstacles which handicap the nanoparticle penetration.44 The interstitial pressure might restrain the interstitial flow into the tumor which arises from the exterior flow. As a result, the effects of higher exterior flow on the penetration became faint. Meanwhile, the exterior flow around the tumor would produce flowing pressure onto the tumor which might also partly counteract the interstitial pressure. These challenges appeal for a more complex in vitro model and method for quantitative study. 4.3. Inspiration to the Real Situation in Tumor. Back to the real situation in tumor, the penetration of NPs in the spheroid-on-chip provides us an insight into the mechanism of NP penetration in tumor. Commonly, the low penetration and accumulation are attributed to high resistance in the tumor. Our study shows that the protein corona, electrostatic interaction, and exterior flow around the tumor might be as important as those reasons. The electrostatic field within the spheroid creates the major resistance for the penetration of positively charged NPs. Serum protein adsorption often occurs in the real biological environment. The protein corona blocks the interaction between NPs and cells, and the low affinity between the protein corona and cells could not hold NPs on the tumor surface under flow condition. The liquid mainly flows around the tumor and the interstitial flow is faint due to the high tumor cell density and the high interstitial fluid pressure. Faint interstitial flow diminishes the convective mass transfer. These effects contribute to the low penetration and accumulation of NPs in the tumor. Based on the results of the present study, we propose some ways to improve the tumor penetration. First, negatively

(1)

where k is mass transfer coefficient for NP penetration (its reciprocal accounting for penetration resistance) through the intercellular pathway. The effects of surface charge, protein corona, and exterior flow on the CS, k, and NL are assessed on the basis of the experimental data and summarized in Table 2. The CS is mainly determined by the surface interaction between the NPs and the superficial cell of the spheroids, which is impacted by the surface charge of NPs, the protein corona, and the flow pattern. Since our previous study demonstrated that the forming corona weakened the cell−NP interaction,22 for both the PSD NPs and PSM NPs, the presence of protein corona results in a lower CS (Figure 4). The fluid flow around the spheroid surface will strip the NPs on the spheroid’s surface. Therefore, for all cases, exterior fluid flow reduces the CS significantly and higher flow rate results in lower CS (Figure 4 and Figure 6). The effect of the surface charge of NPs seems confusing. Since cell surface and extracellular matrix (ECM) both carry negative charge,42 the positively charged NPs are considered to attach more due to the electrostatic interaction,43 but the results are that application of positively charged PSD NPs led to a very low CS. The interaction between NPs and cell is complicated and affected by several factors besides surface charge. Even for the simple condition that NPs interact with single cells, our previous study did not demonstrate the dependence of NP attachment on the electrostatic interaction.22 The surface of a multicellular spheroid is much more complicated than that of a single cell, and the attachment of the NPs might be dominated by other unclear functions rather than the electrostatic interaction. Mass transfer resistance (1/k) for NP penetration into the multicellular spheroid might arise from the interspace diffuse resistance, the entrapment in the superficial layer, and the electric field in the spheroid. Fluid flow around the spheroids gives rise to the interstitial flow across multicellular spheroids.15 Higher flow rate around the spheroids results in a stronger interstitial flow rate inside the spheroid, so that it is easy for the NPs to penetrate deeper in the spheroid and the k becomes larger with the increase of flow rate. Consequently, for both the PSD NPs and PSM NPs, the penetration depth is increased under the flow condition, and higher flow rate leads to higher penetration depth (Figure 7). The entrapment into the superficial layer of the spheroid stands for another obstacle. If the interaction of NPs with the superficial layer of the spheroid is stronger than that with the inner cell and ECM, the NPs tend to stay on the surface and k will become low. On the contrary, if the interaction with inner cell and ECM is stronger, it will become a driven force leading to a larger k. The protein corona weakens the NP affinity with the superficial cells, but does not alter the affinity with the inner cell since protein corona is presumed to collapse in the spheroid. This affinity gradient will drive the penetration. As a result, for both PSD and PSM NPs, the presence of serum proteins leads to a longer penetration depth. As we suppose that protein corona collapse when NPs penetrate into the spheroid, the impacts of surface charge on the mass transfer resistance are mainly dependent on the NP itself. The spheroid displays a layer-like structure comprising a scattered core surrounded by a compact rim, forming an electric charge distribution with more negative charge external but less negative charge internal. This electric charge gradient induces an electric field with the direction from the center to 4625

DOI: 10.1021/acs.molpharmaceut.7b00726 Mol. Pharmaceutics 2017, 14, 4618−4627

Molecular Pharmaceutics

■

ACKNOWLEDGMENTS The present work is supported by grants from the National Basic Research Program of China (No. 2014CB932202) and Natural Science Foundation of China (NSFC 21576148).

charged NPs might be preferable rather than positively charged NPs. Second, the surface should be designed to inhibit the absorption of proteins and the formation of protein corona. Third, the NPs should have appropriate affinity with the tumor surface (cell membrane and ECM), which could reserve the NPs on the tumor surface against the fluid flow around the tumor surface but not so strongly as to entrap the NPs on the surface.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.molpharmaceut.7b00726. CLSM observation and quantitative image analysis, simulation the flow in the chip, and cellular uptake of NPs in the multicellular spheroid-on-chip (PDF)

■

REFERENCES

(1) Lammers, T.; Kiessling, F.; Hennink, W. E.; Storm, G. Drug Targeting to Tumors: Principles, Pitfalls and (Pre-) Clinical Progress. J. Controlled Release 2012, 161, 175−187. (2) Stirland, D. L.; Nichols, J. W.; Miura, S.; Bae, Y. H. Mind the Gap: A Survey of How Cancer Drug Carriers Are Susceptible to the Gap between Research and Practice. J. Controlled Release 2013, 172 (3), 1045−1064. (3) Venditto, V. J.; Szoka, F. C. Cancer Nanomedicines: So Many Papers and so Few Drugs! Adv. Drug Delivery Rev. 2013, 65, 80−88. (4) Kanapathipillai, M.; Brock, A.; Ingber, D. E. Nanoparticle Targeting of Anti-Cancer Drugs That Alter Intracellular Signaling or Influence the Tumor Microenvironment. Adv. Drug Delivery Rev. 2014, 79-80, 107−118. (5) Chauhan, V. P.; Stylianopoulos, T.; Boucher, Y.; Jain, R. K. Delivery of Molecular and Nanoscale Medicine to Tumors: Transport Barriers and Strategies. Annu. Rev. Chem. Biomol. Eng. 2011, 2 (1), 281−298. (6) Minchinton, A. I.; Tannock, I. F. Drug Penetration in Solid Tumours. Nat. Rev. Cancer 2006, 6 (8), 583−592. (7) Ernsting, M. J.; Murakami, M.; Roy, A.; Li, S. D. Factors Controlling the Pharmacokinetics, Biodistribution and Intratumoral Penetration of Nanoparticles. J. Controlled Release 2013, 172, 782− 794. (8) Jain, R. K.; Stylianopoulos, T. Delivering Nanomedicine to Solid Tumors. Nat. Rev. Clin. Oncol. 2010, 7 (11), 653−664. (9) Chauhan, V. P.; Jain, R. K. Strategies for Advancing Cancer Nanomedicine. Nat. Mater. 2013, 12 (11), 958−962. (10) Nel, A. E.; Mädler, L.; Velegol, D.; Xia, T.; Hoek, E. M. V; Somasundaran, P.; Klaessig, F.; Castranova, V.; Thompson, M. Understanding Biophysicochemical Interactions at the Nano-Bio Interface. Nat. Mater. 2009, 8 (7), 543−557. (11) Albanese, A.; Tang, P. S.; Chan, W. C. W. The Effect of Nanoparticle Size, Shape, and Surface Chemistry on Biological Systems. Annu. Rev. Biomed. Eng. 2012, 14, 1−16. (12) Walczyk, D.; Bombelli, F. B.; Monopoli, M. P.; Lynch, I.; Dawson, K. A. What the Cell “sees” in Bionanoscience. J. Am. Chem. Soc. 2010, 132 (16), 5761−5768. (13) Mu, Q.; Jiang, G.; Chen, L.; Zhou, H.; Fourches, D.; Tropsha, A.; Yan, B. Chemical Basis of Interactions between Engineered Nanoparticles and Biological Systems. Chem. Rev. 2014, 114, 7740− 7781. (14) Lee, H.; Fonge, H.; Hoang, B.; Reilly, R. M.; Allen, C. The effects of particle size and molecular targeting on the intratumoral and subcellular distribution of polymeric nanoparticles. Mol. Pharmaceutics 2010, 7 (4), 1195−1208. (15) Albanese, A.; Lam, A. K.; Sykes, E. A.; Rocheleau, J. V.; Chan, W. C. Tumour-on-a-Chip Provides an Optical Window into Nanoparticle Tissue Transport. Nat. Commun. 2013, 4 (10), 2718. (16) Dreher, M. R.; Liu, W.; Michelich, C. R.; Dewhirst, M. W.; Yuan, F.; Chilkoti, A. Tumor Vascular Permeability, Accumulation, and Penetration of Macromolecular Drug Carriers. J. Natl. Cancer Inst. 2006, 98 (5), 335−344. (17) Huang, K.; Ma, H.; Liu, J.; Huo, S.; Kumar, A.; Wei, T.; Zhang, X.; Jin, S.; Gan, Y.; Wang, P. C.; He, S.; Zhang, X.; Liang, X. J. SizeDependent Localization and Penetration of Ultrasmall Gold Nanoparticles in Cancer Cells, Multicellular Spheroids, and Tumors in Vivo. ACS Nano 2012, 6 (5), 4483−4493. (18) Cheng, X.; Tian, X.; Wu, A.; Li, J.; Tian, J.; Chong, Y.; Chai, Z.; Zhao, Y.; Chen, C.; Ge, C. Protein Corona Influences Cellular Uptake of Gold Nanoparticles by Phagocytic and Nonphagocytic Cells in a Size-Dependent Manner. ACS Appl. Mater. Interfaces 2015, 7 (37), 20568−20575.

5. CONCLUSIONS A spheroid-on-chip system was conducted by fixing multicellular spheroids (∼200 μm) into a three-layer PDMS microfluidic chip. It provides a physiological and hydromechanical condition closer to the tumor microenvironment and is a mechanism exploring tool for NP penetration in tumor. The penetration of NPs into the spheroid-on-chip is codetermined by the NP concentration on the spheroid surface and the penetration resistance. The surface charge of the NPs affects both the attachment and penetration resistance arising from the electrostatic field within the multicellular spheroid. Negatively charged NPs attach more on the spheroid surface and penetrate deeper into the spheroids. The protein corona surrounding the NPs, which is formed due to the absorption of serum protein, weakens the NP−cell affinity and affects the NP penetration in two ways. On one hand, it results in the lower NP concentration on the spheroid surface, especially under flow condition. On the other hand, it alleviates the entrapment of NP on the spheroid surface layer and might facilitate deeper penetration. The exterior fluid flow enhances the interstitial flow which benefits the penetration, but it also strips the NPs on the spheroid surface, resulting in a very low NP concentration on the spheroid surface if the NP−spheroid interaction is very weak (e.g., the NPs with protein corona). To improve the penetration, the surface of NPs should be designed to be negatively charged, to inhibit the protein absorption and present appropriate affinity with the tumor cell. We hope the in vitro penetration study, together with a mechanism exploration, will strengthen the understanding into the mass transfer characteristic and spatiotemporal performance of NPs in the tumor, and inform the rational design of NPs to achieve better biodistribution and higher accumulation at the targeted tumor site.

■

Article

AUTHOR INFORMATION

Corresponding Author

*Tel: +8610 62782824. E-mail: [email protected]. ORCID

Guoqiang Jiang: 0000-0001-9066-4828 Notes

The authors declare no competing financial interest. 4626

DOI: 10.1021/acs.molpharmaceut.7b00726 Mol. Pharmaceutics 2017, 14, 4618−4627

Article

Molecular Pharmaceutics (19) Monopoli, M. P.; Walczyk, D.; Campbell, A.; Elia, G.; Lynch, I.; Baldelli Bombelli, F.; Dawson, K. A. Physical−Chemical Aspects of Protein Corona: Relevance to in Vitro and in Vivo Biological Impacts of Nanoparticles. J. Am. Chem. Soc. 2011, 133 (8), 2525−2534. (20) Salvati, A.; Åberg, C.; Dawson, K. A.; Monopoli, M. P.; Åberg, C.; Salvati, A.; Dawson, K. A.; Aberg, C.; Salvati, A.; Dawson, K. A. Biomolecular Coronas Provide the Biological Identity of Nanosized Materials. Nat. Nanotechnol. 2012, 7 (12), 779−786. (21) Tenzer, S.; Docter, D.; Kuharev, J.; Musyanovych, A.; Fetz, V.; Hecht, R.; Schlenk, F.; Fischer, D.; Kiouptsi, K.; Reinhardt, C.; Landfester, K.; Schild, H.; Maskos, M.; Knauer, S. K.; Stauber, R. H. Rapid Formation of Plasma Protein Corona Critically Affects Nanoparticle Pathophysiology. Nat. Nanotechnol. 2013, 8 (10), 772− 781. (22) Huang, K.; Hu, Y.; Yu, C.; Boerhan, R.; Jiang, G. Charged Surface Groups of Nanoparticles and the Adsorbed Proteins Codetermine the Fate of Nanoparticles upon Interacting with Cells. RSC Adv. 2016, 6 (63), 58315−58324. (23) Paul, S. M.; Mytelka, D. S.; Dunwiddie, C. T.; Persinger, C. C.; Munos, B. H.; Lindborg, S. R.; Schacht, A. L. How to Improve R&D Productivity: The Pharmaceutical Industry’s Grand Challenge. Nat. Rev. Drug Discovery 2010, 9 (3), 203−214. (24) Hait, W. N. Anticancer Drug Development: The Grand Challenges. Nat. Rev. Drug Discovery 2010, 9 (4), 253−254. (25) Kim, J. B. Three-Dimensional Tissue Culture Models in Cancer Biology. Semin. Cancer Biol. 2005, 15 (5), 365−377. (26) Debnath, J.; Brugge, J. S. Modelling Glandular Epithelial Cancers in Three-Dimensional Cultures. Nat. Rev. Cancer 2005, 5 (9), 675−688. (27) Lin, R. Z.; Chang, H. Y. Recent Advances in Three-Dimensional Multicellular Spheroid Culture for Biomedical Research. Biotechnol. J. 2008, 3, 1172−1184. (28) Elliott, N. T.; Yuan, F. A Review of Three-dimensional in Vitro Tissue Models for Drug Discovery and Transport Studies. J. Pharm. Sci. 2011, 100 (1), 59−74. (29) Whitesides, G. M. The Origins and the Future of Microfluidics. Nature 2006, 442 (7101), 368−373. (30) Valencia, P. M.; Farokhzad, O. C.; Karnik, R.; Langer, R. Microfluidic Technologies for Accelerating the Clinical Translation of Nanoparticles. Nat. Nanotechnol. 2012, 7 (10), 623−629. (31) Björnmalm, M.; Yan, Y.; Caruso, F. Engineering and Evaluating Drug Delivery Particles in Microfluidic Devices. J. Controlled Release 2014, 190, 139−149. (32) Swartz, M. A.; Fleury, M. E. Interstitial Flow and Its Effects in Soft Tissues. Annu. Rev. Biomed. Eng. 2007, 9 (1), 229−256. (33) Godoy-Gallardo, M.; Ek, P. K.; Jansman, M. M.; Wohl, B. M.; Hosta-Rigau, L. Interaction between Drug Delivery Vehicles and Cells under the Effect of Shear Stress. Biomicrofluidics 2015, 9 (5), 52605. (34) Kwak, B.; Ozcelikkale, A.; Shin, C. S.; Park, K.; Han, B. Simulation of Complex Transport of Nanoparticles around a Tumor Using Tumor-Microenvironment-on-Chip. J. Controlled Release 2014, 194, 157−167. (35) Neto, A. I.; Correia, C. R.; Oliveira, M. B.; Rial-Hermida, M. I.; Alvarez-Lorenzo, C.; Reis, R. L.; Mano, J. F. A Novel Hanging Spherical Drop System for the Generation of Cellular Spheroids and High Throughput Combinatorial Drug Screening. Biomater. Sci. 2015, 3 (4), 581−585. (36) Yuan, Z.; Chen, H.; Tang, J.; Chen, X.; Zhao, D.; Wang, Z. Facile Method to Fabricate Stable Superhydrophobic Polystyrene Surface by Adding Ethanol. Surf. Coat. Technol. 2007, 201 (16−17), 7138−7142. (37) Phung, Y. T.; Barbone, D.; Broaddus, V. C.; Ho, M. Rapid Generation of in Vitro Multicellular Spheroids for the Study of Monoclonal Antibody Therapy. J. Cancer 2011, 2 (1), 507−514. (38) Gao, Y.; Li, M.; Chen, B.; Shen, Z.; Guo, P.; Wientjes, M. G.; Au, J. L.-S. Predictive Models of Diffusive Nanoparticle Transport in 3Dimensional Tumor Cell Spheroids. AAPS J. 2013, 15 (3), 816−831.

(39) Vaupel, P.; Kallinowski, F.; Okunieff, P. Blood Flow, Oxygen and Nutrient Supply, and Metabolic Microenvironment of Human Tumors: A Review. Cancer Res. 1989, 49 (23), 6449−6465. (40) Lundqvist, M.; Stigler, J.; Cedervall, T.; Berggård, T.; Flanagan, M. B.; Lynch, I.; Elia, G.; Dawson, K. The Evolution of the Protein Corona around Nanoparticles: A Test Study. ACS Nano 2011, 5 (9), 7503−7509. (41) Walkey, C. D.; Chan, W. C. W. Understanding and Controlling the Interaction of Nanomaterials with Proteins in a Physiological Environment. Chem. Soc. Rev. 2012, 41 (7), 2780−2799. (42) Stylianopoulos, T.; Poh, M. Z.; Insin, N.; Bawendi, M. G.; Fukumura, D.; Munn, L.; Jain, R. K. Diffusion of Particles in the Extracellular Matrix: The Effect of Repulsive Electrostatic Interactions. Biophys. J. 2010, 99 (5), 1342−1349. (43) Shukla, S.; Ablack, A. L.; Wen, A. M.; Lee, K. L.; Lewis, J. D.; Steinmetz, N. F. Increased tumor homing and tissue penetration of the filamentous plant viral nanoparticle Potato virus X. Mol. Pharmaceutics 2013, 10 (1), 33−42. (44) Heldin, C. H.; Rubin, K.; Pietras, K.; Ostman, A. High interstitial fluid pressure - An obstacle in cancer therapy. Nat. Rev. Cancer 2004, 4 (10), 806−813.

4627

DOI: 10.1021/acs.molpharmaceut.7b00726 Mol. Pharmaceutics 2017, 14, 4618−4627