Passing through the renal clearance barrier: toward ultrasmall sizes with stable ligands for potential clinical applications.

The use of nanoparticles holds promise for medical applications, such as X-ray imaging, photothermal therapy and radiotherapy. However, the in vivo toxicity of inorganic nanoparticles raises some concern regarding undesirable side effects which prevent their further medical application. Ultrasmall sub-5.5 nm particles can pass through the barrier for renal clearance, minimizing their toxicity. In this letter we address some recent interesting work regarding in vivo toxicity and renal clearance, and discuss the possible strategy of utilizing ultrasmall nanomaterials. We propose that small hydrodynamic sized nanoclusters can achieve both nontoxic and therapeutic clinical features.

Dear editor

Biomedical applications of inorganic nanoparticles have been investigated for several years.1–3 It was conceived that nanoparticles hold potential for use as X-ray contrast agents, fluorescence imaging agents, photothermal therapy agents, and radiosensitizers.4–12 However, their toxicity, especially in vivo toxicity, is still a huge challenge for further applications in medicine.13–17 In earlier years, several groups have demonstrated that citric acid-coated nanoparticles are not toxic in vitro, and they demonstrated that human cells were viable and highly active even at millimolar levels.18,19 However, the in vitro toxicity cannot necessarily reflect the in vivo toxicity. The in vivo toxicity of nanoparticles mainly arises from accumulation of nanoparticles in the liver and spleen. Specifically, nanoparticles with sizes ranging from 10 to 250 nm show very high distributions in the liver and spleen due to reticuloendothelial system (RES) absorption.20 Several groups have reported that large nanoparticles accumulate in the liver and spleen, and they are not cleared easily.21–23 Large nanoparticles can be absorbed rapidly by macrophages, because the small naked gold nanoparticles will first react with blood proteins and then form larger nanoparticle–protein complexes, namely protein corona.24–28 The high distribution in the liver and spleen can induce potential liver toxicities.7,27,29

In order to decrease toxicity, one of the most effective options is to decrease the size of the nanoparticles, making them behave as supermolecules which can induce “leakage” of nanoparticles in vivo mediated by the kidney.30,31 Choi et al proposed that nanoparticles can be cleared when the hydrodynamic size of nanoparticles is decreased to 5.5 nm.32 We can define this hydrodynamic size as the size of the renal clearance barrier (Figure 1). Indeed, it has been shown that small quantum dots can present with highly efficient renal clearance.33 However, only being small in size is not enough, because naked nanoparticles can still react with blood proteins and form larger complexes. For example, Hainfeld et al used X-ray imaging of 1.9 nm gold nanoparticles,6 and when ultrahigh doses (2.7 g Au/kg) of gold nanoparticles are used in mice, many were found in the blood vessels. Indeed, these nanoparticles are difficult to clear. Furthermore, toxic effects were not demonstrated, but the high dose was considered to be very dangerous.

Figure 1

Outline of the renal clearance barrier.

Polyethylene glycol (PEG) is widely used in coating of nanoparticles for nanomedical applications.34,35 However, PEG-coated sub-5 nm nanoparticles cannot be cleared by the kidney,27,36,37 even when the hydrodynamic size of PEG-coated gold nanoparticles decreased to 3 nm.36 Ultrasmall BaGdF5-based upconversion nanoparticles also cannot be cleared by the kidney.38 Thus, it is clear that only having a small size is not sufficient for the clearance of nanoparticles.

To overcomes these obstacles, ultrasmall nanoclusters or particles with stable ligands have been proposed. Specifically, nanoparticles with small endogenous ligands would be highly desirable. For example, sub-5 nm gold nanoparticles with glutathione (GSH) ligands, a kind of endogenous small molecule, are highly efficient in renal clearance.36 Meanwhile, the sub-2 nm GSH-protected Au nanoclusters are also highly efficient in renal clearance and they also show low toxicity after injection.39 Even after 30 days, the GSH-protected gold nanoclusters did not show any significant toxicity. Using 1.2 nm gold nanoparticles as cores, the GSH-protected gold nanoparticles can be cleared by the kidney even at the dose of 60 μM.40 In contrast, bovine serum albumin (BSA)-protected Au nanoclusters in a hydrodynamic size of 6 nm cannot pass the renal clearance barrier and show minor liver toxicities. In another independent work, the renal clearance of dithiolated polyaminocarboxylate-coated gold nanoparticles was reported.41 In this work, the hydrodynamic size was as large as 6.6 nm, higher than the proposed value of 5.5 nm for the renal clearance barrier, but the nanoparticles can still induce a highly efficient renal clearance.41 Therefore, this illustrates that stable ligands are as important as the hydrodynamic size in renal clearance of nanoparticles.

With the exception of gold-based nanomaterials as mentioned above, other ultrasmall nanomaterials such as carbon nanomaterials and semiconducting quantum dots have also been widely investigated. For example, Huang et al have shown that amine-functionalized carbon dots in a 4.1 nm hydrodynamic size can be cleared renally.42 However, carboxylated graphene quantum dots in 3–6 nm hydrodynamic sizes cannot go through the renal clearance barrier and thus cause an appreciable distribution in the liver and spleen. It is not yet clear why these graphene quantum dots cannot pass through the renal clearance barrier, but one possible reason could be their instability exogenously. As such, it is still necessary to obtain renally-clearable graphene quantum dots with stable ligands for further medical applications. A similar phenomenon was observed using ultrasmall semiconducting nanoparticles. Polyvinylpyrrolidone-protected Gd2O3 nanoparticles in a 2.9 nm ultrasmall size can achieve efficient renal clearance.43 However, 1.5 nm Ag2Se nanoparticles cannot be metabolized by the kidney.44

Besides this, it is interesting to control the clearance of nanoparticles mediated by macrophages. For example, Chou et al used DNA to control the biological delivery and clearance of inorganic nanoparticles by organizing them into colloidal superstructures.45 The nanoparticles behave as building blocks whose size, surface chemistry and assembly architecture dictate the overall superstructure design. These superstructures are able to interact with cells and tissues as a function of their designs, but subsequently degrade into building blocks that can escape biological sequestrations. Thereby, this design realizes successful clearance of nanoparticles resulting in intact biofunctions.

Despite some significant advances with small nanoparticles, some difficulties still exist. When the core size of the nanoparticles is decreased to sub-3 nm, the physical and chemical properties of some materials might be lost or altered. For gold-based nanomaterials, the surface plasmon resonance of gold disappears when diameters are decreased to sub-5 nm.46 Instead, unique electronic structures and optical properties are introduced.47,48 In this situation, the photothermal efficiency of nanoparticles will be considerably affected due to sharply decreased absorption. As for quantum dots, the quantum dots will be easily quenched or bleached with their sizes decreased to sub-5 nm.49 Meanwhile, as the quantum size effect is induced by small sizes, the band gap of quantum dots will be widened and the wavelength for photoluminescence will shift to the region of shorter wavelengths.50 Taking carbon nanomaterials into account with their size deceased to sub-5 nm, the band gap of graphene nanosheets or graphene quantum dots increases to 2–3 eV according to band gap engineering, while the fluorescence of carbon nanotubes will disappear.51–54 As the diameter is decreased to sub-5 nm levels, magnetic properties of some magnetic materials such as Fe3O4 will be influenced. Meanwhile, it would be more difficult to dope them with other elements. Lots of upconversion materials will lose their fluorescence characteristics with their size decreased to 5 nm. At the size of sub-5 nm, strong surface activities can jeopardize applications, with surface modifications as a typical example. Another challenge lies in how to monitor concentrations of ultrasmall organic nanoparticles, because it is still unclear whether nanoparticles will be broken down in vivo, and in what quantity. To be specific, when nanoparticles are injected into mice, lots of particles will interact with proteins and may then be broken down in vivo. In this case, it will be very difficult to determine how many nanoparticles still stay in the body, making the related clearance complicated. In summary, obtaining ultrasmall particles with good physical and chemical properties for medical applications is still an unmet need and remains a challenge for further research.