The Gap Between Endothelial Cells: Key to The Quick Escape of Nanomaterials?

Keywords::

• endothelial cells
• endothelial permeability
• enhanced permeability and retention
• EPR
• nanomaterials

Nanomaterials sometimes induce surprising and unexpected but interesting biological effects both at the cellular and the pathway level [1–6]. Specifically in nanomedicine, where nanomaterials are intentionally introduced into the vasculature similarly to conventional drugs, potential health concerns have been raised. This brings into question the pharmacokinetics and biodistribution of these nanomaterials, and many animal studies have shown a propensity for these nanomaterials to accumulate in major highly vascularized organs for a reasonable period of time as a consequence of systemic exposure [7,8]. This accumulation would presuppose that some of these nanomaterials may have crossed the endothelium.

The enhanced permeability and retention (EPR) effect is an important concept that is exploited by nanoparticles in order to penetrate deeply into the tumor mass. However, EPR is a tumor-specific effect [9,10]. If EPR alone is the only route that nanoparticles take to escape from the vasculature, one would expect the tumor site to be the sole site of nanoparticles' localization and accumulation. However, there are several studies that have looked at the intravenous introduction of nanoparticles into animals that demonstrate localization and accumulation of nanoparticles in organs other than at the tumor, mostly in highly vascularized organs such as the liver, kidney and spleen [7,8]. External modes of nanoparticle exposures, such as dermal, inhalation and ingestion, also result in nanoparticles accumulation in major internal organs [7,8]. These results suggest that there might be another mode besides EPR by which nanoparticles are able to cross the endothelium. The presence of these nanoparticles in seemingly untargeted or unintended tissues or organs of the body naturally raises questions concerning potential health concerns and their risks and benefits.

Paradigms of crossing the endothelium

There are two prevailing paradigms on how nanoparticles traverse across endothelial barriers [11,12]. The first is a transcellular mechanism. Nanoparticles are endocytosed from the luminal side of the endothelium, transported through the entire depth of the cell and then exocytosed on the basal face of the endothelial cell [13]. Understandably, this strategy presents several hindrances, as it is energetically costly, slow and subject to other endosomal transport mechanisms that might divert the bulk of the nanoparticle population from exocytosis to persisting within the cell instead. This increased persistence may also induce apoptosis in the endothelium in the case of toxic nanoparticles or nanomedicines carrying cell death-inducing drugs. The second prevailing paradigm, aptly named paracellular transport, derives its inspiration from the transport of small molecules, such as glucose and hormones, across cell barriers by diffusion between cells through the cell–cell junctions [11,12]. This paracellular transport does not require any intracellular involvement and logically represents an easier journey for the nanoparticles to cross the endothelium. While this is well understood for small biomolecules, it is not a well-established observation for nanoparticles, with there only being little evidence that this could happen and it appears to be highly dependent on size. Renal clearance rules suggest that nanoparticles of sizes smaller than 6 nm are easily cleared from the vasculature [14]. However, larger-sized nanoparticles that are not cleared by the kidneys still have the propensity to accumulate in other highly vascularized organs [7,8] and paracellular transport is insufficient to explain why larger-sized nanoparticles can still cross the endothelium.

Instead, we observed another phenomenon that appears to supersede both paradigms temporally. We first observed that exposure to nanoparticles very quickly (within 30 min) leads to the development of distinct tens of microns-sized gaps between the cells on a continuous monolayer of endothelial cells [1]. We coined this process to be ‘nanoparticle-induced endothelial leakiness’ (NanoEL). We also tested silver nanoparticles and silica nanoparticles (15–30 nm) and observed the same phenomenon, suggesting that NanoEL may not be restricted to just TiO₂-based nanomaterials [1]. The larger-sized, submicron TiO₂ (680 nm), however, did not demonstrate NanoEL, even at very high treatment doses. We observed that nano-TiO₂ by virtue of its size was able to squeeze into the adherens junctions (~22.5 nm) [15], which are found between endothelial cells and induced leakiness of the otherwise intact endothelial cells monolayer (for a detailed mechanism, see [1]). The gap formed stretched as wide as 10–30 µm, which is almost a 1000-fold difference in magnitude from its original cell–cell gap width of nanometers scale [1]. This then allowed more nanoparticles to pass through these induced gaps. NanoEL is therefore a nanoparticle-induced effect and not a tumor-initiated effect as with EPR.

Bringing this to a physiological in vivo setting, since nanoparticles are in the nanometer size range, they can easily traverse the cellular barrier through these micron-sized gaps. Crossing cell barriers via NanoEL is distinctly different from the paracellular route. In the paracellular paradigm, permeability is increased, but the ability to exclude most molecules is retained. In contrast in NanoEL case, the cell layer is already compromised with the presence of micron sized gaps, which should allow the unimpeded access of up to tens of microns-sized entities from nanoparticles, proteins, viruses, bacteria and even cells. We further showed that leakiness can occur to the extent that micron-sized circulating melanoma cells were able to capitalize on this increased leakiness in the vasculature in order to significantly increase their colonization in a melanoma–lung metastasis mouse model [1]. However, we are not proposing that NanoEL is the only way that endothelial permeability occurs when nanomaterials are involved, because multiwalled carbon nanotubes [16], iron oxide [17] and silica [18] nanoparticles can induce sufficient oxidative stress or apoptosis in order to cause endothelial permeability. Instead, we are proposing not to exclude NanoEL as an alternative paradigm and perhaps a much faster way of inducing endothelial leakiness for at least those nanoparticles that are known to induce benign levels of oxidative stress and exert low cytotoxicity [1].

Pathological implications of NanoEL

The implications of NanoEL are numerous, since the permeability of blood vessels is normally tightly regulated. As with any pathological leakiness of blood vessels, one can expect an increase in stroke, cancer metastasis and infection [1]. As an example on the implication of leakiness, Yamashita et al. observed pregnancy complication and fetotoxicity due to TiO₂ nanoparticles (35–70 nm) crossing the placental barrier from the murine mother to the fetus through an unknown mechanism [3]. We believe that, in light of our evidence, this could be due to NanoEL at the placental capillary bed.

In addition, due to the NanoEL effect, nanomedicine, which is designed to be intravenously introduced, may need to take into account some loss due to NanoEL into unexpected sites. This will ultimately increase the overall dose. While tumor vasculature is already known to be leaky, NanoEL probably will not augment the degree of leakiness; however, other nontargeted and highly vascularized sites, such as the liver and kidneys, might become leaky, and those organs may suffer from unintended effects of these nanomedicine.

We have demonstrated pathological effects of NanoEL, such as an increased chance of metastasis [1], and based on the logic of leaky vasculature, one would expect increased access for blood-borne infectious agents and possibly increased chances of stroke due to increased chances of clot formation. As with all entities that exert a biological effect, dose plays a major role. In our animal study, we observed increased melanoma metastasis at doses that would probably require many years of exposure to the nano materials [1].

Potential positive exploitations of NanoEL

NanoEL may still present potential applications if it could be better controlled. For example, not all tumors at their earlier stages are accessible via the EPR effect. NanoEL may be useful in lowering the threshold for nanomedicine in order to access those tumors in the absence of an EPR effect. From this perspective, the benefits of NanoEL may become apparent in highly resistant barriers, such as the blood–brain barrier [19]. In order for us to minimize the pathological effects of NanoEL and maximize its benefits, we need to understand some pertinent issues about the nanoparticles that cause NanoEL and how the endothelial cells within this context behave.

NanoEL from the nanoparticle perspective

Determining the nanoparticle sizes and charge windows that cause NanoEL will indeed lower the hits and misses of designing NanoEL-capable nanoparticles. Based on our study, in order to cause NanoEL, a nanoparticle should be approximately 20 nm in size, which is approximately the width of the adherens junction [15]. In addition, the charge of the nanoparticles may play a role. In our study, we showed that the nanoparticles that caused NanoEL were negatively charged. A range of sizes and charged nanoparticles would greatly help in clarifying some of our earlier observations. An interesting question pertains to why adherens junctions would be targeted in the first place. It seems highly improbable that the relatively low cell boundary area of the adherens junctions was targeted versus the entire luminal surface of the cell. We alluded to the fact that negatively charged nanoparticles may have landed on the negatively charged glycocalyx-covered surface of the endothelial cells [20] and may have electrostatically ‘bounced’ along the surface until it reached the adherens junction [1]. We showed that the endocytosis of the same nanoparticles was slower than the onset of leakiness, thus further supporting our ‘bouncing nanoparticle’ hypothesis [1].

In conclusion, NanoEL might possibly be an alternative mechanism that affects the vasculature in ways that complement the EPR effect as a potentially exploitable strategy for use in nanomedicine, although this is tempered by nanosafety considerations.

Financial & competing interests disclosure

The authors would like to acknowledge the Ministry of Education, Singapore (R-279-000-350-112 and R-279-000-418-112 to DT Leong and R-279-000-376-112 to CY Tay). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript.

No writing assistance was utilized in the production of this manuscript.

Additional information

Funding

The authors would like to acknowledge the Ministry of Education, Singapore (R-279-000-350-112 and R-279-000-418-112 to DT Leong and R-279-000-376-112 to CY Tay). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript.