Gold Nanoparticles Induced Endothelial Leakiness Depends on

Apr 19, 2017 - ... SetyawatiXianguang DingHui Ling Angie YeoYeong Lan TanDavid Tai ... Nelson Leonel Martínez-Rodríguez , Sara Tavárez , Zaira Isab...
0 downloads 0 Views 2MB Size
Download PDF
Gold Nanoparticles Induced Endothelial Leakiness Depends on Particle Size and Endothelial Cell Origin Magdiel I. Setyawati,*,† Chor Yong Tay,‡,§ Boon Huat Bay,∥ and David T. Leong*,† †

Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, Singapore 117585, Singapore ‡ School of Materials Science and Engineering, Nanyang Technological University, N4.1, 50 Nanyang Avenue, Singapore 639798, Singapore § School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Singapore ∥ Department of Anatomy, National University of Singapore, 4 Medical Drive, Singapore 117594, Singapore S Supporting Information *

ABSTRACT: The endothelium presents a formidable barrier for cancer nanomedicine, as the intravenously introduced nanomedicine needs to leave the blood vessel at the tumor site. Endothelial permeability and retention effect (EPR) is not dependable since it is derived from tumors. Certain nanoparticles with specific characteristics are able to induce micrometer sized gaps between endothelial cells. This effect is called “nanoparticle induced endothelial leakiness” (NanoEL). NanoEL therefore allows the nanotechnology to control access to the tumor even in the absence of any EPR effect. Morever, NanoEL can be applicable to noncancer issues, thereby expanding its usefulness in other subfields of nanomedicine. In this paper, we have shown that Gold (Au) nanoparticles within the range of 10−30 nm are good NanoEL inducing particles. As not all endothelial cells have the same permeability, we found that human mammary endothelial cells and human skin endothelial cells are sensitive to Au induced NanoEL, while human umbilical vein endothelial cells are insensitive, reflective of their innate nature of endothelial permeability. The size window and endothelial cell type sensitivity then helps the nanotechnologists to design future nanoparticles that either exploit NanoEL as a nanotechnology driven strategy to access immature tumors, which do not induce the EPR effect, or avoid NanoEL as a nanotoxic side effect. KEYWORDS: endothelial cell permeability, gold, nanoparticle, nanoparticle induced endothelial leakiness, endothelial permeability and retention

I

In overcoming the EC barrier, these targeting moieties are directed to recognize EC’s specific surface receptors (e.g., tissue plasminogen activator, lung endothelial cell adhesion-1 molecule, glucose transporters and transferrin), allowing its internalization and transcellular transport across the EC barrier.4,10 Nevertheless, targeting the EC surface receptors, which is necessary for transcellular transport to occur, makes the nanomedicine highly vulnerable to the EC cellular processing (i.e., endo/lysosomal digestion). This results with the overall reduction in the nanomedicine dose that reaches the disease site. Moreover, only a select few of EC types (e.g., brain EC, lung EC,

n keeping with its role as the gatekeeper of vascular permeability,1 endothelial barrier also restricts the nanomedicine movement across the vascular bed, impeding the nanomedicine from reaching the target disease site.2−4 Almost all cancer nanomedicine assumes that endothelial permeability and retention (EPR) effect is sufficient and precise enough for the nanomedicine to reach the tumor. However, due to the sole tumor dependence nature of the EPR effect, it is difficult to control escape of the nanomedicine from the endothelium.4,5 There are basically two main strategies in overcoming the endothelial barrier. The first is a transcellular strategy, while the second is by exploiting the lower resistance between endothelial cells (EC). Nanomedicine is commonly formulated with targeting moieties against certain surface receptors to increase recognition and receptor mediated internalization of the nanomedicine.6−9 © 2017 American Chemical Society

Received: March 13, 2017 Accepted: April 19, 2017 Published: April 19, 2017 5020

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030

Article

www.acsnano.org

Article

ACS Nano

Figure 1. Au NP physicochemical characterization. (A) Spherical Au NP with different size were visualized with transmission electron microscope (TEM). Scale bar: 50 nm. (B) The Au NP size difference could also be detected from the red-shift in their absorption spectra. (C) Summary of Au NP physicochemical properties. The primary size was derived from the TEM micrographs while the surface plasmon resonance (SPR) peak was obtained from the UV−vis spectroscopy. The hydrodynamic size and zeta potential measurement of Au NP in ultrapure water (in water) and in supplemented EndoGRO cell culture medium (in medium) were obtained with dynamic light scattering (DLS) analysis. Data are mean ± SD, n = 3.

field.22−25 Moreover, Au NP physicochemical properties are highly tunable, making them an excellent nanoparticle model to prove our hypothesis. At the same time, not every endothelial cells have similar characteristics. Endothelial cells in different tissues vary largely in their morphology and their permeability barrier in accordance to their specific roles in those tissues. Therefore, we hypothesized that the innate propensity to resist permeability also negatively influences NanoEL. We chose three different endothelial cell lines to represent two capillary bed endothelial cell lines (human mammary endothelial cells, HMMEC, and human skin microvascular endothelial cells, HMVEC) and one highly selective endothelial cell line (human umbilical vascular endothelial cells, HUVEC). In addition, HMMEC allows us to investigate the context of mammary blood vessels around a breast cancer tumor. Here, we show that by changing the size of the Au NP, we could not only induce but also modulate the degree of the NanoEL effect. In addition, we also elucidate the mechanism that undergirds the Au NP induction of the paracellular route opening of the EC barrier. We also showed that there is a direct link with the cells inherent state of permeability resistance and the Au NP causing the NanoEL effect.

and heart EC) possess specific markers to be targeted.4 Simultaneously, the EC, damaged by the toxicity of the nanomedicine, may induce acute inflammation,11−13 leading to other complication. Ironically, the endothelial cells are in essence targeted while the cancer cells are spared. As such, nanomedicine that depends on transcellular route is ineffectual. As an alternative to transcellular route, transport across EC barrier could take place in the intercellular space between two EC, known as the paracellular route. The paracellular route is very narrow and obstructed by the various junctional proteins like occludins, claudin and VE-cadherin.1,14 Thus, the native paracellular route is used by small molecules like glucose that crosses the EC barrier via diffusion across the concentration gradient.1,14 In order for nanomedicine to exploit the paracellular route, the gaps need to be widen. We have earlier shown that certain nanomaterials can induced this widening of the paracellular route which we coined as nanoparticle induced endothelial cells barrier leakiness (or NanoEL). We found that NanoEL could be regulated by controlling the NPs’ physicochemical properties.15,16 Nanodiamonds bearing different surface groups, have been demonstrated to govern the paracellular route opening on the EC barrier. This resulted with the control over the amount of doxorubicin being delivered to the cancer cells beyond the barrier.16 Previously, we identified the interaction between nanoparticles and VE-cadherin, the adherens junction (AJ) protein that is pivotal in the barrier integrity maintenance, as the key factor for NanoEL.17 VEcadherin is located along the intercellular gaps on the EC barrier in which its width, as part of the overall EC barrier structural design, varies depending on tissue origin and functionality.4,18−21 Hypothetically, it is possible to target certain EC barrier through a controlled design of NP. Such that NP of a certain size could fit the intended EC barrier, Allowing them to migrate along the gaps, interact with the VE-cadherin and induce the opening of the paracellular route. We, therefore, hypothesized that we could modulate the paracellular route opening of a particular EC bed by directly controlling the NP size through NanoEL. We selected gold NP (Au NP) as our nanoparticle model for their high biocompatibility and wide versatility in the nanomedicine

RESULTS AND DISCUSSION Au NP Characterization. In order to allow a more efficient nanomedicine transport, it is only logical to target the microvasculature that possessed typical intercellular gaps width of 10−25 nm.18,20 As such, we synthesized three different sizes of Au NP (i.e., Au10, Au30, and Au50) that was within the targeted size range and utilized these Au NP to provide evidence to our hypothesis. As depicted in Figure 1A, the synthesized Au NP was spherical in shape with average primary sizes of 11 nm (Au10), 28 nm (Au30), and 48 nm (Au50). As Au NP bearing different sizes possess distinctive optical properties, we further confirmed the different sized Au NP formation with UV−vis spectroscopy analysis. The samples’ comparative plasmonic absorption peak showed a red-shift from 518 nm, which is characteristic to the spherical Au NP with size of ∼10 nm,26 to 523 and 532 nm. As a red-shift on the SPR peak indicates the size increase on the 5021

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030

Article

ACS Nano

Figure 2. Size and cell type dependent induction of Au NanoEL. (A) Immunofluorescence staining showed the formation of observable gaps (magenta arrowheads) on HMVEC and HMMEC monolayer endothelial cells barrier following different size Au NP exposure (100 μM, 1 h). Nucleus (blue) were stained with Hoescht 33342, while the AJ protein, VE-cadherin (green), were visualized with the immunofluorescence technique. Scale bar: 50 μm. Magnified windows were demarcated by the dotted line. (B) Quantification analysis of the formed gaps showed that Au NP induction of NanoEL was dependent on the Au NPs size and the cell type. Data are mean ± SD, n = 5. One-way ANOVA with post hoc Tukey HSD test, p < 0.05; *, significant compared to control. (C) Dose dependent increase FITC-dextran penetration following Au NP exposure (25, 100, and 400 μM) for 1 h. Data are mean ± SD, n = 3. Student’s t-test, p < 0.05; *, significant compared to control. 5022

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030

Article

ACS Nano spherical Au NP,26 this UV−vis spectroscopy results further support our transmission electron microscope (TEM) observation. The size difference between the sample groups was also observed through dynamic light scattering (DLS) analysis. Hydrodynamic diameter in ultrapure water was registered to be slightly increase from the TEM derived NP size. The size increased further, when the Au NP were introduced to the cell culture medium that is rich in protein content. The size increase could be attributed to the protein absorption and subsequent protein corona formation onto the Au NP surface, which was further confirmed by the ζ-potential measurement. The Au NPs was determined to be negatively charged (∼ −16 mV) in ultrapure water due to the presence of carboxylic group from the citrate capping. This registered value changed considerably to approximately −9 mV in cell culture medium, presumably due to the protein absorption and protein corona formation. Au NP Triggered NanoEL on Microvascular Bed in Size Dependent Manner. In this study, EC barriers originating from three different tissues were chosen to represent the structural and functional heterogeneity of the vasculature. HMVEC and HMMEC were chosen to represent the typical microvascular EC barrier. Whereas, HUVEC represents macrovascular bed that is known to be permeability selective. As our aim is to increase the nanomedicine transport to the targeted tissue through modulation of paracellular route, we assessed the Au NP modulating properties on the microvasculature bed. Indeed, we observed the formation of gaps (magenta arrowheads; Figure 2A, Figure S1, S2) on otherwise confluent monolayer microvascular HMVEC and HMMEC barrier following Au NP exposure. As the typical width of intercellular gaps on the nonleaky microvascular EC barrier only ranges between 10 to 25 nm, these micron size gaps (5−30 μm) on the EC barrier confirms our hypothesis that Au NP could induce NanoEL. The NanoEL was observed to occur in time and dose dependent manner with the earliest indication of NanoEL was detected following 30 min exposure with induction concentration as low as 25 μM of Au NP (Figure 2A, Figure S1, S2). More importantly, through our semiquantitative analysis of the gaps area (Figure 2B) indicated that the NanoEL induction was controlled by the Au NP size. This was evident by the reverse relationship between the NanoEL and the Au NP size, in which the degree of NanoEL induction diminished concomitantly with the Au NP size increase (Figure 2B). We further asked whether this Au NP modulated NanoEL could be utilized to increase drug delivery across the EC barrier. In order to answer this question, we chose fluorescein isothiocyanate conjugated dextran (FITC−dextran; 40 kDa) that is comparable in size to common drug molecules as representative molecule. An intact EC barrier restrict transport of drugs and FITC−dextran alike. Therefore, if Au NP modulated NanoEL is able to definitively boost the drug transport across the EC barrier, we could expect an increase of the FITC−dextran penetration across the barrier. To enable us to detect the FITC− dextran penetration, we grew the EC barrier on porous insert and exposed the barrier to the Au NP and FITC−dextran mixture. We detected approximately 3-fold increase in the FITC−dextran transport across the HMVEC and HMMEC barrier following Au10 exposure (100 μM) for 1h (Figure 2C). Consistent with the immunofluorescence analysis, we observed that the FITC− dextran transport across the treated microvascular EC barrier decreased proportionally with the increase of Au NP diameter (Figure 2C, Figure S4). As the Au NP were produced through a

sequential growth method with the same capping ligand, any difference in the NanoEL induction and increase transport across the barrier then is mostly likely attributed to the size difference of the Au NP. Taken together, these results reinforce our hypothesis that the different dimensions of the Au NP could be used to regulate the vascular barrier opening and consequently drug transport across the EC barrier. In contrast to the microvascular barriers, the Au NP were not able to induce NanoEL on the more selective HUVEC barrier, as evident from the absence of the gap formation on the barrier (Figure 2A, Figure S3). Semi qualitative analysis detected no significant change on the gap area on the HUVEC barrier following Au NP treatment (Figure 2B). The absence of NanoEL induction on HUVEC barrier was evident from the FITC− dextran penetration that was found to be marginal even after 3 h of Au NP exposure (Figure 2C and Figure S4). These results indicate that NanoEL induction was not only modulated by the Au NP dimensionality but also governed by the structural difference on the EC barrier. NanoEL Induction Is Independent of Au NP Internalization. To fully utilize the NanoEL induction by Au NP, it is pivotal to understand the underlying mechanism. Gap formation culminating to lowering of the EC barrier can be caused by NPinduced reactive oxygen species (ROS) production.16,27 To evaluate whether there ROS is involved in the NanoEL induction, we assessed the intracellular ROS production of the endothelial cells. We noted no significant changes on the intracellular ROS production following short exposure of the Au NP. Significant increase on the ROS level production was detected consistently on all three EC barriers only after 12 h exposure (Figure S5) which is exceedingly later than the early onset of NanoEL at 30 min. This result allows us to eliminate the ROS involvement in initiating the NanoEL. The paracellular route opening could also occur as an inflammatory response to a foreign substance, including nanomaterials.28,29 To determine whether the NanoEL was initiated as part of the EC barrier’s inflammation response, we assessed the nuclear factor kappa B (NF-κB) expression level. As NF-κB regulates the cytokines production, its upregulation would signify the upregulation of the cytokines production and activation of the inflammation response.29,30 We observed no significant change on the NF-κB expression level following 1 h of Au NP exposure (Figure S6), suggesting that the Au NP activated NanoEL was not part of the inflammation response. Cell−cell separation that becomes the highlights of the paracellular route opening on the EC barrier is also observed in cell apoptosis event. Therefore, we further investigated the involvement of apoptosis event in the NanoEL induction by assessing the activation of tumor suppressor p53 the main regulator of apoptosis event. Our immunoblotting data show no apparent phosphorylation at serine 15 on the p53 protein (Pp53(S15)), indicating that the apoptosis pathway was not activated by the Au NP exposure (Figure S6). Consistent with the immunoblotting data, we observed no significant change in the cell viability profile on the EC barrier even after a prolonged 24 h Au NP exposure (Figure S7). These results are in line with the reported studies24,25,31,32 attesting to the Au NP general biocompatibility. More importantly, these results proved little if not no involvement of apoptosis in the formation of the intercellular gaps on the EC barrier. Considering intracellular event, in general, requires sufficiently long processing time prior to it fully manifests, these intracellular events could not be attributed to initiate the NanoEL that 5023

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030

Article

ACS Nano

Figure 3. Au NanoEL induction was independent of their internalization. (A) ICP-MS analysis detected reduction in the Au amount (100 μM, 30 min) in the presence of endocytosis inhibitors cocktail (5 mM MβCD and 10 μM MDC) that blocks the major internalization pathways. Data are mean ± SD, n = 3. Student’s t-test, p < 0.05; *, significant compared to no inhibitors cocktail treatment. (B) Blocking the internalization of Au NP (100 μM, 30 min) with endocytosis inhibitors, 5 mM MβCD, 10 μM MDC, and cocktail of 5 mM MβCD and 10 μM MDC did not affect the FITC−dextran transport across the EC barrier. Data are mean ± SD, n = 3. Student’s t-test, p > 0.05 when compared to no inhibitors group.

process that required hours (∼3 h) to complete.36 To ensure that Au NP internalization process was necessary to the NanoEL induction, we treated the EC barrier with these endocytosis inhibitor cocktail and analyzed the FITC-dextran penetration profile across the Au NP treated EC barrier. If the internalization process were necessary for NanoEL to occur, blocking the endocytosis pathway would reduce Au NP internalization, subsequently the degree of NanoEL induction. Pretreating the cells with the endocytosis inhibitors cocktail did not significantly reduce the Au NP induced NanoEL. We observed no perceivable difference in NanoEL induction between those microvascular EC barriers that received the endocytosis inhibitors treatment and those that did not (Figure 3B). These results suggest that Au NP internalization was not required for the NanoEL induction to occur. Thus, far, our collective data suggest that the NanoEL was initiated from the extracellular domain of the EC barrier. The opening of the paracelullar route is known to be initiated through vascular endothelial growth factor (VEGF) binding to its receptor (VEGFR) that was located on the apical surface of the EC. Considering that majority of the Au NP were still on the EC extracellular portion, it is possible that the Au NP acted as a VEGF analogue, interacted with VEGFR and activated the receptor to bring about the NanoEL. Our ICP-MS analysis (Figure 3A) detected the size dependent deposition of Au NP on the EC surface consistently throughout the three tested EC types. It could be argued, if the NanoEL occurred due to the interaction between the deposited Au NP and the VEGFR on the cell surface, we should similarly observe a consistent size dependent trend on the NanoEL induction throughout the three tested EC barriers. Yet, the size dependent induction only occurred on the microvascular beds of the HMVEC and HMMEC but not the macrovascular bed formed by the HUVEC (Figure 2). This suggested that the NanoEL induction

occurred within a very short exposure of Au NP (30 min). Hence, we shifted our focus to the extracellular domain of the EC barrier and hypothesized that the NanoEL event induction was originated from this domain. To validate the hypothesis, we first determined whether an appreciable Au NP internalization had occurred. In line with our NanoEL observation (Figure 2), ICP MS analysis detected less Au amount as the EC barrier treated with the smaller Au NP size (Figure 3A). This was also in line with our in vitro sedimentation, diffusion and dosimetry (ISDD) modeling that showed Au NP with smaller size settled on the cell surface in higher degree when compared to their counterpart with bigger particle size (Figure S8). Au NP internalization, however, depends not only on the NP deposited amount but also on the degree of interaction between the NP and the surface receptors.33−35 Chithrani et al. observed an optimum internalization of medium size range of Au NP (∼50 nm) when compared to their smaller (14 and 30 nm) and bigger (74 and 100 nm) counterparts.34 As such, if the ICP-MS detection (Figure 3A) was a true measurement of Au NP internalization by the EC, we would expect Au content on the Au50 treated EC group to be the highest when compared to other size groups. Due to the short exposure duration (∼30 min), it is possible that ICPMS analysis was detecting the Au NP being deposited on the EC surface rather than the actual internalization process into the EC. To ascertain the occurrence of Au NP internalization, we employed endocytosis inhibitors cocktail comprising of monodansycadaverine (MDC) and methyl-β-cyclodextrin (MβCD) that blocks Au NP internalization process into the EC. We noted that the endocytosis inhibitors treatment successfully reduced the Au NP internalization by ∼20%, suggesting majority of the deposited Au NP (∼80%) were still on the extracellular domain. This in line with the visual observation made by Zarka et al., who reported that though Au NP association with the cell membrane was fast, complete internalization was a comparatively slow 5024

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030

Article

ACS Nano

Figure 4. Au NanoEL required VE-cadherin signaling. (A) Immunoblotting (left panel) and its semiquantitative analysis (right panel) showed increased phosphorylation of VE-cadherin at tyrosine 731 site (P-VEC(Y731)) in conjunction with downregulation of VE-cadherin expression (VEC). Inhibiting VE-cadherin activation with Src kinase inhibitor, PP1, effectively repressed the VEC signaling activation. Data are mean ± SD, n = 3. Student’s t-test, p < 0.05; *, significant compared to control; #, significant compared to no PP1 treatment. (B) Quantitative analysis with transwell insert permeability assay showed significant reduction of NanoEL following the PP1 treatment. Data are mean ± SD, n = 3. Student’s ttest, p < 0.05; *, significant compared to control; #, significant compared to no PP1 treatment.

was not initiated through the interaction between the Au NP and the VEGFR found on the apical surface of the EC barrier. Au NanoEL Induction Requires VE-Cadherin Signaling and Actin Remodeling. The width of intercellular junction is maintained by pairs of interacting transmembrane VE-cadherin that is located along the junction.1,37 The disruption of this interaction among pairs of VE-cadherin directly decreased the EC barrier integrity.37−39 We have previously showed that titanium dioxide, silica and silver NP brought about NanoEL by disrupting the VE-cadherin homologue interaction through their interaction with the VE-cadherin that leads to the activation of VE-cadherin signaling.17 To validate our hypothesis, we analyzed whether Au NP activated the VE-cadherin signaling. Au NP induced significant increase in the level of phosphorylated VEcadherin at tyrosine 731 (P-VEC(Y731)) residue, the docking site of β-catenin. The β-catenin binding to VE-cadherin results in the formation of a cadherin-catenin-actin ternary complex that connects VE-cadherin to the actin cytoskeleton; thus further anchoring VE-cadherin to the interior of the cell. The direct impact of phosphorylation at the Y731 residue is the disengagement of VE-cadherin to the actin cytoskeleton, resulting with the VE-cadherin intracellular retraction and degradation.1,37,40,41

Indeed, a significant reduction on the level of total VE-cadherin expression was observed following the Au NP treatment (Figure 4A). This confirms the feedback degradation process that decreases the amount of transmembrane VE-cadherin available to immediately reform an intact cell junction. To conclusively prove that the VE-cadherin activation was involved in NanoEL induction, we blocked the VE-cadherin activation by inhibiting the Src kinase activity with PP1. We noted that PP1 successfully abrogated VE-cadherin signaling in both HMMEC and HMVEC groups. This was evident from the phosphorylated VE-cadherin level that was almost the same as the basal control with PP1 and Au NPs treatment in those cell lines. Conversely, the total VE-cadherin level was also normalized back when the PP1 was added together with the Au NP treatment (Figure 4A). More importantly, we noted that the PP1 treatment helped to significantly reduced NanoEL induction. Overall, the data suggest that VE-cadherin signaling activation was necessary in the Au NanoEL induction. In addition to leaving the VE-cadherin unanchored, the connection loss between the VE-cadherin and actin cytoskeleton also results with the untethered actin cytoskeleton primed for remodeling. This cytoskeleton remodeling results in cellular 5025

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030

Article

ACS Nano

Figure 5. Au NanoEL required actin remodeling process. (A) Blocking RhoA kinase actin remodeling process with Y27632 (10 μM) significantly reduced the gaps formation induce by Au NP exposure (white arrowheads). Green and blue fluorescent signals represent the VE-cadherin and nucleus, respectively. Scale bar: 50 μm. (B) Transwell insert assay showed Y27632 significantly reduced the FITC-dextran penetration across the EC barrier, suggesting it hampered Au NPs capability to induce the NanoEL. Data are mean ± SD, n = 3. Student’s t-test, p < 0.05; *, significant compared to control; #, significant compared to no Y-27632 treatment.

contraction in which the cells pull from each other and cause the widening of the paracellular route. To ensure the involvement of cytoskeleton remodeling in the NanoEL induction, we inhibit the

RhoA kinase activity with Y27632 inhibitor. The RhoA kinase is involved in the ROCK signaling that is responsible for the actin cytoskeleton process.42,43 As such, suppressing the Rho kinase 5026

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030

Article

ACS Nano

Figure 6. Proposed mechanism of Au NanoEL. BEFORE: the paracellular route on microvascular barrier is maintained by the VE-cadherin homologue interaction that is buttressed with cadherin-catenin-actin tertiary complex. AFTER: Au NPs interaction with VE-cadherin activates the VE-cadherin signaling that results with the VE-cadherin being unanchored from the actin cytoskeleton. This leads to VE-cadherin internalization and degradation. The untethered actin becomes vulnerable to remodeling process that leads to cell contraction and subsequently results with the opening of paracellular route. VE-cadherin, VEC; β-catenin, β-cat; α-catenin, α-cat.

along the junction, and induce the occurrence of NanoEL. It is well established that the paracellular route of the highly specialized ECs barrier of the blood-brain barrier (BBB) and the macrovascular barrier like the placenta is built with spatially ordered architecture in which the tight junction complex is located on the most apical side of the junction followed by the adherens junction complex. This architecture constricts the paracellular route on these barriers to a mere 1−3 nm.37,44 In contrast, the microvascular barrier (e.g., the breast mammary glands and dermis and hypodermis of the skin) is built differently in structure to allow for solute exchange with their surround tissue. The intercellular junction of this microvascular barrier is not formed in the same orderly manner as its macrovascular counterpart. Tight junction proteins on the dermal microvascular barrier are loosely assembled.14 Moreover, these proteins are not assembled into protein complex separated from the adherens junction but coassembled together with the adherens junction protein.44 These junction structural built results with a more permissive microvascular barrier when compared to the macrovascular barrier. Clearly, the difference on the structural design between these barriers plays a major role in the induction of the NanoEL. Macrovascular barrier with gap size of 3 nm would restrict the Au NP (with size 10−50 nm) from migrating along the gap junction, consequentially, prohibiting the Au NP interaction with the VE-cadherin. In contrast, the microvascular barrier with gap size of 10−25 nm would be more permissive to the Au NP, allowing these NP to interact with the VE-cadherin, activate the VE-cadherin signaling, and induce the NanoEL (Figure 6). The purpose of accessing tumors through their vasculature with NanoEL, especially in cases where the EPR is not reliable and definitive, is an important consideration. First, the model of

activity would halt the ROCK signaling, suppress the actin rearrangement process, and subsequently reduce the cell contraction. If the actin remodeling process were necessary for the NanoEL to occur, we would expect to observe reduction on the NanoEL when the actin remodeling process was suppressed. In line with previous reports, Y27632 curbed the stress fiber formation on the EC barrier (Figure 5A; + Y27632 panels), resulting with the removal of the intracellular contractile force.42,43 Our immunofluorescence showed much reduced occurrence of NanoEL when the Y27632 was used on the HMMEC and HMVEC groups. The result of transwell permeability assay (Figure 5B) that we conducted also agrees well with our immunofluorescence observation. We noted at the concentration of Y27632 was able to significantly reduce the degree of gap formation. These results suggest that the actin remodeling process was necessary for the Au NanoEL induction. We noted that both VE-cadherin activation and actin remodeling process were size dependently triggered on the microvascular EC barrier. This is in good agreement with our NanoEL observation (Figure 2). We posited that this size dependent induction of the NanoEL occurred due to the difference in the amount of the Au NP available to induce the VEcadherin activation. On the basis of ISSD model (Figure S8A) with consideration of the internalization that occurred, we estimated close to ∼6.7 × 109, 2.3 × 109, and 0.5 × 109 particles of Au10, Au30 and Au50, respectively, were available to induce the NanoEL. Interestingly, both VE-cadherin activation and actin remodeling process were markedly absent on the more selective barrier formed by the HUVEC (Figure 4, 5). This further supports our hypothesis that the Au NP could migrate through the intercellular junction, interact with the VE-cadherin presence 5027

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030

Article

ACS Nano

from the Sciencell (USA) and ATCC (USA), respectively. Passage 7−10 of HMVEC and passage 5−9 of HMMEC and HUVEC were used in the study. Cell culture was conducted following a standard culture condition (37 °C, 5% CO2) in which the cells were grown in EndoGRO complete culture media kit (Merck Millipore, USA) supplemented with 50 μg/mL of Gentamicin (Sigma-Aldrich, USA) and 50 ng/mL of Ampothericin-B (Sigma-Aldrich, USA). Throughout the study the cell seeding density was maintained at 40 000 cells/cm2 in which the cells were left to grow and form the necessary endothelial cells monolayer barrier. Transwell Insert Assay. The monolayer endothelial cells were cultured on Transwell insert (with polycarbonate membrane, 0.4 μm pore; Corning Costar, USA) and exposed to different size Au NPs for 0.5, 1, and 3 h. Fresh cell culture medium were used as control. FITC− dextran (1 mg/mL, 40 kDa; Sigma-Aldrich) was supplemented in all treatment group. Following the treatment, the supernatant at lower compartment of the well was collected and the FITC−dextran fluorescence signal was quantified with a microplate reader (Biotek H4, USA) at excitation/emission wavelength of 490/520 nm. The degree of FITC−dextran transport was determined by normalizing the fluorescent signal from the treatment group to the control group. Immunofluorescence. Monolayer endothelial cells formed in 8well chamber slide were exposed to Au NPs. Following the Au NPs exposure, the cells were fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.2% Triton X-100 for 15 min, and blocked with solution containing 2% BSA and 0.1% Triton X-100 for 1 h. Then, the cells were incubated with primary antibody solution overnight at 4 °C. Afterward, the cells were incubated for 1 h with the appropriate Alexa conjugated secondary antibody solution supplemented with 1 μg/mL Hoescht 33342. To detect the actin filament, the secondary antibody solution was supplemented with phalloidin stain. Following secondary antibody incubation, the Prolong gold antifade reagent (Life Technologies) was added. Immunofluorescence images were obtained with inverted Leica Epifluorescence microscope (Leica DMI 6000, Germany). PBS was used for general washing and solutions preparation. Antibody solutions were supplemented with 0.2% BSA, 0.1% Triton X100. The primary antibody and the phalloidin stain were used at dilution of 1:200, and the Alexa conjugated secondary antibody was used at dilution of 1:400. Semiquantitative analysis of the NanoEL was conducted with ImageJ, in which intercellular gaps depicted on the immunofluorescences images was measured and normalized to the total image area. The complete list of antibody used could be found in the Supporting Information.

endothelial cells should be carefully selected because not all endothelial cells have the same permeability. The pervasive use of HUVEC as the endothelial cell model in cancer nanomedicine studies might have created an unrealistically high barrier resistance due to its inherently high resistance to leak, even via nanoEL as shown in this paper. Second, through this paper, we have suggested that the size window for nanoEL in HMMEC and HMVEC using gold nanoparticles lies within 10 nm to 30 nm. This window may provide useful information on the design of nanoparticles that exploit NanoEL. Noteworthy that throughout the entire study, there was no involvement of cancer cells and we are observing leakiness in the endothelial layer of HMMEC and HMVEC. Thus, the EPR effect is not at play here. Hence, NanoEL is distinctly different from EPR in the instrument for inducing leakiness between endothelial cells. NanoEL leakiness is induced by certain nanoparticles. EPR leakiness is induced by the mature tumor’s intrinsic neo-angiogenic program. We therefore suggest that the nanomedicine community can reframe the tumor target to try to access immature tumors where there are no EPR effect yet but relatively having less mutations that are in turn more susceptible to cancer drugs. Accessing these immature tumors with NanoEL will therefore complement current focus on EPR available in mature tumors. This way, the nanomedicine field can cover the entire spectrum of tumors with nanotechnology. The innate sensitivity of the endothelial cells source toward NanoEL is in essence some form of targeting. As solid tumors started out as epithelial cells and many of these cancers are also fed through microvascular type of blood vessels, these endothelial cells are more prone to NanoEL effects over the others.

CONCLUSIONS In this study, we have shown that Au nanoparticles of sizes between 10 to 30 nm were able to induce micrometers sized gaps between endothelial cells within a mere 30 min of exposure. These gaps are so large that it allows drugs and nanomedicine to cross the endothelial cell barrier which needs to be traverse in any clinically relevance nanomedicine. This gap forming effect is coined as nanoparticles induced endothelial leakiness, NanoEL. The NanoEL occurs via a disruption of VE-cadherin-VEcadherin interactions such that it activates the intracellular pathway that brings about actin remodeling. It does not depend on endocytosis which takes a much longer time. The tested endothelial cells that are sensitive to the NanoEL effect are human mammary endothelial cells and human skin endothelial cells. The more permeability stringent and selective human umbilical vein endothelial cells are not sensitive to the NanoEL effect. These findings are a step forward to place the control of accessing the tumor via endothelial leakiness in the hands of the nanotechnologist without reliance on the endothelial permeability and retention (EPR) effect. Overreliance on the EPR effect effectively only targets the mature tumors, which intrinsically are more resistant to cancer therapy. In the future, through rational design of NanoEL inducing nanomedicine, we can now target the entire spectrum of tumors, from immature to mature tumors.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.7b01744. Additional experimental method; Complete data set for Au induced NanoEL, ROS production analysis, inflammation and apoptosis marker analysis, cell viability analysis, and ISSD modeling (PDF)

AUTHOR INFORMATION Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

David T. Leong: 0000-0001-8539-9062 Author Contributions

M.I.S. and D.T.L. conceptualized the study and the design of experiments. M.I.S. performed all the experiments. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

MATERIALS AND METHODS Cell Culture. This study employed primary human endothelial cells from different representative tissue. Human skin microvascular endothelial cells (HMVEC) were obtained from Thermoscientific (USA). Human mammary microvascular endothelial cells (HMMEC) and human umbilical vein endothelial cells (HUVEC) were purchased

Notes

The authors declare no competing financial interest. 5028

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030

Article

ACS Nano

(20) Taveau, J. C.; Dubois, M.; Le Bihan, O.; Trepout, S.; Almagro, S.; Hewat, E.; Durmort, C.; Heyraud, S.; Gulino-Debrac, D.; Lambert, O. Structure of Artificial and Natural VE-cadherin-Based Adherens Junctions. Biochem. Soc. Trans. 2008, 36, 189−193. (21) Yuan, S. Y.; Rigor, R. R. The Endothelial Barrier. In Regulation of Endothelial Barrier Function; Granger, D. N., Granger, J. P., Eds.; Morgan & Claypool: San Rafael, 2010; http://www.ncbi.nlm.nih.gov/books/ NBK54123/. (22) Boisselier, E.; Astruc, D. Gold Nanoparticles in Nanomedicine: Preparations, Imaging, Diagnostics, Therapies and Toxicity. Chem. Soc. Rev. 2009, 38, 1759−1782. (23) Peng, H.; Wang, C.; Xu, X.; Yu, C.; Wang, Q. An Intestinal Trojan Horse for Gene Delivery. Nanoscale 2015, 7, 4354−4360. (24) Zhang, X.-D.; Luo, Z.; Chen, J.; Shen, X.; Song, S.; Sun, Y.; Fan, S.; Fan, F.; Leong, D. T.; Xie, J. Ultrasmall Au10−12(SG)10−12 Nanomolecules for High Tumor Specificity and Cancer Radiotherapy. Adv. Mater. 2014, 26, 4565−4568. (25) Zhang, X.-D.; Luo, Z.; Chen, J.; Wang, H.; Song, S.-S.; Shen, X.; Long, W.; Sun, Y.-M.; Fan, S.; Zheng, K.; Leong, D. T.; Xie, J. Storage of Gold Nanoclusters in Muscle Leads to Their Biphasic In vivo Clearance. Small 2015, 11, 1683−1690. (26) Bastús, N. G.; Comenge, J.; Puntes, V. Kinetically Controlled Seeded Growth Synthesis of Citrate-Stabilized Gold Nanoparticles of up to 200 nm: Size Focusing versus Ostwald Ripening. Langmuir 2011, 27, 11098−11105. (27) Apopa, P. L.; Qian, Y.; Shao, R.; Guo, N. L.; Schwegler-Berry, D.; Pacurari, M.; Porter, D.; Shi, X.; Vallyathan, V.; Castranova, V.; Flynn, D. C. Iron Oxide Nanoparticles Induce Human Microvascular Endothelial Cell Permeability Through Reactive Oxygen Species Production and Microtubule Remodeling. Part. Fibre Toxicol. 2009, 6, 1. (28) Brun, E.; Carriere, M.; Mabondzo, A. In vitro Evidence of Dysregulation of Blood-Brain Barrier Function after Acute and Repeated/Long-Term Exposure to TiO2 Nanoparticles. Biomaterials 2012, 33, 886−896. (29) Pober, J. S.; Sessa, W. C. Evolving Functions of Endothelial Cells in Inflammation. Nat. Rev. Immunol. 2007, 7, 803−815. (30) Kempe, S.; Kestler, H.; Lasar, A.; Wirth, T. NF-κB Controls the Global Pro-Inflammatory Response in Endothelial Cells: Evidence for the Regulation of a Pro-Atherogenic Program. Nucleic Acids Res. 2005, 33, 5308−5319. (31) Zhang, X. D.; Luo, Z.; Chen, J.; Song, S.; Yuan, X.; Shen, X.; Wang, H.; Sun, Y.; Gao, K.; Zhang, L.; Fan, S.; Leong, D. T.; Guo, M.; Xie, J. Ultrasmall Glutathione-Protected Gold Nanoclusters as Next Generation Radiotherapy Sensitizers with High Tumor Uptake and High Renal Clearance. Sci. Rep. 2015, 5, 8669. (32) Shukla, R.; Bansal, V.; Chaudhary, M.; Basu, A.; Bhonde, R. R.; Sastry, M. Biocompatibility of Gold Nanoparticles and Their Endocytotic Fate Inside the Cellular Compartment: A Microscopic Overview. Langmuir 2005, 21, 10644−10654. (33) Tay, C. Y.; Setyawati, M. I.; Xie, J.; Parak, W. J.; Leong, D. T. Back to Basics: Exploiting the Innate Physico-chemical Characteristics of Nanomaterials for Biomedical Applications. Adv. Funct. Mater. 2014, 24, 5936−5955. (34) Chithrani, B. D.; Ghazani, A. A.; Chan, W. C. Determining the Size and Shape Dependence of Gold Nanoparticle Uptake into Mammalian Cells. Nano Lett. 2006, 6, 662−668. (35) Nel, A. E.; Madler, 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, 543−557. (36) Zarska, M.; Novotny, F.; Havel, F.; Sramek, M.; Babelova, A.; Benada, O.; Novotny, M.; Saran, H.; Kuca, K.; Musilek, K.; Hvezdova, Z.; Dzijak, R.; Vancurova, M.; Krejcikova, K.; Gabajova, B.; Hanzlikova, H.; Kyjacova, L.; Bartek, J.; Proska, J.; Hodny, Z. Two-Step Mechanism of Cellular Uptake of Cationic Gold Nanoparticles Modified by (16Mercaptohexadecyl)trimethylammonium Bromide. Bioconjugate Chem. 2016, 27, 2558−2574. (37) Mehta, D.; Malik, A. B. Signaling Mechanisms Regulating Endothelial Permeability. Physiol. Rev. 2006, 86, 279−367.

ACKNOWLEDGMENTS This study was supported by Ministry of Education Academic Research grant (R-279-000-418-112 to DTL) REFERENCES (1) Dejana, E. Endothelial Cell-Cell Junctions: Happy Together. Nat. Rev. Mol. Cell Biol. 2004, 5, 261−270. (2) Howard, M.; Zern, B. J.; Anselmo, A. C.; Shuvaev, V. V.; Mitragotri, S.; Muzykantov, V. Vascular Targeting of Nanocarriers: Perplexing Aspects of the Seemingly Straightforward Paradigm. ACS Nano 2014, 8, 4100−4132. (3) von Roemeling, C.; Jiang, W.; Chan, C. K.; Weissman, I. L.; Kim, B. Y. Breaking Down the Barriers to Precision Cancer Nanomedicine. Trends Biotechnol. 2016, 35, 159−171. (4) Setyawati, M. I.; Tay, C. Y.; Docter, D.; Stauber, R. H.; Leong, D. T. Understanding and Exploiting Nanoparticles’ Intimacy with the Blood Vessel and Blood. Chem. Soc. Rev. 2015, 44, 8174−8199. (5) Bae, Y. H.; Park, K. Targeted Drug Delivery to Tumors: Myths, Reality and Possibility. J. Controlled Release 2011, 153, 198−205. (6) Cao, J.; Wang, R.; Gao, N.; Li, M.; Tian, X.; Yang, W.; Ruan, Y.; Zhou, C.; Wang, G.; Liu, X.; Tang, S.; Yu, Y.; Liu, Y.; Sun, G.; Peng, H.; Wang, Q. A7RC Peptide Modified Paclitaxel Liposomes Dually Target Breast Cancer. Biomater. Sci. 2015, 3, 1545−1554. (7) Wang, Q.; Cheng, H.; Peng, H.; Zhou, H.; Li, P. Y.; Langer, R. NonGenetic Engineering of Cells for Drug Delivery and Cell-Based Therapy. Adv. Drug Delivery Rev. 2015, 91, 125−140. (8) Jain, R. K.; Stylianopoulos, T. Delivering Nanomedicine to Solid Tumors. Nat. Rev. Clin. Oncol. 2010, 7, 653−664. (9) Kutty, R. V.; Chia, S. L.; Setyawati, M. I.; Muthu, M. S.; Feng, S.-S.; Leong, D. T. In vivo and Ex vivo Proofs of Concept that Cetuximab Conjugated Vitamin E TPGS Micelles Increases Efficacy of Delivered Docetaxel against Triple Negative Breast Cancer. Biomaterials 2015, 63, 58−69. (10) Du, D.; Chang, N.; Sun, S.; Li, M.; Yu, H.; Liu, M.; Liu, X.; Wang, G.; Li, H.; Liu, X.; Geng, S.; Wang, Q.; Peng, H. The Role of Glucose Transporters in the Distribution of p-aminophenyl-alpha-D-mannopyranoside Modified Liposomes within Mice Brain. J. Controlled Release 2014, 182, 99−110. (11) Kennedy, I. M.; Wilson, D.; Barakat, A. I. Uptake and Inflammatory Effects of Nanoparticles in a Human Vascular Endothelial Cell Line. Res. Rep. - Health Eff. Inst. 2009, 136, 3−32. (12) Khanna, P.; Ong, C.; Bay, B.; Baeg, G. Nanotoxicity: An Interplay of Oxidative Stress, Inflammation and Cell Death. Nanomaterials 2015, 5, 1163. (13) Liu, X.; Sun, J. Endothelial Cells Dysfunction Induced by Silica Nanoparticles Through Oxidative Stress via JNK/P53 and NF-κB Pathways. Biomaterials 2010, 31, 8198−8209. (14) Bazzoni, G.; Dejana, E. Endothelial Cell-to-Cell Junctions: Molecular Organization and Role in Vascular Homeostasis. Physiol. Rev. 2004, 84, 869−901. (15) Tay, C. Y.; Setyawati, M. I.; Leong, D. T. Nanoparticle Density: A Critical Biophysical Regulator of Endothelial Permeability. ACS Nano 2017, 11, 2764−2772. (16) Setyawati, M. I.; Mochalin, V. N.; Leong, D. T. Tuning Endothelial Permeability with Functionalized Nanodiamonds. ACS Nano 2016, 10, 1170−1181. (17) Setyawati, M. I.; Tay, C. Y.; Chia, S. L.; Goh, S. L.; Fang, W.; Neo, M. J.; Chong, H. C.; Tan, S. M.; Loo, S. C.; Ng, K. W.; Xie, J. P.; Ong, C. N.; Tan, N. S.; Leong, D. T. Titanium Dioxide Nanomaterials Cause Endothelial Cell Leakiness by Disrupting the Homophilic Interaction of VE-cadherin. Nat. Commun. 2013, 4, 1673. (18) Oda, H.; Takeichi, M. Evolution: Structural and Functional Diversity of Cadherin at the Adherens Junction. J. Cell Biol. 2011, 193, 1137−1146. (19) Satchell, S. C.; Braet, F. Glomerular Endothelial Cell Fenestrations: An Integral Component of the Glomerular Filtration Barrier. Am. J. Physiol. Renal Physiol. 2009, 296, F947−956. 5029

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030

Article

ACS Nano (38) Birukova, A. A.; Tian, Y.; Dubrovskyi, O.; Zebda, N.; Sarich, N.; Tian, X.; Wang, Y.; Birukov, K. G. VE-cadherin Trans-Interactions Modulate Rac Activation and Enhancement of Lung Endothelial Barrier by Iloprost. J. Cell. Physiol. 2012, 227, 3405−3416. (39) Huang, R. L.; Teo, Z.; Chong, H. C.; Zhu, P.; Tan, M. J.; Tan, C. K.; Lam, C. R.; Sng, M. K.; Leong, D. T.; Tan, S. M.; Kersten, S.; Ding, J. L.; Li, H. Y.; Tan, N. S. ANGPTL4 Modulates Vascular Junction Integrity by Integrin Signaling and Disruption of Intercellular VEcadherin and Claudin-5 Clusters. Blood 2011, 118, 3990−4002. (40) Adam, A. P.; Sharenko, A. L.; Pumiglia, K.; Vincent, P. A. Srcinduced Tyrosine Phosphorylation of VE-cadherin Is Not Sufficient to Decrease Barrier Function of Endothelial Monolayers. J. Biol. Chem. 2010, 285, 7045−7055. (41) Hatanaka, K.; Simons, M.; Murakami, M. Phosphorylation of VEcadherin Controls Endothelial Phenotypes via p120-Catenin Coupling and Rac1 Activation. Am. J. Physiol.: Heart Circ. Physiol. 2011, 300, H162−H172. (42) Maekawa, M.; Ishizaki, T.; Boku, S.; Watanabe, N.; Fujita, A.; Iwamatsu, A.; Obinata, T.; Ohashi, K.; Mizuno, K.; Narumiya, S. Signaling from Rho to the Actin Cytoskeleton through Protein Kinases ROCK and LIM-Kinase. Science 1999, 285, 895−898. (43) Kaunas, R.; Nguyen, P.; Usami, S.; Chien, S. Cooperative Effects of Rho and Mechanical Stretch on Stress Fiber Organization. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 15895−15900. (44) Rüffer, C.; Strey, A.; Janning, A.; Kim, K. S.; Gerke, V. Cell−Cell Junctions of Dermal Microvascular Endothelial Cells Contain Tight and Adherens Junction Proteins in Spatial Proximity. Biochemistry 2004, 43, 5360−5369.

5030

DOI: 10.1021/acsnano.7b01744 ACS Nano 2017, 11, 5020−5030