bridge over troubled waters: understanding the synthetic and ......overview bridge over troubled...

Overview

Bridge over troubled waters:understanding the synthetic andbiological identities of engineerednanomaterialsBengt Fadeel,1∗ Neus Feliu,1 Carmen Vogt,1 Abuelmagd M.Abdelmonem2 and Wolfgang J. Parak2

Engineered nanomaterials offer exciting opportunities for ‘smart’ drug deliveryand in vivo imaging of disease processes, as well as in regenerative medicine. Theability to manipulate matter at the nanoscale enables many new properties thatare both desirable and exploitable, but the same properties could also give riseto unexpected toxicities that may adversely affect human health. Understandingthe physicochemical properties that drive toxicological outcomes is a formidablechallenge as it is not trivial to separate and, hence, to pinpoint individualmaterial characteristics of nanomaterials. In addition, nanomaterials that interactwith biological systems are likely to acquire a surface corona of biomoleculesthat may dictate their biological behavior. Indeed, we propose that it is thecombination of material-intrinsic properties (the ‘synthetic identity’) and context-dependent properties determined, in part, by the bio-corona of a given biologicalcompartment (the ‘biological identity’) that will determine the interactions ofengineered nanomaterials with cells and tissues and subsequent outcomes. Thedelineation of these entwined ‘identities’ of engineered nanomaterials constitutesthe bridge between nanotoxicological research and nanomedicine. © 2013 WileyPeriodicals, Inc.

How to cite this article:WIREs Nanomed Nanobiotechnol 2013, 5:111–129. doi: 10.1002/wnan.1206

INTRODUCTION

Nanomaterials are in the same size range asbiomolecules and cellular structures; this factlies at the very heart of nanomedicine, a field in whichmany applications rely on nanoscale interactions.However, this is also the reason for the currentconcern surrounding nanomaterials: the interferenceof man-made nanomaterials with biological systemscould also lead to hazardous effects on humanhealth.1 While the interest in nanoscale materials hasincreased tremendously in recent years, importantobservations on their interactions with biological

∗Correspondence to: [email protected] of Molecular Toxicology, Institute of EnvironmentalMedicine, Karolinska Institutet, Stockholm, Sweden2Fachbereich Physik and Wissenschaftlichen Zentrum für Material-wissenschaften, Philipps Universität Marburg, Marburg, Germany

systems were reported much earlier. For instance,the fact that nanoparticles are typically incorporatedby cells via endocytosis was known for decades.2

In addition, colloidal nanoparticles were shown toinduce alterations in the blood–air barrier in themouse lung more than half a century ago.3 Numerousstudies have been published more recently in whichexposure to engineered nanoparticles has been linkedto toxicity.4–6 However, understanding which ofthe physicochemical properties of nanomaterials thatare driving toxicity remains a challenge; if onecould connect material properties (size, shape, surfacecharge, porosity, colloidal stability, purity/degree ofcontamination, etc.) with toxicity, then this wouldenable prediction of potential hazards and could alsolead to the design of nanomaterials with minimaltoxicity.7 In addition, a thorough understanding of theproperties of nanomaterials that determine biological

Volume 5, March/Apr i l 2013 © 2013 Wiley Per iodica ls, Inc. 111

Overview wires.wiley.com/nanomed

responses would also facilitate the design of betternanomedicines for the treatment of human disease.

In this review, we discuss the bridgingof nanotoxicological research and nanomedicine.We suggest that a careful understanding ofnanomaterial physicochemical properties, i.e., the‘synthetic identity’, constitutes the bridge betweenthese two disciplines. Moreover, we propose that the‘biological identity’ of nanomaterials is determined,in part, by the adsorption of biomolecules onto thenanomaterial surface upon introduction into a livingsystem. In fact, as pointed out by Walkey and Chan8

in their excellent review on the protein corona asit applies to nanomaterials, ‘once fully mapped, therelationships between synthetic identity, biologicalidentity, and physiological response will enableresearchers to predict the physiological responseof a nanomaterial by characterizing its syntheticidentity’. This statement points toward a predictivenanotoxicology, the ultimate goal of which is todecode nanomaterial properties to enable redesignof materials that are both useful and safe.9

NANOMATERIALS IN MEDICINE

Engineered nanomaterials offer great potential inmedical applications.10 The variety of possibleapplications is very broad. Here, we providesome highlights, and we attempt to emphasize thephysicochemical properties that make nanomaterialsso favorable, in particular for medical imaging anddrug or gene delivery.

Medical ImagingMedical imaging is typically based on the useof contrast agents, which facilitate visualizationof tissues and organs. For imaging, two basicproperties are required. First, the contrast agentshould provide contrast compared to the localenvironment, and second, the contrast agent shouldbe specifically localized at the region of interest. Whatmay nanoparticles contribute in this direction? Thefirst answer to this question is relatively obvious:nanoparticles can provide higher contrast because oftheir larger size compared to individual molecules.Furthermore, instead of having only one fluorophorefor fluorescence imaging, or just one chelated ion suchas Gd2+ to provide contrast for magnetic resonanceimaging (MRI), several fluorophores or Gd2+ ions canbe combined in one nanoparticle, and this multivalentdisplay may provide higher contrast. In fact, thecombination of different contrast agents in a singlenanoparticle allows for multimodal imaging.11 The

combination of diagnostic and therapeutic functionsin a single ‘theranostic’ platform has also beenattempted. Yang et al.12 functionalized a reducedgraphene oxide–iron oxide nanoparticle (RGO-IONP)complex with poly(ethylene glycol) (PEG), obtaininga RGO-IONP-PEG theranostic nanoprobe that wasused for in vivo trimodal fluorescence, photoacoustic,and MR imaging, uncovering high passive tumortargeting, which was further exploited for thermalablation of tumors in mice.

The second answer to the aforementioned ques-tion is less obvious. Nonetheless, because of theirlarger size compared to molecules, nanoparticlesare passively trapped in tumors; this phenomenonis known as the enhanced permeability and reten-tion (EPR) effect. Nanoparticles are small enoughto leak out from the bloodstream into tumors, yetbig enough to be trapped in the tumor vascula-ture. Hence, nanoparticles with many fluorophorescan passively accumulate in tumors more efficientlythan the same individual fluorophores.13 However,subsequent penetration into the tumor itself is notreadily achieved. Wong et al.14 generated a mul-tistage nanoparticle delivery system for deep pen-etration into tumors. Hence, the gelatin core of100-nm nanoparticles was degraded by proteasespresent in the tumor microenvironment thereby releas-ing 10-nm quantum dots (QDs) after extravasation.Chauhan et al.15 investigated how vascular normaliza-tion affects nanoparticle delivery by studying whethera vascular endothelial growth factor receptor-2-blocking antibody modulates nanoparticle penetrationrates in mammary tumors in vivo. The authors demon-strated that 12-nm particles penetrate tumors betterthan larger particles (125 nm) once abnormal vesselsare repaired, suggesting that small nanoparticles lessthan 12 nm are superior because of higher tumorpenetration.

Kim et al.16 provided a particularly relevantexample of nanoparticle-based imaging involving QD-based fluorescence labeling, allowing for sentinellymph node mapping in large animals under imageguidance. This approach could have significant impacton such surgical procedures in cancer patients,provided that toxicity of QDs is controlled. Thisis a nontrivial question, as QDs are typically madefrom inherently toxic components such as cadmium,a heavy metal with known adverse effects onhuman health. A recent study in nonhuman primatessuggested that phospholipid micelle-encapsulatedCdSe/CdS/ZnS QDs do not induce major signsof toxicity up to 90 days postexposure; however,chemical analysis revealed that most of the initial

112 © 2013 Wiley Per iodica ls, Inc. Volume 5, March/Apr i l 2013

WIREs Nanomedicine and Nanobiotechnology Bridging nanotoxicology and nanomedicine

dose of cadmium remained in the liver, spleen, andkidneys.17

Finally, it is noteworthy that nanoparticles alsoafford label-free detection. Hence, carbon nanotubeshave been shown to be useful for photoacousticimaging, an approach that offers higher spatialresolution and allows deeper tissues to be imagedcompared with most optical imaging techniques.18

In a more recent study, Tong et al.19 reported thattransient absorption microscopy offers an alternative,label-free method to image both semiconducting andmetallic single-walled carbon nanotubes (SWCNTs)in vitro and in vivo, in real time, with submicrometerresolution.

Drug DeliveryNanoparticles clearly offer novel features for ‘smart’drug delivery.20 First of all, nanoparticles offerpotential as passive carrier systems for delivery. Thisis due to the fact that drug-loaded nanoparticlesinteract differently with cells than the correspondingdrug alone.13 Nanoparticles can also be loaded withdrugs in a way that allows for their slow release.Fine tuning of the surface of nanoparticles allowsfor regulation of nanoparticle interactions with cellsand thus the mode of delivery.21,22 Nanoparticles canbe used to increase the local concentration of drugsin, for instance, cancer cells. Ashley et al.23 designedporous nanoparticle-supported lipid bilayers termedprotocells that synergistically combined properties ofliposomes and nanoporous particles. The protocellscan be loaded with combinations of therapeuticagents, e.g., drugs or small interfering RNAs. The veryhigh capacity of the high-surface area nanoporouscore combined with the enhanced targeting efficacytoward cancer cells enabled by the fluid-supportedlipid bilayer enabled a single protocell loaded witha drug cocktail to kill a drug-resistant humanhepatocellular carcinoma cell, representing a million-fold improvement over comparable liposomes. Furtherin vivo studies are certainly warranted. Davis et al.24

administered nanoparticles functionalized with atargeting ligand (transferrin) systemically to a smallnumber of cancer patients and were able todemonstrate successful RNA interference, i.e., specificinhibition of gene expression.

Notably, in a recent landmark study, the gapbetween preclinical development and clinical transla-tion was bridged using targeted doxorubicin-loadednanoparticles.25 The nanoparticles were developedfrom a combinatorial library of more than 100 tar-geted nanoparticle formulations varying with respectto particle size, targeting ligand density, surface

hydrophilicity, drug loading, and drug release prop-erties. In tumor-bearing mice, rats, and nonhumanprimates, doxorubicin-loaded nanoparticles displayedpharmacokinetic characteristics consistent with pro-longed circulation of nanoparticles in the vascularcompartment and controlled release of the drug. Inaddition, clinical data in patients with advanced solidtumors indicated a pharmacokinetic profile consis-tent with the preclinical data as well as some casesof tumor shrinkage at doses below the solvent-baseddoxorubicin formulation dose typically used in theclinic.25 This study shows that the ‘valley of death’between preclinical research and clinical applicationscan be bridged through a rational design approach(Figure 1).

Moreover, inorganic nanoparticles can be usedfor introducing new functionalities. Here, twofascinating examples are given. First, magneticnanoparticles can be used for locally trappingdrugs (which are attached to the nanoparticles) byapplication of magnetic field gradients.26 Magnetictargeting has been applied both in vitro27 and invivo.28 Although clinical applications so far arelimited to pets, this technology has the potentialfor being applied to humans in the future, inparticular in cases of tumors close to the skin,as sufficiently high magnetic field gradients can bedirected to the body surface.28 Second, plasmonicnanoparticles, in particular those based on gold,can be used for light-controlled release of drugs.29

Upon optical excitation at the plasmon resonancefrequency, collective motion of electrons ultimatelyleads to dissipation and thus local heating of theenvironment of the nanoparticle surface.30 Initially,gold nanoparticles directed to tumor tissue have beenused for local tissue destruction by light-inducedheating, also referred to as hyperthermal ablation.31

Hence, nanoparticles may not only deliver drugs butcan also act as therapeutic agents per se (furtherexamples of such nanomaterial-intrinsic effects arediscussed below). The same phenomenon, however,can be employed for controlled delivery. Upon heatformation on the nanoparticle surface, molecularbonds can be broken and attached molecules canthus be released based on light triggers. Double-stranded DNA is a good linker, as it can be moltenat temperatures well below the boiling point of water.Light-controlled heating can also be used for theregulated opening of the nanoscale containers.32 Thismay thus enable release of drugs by local illumination.Because of absorption of light by tissue [even inthe near-infrared (NIR)], the most likely in vivoapplications will be for tumors located close to theskin.



phys-chem

characterization

Nanomedicine

Nanotoxicology

Synthetic andbiological identity

Clinical trials

Pre-clinical studies

(a)

(b)

FIGURE 1 | Bridging nanotoxicology and nanomedicine. We positthat careful assessment of the physicochemical properties of engineerednanomaterials constitutes the ‘bridge’ between nanotoxicologicalresearch and nanomedicine insofar as a detailed understanding ofmaterial properties, i.e., the ‘synthetic identity’ is critical both fortoxicological assessment of nanomaterials and for the development ofnovel nanomedicines (a). Furthermore, understanding the ‘synthetic’and ‘biological’ identities of nanomaterials will facilitate the bridging ofpreclinical studies and the use of nanomaterials in medical imaging,drug delivery, and regenerative medicine (b). The ‘biological’ identity ofa nanomaterial is largely determined by the ‘corona’ of biomoleculesthat forms in a biological environment; see text for details.

Regenerative MedicineNanomaterials may also have considerable impacton regenerative medicine, i.e., the replacement orregeneration of human cells, tissues, or organs.33 Forinstance, magnetic nanoparticles can be used to imageand guide stem cells to their target in stem cell-basedtherapies.34,35 Cells interact with the surroundingenvironment by making nanoscale interactions withextracellular signals and nanomaterials can beemployed as biomimetic scaffolds to stimulate tissuegrowth. Intriguingly, supramolecular nanostructuresthat mimic, for instance, a growth factor canbe used as a strategy for tissue regeneration andrepair.36,37 Furthermore, in a recent clinical study,Jungebluth et al.38 reported the first transplantationof an artificial trachea in a cancer patient. Aftercomplete tumor resection, the patient’s airway wasreplaced with a tailored bioartificial nanocompositepreviously seeded with autologous bone marrowmononuclear cells in a bioreactor. The cellsdifferentiated into appropriate cell types. There areseveral advantages to this approach. For instance, byusing the patient’s own stem cells to populate the

scaffold, there are no concerns over rejection of thetransplant.39

SAFETY ASSESSMENTOF NANOMATERIALS

Rational design of ‘nanomedicines’ began almost halfa century ago, and several products including lipo-somes (i.e., passive nanoscale carriers) have enteredinto routine clinical use (see Duncan and Gaspar40 foran excellent historical perspective). However, count-less other, more sophisticated nanomedicines are inthe pipeline and the potential risks to human healthof these novel entities need to be seriously considered.The latter is certainly true for all pharmaceutical prod-ucts. Nanotoxicology attempts to investigate the inter-actions of nanomaterials with biological systems.41

However, there are several important and complicat-ing aspects to address in nanotoxicological studiesincluding not only the need for standardized assaysand reference materials9 but also the issue of themost appropriate dose metric to use (surprisingly, thisremains largely unresolved), and it may as yet be tooearly to draw general conclusions regarding toxicityof nanomaterials; the prevailing view today is thatnanomaterials should be studied on a case-by-casebasis.42 Nevertheless, some lessons can be garneredfrom studies conducted over the past several years. Inthe following sections, we will discuss why engineerednanoparticles are potentially hazardous, with the aimto elucidate physicochemical properties that have beenlinked to toxicity. We also provide an overview ofemerging trends in nanotoxicology including high-throughput screening (HTS) and in silico modelingapproaches.

High-Throughput ScreeningAs pointed out recently,43 results of toxicological stud-ies using extraordinarily high doses of nanomaterialshave to be interpreted with caution. Indeed, while invitro tests may prove useful for hazard identification,in vivo studies are needed to bridge the gap betweencell culture model systems and the human exposuresituation, in order to understand whether nanomate-rials pose any risk to human health. At the same time,it is not ethically, economically, or practically feasi-ble or reasonable to screen all nanomaterials usinganimal models. Moreover, a model is only a model(and ‘essentially, all models are wrong, but some areuseful’, as the statistician George Box famously wrote)and we would be amiss to assume that results com-ing from animal studies are always relevant.44 How,then, do we move forward? Lai45 has proposed a



nanotoxicity testing strategy based on short-term invivo animal studies (i.e., shorter than a conventional90-day study) in conjunction with HTS and mecha-nistic in vitro studies, and comparing the data withthose of reference nanomaterials for the specific sub-class in question—an approach in concordance withthe ‘Toxicity Testing in the 21st Century’ strategy forchemicals.46

To this end, more advanced in vitro modelsare needed, in particular, assays that can be adaptedfor HTS. Huh et al.47 reported on a biomimeticmicrosystem that reconstitutes the critical functionalalveolar-capillary interface of the human lung. This‘lung mimic’ revealed that cyclic mechanical strainaccentuates toxic and inflammatory responses of thelung to silica nanoparticles. The authors concludedthat mechanically active ‘lung-on-a-chip’ microdevicesthat reconstitute tissue–tissue interfaces critical toorgan function may provide low-cost alternatives toanimal studies for toxicity testing. The ‘lung mimic’might also be amenable to HTS.47

Naturally, it is important to validate in vitroassays. Han et al.48 administered doses of titaniumdioxide nanoparticles of different sizes (3–100 nm)to a rat alveolar epithelial cell line in vitro andthe same nanoparticles by intratracheal instillationin rats in vivo to examine the correlation betweenin vitro and in vivo responses. The in vivo endpointwas the number of neutrophils in bronchoalveolarlavage fluid following exposure to nanoparticles.The correlations were based on toxicity rankingsof nanoparticles after adopting surface area as dosemetric and response per unit surface area as responsemetric. Slope analyses of the dose response curvesshowed that in vitro and in vivo responses werewell correlated. This study underlines the importanceof determining the appropriate dose metric innanotoxicity studies. Shaw et al.49 applied a high-content approach, i.e., a battery of test for multipleendpoints using multiple cell lines to test nanoparticlesand derived detailed structure–activity relationshipsfor the various nanomaterials tested. Importantly,nanoparticles with similar activity profiles in vitroexerted similar effects on monocyte numbers invivo.

HTS is a method for scientific experimentationthat comprises the screening of large chemical librariesfor activity against biological targets via the useof automation, miniaturized assays, and large-scaledata analysis.50 HTS techniques have emerged asa potentially useful tool to predict the possiblehazards of nanomaterials.51,52 However, the fact thatnanomaterials may interfere with commonly used invitro assays needs to be taken into account.4 Indeed,

novel nanotoxicity assays based on label-free detectionof cellular responses are needed.53

Mortimer et al.54 demonstrated that the so-called kinetic Vibrio fischeri luminescence inhibitiontest is a potentially useful tool for screening of thetoxicity of nanomaterials that can be adapted forHTS of ecotoxicological effects of nanomaterials.Jan et al.55 reported on high-content screening for‘fingerprinting’ of nanomaterials using cancer celllines of neuronal and hepatic origin. George et al.56

provided evidence that an in vitro-based HTSapproach combined with in silico data handling andzebrafish testing may constitute a paradigm for rapidscreening of nanomaterials.

In Silico (Modeling) ApproachesToxicology assessment of nanomaterials is expensiveand time-consuming. Therefore, in addition toexperimental approaches for hazard assessment, thereis a need for in silico methods in order to developstructure–activity relationships that correlate toxicityendpoints. These structure–activity relationships canbe quantitative or qualitative in nature and theycan predict toxicological effects directly from thephysicochemical properties of the entities, e.g.,nanoparticles of interest.57

There are currently only a handful of nano-QSAR modeling studies. In one recent study, theauthors developed a model to describe the cytotoxicityof 17 different types of metal oxide nanoparticles toEscherichia coli. The model was found to reliablypredict the toxicity of metal oxide nanoparticles.58

Using a more extensive dataset of 109 nanoparticlespossessing the same metal core but different organicmolecules on their surface, Fourches et al.59 foundthat the cellular uptake of nanoparticles can bepredicted by taking into account the chemicalstructure of the coating molecules. The chemical orstructural properties of nanomaterials are representedby mathematical objects called descriptors, manyof which can be calculated rather than measured.Examples of descriptors suitable for nanomaterialsinclude particle size, shape, and surface area,ionization potentials of metals, zeta potentials, andphysicochemical properties of molecules covalentlybound to nanoparticle surfaces.57 In a physiologicalenvironment, nanoparticles selectively absorb proteinsto form a nanoparticle ‘corona’, a process governedby molecular interactions between chemical groups onthe nanoparticle surfaces and the amino acid residuesof the proteins (see below). Recently, a biologicalsurface adsorption index (BSAI) was developed basedon the competitive adsorption onto nanoparticles of



a set of small-molecule probes that mimic aminoacid residues.60 By assuming that the adsorptionwas governed by five basic molecular forces, themeasured adsorption coefficients were used to developdescriptors, which, in turn, could be used to predict theadsorption of small molecules to other nanomaterials.In a subsequent study of a panel of 16 differentnanomaterials, the nanomaterials were classified intodistinct clusters according to their surface adsorptionproperties.61 It will be of interest to see whether theBSAI could be used to predict the formation of aprotein corona in a physiological setting.

Linking Toxicity to Material PropertiesTo systematically investigate toxic effects of thenanoparticles, it would be highly desirable tocorrelate their toxic effects with their physicochemicalproperties.5,62 However, unfortunately, this approachis not straightforward, as many physicochemicalproperties are strongly entangled and are difficult tocontrol independently.7 Nevertheless, in the followingsection, we discuss selected studies showing howmaterial properties may be linked to toxicity. Carefulassessment of material properties serves as thebridge between nanotoxicology and nanomedicine(Figure 1).

Size matters, in particular, for cellular uptakeof nanoparticles. Moreover, in general, the greaterthe intracellular dose of nanoparticles, the more thetoxic effects they generate. Chan and coworkersdemonstrated in a series of experiments that therecan be an optimal size for nanoparticle uptake.63,64

Similar claims have been made by several otherauthors, but the latter work stands out, as sizewas controlled in an exclusive way. Indeed, thenanoparticles were colloidally stable and thus werenot agglomerated, i.e., they did not have the effectivelylarger diameter of an agglomerate of nanoparticles,and surface chemistry was the same for all sizes. In thisway, the size dependence of nanoparticle uptake andcytotoxicity could be investigated. However, cellularuptake is not mandatory for cytotoxicity to occur:cobalt–chromium nanoparticles can damage humanfibroblasts across an intact cellular barrier withouthaving to cross the barrier. The outcome, whichincludes DNA damage without significant cell death,is different from that observed in cells subjected todirect exposure to nanoparticles.65

Also, shape can be important, though it isprobably overrated. The classical example is carbonnanotubes, which are thought to exert toxicity byvirtue of their ‘needle-like’ shape, i.e., an extremelyhigh aspect ratio enabling these materials to piercecell membranes. This may be relevant at least for

multiwalled carbon nanotubes (MWCNTs) with highwidth and, therefore, high rigidity.66 Interestingly, inthe latter study, thin and thick nanotubes similarlyaffected macrophages, while the deleterious effects ofcarbon nanotubes on human mesothelial cells werediameter-dependent. However, it is important to askwhen a fiber is a fiber, and when is it, effectively, aparticle? Murray et al.67 have recently shown thatit is important to factor in agglomeration whenassessing the in vivo toxicity of SWCNTs. Shape caninfluence the mode of cellular uptake. Consider a rod-shaped nanoparticle and a spherical nanoparticle ofthe same volume: the leading edge of the rod-shapednanoparticle has a much smaller cross-section and maytherefore penetrate cell membranes more effectively.However, in many studies, in particular, in theoreticalsimulations, aspect ratios are calculated for thenanoparticle cores, neglecting the surface coating andthe adsorbed protein corona (discussed below), whichreduces the effective aspect ratio and thus nullifiespotential shape effects. Furthermore, agglomerationin physiological media may rule out effects ofthe shape of individual nanoparticles.68 Schaeublinet al.69 investigated two gold nanoparticles withdifferent aspect ratios using a keratinocyte cell line andfound that gold nanospheres were nontoxic, whereasthe gold nanorods induced apoptosis. Notably, bothnanoparticles formed agglomerates in cell culturemedium, but the spherical particles had a large fractaldimension (i.e., tightly bound and densely packed)while the nanorod agglomerates had a small fractaldimension (i.e., loosely bound).

Surface charge strongly influences uptakeof nanoparticles. In general, positively chargednanoparticles are incorporated faster by cells thannegatively charged ones, which is typically explainedby the overall net negative charge of cellular surfaces.Although studies exist that demonstrate that insome cases positively charged nanoparticles interactwith cells differently when compared to negativelycharged ones, resulting in different mechanisms ofcytotoxicity, the higher toxicity of positively chargednanoparticles is generally correlated to their enhancedcellular uptake.70 To elucidate surface charge-dependent toxicity, nanoparticles with differentsurface charge, but with other physicochemicalparameters constant are required, which often isexperimentally complicated to achieve.71,72 However,as pointed out by Walkey and Chan,8 the proteincorona tends to give nanomaterials a zeta potentialof about −10 to −20 mV irrespective of thenanomaterial chemistry; this ‘normalization’ of zetapotentials is related to the fact that most plasmaproteins carry a net negative charge at physiological



pH. In other words, the ‘biological’ identity mayoverride the ‘synthetic’ identity.

Perhaps, the most important physicochemicalparameter that interferes with most others is colloidalstability. Obviously, agglomerated nanoparticles donot have the size of the individual nanoparticles butthe size of the agglomerate. This means, therefore,that unless nanoparticles are very well dispersed, anystatement about size- or shape-dependent uptake orcytotoxicity is not sound, as the cell would interactwith the agglomerates and not with the individualnanoparticles. Besides the fact that agglomerationmasks effects of other physicochemical parameters, itcan also directly affect interaction with cells. ‘Sticky’agglomerates of nanoparticles tend to precipitate ontop of cells and thus can cause cytotoxic effects.73

Many metal and metal oxide nanoparticlescan undergo dissolution within acidic compartments(lysosomes) in the cell which could drive toxicity.This phenomenon, sometimes referred to as aTrojan horse-type uptake mechanism because itcircumvents the plasma membrane barrier and allowstoxic ions to ‘sneak’ into cells, has been shown,for instance, for oxides of zinc, iron, manganese,and cobalt.74 Cho et al.75 evaluated the pulmonaryinflammogenicity of 15 different metal/metal oxidenanoparticles and showed that toxicity of thenanomaterials displayed a significant correlationwith one of two physicochemical parameters: zetapotential under acid conditions for low-solubilitynanoparticles and solubility (degree of dissolution) forhigh-solubility nanoparticles. The authors suggestedthat in the case of high-solubility nanoparticles,inflammogenicity depends on the ions that areproduced during dissolution of nanoparticles insidethe acidic phagolysosomes of the cells.

Catalytic effects at the nanoparticle surfaceplay an important role in the generation of reactiveoxygen species (ROS).76 Sayes et al.77 studied theeffects of titanium dioxide nanoparticles in cellculture and found that the extent to which nanoscaletitania affected cellular behavior was not dependenton surface area; what did correlate strongly tocytotoxicity, however, was the phase composition ofthe nanoscale titania insofar as anatase TiO2 was 100times more toxic than rutile TiO2. The most cytotoxicnanoparticle samples were also the most effective atgenerating ROS.

In synopsis, it may seem disappointing thatone cannot pinpoint how a certain physicochemi-cal parameter influences the toxicity of (all) nano-materials. This is due, in part, to the fact thatmany studies published to date are based on poorlydefined nanoparticles, in which many physicochemical

parameters are entangled. In fact, it is nontrivial tochange only one physicochemical parameter, withoutaffecting others. In addition, not all nanomaterials arecreated equal. Thus, a conclusive picture remains elu-sive. To be more conclusive, toxicity studies should beperformed with well-defined model nanoparticles, inwhich specific particle properties can be independentlyvaried. Advanced synthesis approaches are pointingin this direction, for instance by creating nanoparti-cles in which surface charge can be tuned (almost)independently from other particle properties.72 How-ever, most studies are performed with nanoparticlesof poor definition and/or agglomerated nanoparticlesystems. To remedy this situation, enhanced commu-nication between material scientists and toxicologistsis needed.

THE NANO-BIO-CORONA CONCEPT

To go one step further in terms of understandingthe interactions of engineered nanomaterials withliving systems, we need to consider the fact thatnanomaterials may adopt a ‘new’ identity throughthe adsorption of biomolecules, a phenomenon that,in turn, is linked to nanomaterial-intrinsic properties,e.g., size (surface curvature) and hydrophobicity.Indeed, as stated recently by Mahon et al.,78 ‘pristinenanoparticles in biological fluids act as a scaffoldfor biomolecules, which adsorb rapidly to thenanoparticle surface, conferring a new biologicalidentity’. Furthermore, the formation of a ‘bio-corona’ on nanoparticles is an inherently bilateralphenomenon, as proteins that adsorb to nanoparticlesurfaces may also alter their behavior as a resultof unfolding79 or fibrillation.80 The opsonization ofparticles with serum proteins is, however, not really a‘new’ phenomenon81 even though recent research hasprovided new insights into the parameters that controlthis process.

It is generally believed that immediately aftercontact with biological media, an initial corona isformed on nanoparticles by loosely bound, low-affinity proteins. Prolonged incubation in plasmaallows the formation of a denser, irreversibly attached‘hard’ corona with high-affinity proteins82,83 anda satellite ‘soft’ corona that undergoes intensiveexchange with the surrounding media.84 The ‘hard’corona is formed because of the direct interactionof proteins with the surface of the nanoparticles,whereas protein–protein interactions dominate theinteractions of the ‘soft’ corona with the ‘hard’corona.85 The time scale of the process probably isvery short. The ‘hard’ protein corona that is stronglyattached to the surface of nanoparticles is likely



the most relevant one for the in vivo fate of long-circulating nanoparticles.86 Furthermore, changes inthe ‘hard’ corona may occur when nanoparticlesare transferred to a new biological compartment,e.g., upon translocation of nanoparticles across theplasma membrane.86 Recent studies have shown thatthe corona of biomolecules attached to nanoparticlesis degraded by the protease cathepsin L within theendosomal compartment following endocytosis ofnanoparticles.87 This needs to be taken into accountwhen designing nanomaterials for intracellularapplications. Sund et al.88 noted that the binding ofcytoplasmic proteins depends on the surface chemistryof the nanoparticles. Hence, uncoated anatase andrutile phases of TiO2 nanoparticles adsorbed proteinssimilarly, whereas alumina and silicone-coated rutileforms of TiO2 bound only a few proteins.

Walkey and Chan8 recently provided acompilation of 26 published studies on the plasma-derived protein corona, and concluded that ‘theprotein corona is complex, that there is noone ‘universal’ plasma protein corona for allnanomaterials, and that the relative densities ofthe adsorbed proteins do not, in general, correlatewith their relative abundances in plasma’ (in otherwords, there is a degree of specificity). Instead,it is suggested that the protein corona dependson the ‘synthetic identity’ of each nanomaterial.8

Indeed, the adsorption of biomolecules is drivenby surface charge, hydrophobicity/hydrophilicity, andparticle size.83,89,90 Our recent studies show thatsuperparamagnetic iron oxide nanoparticles (SPIONs)with different surface coating display distinct plasmaprotein corona compositions (Vogt et al., manuscriptin preparation). Does the bio-corona cover thenanoparticle surfaces completely or will targetingligands remain accessible? Simberg et al.91 reportedthat both the dextran coat and the iron oxide core ofdextran-coated SPIONs remained accessible to specificprobes after incubation in plasma, suggesting that thenanoparticle surface could be available for recognitionby cells despite the bio-corona.

The majority of bio-corona studies have beenperformed with plasma proteins,92 which is certainlyrelevant in cases when nanoparticles are administeredinto the bloodstream. Nevertheless, it is important toalso consider other portals of entry of nanomaterialsinto the body, e.g., via inhalation or through theskin or via the gastrointestinal tract as the coronacomposition is likely to change as a function of theanatomical site and the specific biofluids encounteredat each of these sites. Kapralov et al.93 reported onthe in vivo formation of a lipid–protein corona onthe surface of SWCNTs following administration

by pharyngeal aspiration in mice. The bio-coronawas identical to lung surfactant and subsequent invitro studies demonstrated a role for the surfactantcorona of lipids + proteins in macrophage uptake ofcarbon nanotubes. Of note, plasma protein adsorptionto MWCNTs is influenced by prior adsorption ofpulmonary surfactant lipids.94

There are, overall, few studies on long-termeffects of nanomaterials and few, if any, of thesestudies have addressed the potential role of the‘intrinsic’ versus the ‘biological’ identity of thenanomaterials in question. Nevertheless, it may beuseful to consider whether the bio-corona plays a roleunder such conditions. In a recent study, Ruge et al.95

studied the impact of lung surfactant componentson macrophage clearance of nanoparticles andthey concluded that because of the interplay ofboth surfactant lipids and proteins, the alveolarmacrophage clearance of nanoparticles is essentiallythe same, regardless of different intrinsic surfaceproperties. The latter study thus suggests that the‘biological’ identity may override the ‘synthetic’identity of nanoparticles (at least in the short term).However, we postulate that in the long term, material-intrinsic properties (i.e., the ‘synthetic’ identity) willcome into play and the long-term fate of nanoparticleswill depend largely on whether the nanoparticlesundergo dissolution and/or are susceptible tobiodegradation, or whether they escape clearance bythe reticuloendothelial system and are subsequentlycleared from the body. Indeed, in the chronic phase,at which point the nanoparticles have left the systemiccirculation and have been uptaken by cells, thebody’s own responses to the nanoparticles maypredominate. For instance, inhalation of SWCNTsin mice will trigger a cascade of pathological eventsrealized through early inflammatory responses andthe induction of oxidative stress culminating in thedevelopment of multifocal granulomatous pneumoniaand interstitial fibrosis.96 Thus, while the carbonnanotubes represent the initial offending trigger, thelong-term effects (including potential carcinogeniceffects) are manifested through subsequent cellularresponses to this trigger; moreover, such organ andtissue responses may follow a common pattern ofhost defense reactions (oxidative stress, inflammation,etc.) toward foreign intrusion. In another recentstudy, Mahler et al.97 showed that chronic oralexposure to polystyrene nanoparticles can influenceiron uptake and iron transport in an in vivo chickenintestinal loop model. Importantly, chronic exposurecaused remodeling of the intestinal villi in exposedanimals, which increased the surface area availablefor iron absorption. In other words, the physiological



responses triggered by the nanoparticles (in thiscase, tissue remodeling) may determine long-termoutcomes, not the nanoparticles per se and not thebio-corona.

Controlling the Bio-CoronaFrom a nanomedicine point of view, it may bedesirable to avoid ‘nonspecific’ protein adsorption,i.e., bio-corona formation. This is commonly achievedby grafting PEG onto nanoparticles; this may preventnonspecific uptake of nanoparticles by cells of theimmune system, thereby prolonging their half-life incirculation.10 Modifying the surface of nanoparticleswith an antifouling polymer makes the proteinadsorption thermodynamically unfavorable, whilethe high-molecular-weight polymeric chains induceprotein ‘repulsion’ because of their conformationalflexibility and induced steric hindrance. PEGylationdoes not, however, prevent protein adsorptionaltogether.78 Increased PEG grafting density on thesurface of gold nanoparticles positively correlates witha decrease in total protein adsorption and reduceduptake in J774A.1 murine macrophages.98

On the other hand, one may considerto exploit the bio-corona phenomenon for tar-geting purposes. PEG-polyhexadecylcyanoacrylate(PEG-PHDCA) nanoparticles have been shown totranslocate into the brain after intravenous injectionin rats, whereas PHDCA nanoparticles do not. Kimet al.99 found that, after incubation with rat serum,apolipoprotein E (ApoE) adsorbed more onto PEG-PHDCA than onto PHDCA nanoparticles. Moreover,ApoE or ApoB-100 preadsorption onto PEG-PHDCAnanoparticles was required for efficient penetrationinto rat brain endothelial cells. These data suggestthe involvement of apolipoproteins in the transportof PEG-PHDCA nanoparticles across the blood–brainbarrier, which could be deployed for delivery of drugsinto the brain. Prapainop et al.100 attempted cell-specific uptake of nanomaterials by ‘reprogramming’of the behavior of the protein corona on nanomateri-als. Specifically, the surface of CdSe/ZnS QDs possess-ing an amino-functionalized, PEGylated hydrophilicsurface was decorated with the inflammatory metabo-lite, cholesterol 5,6-secosterol atheronal-B, and theresulting nanoparticles were shown to bind to andinduce the misfolding of apolipoprotein B leadingto uptake by RAW264.7 murine macrophages. Aspointed out by the authors, the ability to programthe bio-corona on nanoparticles with small moleculescould be developed to direct nanoparticles into celltypes that they may not have been able to reachbefore.100

Impact of Bio-Corona on Cellular FunctionsThe protein corona has been shown to play animportant role in modulating uptake and toxicityof SWCNTs.101,102 However, it remains to be firmlyestablished whether the biological identity of nanopar-ticles is the result of a specific protein(s) in thenanoparticle corona or a nonspecific effect relatedto the fact that proteins may alter the agglomerationbehavior of nanoparticles leading to a difference incellular uptake, which, in turn, has an impact on cyto-toxicity. Ehrenberg et al.103 reported that the capacityof polystyrene nanoparticle surfaces to adsorb pro-tein is indicative of their tendency to associate withcells. However, removal of the most abundant pro-teins from cell culture media did not affect the levelof cell association, and the authors concluded thatcellular association is not dependent on the identityof adsorbed proteins. Lartigue et al.104 studied theadsorption of proteins on biomedically relevant ironoxide nanoparticles by magneto-optical birefringence;the effect of plasma at different concentrations rang-ing from 1 to 100% on nanoparticle behavior wasassessed. It was noted that at low plasma concen-trations (representative of most in vitro conditions),the nanoparticles tended to form clusters triggered byproteins such as fibrinogen, whereas at high plasmaconcentrations (closer to the physiological situation)other proteins such as apolipoproteins tended to coatand subsequently to stabilize individual nanoparticles.This, in turn, affected in vitro uptake by macrophages.Lesniak et al.105 reported that silica nanoparticlesincubated with A549 cells in the absence of serumhave a stronger adhesion to the cell membrane andhigher internalization efficiency when compared withnanoparticles with a preformed surface corona.

In a key study of the bio-corona phenomenon,Deng et al.79 demonstrated that negatively chargedpoly(acrylic acid)-conjugated gold nanoparticles bindto and induce unfolding of fibrinogen, whichpromotes interaction with integrin receptors onmacrophage-like THP.1 cells, resulting in the releaseof inflammatory cytokines (Figure 2). In a follow-up study, the authors showed that fibrinogen boundwith high affinity to positively and negatively chargedgold nanoparticles.106 However, only the negativelycharged nanoparticles triggered cytokine release inTHP.1 cells, perhaps because of a different orientationof the protein on the different particles. Thus, whilecommon proteins may bind to different nanoparticles,the physiological response may not be the same.

The complement system constitutes an impor-tant barrier to infection or other foreign intru-sion. Nanoparticles may also activate complement;



00 2 4 6 8 10

50

100

150Size (nm)5 10 20

43557295

PAA-GNP (μg)

Unb

ound

fibr

inog

en (

%)

Mol

ecul

ar w

eigh

t (kD

a)

Unb

ound

fibr

inog

en (

%)

00 500Surface area (mm2)

1,000 1,500

50

100

150

D domain E domain D domain

~45 nm

C terminus ofα chain

C terminus ofα′ chain

Misfolded fibrinogen

NP NP THP.1 TNF-α

Bio-coronaformation

Mac-1

(a)

(c)

(d)

(b)

FIGURE 2 | Protein corona: role in proinflammatory responses.Fibrinogen is the major human plasma protein bound by poly(acrylicacid)-coated gold nanoparticles (PAA–GNP). (a) SDS–PAGE of humanplasma proteins bound to PAA–GNP with diameters of 5, 10, and20 nm. Three major protein bands were observed at 65, 55, and 45 kDa.(b) Unbound fibrinogen following pull-down with PAA–GNP withdiameters of 5 nm (blue) or 20 nm (red). Purified fibrinogen (0.6 mg)was incubated with increasing amounts of PAA–GNP. Inset: unboundfibrinogen is plotted against total surface area for the twonanoparticles. (c) Crystal structure of fibrinogen. The protein was drawnusing Swiss-PdbViewer and coordinates for PDB entry 3GHG. Commondomains are shown. Inset: the C-terminus of the g chain (purple) thatinteracts with the Mac-1 receptor. (Reprinted with permission from Ref79. Copyright 2011 Macmillan Publishers Ltd.) (d) The schematicdiagram illustrates how unfolding of fibrinogen on the surface ofPAA–GNP leads to interaction with the integrin receptor, Mac-1, on thesurface of THP.1 monocytes, which in turn increases NF-κB signalingleading to secretion of tumor necrosis factor-α. It is pertinent to notethat fibrinogen, which has a length of 45 nm and a diameter of5 nm, is much larger than the 5-nm PAA–GNP. Deng et al.79 showedthat the maximum protein binding was 2 μg for the 5-nm PAA–GNP,which represents one to two nanoparticles per fibrinogen molecule.

this may be viewed as a special case of bio-corona formation and one that is of particularrelevance in nanomedicine. Nanomaterial interac-tion with the complement system is complex andregulated by interrelated physicochemical factorssuch as size, morphology, and surface properties.107

Hamad et al.108 investigated polystyrene nanoparti-cles with surface-projected polyethylene oxide chains

in ‘mushroom-brush’ and ‘brush’ configurations andfound that distinct polymer architectures mediateswitching of complement activation pathways. Aspointed out by the authors, these studies suggest arational basis for the design of targetable nanosystemsfor nanomedicine applications.

THE IN VIVO FATEOF NANOPARTICLES

In addition to understanding the synthetic and bio-logical identities of nanomaterials, it is important totake into consideration the context-dependent behav-ior of a nanomaterial. In other words, to consider howthe biological identity of a nanomaterial may changedepending on the specific biological compartment (inthe body or within a cell). Indeed, as noted previ-ously, ‘one of the key features of nanoscale materials,and the one that may suggest novel and unantici-pated health risks, may very well be the propensity ofsuch materials to cross biological barriers in a man-ner not predicted from studies of larger particles ofthe same chemical composition’.9 Here, we discusssome studies illustrating how nanoparticles may crossbiological barriers, and how material-intrinsic proper-ties may dictate such interactions. We will also touchon factors that regulate nanoparticle pharmacokinet-ics. Understanding the in vivo fate and behavior ofnanomaterials is another area of common interest innanotoxicology and nanomedicine.

Crossing Biological BarriersNanoparticles can cross biological barriers and enterand distribute within cells by different pathwaysand for this reason they are considered a primaryvehicle for targeted therapies. In the body, we findcellular barriers that include the cell membrane,and endosomal–lysosomal and nuclear membranes,and physiological barriers that prevent extravasationof foreign substances from the blood such as theblood–brain barrier. The skin is the main barrier thatprotects our body from the external environment.Understanding the barriers imposed by a biologicalsystem is critical to the design of nanomaterials forbiomedical applications (see Kievit and Zhang109

for an excellent review). It is also important toconsider whether one should attempt to breachbiological barriers between bodily compartments withnanoparticles as this may trigger unexpected toxicitiesand disease processes.110

Yamashita et al.111 showed that silica andtitanium dioxide nanoparticles with diameters of70 and 35 nm, respectively, can cross the placenta



and cause pregnancy complications when injectedintravenously into pregnant mice. Larger (300 and1000 nm) silica particles did not induce suchcomplications. It remains unclear if the fetotoxicitywas caused by direct exposure to the nanoparticlesor by the damage to the placenta. Nonetheless, thedetrimental effects were abolished when the surfacesof the silica nanoparticles were modified with carboxyland amine groups. Hence, size and surface chargeboth may impact on the propensity of nanoparticlesto cause damage to the unborn fetus. Similarly, Schlehet al.112 demonstrated that size and surface chargeof gold nanoparticles determine absorption acrossintestinal barriers and accumulation in secondarytarget organs after oral administration in a ratmodel. Choi et al.113 determined that nanoparticleswith hydrodynamic diameter less than 34 nm withnoncationic surface charge translocate rapidly fromthe lungs to regional lymph nodes in rats followingintratracheal instillation. Furthermore, nanoparticleswith a hydrodynamic diameter less than 6 nm werefound to traffic rapidly from the lungs to lymph nodesand the bloodstream, ultimately being cleared fromthe body through the kidneys. Moreover, as discussedin further detail below, nanoparticle behavior wasfound to depend strongly on surface coating. Thesefindings suggest strategies for the rational design ofnanoparticles for drug delivery via lung inhalation.Kannan et al.114 recently devised a prodrug approachto treat cerebral palsy, a developmental disorderresulting from an insult to a growing fetal or infantbrain. In this preclinical study, N-acetyl-cysteine(NAC) was linked to polyamidoamine dendrimersthat enabled NAC to cross the blood–brain barrier andreach microglia and astrocytes. This nanoformulation(D-NAC) was administered within 6 h of birth withimprovement in motor performance and ameliorationof inflammation in newborn animals.

However, nanoparticles may disrupt or evenremodel biological barriers. Mahler et al.97 reportedthat chickens acutely exposed to carboxylatedpolystyrene nanoparticles had a lower iron absorptionthan unexposed or chronically exposed birds.As mentioned earlier, Chronic exposure causedremodeling of the intestinal villi, which increased thesurface area available for iron absorption, and thisincrease in intestinal surface area compensated forthe lowered iron transport caused by nanoparticleexposure.

Biodistribution and Tumor TargetingPharmacokinetics is concerned with quantifying theadsorption, distribution, metabolism, and elimination(ADME) of chemicals and drugs in the body; the

aim is to relate drug dose or chemical exposure tobiological effects.115 Evaluation of ADME propertiesof nanomaterials is crucial for the medical imple-mentation of these materials. To this end, in vivomodel systems are certainly needed. Riviere115 hasprovided a concise overview of studies on the invivo disposition of fullerenes, carbon nanotubes, andQDs after parenteral administration. Functionalized,water-soluble SWCNT and MWCNT may negotiatethe glomerular filtration barrier and undergo renalexcretion without extensive accumulation in the body,in a manner dependent upon the degree of individu-alization of the nanotubes.116,117 Notably, pristineSWCNTs may undergo enzymatic biodegradation invitro118and in vivo119; biodegradation by neutrophilsis promoted when the carbon nanotubes are coatedwith a corona of immunoglobulins, which leads toenhanced cellular uptake via Fc receptors expressed onneutrophils.118

How about the disposition of biomedically rel-evant nanomaterials? Schädlich et al.120 investigatedthe influence of the size of biodegradable PEG-PLAnanoparticles both in vivo and ex vivo and foundthat nanoparticles of 111 and 141 nm accumulatedin human xenograft tumor tissue while slightly biggernanoparticles (166 nm) were rapidly eliminated by theliver. These studies demonstrate how different biodis-tribution may occur because of small nanoparticlesize differences. The importance of further miniatur-izing nanocarrier size to optimize tumor accumula-tion and penetration was recently shown121 (and seeabove, section on Medical Imaging, for additionalexamples).

The EPR effect and/or targeting approaches mayenable nanoscale carriers to reach a tumor, but thisdoes not necessarily mean that the nanoparticles willalso penetrate into the tumor and deliver their payloadof anticancer drugs. Cabral et al.122 compared theaccumulation and effectiveness of different sizes(30, 50, 70, and 100 nm) of long-circulating, drug-loaded polymeric micelles in highly versus poorlypermeable tumors in a preclinical model, and foundthat only the 30-nm micelles could penetrate poorlypermeable, hypovascular pancreatic tumors to achievean antitumor effect. Interestingly, the penetration andefficacy of the larger nanoparticles could be enhancedby pharmacologically increasing the permeability ofthe tumors.

Choi et al.113 followed the fate of intratra-cheally instilled NIR fluorescent nanoparticles thatwere varied systematically in size, surface modi-fication, and core composition and showed thatnanoparticle behavior depends strongly on the sur-face coating, which affects protein adsorption in body



fluids; hence, for charged nanoparticles, nonspecificadsorption of endogenous proteins, mostly albumins,resulted in a large increase in hydrodynamic size ofthe nanoparticles, and this affected the biodistributionof the nanoparticles following their uptake in this ratmodel.

Finally, von Maltzahn et al.123 have provideda fascinating example of ‘communicating’ nanopar-ticle systems based on nanotechnological mimicry ofthe recruitment of immune cells to an inflammatorylesion to improve in vivo tumor-targeting efficiency.Hence, the authors designed multifunctional systemswhereby the coagulation cascade in tumors is activatedby photothermal heating of gold nanorods in order to‘broadcast’ tumor location to clot-targeted nanopar-ticles, i.e., doxorubicin-loaded liposomes coated withFactor XIII, a component of the coagulation cas-cade. This approach, which thus takes advantage ofthe endogenous coagulation cascade, yielded over 40times higher doses of doxorubicin in tumors whenthe drug is loaded with Factor XIII-covered liposomeswhen compared to plain liposomes.123

A CASE OF STOLEN IDENTITY

Nature may inspire the design of syntheticnanoparticles. Bertram et al.124 developed ‘artificialplatelets’ based on Arg-Gly-Asp (RGD)-functionalizednanoparticles, which halved the bleeding time afterintravenous administration in a rat model ofmajor trauma. The synthetic platelets consistingof poly(lactic-co-glycolic acid)–poly-l-lysine blockcopolymer cores conjugated with PEG chains termi-nated with RGD functionalities were cleared within24 h, and no toxicity was seen up to 7 days postin-fusion. Hu et al.125 presented a novel approach inparticle functionalization by coating biodegradablepolymeric nanoparticles with a corona of naturalmembranes derived from red blood cells, includingboth membrane lipids and associated membrane pro-teins, in order to achieve ‘stealthy’, long-circulatingnanoparticles for drug delivery. The latter study rep-resents an example of ‘borrowed identity’ of nanoma-terials. Indeed, biomimetic design could be exploitedfor drug delivery.126 Interestingly, the immune systemutilizes its very own nanoparticles (exosomes) to trans-mit information between cells. Exosomes may containboth mRNA and microRNA, and the transferred exo-somal mRNA has been shown to be translated inthe recipient cell.127 More recent studies confirmedthat the transmitted microRNA is also functional.128

Hence, exosomes serve as a template for the delivery ofshort RNAs for modulation of gene expression using

Physiological responses

Synthetic identity

NP NP

Biological identity

- tissue targeting

- cellular uptake

- cytotoxicity- cytokine secretion- immunogenicity- degradation, excretion- etc

Bio-corona

Biomoleculeseg. protein, lipids

Material-intrinsicproperties

- size, shape, aspect-ratio- surface charge- colloidal stability- stability / disssolution- catalytic properties- etc

Spatial determinants:- portal of entry- body fluid / organ- subcellular compartmenttemporal determinants:- acute or long-term effect

Context-dependentproperties

FIGURE 3 | Synthetic and biological identities of nanomaterials.Schematic view of the ‘synthetic’ identity of nanomaterials that isdetermined by material-intrinsic properties and the ‘biological’ identitythat is manifested in a living system and can be viewed as the sum ofthe context-dependent properties of the nanomaterial. As discussed inthis review, the biological identity is shaped, in part, by the adsorptionof biomolecules (proteins and lipids) that form a ‘corona’ on the surfaceof nanoparticles; the composition of the bio-corona depends on theparticular biofluid (e.g., blood, lung fluid, and gastrointestinal fluid) andmay exhibit dynamic changes as the nanoparticle crosses from onebiological compartment to another. The physiological responses tonanomaterials are dictated by the synthetic and biological identities; apartial list of possible biological/toxicological outcomes is shown in thisfigure.

nanoscale delivery vehicles that are, by definition,biocompatible.

Stark has pointed out that nanoparticles dif-fer from molecules in several respects; nevertheless, heconcludes that ‘from a functional point of view, chem-ically well-defined nanoparticles are an extension ofthe classical concept of the molecule’, as they combinethe properties of solids with mobility (a property ofmolecules).74 Indeed, certain nanoparticles appear tobridge the gap between molecules and particles. Den-drimers are polymeric nanoparticles with perfectlydefined structure and molecular weight.129 Hayderet al.130 reported recently that azabisphosphonate



(ABP)-capped dendrimers selectively target monocytesand direct them toward anti-inflammatory activation.The dendrimers also exhibited antiosteoclastic activ-ity, thus preventing bone erosion. Intravenous injec-tions of ABP-capped dendrimers inhibited the devel-opment of inflammatory arthritis in two animalmodels. This exciting study suggests that dendrimerscould function as novel therapeutics for rheumatoidarthritis. Moreover, dendrimers conjugated to glu-cosamine and glucosamine 6-sulfate were shown topossess immunomodulatory and antiangiogenic prop-erties, respectively, and when administered together,the nanoparticles increased the long-term success ofglaucoma surgery in an animal model by prevent-ing scar tissue formation.131 Thus, in some cases, thesynthetic and biological identities of a nanomaterialappear to blend into one: dendrimers may functionas drugs per se by virtue of their unique physico-chemical properties, i.e., size and multivalent surfacefunctionalities, which allow these nanoparticles todirectly engage biological receptors and modulate cellfunction.

CONCLUSIONS AND PERSPECTIVESThe Stone Age did not end because they ran outof stones. New technologies inevitably replace oldones. We are now at the dawn of a nanotech-nological revolution with far-reaching implicationsfor society and it is crucial that we ensure thesafety of these novel materials while not imped-ing their implementation in important areas suchas in medicine. In this review, we have attemptedto highlight the role of physicochemical propertiesof engineered nanomaterials and their impact onnanomaterial behavior in biological systems. Impor-tantly, a growing body of evidence indicates that theadsorption of biomolecules onto nanoparticle surfacesmay bestow a new ‘biological identity’ onto thesematerials.8,78 This has considerable ramifications notonly for nanotoxicological assessment of syntheticnanoscale materials but also for their implementa-tion in medicine. Of note, the bio-corona of serumproteins should not necessarily be viewed as an unde-sirable biological phenomenon; the bio-corona canbe controlled100 and may even be exploited for drugdelivery.132 In addition, we would be amiss to ignorefundamental physicochemical properties of nanoma-terials: the cells may also ‘see’ what is beneath thecorona.

Careful assessment of material-intrinsic proper-ties and how these properties are linked to physio-logical responses is thus essential both in nanotoxi-cology and in nanomedicine. Notably, the very sameproperty may be highly desirable for certain clini-cal applications (for instance, the delivery of smallparticles to exploit the EPR effect) but could alsoyield unwanted hazardous effects. Taking into con-sideration not only the synthetic identity but alsothe biological identity of nanomaterials, and howthese identities may evolve over time and as afunction of different biological compartments in thebody or at the subcellular level may enable a betterunderstanding of nanomaterial-induced physiologicalresponses (Figure 3). More studies are needed on thelong-term effects of nanomaterials and on the rel-ative importance of surface-adsorbed biomolecules(the bio-corona) versus material-intrinsic propertiesof nanomaterials under such conditions; understand-ing how common physiological reactions (oxidativestress, inflammation, etc.) are triggered is also ofimportance in order to mitigate adverse effects follow-ing nanomaterial exposure. Furthermore, more refinedtechniques to study the corona of adsorbed proteins,lipids, and other biomolecules on nanomaterial sur-faces are warranted along with a greater emphasis onthe potential impact of individual components of thecorona on physiological responses. Bioinformatics-based approaches may prove helpful when decipheringthe bio-corona data. New approaches including HTSfor the rapid screening and ranking of the hazardpotential of vast numbers of engineered nanomateri-als and mathematical modeling of structure–activityrelationships of nanomaterials may also facilitate thedevelopment of safe and useful nanomaterials for invivo imaging, drug delivery, and other clinical appli-cations.

ACKNOWLEDGMENTS

The authors are supported, in part, through grantsfrom the Swedish Research Council for Environ-ment, Agricultural Sciences and Spatial Planning(FORMAS), the Swedish Cancer and Allergy Foun-dation, the Seventh Framework Programme of theEuropean Commission (FP7-MARINA-263215 andFP7-NANOGNOSTICS-242264), and BMFM Ger-many (Umsicht). BF holds a Senior Investigator Awardfrom the Swedish Research Council.



REFERENCES1. Shvedova AA, Kagan VE, Fadeel B. Close encoun-

ters of the small kind: adverse effects of man-madematerials interfacing with the nano-cosmos of bio-logical systems. Annu Rev Pharmacol Toxicol 2010,50:63–88.

2. Harford CG, Hamlin A, Parker E. Electron microscopyof HeLa cells after the ingestion of colloidal gold.J Biophys Biochem Cytol 1957, 3:749–756.

3. De Groodt N, Lagasse A, Sebruyns M. Alterations inthe ultrastructure of the blood-air barrier in the mouselung after inhalation of colloidal gold particles. Nature1958, 181:1418–1419.

4. Lewinski N, Colvin V, Drezek R. Cytotoxicity ofnanoparticles. Small 2008, 4:26–49.

5. Fadeel B, Garcia-Bennett AE. Better safe than sorry:understanding the toxicological properties of inor-ganic nanoparticles manufactured for biomedicalapplications. Adv Drug Deliv Rev 2010, 62:362–374.

6. Shvedova AA, Pietroiusti A, Fadeel B, Kagan VE.Mechanisms of carbon nanotube-induced toxicity:focus on oxidative stress. Toxicol Appl Pharmacol2012, 261:121–133.

7. Rivera-Gil P, De Aberasturi DJ, Wulf V, Pelaz B, DelPino P, Zhao Y, De La Fuente J, De LarramendiIR, Rojo T, Liang XJ, et al. The challenge to relatephysico-chemical properties of colloidal nanoparticlesto their cytotoxicity. Acc Chem Res (Epub ahead ofprint; July 11, 2012).

8. Walkey CD, Chan WC. Understanding and control-ling the interaction of nanomaterials with proteins ina physiological environment. Chem Soc Rev 2012,41:2780–2799.

9. Nyström AM, Fadeel B. Safety assessment of nano-materials: implications for nanomedicine. J ControlRelease 2012, 161:403–408

10. Riehemann K, Schneider SW, Luger TA, GodinB, Ferrari M, Fuchs H. Nanomedicine-challengeand perspectives. Angew Chem Int Ed Engl 2009,48:872–897.

11. Ali Z, Abbasi AZ, Zhang F, Arosio P, Lascialfari A,Casula MF, Wenk A, Kreyling W, Plapper R, Seidel M,et al. Multifunctional nanoparticles for dual imaging.Anal Chem 2011, 83:2877–2882.

12. Yang K, Hu L, Ma X, Ye S, Cheng L, Shi X, Li C,Li Y, Liu Z. Multimodal imaging guided photother-mal therapy using functionalized graphene nanosheetsanchored with magnetic nanoparticles. Adv Mater2012, 24:1868–1872.

13. Altinoglu EI, Russin TJ, Kaiser JM, Barth BM, EklundPC, Kester M, Adair JH. Near-infrared emittingfluorophore-doped calcium phosphate nanoparticlesfor in vivo imaging of human breast cancer. ACSNano 2008, 2:2075–2084.

14. Wong C, Stylianopoulos T, Cui J, Martin J, ChauhanVP, Jiang W, Popovic Z, Jain RK, Bawendi MG, Fuku-mura D. Multistage nanoparticle delivery system fordeep penetration into tumor tissue. Proc Natl Acad SciU S A 2011, 108:2426–2431.

15. Chauhan VP, Stylianopoulos T, Martin JD, PopovicZ, Chen O, Kamoun WS, Bawendi MG, Fuku-mura D, Jain RK. Normalization of tumour bloodvessels improves the delivery of nanomedicines ina size-dependent manner. Nat Nanotechnol 2012,7:383–388.

16. Kim S, Lim YT, Soltesz EG, Grand AMD, Lee J,Nakayama A, Parker JA, Mihaljevic T, LaurenceRG, Dor DM, et al. Near-infrared fluorescent typeII quantum dots for sentinel lymph node mapping.Nat Biotechnol 2004, 22:93–97.

17. Ye L, Yong KT, Liu L, Roy I, Hu R, Zhu J, Cai H, LawWC, Liu J, Wang K, et al. A pilot study in non-humanprimates shows no adverse response to intravenousinjection of quantum dots. Nat Nanotechnol 2012,7:453–458.

18. de la Zerda A, Zavaleta C, Keren S, Vaithilingam S,Bodapati S, Liu Z, Levi J, Smith BR, Ma TJ, OralkanO, et al. Carbon nanotubes as photoacoustic molec-ular imaging agents in living mice. Nat Nanotechnol2008, 3:557–562.

19. Tong L, Liu Y, Dolash BD, Jung Y, Slipchenko MN,Bergstrom DE, Cheng JX. Label-free imaging of semi-conducting and metallic carbon nanotubes in cellsand mice using transient absorption microscopy. NatNanotechnol 2011, 7:56–61.

20. Lehner R, Wang X, Wolf M, Hunziker P. Design-ing switchable nanosystems for medical application.J Control Release 2012, 161:307–316.

21. Verma A, Uzun O, Hu YH, Hu Y, Han HS, WatsonN, Chen SL, Irvine DJ, Stellacci F. Surface-structure-regulated cell-membrane penetration by monolayer-protected nanoparticles. Nat Mater 2008, 7:588–595.

22. Leduc C, Jung JM, Carney RR, Stellacci F, Lounis B.Direct investigation of intracellular presence of goldnanoparticles via photothermal heterodyne imaging.ACS Nano 2011, 5:2587–2592.

23. Ashley CE, Carnes EC, Phillips GK, Padilla D, Dur-fee PN, Brown PA, Hanna TN, Liu J, Phillips B,Carter MB, et al. The targeted delivery of multi-component cargos to cancer cells by nanoporousparticle-supported lipid bilayers. Nat Mater 2011,10:389–397.

24. Davis ME, Zuckerman JE, Choi CH, Seligson D,Tolcher A, Alabi CA, Yen Y, Heidel JD, Ribas A. Evi-dence of RNAi in humans from systemically adminis-tered siRNA via targeted nanoparticles. Nature 2010,464:1067–1070.

25. Hrkach J, Von Hoff D, Ali MM, Andrianova E, AuerJ, Campbell T, De Witt D, Figa M, Figueiredo M,



Horhota A, et al. Preclinical development and clinicaltranslation of a PSMA-targeted docetaxel nanoparti-cle with a differentiated pharmacological profile. SciTransl Med 2012, 4:128ra39.

26. Colombo M, Carregal-Romero S, Casula MF,Gutiérrez L, Morales MP, Böhm IB, HeverhagenJT, Prosperi D, Parak WJ. Biological applicationsof magnetic nanoparticles. Chem Soc Rev 2012,41:4306–4334.

27. Child HW, Del Pino PA, De La Fuente JM, Hurst-house AS, Stirling D, Mullen M, McPhee GM, NixonC, Jayawarna V, Berry CC. Working together: thecombined application of a magnetic field and pene-tratin for the delivery of magnetic nanoparticles tocells in 3D. ACS Nano 2011, 5:7910–7919.

28. Alexiou C, Arnold W, Klein RJ, Parak FG, Hulin P,Bergemann C, Erhardt W, Wagenpfeil S, Lübbe AS.Locoregional cancer treatment with magnetic drugtargeting. Cancer Res 2000, 60:6641–6648.

29. Ray PC. Size and shape dependent second order non-linear optical properties of nanomaterials and theirapplication in biological and chemical sensing. ChemRev 2010, 110:5332–5365.

30. Sperling RA, Rivera Gil P, Zhang F, Zanella M,Parak WJ. Biological applications of gold nanopar-ticles. Chem Soc Rev 2008, 37:1896–1908.

31. Hirsch LR, Stafford RJ, Bankson JA, Sershen SR,Rivera B, Price RE, Hazle JD, Halas NJ, West JL.Nanoshell-mediated near-infrared thermal therapy oftumors under magnetic resonance guidance. Proc NatlAcad Sci U S A 2003, 100:13549–13554.

32. Carregal Romero S, Ochs M, Rivera-Gil P, Gana C,Pavlov AM, Sukhorukov GB, Parak WJ. NIR-lighttriggered delivery of macromolecules into the cytosol.J Control Release 2012, 159:120–127.

33. Fadeel B, Kasemo B, Malmsten M, Strømme M.Nanomedicine: reshaping clinical practice. J InternMed 2010, 267:2–8.

34. Mahmoudi M, Hosseinkhani H, Hosseinkhani M,Boutry S, Simchi A, Journeay WS, Subramani K, Lau-rent S. Magnetic resonance imaging tracking of stemcells in vivo using iron oxide nanoparticles as a toolfor the advancement of clinical regenerative medicine.Chem Rev 2011, 111:253–280.

35. Pickard MR, Barraud P, Chari DM. The transfectionof multipotent neural precursor/stem cell transplantpopulations with magnetic nanoparticles. Biomaterials2011, 32:2274–2284.

36. Shah RN, Shah NA, Del Rosario Lim MM, HsiehC, Nuber G, Stupp SI. Supramolecular design of self-assembling nanofibers for cartilage regeneration. ProcNatl Acad Sci U S A 2010, 107:3293–3298.

37. Webber MJ, Tongers J, Newcomb CJ, Marquardt KT,Bauersachs J, Losordo DW, Stupp SI. Supramolecu-lar nanostructures that mimic VEGF as a strategy for

ischemic tissue repair. Proc Natl Acad Sci U S A 2011,108:13438–13443.

38. Jungebluth P, Alici E, Baiguera S, Le Blanc K,Blomberg P, Bozoky B, Crowley C, Einarsson O, Grin-nemo KH, Gudbjartsson T, et al. Tracheobronchialtransplantation with a stem-cell-seeded bioartificialnanocomposite: a proof-of-concept study. Lancet2011, 378:1997–2004.

39. Badylak SF, Weiss DJ, Caplan A, Macchiarini P. Engi-neered whole organs and complex tissues. Lancet2012, 379:943–952.

40. Duncan R, Gaspar R. Nanomedicine(s) under themicroscope. Mol Pharm 2011, 8:2101–2141.

41. Oberdörster G, Oberdörster E, Oberdörster J. Nan-otoxicology: an emerging discipline evolving fromstudies of ultrafine particles. Environ Health Perspect2005, 113:823–839.

42. Krug HF, Wick P. Nanotoxicology: an interdisci-plinary challenge. Angew Chem Int Ed Engl 2011,50:1260–1278.

43. Oberdörster G. Safety assessment for nanotechnol-ogy and nanomedicine: concepts of nanotoxicology.J Intern Med 2010, 267:89–105.

44. Hartung T, Sabbioni E. Alternative in vitro assaysin nanomaterial toxicology. Wiley Interdiscip RevNanomed Nanobiotechnol 2011, 3:545–573.

45. Lai DY. Toward toxicity testing of nanomaterialsin the 21st century: a paradigm for moving for-ward. Wiley Interdiscip Rev Nanomed Nanobiotech-nol 2012, 4:1–15.

46. Collins FS, Gray GM, Bucher JR. Toxicology. Trans-forming environmental health protection. Science2008, 319:906–907.

47. Huh D, Matthews BD, Mammoto A, Montoya-Zavala M, Hsin HY, Ingber DE. Reconstitutingorgan-level lung functions on a chip. Science 2010,328:1662–1668.

48. Han X, Corson N, Wade-Mercer P, Gelein R, JiangJ, Sahu M, Biswas P, Finkelstein JN, Elder A,Oberdörster G. Assessing the relevance of in vitrostudies in nanotoxicology by examining correlationsbetween in vitro and in vivo data. Toxicology 2012,297:1–9.

49. Shaw SY, Westly EC, Pittet MJ, Subramanian A,Schreiber SL, Weissleder R. Perturbational profilingof nanomaterial biologic activity. Proc Natl Acad SciU S A 2008, 105:7387–7392.

50. Mayr LM, Bojanic D. Novel trends in high-throughputscreening. Curr Opin Pharmacol 2009, 9:580–588.

51. Feliu N, Fadeel B. Nanotoxicology: no small matter.Nanoscale 2010, 2:2514–2520.

52. Damoiseaux R, George S, Li M, Pokhrel S, Ji Z,France B, Xia T, Suarez E, Rallo R, Madler L, et al.No time to lose—high throughput screening to assessnanomaterial safety. Nanoscale 2011, 3:1345–1360.



53. Seiffert JM, Baradez MO, Nischwitz V, LekishviliT, Goenaga-Infante H, Marshall D. Dynamic moni-toring of metal oxide nanoparticle toxicity by labelfree impedance sensing. Chem Res Toxicol 2012,25:140–152.

54. Mortimer M, Kasemets K, Heinlaan M, Kurvet I,Kahru A. High throughput kinetic Vibrio fischeribioluminescence inhibition assay for study of toxiceffects of nanoparticles. Toxicol In Vitro 2008,22:1412–1417.

55. Jan E, Byrne SJ, Cuddihy M, Davies AM, Volkov Y,Gun’ko YK, Kotov NA. High-content screening asa universal tool for fingerprinting of cytotoxicity ofnanoparticles. ACS Nano 2008, 2:928–938.

56. George S, Xia T, Rallo R, Zhao Y, Ji Z, Lin S, WangX, Zhang H, France B, Schoenfeld D, et al. Use ofa high-throughput screening approach coupled within vivo zebrafish embryo screening to develop haz-ard ranking for engineered nanomaterials. ACS Nano2011, 5:1805–1817.

57. Winkler DA, Mombelli E, Pietriusti A, Tran L,Worth A, Fadeel B, McCall MJ. Applying quan-titative structure-activity relationship approaches tonanotoxicology: current status and future potential.Toxicology. (Epub ahead of print; Nov 16, 2012).

58. Puzyn T, Rasulev B, Gajewicz A, Hu X, Dasari TP,Michalkova A, Hwang HM, Toropov A, Leszczyn-ska D, Leszczynski J. Using nano-QSAR to predictthe cytotoxicity of metal oxide nanoparticles. NatNanotechnol 2011, 6:175–178.

59. Fourches D, Pu D, Tassa C, Weissleder R, Shaw SY,Mumper RJ, Tropsha A. Quantitative nanostructure-activity relationship modeling. ACS Nano 2010,4:5703–5712.

60. Xia XR, Monteiro-Riviere NA, Riviere JE. An indexfor characterization of nanomaterials in biological sys-tems. Nat Nanotechnol 2010, 5:671–675.

61. Xia XR, Monteiro-Riviere NA, Mathur S, Song X,Xiao L, Oldenberg SJ, Fadeel B, Riviere JE. Map-ping the surface adsorption forces of nanomaterials inbiological systems. ACS Nano 2011, 5:9074–9081.

62. Rivera-Gil P, Oberdörster G, Elder A, Puntes V, ParakWJ. Correlating physico-chemical with toxicologicalproperties of nanoparticles: the present and the future.ACS Nano 2010, 4:5527–5531.

63. Chithrani BD, Ghazan AA, Chan CW. Determiningthe size and the shape dependence of gold nanopar-ticle uptake into mammalian cells. Nano Lett 2006,6:662–668.

64. Chithrani BD, Chan WCW. Elucidating the mecha-nism of cellular uptake and removal of protein-coatedgold nanoparticles of different sizes and shapes. NanoLett 2007, 7:1542–1550.

65. Bhabra G, Sood A, Fisher B, Cartwright L, Saun-ders M, Evans WH, Surprenant A, Lopez-Castejon G,Mann S, Davis SA, et al. Nanoparticles can cause DNA

damage across a cellular barrier. Nat Nanotechnol2009, 4:876–883.

66. Nagai H, Okazaki Y, Chew SH, Misawa N, YamashitaY, Akatsuka S, Ishihara T, Yamashita K, YoshikawaY, Yasui H, et al. Diameter and rigidity of multiwalledcarbon nanotubes are critical factors in mesothelialinjury and carcinogenesis. Proc Natl Acad Sci U S A2011, 108:E1330–E1338.

67. Murray AR, Kisin ER, Tkach AV, Yanamala N, Mer-cer RR, Young SH, Fadeel B, Kagan VE, ShvedovaAA. Factoring-in agglomeration of carbon nanotubesand nanofibers for better prediction of their toxicityversus asbestos. Part Fibre Toxicol 2012, 9:10.

68. Liu S, Wei L, Hao L, Fang N, Chang MW, Xu R, YangY, Chen Y. Sharper and faster ‘‘nano darts’’ kill morebacteria: a study of antibacterial activity of individu-ally dispersed pristine single-walled carbon nanotube.ACS Nano 2009, 3:3891–3902.

69. Schaeublin NM, Braydich-Stolle LK, Maurer EI, ParkK, MacCuspie RI, Afrooz AR, Vaia RA, Saleh NB,Hussain SM. Does shape matter? Bioeffects of goldnanomaterials in a human skin cell model. Langmuir2012, 28:3248–3258.

70. Oh WK, Kim S, Choi M, Kim C, Jeong YS, ChoBR, Hahn JS, Jang J. Cellular uptake, cytotoxicity,and innate immune response of silica-titania hollownanoparticles based on size and surface functionality.ACS Nano 2010, 4:5301–5313.

71. Sayes CM, Fortner JD, Guo W, Lyon D, Boyd AM,Ausman KD, Tao YJ, Sitharaman B, Wilson LJ,Hughes JB, et al. The differential cytotoxicity of water-soluble fullerenes. Nano Lett 2004, 4:1881–1887.

72. Geidel C, Schmachtel S, Riedinger A, Pfeiffer C,Müllen K, Klapper M, Parak WJ. A general syntheticapproach for obtaining cationic and anionic inorganicnanoparticles via encapsulation in amphiphilic copoly-mers. Small 2011, 7:2929–2934.

73. Kirchner C, Liedl T, Kudera S, Pellegrino T, MunozJavier A, Gaub HE, Stölzle S, Fertig N, Parak WJ.Cytotoxicity of colloidal CdSe and CdSe/ZnS nanopar-ticles. Nano Lett 2005, 5:331–338.

74. Stark WJ. Nanoparticles in biological systems. AngewChem Int Ed Engl 2011, 50:1242–1258.

75. Cho WS, Duffin R, Thielbeer F, Bradley M, MegsonIL, Macnee W, Poland CA, Tran CL, Donaldson K.Zeta potential and solubility to toxic ions as mech-anisms of lung inflammation caused by metal/metaloxide nanoparticles. Toxicol Sci 2012, 126:469–477.

76. Nel A, Xia T, Madler L, Li N. Toxic potential of mate-rials at the nanolevel. Science 2006, 311:622–627.

77. Sayes CM, Wahi R, Kurian PA, Liu Y, West JL,Ausman KD, Warheit DB, Colvin VL. Correlatingnanoscale titania structure with toxicity: a cytotox-icity and inflammatory response study with humandermal fibroblasts and human lung epithelial cells.Toxicol Sci 2006, 92:174–185.



78. Mahon E, Salvati A, Baldelli Bombelli F, Lynch I,Dawson KA. Designing the nanoparticle-biomoleculeinterface for ‘‘targeting and therapeutic delivery’’.J Control Release 2012, 161:164–174.

79. Deng ZJ, Liang M, Monteiro M, Toth I, MinchinRF. Nanoparticle-induced unfolding of fibrinogen pro-motes Mac-1 receptor activation and inflammation.Nat Nanotechnol 2011, 6:39–44.

80. Linse S, Cabaleiro-Lago C, Xue W-F, Lynch I, Lind-man S, Thulin E, Radford SE, Dawson KA. Nucleationof protein fibrillation by nanoparticles. Proc Natl AcadSci U S A 2007, 104:8691–8696.

81. Saba TM, Di Luzio NR. Kupffer cell phagocytosisand metabolism of a variety of particles as a func-tion of opsonization. J Reticuloendothel Soc 1965,2:437–453.

82. Cedervall T, Lynch I, Lindman S, Berggård T, ThulinE, Nilsson H, Dawson KA, Linse S. Understand-ing the nanoparticle–protein corona using methodsto quantify exchange rates and affinities of proteinsfor nanoparticles. Proc Natl Acad Sci U S A 2007,104:2050–2055.

83. Casals E, Pfaller T, Duschl A, Oostingh GJ, PuntesVF. Hardening of the nanoparticle–protein corona inmetal (Au, Ag) and oxide (Fe3O4, CoO, and CeO2)nanoparticles. Small 2011, 7:3479–3486.

84. Milani S, Baldelli Bombelli F, Pitek AS, Dawson KA,Rädler J. Reversible versus irreversible binding oftransferrin to polystyrene nanoparticles: soft and hardcorona. ACS Nano 2012, 6:2532–2541.

85. Walczyk D, Bombelli FB, Monopoli MP, Lynch I,Dawson KA. What the cell ‘‘sees’’ in bionanoscience.J Am Chem Soc 2010, 132:5761–5768.

86. Lundqvist M, Stigler J, Cedervall T, Berggård T, Flana-gan MB, Lynch I, Elia G, Dawson K. The evolution ofthe protein corona around nanoparticles: a test study.ACS Nano 2011, 5:7503–7509.

87. See V, Free P, Cesbron Y, Nativo P, Shaheen U, Rig-den DJ, Spiller DG, Fernig DG, White MRH, PriorIA, et al. Cathepsin L digestion of nanobioconjugatesupon endocytosis. ACS Nano 2009, 3:2461–2468.

88. Sund J, Alenius H, Vippola M, Savolainen K, PuustinenA. Proteomic characterization of engineered nano-material–protein interactions in relation to surfacereactivity. ACS Nano 2011, 5:4300–4309.

89. Lundqvist M, Stigler J, Elia G, Lynch I, Cedervall T,Dawson KA. Nanoparticle size and surface propertiesdetermine the protein corona with possible implica-tions for biological impacts. Proc Natl Acad Sci U S A2008, 105:14265–14270.

90. Tenzer S, Docter D, Rosfa S, Wlodarski A, KuharevJ, Rekik A, Knauer SK, Bantz C, Nawroth T, Bier C,et al. Nanoparticle size is a critical physicochemicaldeterminant of the human blood plasma corona: acomprehensive quantitative proteomic analysis. ACSNano 2011, 5:7155–7167.

91. Simberg D, Park JH, Karmali PP, Zhang WM,Merkulov S, McCrae K, Bhatia SN, Sailor M,Ruoslahti E. Differential proteomics analysis of thesurface heterogeneity of dextran iron oxide nanopar-ticles and the implications for their in vivo clearance.Biomaterials 2009, 30:3926–3933.

92. Dufort S, Sancey L, Coll JL. Physico-chemical param-eters that govern nanoparticles fate also dictate rulesfor their molecular evolution. Adv Drug Deliv Rev2012, 64:179–189.

93. Kapralov AA, Feng WH, Amoscato AA, Yanamala N,Balasubramanian K, Winnica DE, Kisin ER, KotcheyGP, Gou P, Sparvero LJ, et al. Adsorption of sur-factant lipids by single-walled carbon nanotubes inmouse lung upon pharyngeal aspiration. ACS Nano2012, 6:4147–4156.

94. Gasser M, Rothen-Rutishauser B, Krug HF, GehrP, Nelle M, Yan B, Wick P. The adsorption ofbiomolecules to multi-walled carbon nanotubes isinfluenced by both pulmonary surfactant lipids andsurface chemistry. J Nanobiotechnol 2010, 8:31.

95. Ruge CA, Schaefer UF, Herrmann J, Kirch J,Cañadas O, Echaide M, Pérez-Gil J, Casals C, MüllerR, Lehr CM. The interplay of lung surfactant proteinsand lipids assimilates the macrophage clearance ofnanoparticles. PLoS One 2012, 7:e40775.

96. Shvedova AA, Kisin E, Murray AR, Johnson VJ,Gorelik O, Arepalli S, Hubbs AF, Mercer RR, Keo-havong P, Sussman N, et al. Inhalation vs. aspirationof single-walled carbon nanotubes in C57BL/6 mice:inflammation, fibrosis, oxidative stress, and mutage-nesis. Am J Physiol Lung Cell Mol Physiol 2008,295:L552–L565.

97. Mahler GJ, Esch MB, Tako E, Southard TL,Archer SD, Glahn RP, Shuler ML. Oral exposure topolystyrene nanoparticles affects iron absorption. NatNanotechnol 2012, 7:264–271.

98. Walkey CD, Olsen JB, Guo H, Emili A, Chan WCW.Nanoparticle size and surface chemistry determineserum protein adsorption and macrophage uptake.J Am Chem Soc 2012, 134:2139–2147.

99. Kim HR, Andrieux K, Gil S, Taverna M, Cha-cun H, Desmaële D, Taran F, Georgin D, CouvreurP. Translocation of poly(ethylene glycol-co-hexadecyl)cyanoacrylate nanoparticles into rat brainendothelial cells: role of apolipoproteins in receptor-mediated endocytosis. Biomacromolecules 2007,8:793–799.

100. Prapainop K, Witter DP, Wentworth P. A chemicalapproach for cell-specific targeting of nanomateri-als: small-molecule-initiated misfolding of nanopar-ticle corona proteins. J Am Chem Soc 2012,134:4100–4103.

101. Dutta D, Sundaram SK, Teeguarden JG, Riley BJ,Fifield LS, Jacobs JM, Addleman SR, Kaysen GA,



Moudgil BM, Weber TJ. Adsorbed proteins influ-ence the biological activity and molecular targetingof nanomaterials. Toxicol Sci 2007, 100:303–315.

102. Ge C, Du J, Zhao L, Wang L, Liu Y, Li D, Yang Y,Zhou R, Zhao Y, Chai Z, et al. Binding of blood pro-teins to carbon nanotubes reduces cytotoxicity. ProcNatl Acad Sci U S A 2011, 108:16968–16973.

103. Ehrenberg MS, Friedman AE, Finkelstein JN, Ober-dorster G, McGrath JL. The influence of proteinadsorption on nanoparticle association with culturedendothelial cells. Biomaterials 2009, 30:603–610.

104. Lartigue L, Wilhelm C, Servais J, Factor C, DencausseA, Bacri JC, Luciani N, Gazeau F. Nanomagneticsensing of blood plasma protein interactions with ironoxide nanoparticles: impact on macrophage uptake.ACS Nano 2012, 6:2665–2678.

105. Lesniak A, Fenaroli F, Monopoli MP, ÅbergC, Dawson KA, Salvati A. Effects of the presenceor absence of a protein corona on silica nanopar-ticle uptake and impact on cells. ACS Nano 2012,6:5845–5857.

106. Deng ZJ, Liang M, Toth I, Monteiro M, Minchin RF.Plasma protein binding of positively and negativelycharged polymer-coated gold nanoparticles elicits dif-ferent biological responses. Nanotoxicology (Epubahead of print; March 6, 2012).

107. Moghimi SM, Andersen AJ, Ahmadvand D, WibroePP, Andresen TL, Hunter AC. Material properties incomplement activation. Adv Drug Deliv Rev 2011,63:1000–1007.

108. Hamad I, Al-Hanbali O, Hunter AC, Rutt KJ,Andresen TL, Moghimi SM. Distinct polymer archi-tecture mediates switching of complement activationpathways at the nanosphere-serum interface: implica-tions for stealth nanoparticle engineering. ACS Nano2010, 4:6629–6638.

109. Kievit FM, Zhang M. Cancer nanotheranostics:improving imaging and therapy by targeted deliv-ery across biological barriers. Adv Mater 2011,23:H217–H247.

110. Moghimi SM, Hunter AC, Andresen TL. Factor

bridge over troubled waters: understanding the synthetic and ......overview bridge over troubled...

Documents