REVIEW

Exposure to airborne gold nanoparticles: a reviewof current toxicological data on the respiratory tract

Barbara De Berardis & Magda Marchetti &Anna Risuglia & Federica Ietto & Carla Fanizza &Fabiana Superti

Received: 20 March 2020 /Accepted: 28 July 2020# The Author(s) 2020

Abstract In recent years, the introduction of innovativelow-cost and large-scale processes for the synthesis ofengineered nanoparticles with at least one dimensionless than 100 nm has led to countless useful and exten-sive applications. In this context, gold nanoparticlesstimulated a growing interest, due to their peculiar char-acteristics such as ease of synthesis, chemical stability

and optical properties. This stirred the development ofnumerous applications especially in the biomedicalfield. Exposure of manufacturers and consumers to in-dustrial products containing nanoparticles poses a po-tential risk to human health and the environment. De-spite this, the precise mechanisms of nanomaterial tox-icity have not yet been fully elucidated. It is well knownthat the three main routes of exposure to nanomaterialsare by inhalation, ingestion and through the skin, withinhalation being the most common route of exposure toNPs in the workplace. To provide a complete picture ofthe impact of inhaled gold nanoparticles on humanhealth, in this article, we review the current knowledgeabout the physico-chemical characteristics of thisnanomaterial, in the size range of 1–100 nm, and itstoxicity for pulmonary structures both in vitro andin vivo. Studies comparing the toxic effect of NPs largerthan 100 nm (up to 250 nm) are also discussed.

Keywords Nanomaterials . Gold nanoparticles .

Physico-chemical properties . Respiratory exposure .

In vitro toxicity . In vivo toxicity . Environmental andhealth effects

Introduction

Nanotechnology enabled the development ofengineered materials at the nanometre scale with specif-ic properties related to their size, morphology and sur-face reactivity. Due to their peculiar physico-chemicalcharacteristics, nanoparticles raised great interest for

J Nanopart Res (2020) 22: 235https://doi.org/10.1007/s11051-020-04966-9

Barbara De Berardis and Magda Marchetti contributed equally tothis work.

B. De Berardis :M. Marchetti : F. Superti (*)National Centre for Innovative Technologies in Public Health,National Institute of Health, Viale Regina Elena 299,00161 Rome, Italye-mail: fabiana.superti@iss.it

B. De Berardise-mail: barbara.deberardis@iss.it

M. Marchettie-mail: magda.marchetti@iss.it

A. Risuglia : F. Ietto : C. FanizzaDepartment of Technological Innovations and Safety of Plants,Products and Anthropic Settlements, INAIL, Via R. Ferruzzi38-40, 00143 Rome, Italy

A. Risugliae-mail: a.risuglia@inail.it

F. Iettoe-mail: f.ietto@inail.it

C. Fanizzae-mail: c.fanizza@inail.it

http://crossmark.crossref.org/dialog/?doi=10.1007/s11051-020-04966-9&domain=pdfhttps://orcid.org/0000-0003-4909-7598

their potential applications both in consumer productsand in the biomedical field. Gold nanoparticles (AuNPs)are attractive in several areas such as electronics, sensors(Han et al. 2015a), solar cells (Chen et al. 2015) andcatalysis (Nita et al. 2016). They are also used in cos-metics and as dietary supplements.

According to the inventory of the Woodrow WilsonInternational Center for Scholars, products containingAuNPs rank sixth among nanotechnology consumerproducts (Vance et al. 2015). In particular, the increasedinterest in AuNP biomedical applications is due to theirinteresting optical properties derived from their uniqueinteraction with light. The conduction electrons presenton the surface of AuNPs are mobile and create oscillationsthat propagate as waves, called surface plasmons. Whenthe wave vector of incident light is coupled with the wavevector of the surface plasmons, a collective coherentoscillation of conduction electrons occurs, called surfaceplasmon resonance (SPR). The plasmon resonance caneither radiate light, a process used in optical and imagingfields, or it can be absorbed and converted to heat, aphenomenon used in photothermal cancer therapy(Bodelón et al. 2017). All these properties, together witheasy surface functionalisation, makeAuNPs versatile plat-forms for diagnosis, therapy, drug delivery and tissueengineering. Recently, in order to reduce the risk associ-ated with systemic exposure, the use of inhaled AuNPs asdrug carrier and radiosensitiser (Hamzawy et al. 2017;Hao et al. 2015) has been investigated.

The increasing use of AuNPs in industrial processesand biomedical research involves the exposure of ahigher number of workers. Although no exact data areavailable on the number of workers exposed to NPs, in2012, the European Commission declared about400,000 workers involved in the nanotechnology sectorin the EuropeanUnion alone (EEC 2012), with 6millionworkers estimated to be potentially exposed in 2020(Roco 2011). Research laboratories using engineerednanomaterials (ENMs) represent the workplaces withthe highest potential exposure (NIOSH 2012). Workerexposure to AuNPsmight occur during production (syn-thesis, harvesting, dumping, mixing, reactor cleaning,transferring), preparation of suspensions or incorpora-tion into products.

Due to the growing number of workers exposed toNPs, measurement strategies, techniques and methodsto evaluate exposure in workplaces have been devel-oped (Brenner et al. 2016; Garcia et al. 2017; Asbachet al. 2017).

Currently, there are no occupational exposure stan-dards for NPs. Moreover, the choice of the most appro-priate metric to evaluate exposure levels to NPs is stillwidely debated and complicated due to their high num-bers and negligible masses. The US National Institutefor Occupational Safety and Health (NIOSH) proposedparticle number concentration as an alternative metric tothe traditional mass concentration (Methner et al. 2010).

Several unofficial occupational exposure limits(OELs) have been recommended by national organisa-tions in the UK, Germany, the Netherlands, the USAand Australia. These organisations categorised NPs intofive groups: insoluble fibre-like NPs with high aspectratio; biopersistent NPs with density lower than6000 kg/m3; biopersistent NPs with density higher than6000 kg/m3; NPs already classified in their larger size ascarcinogenetic, mutagenic, asthmagenic or reproductivetoxin; and soluble NPs. Cerium oxide, gold, iron, ironoxide, silver, cobalt, lanthane, lead and antimony oxidebelong to the third group. For NPs in this group, whichincludes AuNPs, an OEL equal to 2 × 104 particles/cm3

has been proposed by the UK, Germany and the Neth-erlands (van Broekhuizen et al. 2012).

Very few data are available on AuNP exposure in theworkplaces. Zimmermann et al. (2012) showed thatAuNP release can occur during the maintenance andthe cleanout of deposition equipment. In a case studyon AuNP exposure assessment in a pilot-scale facility,the authors measured particle emission during the prep-aration and the synthesis and post-synthesis phases of14 nm AuNPs, using a combination of particle numberconcentration counters and filter-based air sampling forparticle characterisation. They found a significant in-crease of particle number concentration up to10,295 particles/cm3 compared to background, belowthe proposed OEL for AuNPs, in the size range between7 and 300 nm during the synthesis phase, with a peak inthe range 7–35 nm (OECD 2016a). This does not meanthat exposure to AuNPs of workers, engaged in indus-trial processes or laboratory activities, is not a potentialhealth risk. In fact, they are subjected to repeated expo-sures that may involve a risk, especially for workerswith lung and cardiovascular diseases, as alreadyhighlighted for exposure to airborne ultrafine particles(Miller et al. 2012).

Inhalation is one of the main routes of exposure toNPs and many studies showed how NP exposure mayimpact on pulmonary response; however, only few in-vestigations on pulmonary toxicity of AuNPs have been

Page 2 of 41 J Nanopart Res (2020) 22: 235235

performed. In particular, several toxicological studiesand biodistribution of AuNPs after intravenous, oraland peritoneal administration are available in the litera-ture (De Jong et al. 2008; Balasubramanian et al. 2010;Zhang et al. 2010; Schleh et al. 2012; Rambanapasi et al.2016), while there is a lack of information about inha-lation toxicity, especially as in vivo studies. Most ofthese investigations have used intratracheal instillationto deliver AuNPs to the lung and very few studies havebeen performed with a whole-body inhalation chamber(Semmler-Behnke et al. 2008; Jacobsen et al. 2009;Sadauskas e t a l . 2009; Lipka et a l . 2010;Balasubramanian et al. 2013; Han et al. 2015b; Sunget al. 2011).

In this manuscript, we illustrate the state of the art onAuNP in vitro and in vivo toxicity due to inhalationexposure. We also review the principal methods forAuNP synthesis and surface functionalisation and themost important shapes and types of functionalisationused in consumer products and in biomedical research,as well as the principal techniques used to determinetheir physico-chemical properties. The aim of this re-view is to analyse in-depth the literature available todetect strengths and possible deficiencies to identify oneor more ideal approaches for the risk assessment ofinhalation exposure to AuNPs.

Gold nanoparticle synthesis, shapeand functionalisation

Gold nanoparticle synthesis and nanoparticle solutionstabilisation

The methods for AuNP synthesis can be divided intophysical, chemical and biological. The physical methodconsists in laser ablation of a solid target placed in aliquid medium (Sylvestre et al. 2004), whereas chemicalsynthesis of metal NPs consists in the reduction of metalsalts. The most used chemical methods to synthesiseAuNPs are due to Turkevich et al. (1951) and Brustet al. (1994). Using the Turkevich method, AuNP-stabilised suspensions of negatively charged and spher-ical shape with sizes between 15 and 100 nm are syn-thesised, depending on the ratio of sodium citrate to Au.As sodium citrate concentration increases, the AuNPdiameter decreases (Nicol et al. 2015). The Brust-Schiffrin method is used to synthetise thiolate-

stabilised AuNPs with diameter less than 5 nm (Zhaoet al. 2013).

For the biological synthesis, several procedures areavailable to produce NPs with different sizes, shapesand surface properties (Zhao et al. 2013). The synthesisconsists in the incubation, in an appropriate medium, ofmetal salt solution with a biological agent, as reducingagent. The biological agent used is part of a bacterial,fungal, viral or plant system, like enzymes, biomole-cules, terpenoids, phenols and alkaloids. Plants or algaeare the most used systems for the production of NPsbecause they are efficient, safer and less time-consuming (Chugh et al. 2018). Shape, size and prop-erties of NPs can be modified by altering pH, tempera-ture, pressure, time and biological reducing agent(Chugh et al. 2018). The main advantage of greensynthesis is the production of pure NPs, whereas NPsobtained by chemical methods are contaminated byproducts used during the process. Once the NP synthesisprocess is complete, it is necessary to perform astabilisation step of the obtained NP solution, with adispersant layer of appropriate thickness. This is neededin order to avoid agglomeration and aggregation phe-nomena and to allow their effective application in var-ious fields. Two methods can be used to stabilise thesolution: electrostatic stabilisation, leveraging the samecharge on the NP surface, causing the repulsion ofparticles; or steric stabilisation, using a polymeric coat-ing that does not allow the NPs to reach the minimumdistance for which the van der Waal forces can causetheir aggregation.

Shape of gold nanoparticles in biomedical applications

The shape and size of AuNPs can be finely tuned andeasily controlled during the synthesis process in therange 1–150 nm in order to obtain NPs with specificchemical, electrical and optical properties (Jain et al.2008) used in different fields of nanotechnologyapplications.

AuNPs with different shapes, as nanospheres, nano-rods, nanostars, nanocages, nanoshells, hollow nano-shells and nanomatryoshkas, have been synthetised(Fig. 1).

Each of these shapes exhibits a specific maximumexcitation of surface electrons and a unique conversionefficiency of light to heat, depending on size and aggre-gation state. Au nanorods are particles with a cylindricalshape and a length to diameter ratio (aspect ratio) greater

J Nanopart Res (2020) 22: 235 Page 3 of 41 235

than 1. The sizes of nanorods have a strong influence onthe scattering efficiency of NPs because the longitudinalSPR along the long axis of these NPs depends stronglyon aspect ratio. It has been shown (Lohse and Murphy2013) that larger Au nanorods scatter light more effi-ciently at longitudinal SPR than smaller Au nanorods.As a result, larger Au nanorods can find better applica-tions in the field of optical imaging, whereas smaller Aunanorods can be used in photothermal therapy due totheir higher absorption efficiency (Lohse and Murphy2013). Au nanostars exhibit a spherical core with mul-tiple sharp edges. These protruding tips allow to obtain ared shift in the absorption spectra of AuNPs, used inphotothermal therapy (Vines et al. 2019). Au nanocagesare nanostructures that exhibit hollow interiors or po-rous walls. Due to this morphology, these NPs havegained an increasing interest in biomedical applications,as this morphology allows to load drugs and theranosticagents in the cavity of NPs (Kohout et al. 2018). Aunanoshells have a layer of gold surrounding a core madeof silica that causes a red shift of the characteristicplasmon absorption spectrum of gold to the NIR regionof the electromagnetic spectrum. The plasmon reso-nance frequency is tunable by altering the core to shellratio. These NPs are biologically non-reactive, particu-larly if coated with a layer of polyethylene glycol (PEG).Au hollow nanoshells show a thin outer Au shell and a

hollow core, used for drug delivery (Vines et al. 2019).Au nanomatryoshkas have a gold core coupled to a thinsilica shell with a gold epilayer around them. In thisway, a strong coupling between the plasmons of thegold core and the shell is established, which makes theAuNPs intensely photothermically activable. Their plas-mon resonance is tunable to NIR wavelengths by alter-ing the relative thicknesses of the core, the shell and theepilayer (Kaur et al. 2016).

Functionalisation of gold nanoparticles

The biomedical applications of AuNPs require a modi-fication of their surface to achieve high efficiency andlimit the treatment toxicity. AuNPs possess tunablesurface chemistry so that several functional groups canbe conjugated on their surface. Taking advantage of theability of AuNP to bind thiol and amine groups, togetherwith SPR, their surface can be modified in order toenhance cellular uptake and achieve more targeted ther-apy (Nicol et al. 2015; Vines et al. 2019). By replacingthe stabilising surfactant surrounding AuNP surfacewith PEG, a long circulation can be achieved (Choiet al. 2003). Several functionalised groups can be at-tached to NPs, including low molecular weight ligands,peptides, proteins, polysaccharides, polyunsaturatedand saturated fatty acids, DNA, antibodies, plasmids

Fig. 1 Schematic representationof different shapes of AuNPs

Page 4 of 41 J Nanopart Res (2020) 22: 235235

and siRNA. For conjugation of antibodies orfunctionalised groups to AuNPs, both covalent andnon-covalent immobilisation modes can be used(Fig. 2). These techniques take advantage of the physi-cal and chemical interactions between AuNPs and anti-bodies or other molecules. The non-covalent mode isbased on two types of physical interactions: the electro-static attraction between positively charged groups inantibodies and the AuNP negatively charged surfaceand the hydrophobic attraction between the AuNP sur-face and the antibody. The main limitations of thismethod are that it is suitable only for appropriatelycharged ligands, and the binding is influenced by pHchanges (Nicol et al. 2015). The covalent mode is basedon the dative binding between the Au conducting elec-trons and amino acid sulphur atoms of the antibody andthe chemical interactions between antibodies andAuNPs. These interactions are achieved by chemisorp-tion via thiol derivatives, through the use of bifunctionallinkers and through the use of adapter molecules likeavidin and biotin (Jazayeri et al. 2016; Kumar et al.2008). Using these techniques, AuNPs can be co-functionalised with different ligands in the form ofmixed monolayer like PEG and peptides, PEG andfluorescing dye, or PEG and heavy metal atoms(Kumar et al. 2012; Huang et al. 2010; Kirui et al.

2010). An alternative approach to attach a ligand onAuNP surface consists in the use of a linker molecule,like a polymer chain, in order to achieve AuNP stability.At the other end of the chain, the functional ligand isattached and it is available for biological interaction(Nicol et al. 2015).

Characterisation of gold nanoparticles

The increasing number of NP-based products is a matterof concern among the scientific community and inter-national bodies, such as the Organisation for EconomicCo-operation and Development (OECD), EuropeanFood Safety Authority (EFSA) and Scientific Commit-tee on Emerging and Newly Identified Health Risks(SCENIHR), due to the potential human health effectsof NPs, related to their increased surface chemical reac-tivity. Although the mechanisms underlying NP toxicityhave not yet been fully elucidated, regarding AuNPs,toxicological data point out that the main characteristicsinvolved in AuNP toxicity mechanisms are primarysize, shape, specific surface area, surface charge andsurface chemistry. As expected, the size plays an impor-tant role in the cellular uptake process (Chithrani et al.2006; Arnida et al. 2010). The major pathway for cel-lular uptake of most AuNPs is receptor-mediated orabsorptive endocytosis. This mechanism is probablyenabled by the protein layer that coats the NP surfacein the presence of biological fluids: the protein corona(Chithrani et al. 2006; Arnida et al. 2010; Chithrani et al.2010; Chithrani 2010; Ng et al. 2015). However, AuNPssmaller than 10 nm pass into cells directly through thecell membrane and enter the cell nucleus probably be-cause they can pass through the nuclear pores (Chithraniet al. 2006; Huo et al. 2014; Pan et al. 2007). Also,in vivo, a size-dependent effect on the translocation ofAuNPs to other parts of the body has been observed(Sonavane et al. 2008; Hirn et al. 2011). Chithrani et al.(2006) showed that single AuNPs of 50 nm enter cellsmore efficiently than 14 nm AuNPs of the same chem-ical composition. It is difficult to draw meaningful con-clusions about the relationship between size and cyto-toxicity as several studies showed a size-dependenttoxicity effect of AuNPs (Pan et al. 2007; Senut et al.2016), while others consider more important the contri-bution of concentration, surface modification or shape(Chithrani et al. 2006; Mironava et al. 2010; Pan et al.2009; Wang et al. 2013). Chithrani et al. (2006) have

Fig. 2 Non-covalent and covalent conjugation modes of antibod-ies to AuNPs: schematic representation

J Nanopart Res Page 5 of 41(2020) 22: 235 235

shown a role of NP shape and surface chemistry incellular uptake and cytotoxicity induction. SphericalAuNPs with size between 10 and 100 nm showed highercellular uptake than Au nanorods in the same size range.Moreover, rod-shaped AuNPs with aspect ratio (1:3)showed cellular uptake greater than NPs with aspectratio (1:5) (Chithrani et al. 2006). Wang et al. (2013)compared the in vitro and in vivo capabilities of Aunanohexapods, nanorods and nanocages forphotothermal cancer treatment and found that Aunanohexapods exhibit the highest cellular uptake andthe lowest cytotoxicity in vitro. Wang et al. (2016)analysed the impact of AuNP shape on the cytotoxicityinduction in a zebrafish model showing that sphericalparticles are more toxic than rod- and polyhedron-shapedNPs. In contrast, Karakoçak et al. (2016) showedthat nanorods are more cytotoxic than spherical NPs in aretinal pigment epithelial (ARPE-19) cell line. Finally,Wan et al. (2015) demonstrated that NP cytotoxic effectis related to their surface chemistry but not to theiraspect ratio, whereas Qiu et al. (2010) did not excludea role to this characteristic. Surface charge also plays animportant role in NP cellular uptake due to the electro-static interactions between the negatively charged cellmembrane glycoproteins and the NPs. An uptake mech-anism different from endocytosis has been demonstratedfor ligand-free positively charged AuNPs, and diffusionhas been proposed as a possible entry pathway (Tayloret al. 2010). It has been demonstrated that the uptakeefficiency of the positively charged AuNPs is greaterthan the neutral or negatively charged ones (Hauck et al.2008; Alkilany et al. 2009; Cho et al. 2009; Liang et al.2010; Liu et al. 2013). Moreover, the cells take upnegatively charged AuNPs with greater efficiency thanthe neutral AuNPs, most probably because of a betterbinding with membrane surface proteins (Liang et al.2010). Positively charged AuNPs not only penetratemore effectively the cell membrane but also showhigher toxicity compared to negatively NPs (Liu et al.2013). Finally, it has been shown that the charge ofAuNPs plays a greater role in the uptake of AuNPs withrespect to their size (Liang et al. 2010). All these dataavailable in the literature show that numerous physico-chemical properties of NPs are simultaneously involvedin NP–cell interactions. Therefore, it is essential to studyin-depth NP characteristics in order to identify andclarify the mechanism of their potential toxicity.

NP characterisation is complex and the OECD andother international bodies have provided a list of

analytical techniques to be used for each parameter ofinterest in NP characterisation (OECD 2016b; ECHA2017; EFSA 2018). Multiple techniques should be usedwhere possible in order to better understand NP proper-ties and NP–cell interaction mechanisms. Examples ofanalytical techniques for the most important character-istics of NPs are shown in Table 1.

Two types of NP samples are used in the in vivostudy for inhalation exposure: NP suspensions and NPaerosol. The principal techniques for the characterisa-tion of AuNP powders and AuNP suspension as well asAuNP aerosol are described below.

Principal techniques for the characterisation of AuNPpowders and suspensions

Size/size distribution evaluation

In recent years, various analytical techniques have beendeveloped for determining the size and size distributionof ENMs, based on physical phenomena observed inparticle interaction or on nanometric scale forces. Sincethere is no single analytical technique capable to deter-mine the size distribution in the range 1–100 nm satis-factorily, the Working Party of ManufacturedNanomaterial of OECD, EFSA and other internationalbodies declared that different techniques should be usedfor NP size determination. The choice of techniquedepends on the state in which ENMs occur, such as inpowder form, dispersed in a liquid or incorporated in asolid material or in a matrix. One of the techniques usedto determine both size and size distribution and thereforeto identify the NPs in a consumer product must beelectron microscopy (EM), such as transmission elec-tron microscopy (TEM) and scanning electron micros-copy (SEM). The main advantage of EM consists in thedetermination of size, shape and crystal structure of eachindividual particle. Moreover, unlike other techniques, itallows to characterise samples containing mixtures ofNPs with different sizes and shapes. Finally, if com-bined with an energy-dispersive X-ray spectrometer, itallows to determine ENM chemical composition (DeBerardis et al. 2010). The main limitation is that EMdoes not permit a quantitative analysis of the sample andrequires long analysis times and experienced operators.In addition, this technique requires sample preparationsto remove any solvent or liquid medium for NP disper-sions leading to artefacts.

Page 6 of 41 J Nanopart Res (2020) 22: 235235

Table 1 Examples of analytical techniques for the most important characteristics of NPs

NP properties Techniques/principles Main disadvantages Available guidancea

Size/size distribution SEM: secondary electrons(SE) or backscatteredelectrons (BSE) to formNP image

High vacuum condition requiredConductive material coating requiredOperator experience dependentLong analysis timesPossible artefacts due to sample preparation

ISO/WD 19749ISO 13322-1

TEM: transmitted electronsthrough the sample toform the image

High vacuum condition requiredHigh voltage electron beamExtremely thin sample to create imageOperator experience dependentLong analysis times

ISO/WD 21363ISO 13322-1

DLS: fluctuations in theintensity of scatteredlight caused by Brownianmotion of particles

Scattering intensity dependence by thesixth power of the NP radius leading tomisleading results for suspension withNPs of different sizes

Limitations in polydisperse suspensionsAssumption of spherical shape NPsMisleading results for non-spherical NP

suspensions

ISO 22412

PTA: observation andregistration on a videocamera of light scatteredby single NPs or NPagglomerates

Difficult detection of NPs smaller than 20 nm ISO 19430

Shape Analysis systems of SEimages obtained by SEM

Operator experience dependentLong analysis timesPossible artefacts due to sample preparation

ISO/WD19749 ISO 13322-1

TEM: analysis systems ofimage obtained by TEM

Extremely thin sample to create imageOperator experience dependentLong analysis times

ISO/WD 21363ISO 13322-1

Surface area BET: adsorption isothermtheory to calculatesurface area

Dry solid sample requiredOnly for NPs that do not absorb gasLimitation for porous NP application

ISO 9277ISO 15901-2/−3 ISO 18757

Surface charge Zeta potential: electrophoreticlight scattering

Dependent on pH suspensionParticle concentrationNon-suitable for fluorescing samples

ISO 13099 series

Chemical composition EDX: characteristic X-rayemitted by sample uponinteraction with an incidentelectron beam

High vacuum condition requiredConductive material coating requiredSemi-quantitative analysisDetection of element with Z > 5Thermal damage to the samplePoor accuracy of semi-quantitative

analysis for element at low concentrations

ISO 22489

Surface chemistry XPS: kinetic energymeasurements of electronsescaped from samplesurface upon incidentX-ray beam

High vacuum condition requiredSuitable only for elemental composition

of a 0–10-nm surface layerLong collection time of acquisition

ISO/TR 14187ISO 18118

Raman: measurement ofvibrational energy levelsassociated with the chemicalbonds in the sample

Interference of fluorescenceExtremely small cross sectionIntense laser excitationLarge amount of sample to provide

sufficient signals

NA

SEM, scanning electron microscopy; TEM, transmission electron microscopy; DLS, dynamic light scattering; BET, Brunauer, Emmet andTeller; EDX, energy-dispersive X-ray spectroscopy; XPS, X-ray photoelectron spectroscopy; NA, not availablea Guidance available on “Guidance on risk assessment of the application of nanoscience and nanotechnologies in the food and feed chain:Part 1, human and animal health” (EFSA 2018)

J Nanopart Res Page 7 of 41(2020) 22: 235 235

Other methods widely used to determine the size andsize distribution of NPs are based on laser light scatter-ing techniques, such as dynamic light scattering (DLS)or particle tracking analysis (PTA) (Calzolai et al.2012). DLS enables a rapid in situ characterisation ofsize and size distribution of particles dispersed in aliquid, illuminating an optical cell containing a dilutesolution of particles with a laser light. The hydrodynam-ic diameter is calculated by measuring fluctuations inthe intensity of scattered light caused by Brownianmotion. DLS also provides a number indicating thepolydispersity of the particle population, called polydis-persity index (PDI). This technique is ideal formonodispersed samples (PDI < 0.1) containing spheri-cal particles. When particles with different sizes or non-spherical particles are present in the sample, DLS cangive misleading results due to scattering intensity de-pendence on the 6th power of radius (Calzolai et al.2012) and on the assumption of spherical shape for thedetermination of hydrodynamic diameter. In the case ofnon-spherical particles, DLS cannot provide any reliablemeasure of size without prior knowledge of particleshape data that can only be determined by EM. ThePTA observes and records, on a highly sensitive videocamera, the rapid movement of each particle or aggre-gate in a suspension illuminated by a laser source. Thehydrodynamic diameter of each particle is calculatedthrough the Stokes–Einstein equation. Unlike DLS,PTA calculates particle size on a particle-by-particlebasis and is highly sensitive since it detects particles atconcentrations as low as 106 particles/cm3. The biggestdisadvantage of this technique is the limitation on thelowest particle size that can be detected. As the lightscattering of particles in solutions depends not only onthe refractive indices of the material and liquid but alsoon the particle size, it is difficult to detect particlessmaller than 20 nm, except for those with a high refrac-tive index such as Au (Calzolai et al. 2012).

Shape evaluation

NP shape is variable, and to characterise this feature,shape factors such as sphericity, circularity, aspect ratio,convexity and fractal dimension are used (ECHA 2017).

Particle shape is usually determined by EM, the onlytechnique able to identify the presence of particles withdifferent morphologies in a sample. It includes manyqualitative and semi-quantitative techniques, based onimage-analysis systems to investigate the morphology

and the aggregation state. Two different morphologiesof commercial AuNPs (Nanocs Inc., USA) are shown inFig. 3: spherical (40 nm) and rod-like shape (10 ×60 nm). NPs with irregular shape were observed in theAu nanorod sample.

Specific surface area evaluation

The specific surface area is the area of the exposedsurface of a certain amount of material divided by itsmass. It plays an important role in the physical andchemical interactions because molecules detachedfrom the surface or located at the surface of the ENMsinteract with the biological fluids and tissue. Jacobsenet al. (2009) in a study on pulmonary toxicity ofintratracheally instilled AuNP in the lungs of mice re-ported that NPs with high specific surface area exhibitedhigher reactivity than particles with low specific surfacearea. This characteristic determines the surface charge infunctionalised NPs and has direct consequences on NPagglomeration and NP–cell interactions.

The most common technique for measurement of thesurface area of particles is by gas absorption measure-ments using Brunauer, Emmet and Teller (BET) adsorp-tion isotherm theory (Brunauer et al. 1938). Nitrogen isthe most used, but many other gases such as argon,carbon dioxide or krypton are also employed. The maindisadvantages of this technique are in its applicationonly to dry ENMs and in the assumption of amonodispersed spherical sample. Moreover, this tech-nique is not optimal for porous NPs (OECD 2016b).

Surface charge evaluation

In order to determine the surface charge, the surfacepotential is evaluated by measuring the zeta potential(OECD 2016b; ECHA 2017; EFSA 2018). This is de-fined as the potential difference between the NP disper-sion medium and the stationary double ionic layer offluid attached to NPs. Measurement of zeta potentialgives information on the stability of NP suspension.High positive (>+ 30 mV) or negative (

Surface chemistry evaluation

The expression surface chemistry includes coatings,functional groups and potential surface reactions ofNPs in different media. In most NP applications, it isessential to control the surface chemistry of ENMs;therefore, high sensitivity spectroscopic methods areused and developed to characterise their surface chem-istry in order to gain a better insight into the elementaldistribution, such as X-ray photoelectron spectroscopy(XPS) and Raman spectroscopy.

XPS is an ultrahigh-vacuum surface analysis tech-nique widely used for the analysis of chemical compo-sition, chemical state and electronic state of the elementson the surface material and for the determination ofcontaminations or functional group on it. It is based onthe ejection of valence electrons (photoelectrons) byatoms from the top 1–10 nm of the sample when a beamof X-ray irradiates its surface. Since these photoelec-trons have energies that are specific for the emittingelements and sensitive for the chemical states of theelements, the energies and the peak intensities in theXPS spectrum give information on elemental composi-tion and on chemical bonding identified on the samplesurface.

Raman spectroscopy is based on the detection ofradiation scattered from the molecules in a sample whenirradiated by a monochromatic light. This analyticaltechnique, providing a characteristic spectrum of thespecific vibrations of a molecule, allows the identifica-tion of the substance (Smith and Dent 2005). Moreover,it gives information on the structure of molecules andhow they interact with that surrounding. Raman

spectroscopy is a useful technique to describe the bind-ing interactions of ligand-AuNPs since, unlike otherENMs, AuNPs enhance vibrational Raman bands dueto surface plasmon effects (Smith and Dent 2005). XPSand Raman spectroscopy were used to characterise thesurface of functionalised AuNPs (Debouttière et al.2006; Techane et al. 2011; Lam et al. 2014).

Principal techniques for aerosol characterisation

The aerosol consists of airborne particles with dimen-sions in the range 10−9 m to 100 μm (Kulkarni et al.2012). Due to their different sizes, these particles behavedifferently in the air because different physical lawsgovern their motion. The inertia dominates the motionof particles larger than 0.3–0.5 μm that is described byaerodynamic equivalent diameter, while the Brownianmotion governs prevalently the motion of particles withthe size of a few nanometres. It is described by mobilityequivalent diameter, defined as the diameter of a spher-ical particle with the same velocity (mobility) producedby a unit external electrical force as the particle inquestion (Kulkarni et al. 2012).

These principles are used in the various tools devel-oped for aerosol characterisation.

There are two distinct categories of measurementtechniques to characterise the aerosol; the first is basedon the collection of aerosol particles on a substrate, andthe second is on in situ, near real-time measurement ofaerosol. The first approach allows a deep characterisa-tion of aerosol to determine primary size, morphology,agglomeration and chemical composition of NP sam-pled in the aerosol. However, its main disadvantages

Fig. 3 TEM images of a spherical AuNPs and b Au nanorods

J Nanopart Res Page 9 of 41(2020) 22: 235 235

consist in that the collection processes may lead toparticle modification and the measurements are time-averaged (Kulkarni et al. 2012). The second approachallows amore limited degree of particle characterisation.In studies to assess the in vivo inhalation toxicity, thesecond approach is used because it is critical in real-timemonitoring of system exposure. The most importantcharacteristics of aerosol are the size distribution andthe particle number concentration.

The size distribution of aerosol used in studies toassess the biodistribution of NPs after inhalation expo-sure ranges between a few nanometres and 1 μm. Re-cently, different sensors for size distribution evaluationhave been developed, based on the electrical mobilitydetermination. They consist of a corona charger, a mo-bility classifier and a particle detector. The chargedaerosol, via ion–particle collisions by the corona char-ger, passes into the coaxially cylindrical electrodes ofthe electrical mobility classifier (Intra and Tippayawong2007). An electric field, applied between the electrodes,deflects the particles radially outward, and particleshaving specific mobility are collected. The instrumentmeasures the currents corresponding to the number con-centration of particles of a given mobility, related toparticle size. A data acquisition system processed thesignal currents and the particle mobility diameter iscalculated (Intra and Tippayawong 2007). One of themost used aerosol sizer, based on the electrical mobilitymeasurement, is the scanning mobility particle sizer(SMPS). It allows to determine the particle numberconcentration in the range of 3–1000 nm separatingthe particles from a polydisperse aerosol in narrowintervals of sizes and counting the particles within thatsize range by a detector (Flagan 2008). The SMPS scansacross a voltage range. For a given applied voltage,particles with higher electrical mobility are collectedsooner than those with lower electrical mobility. Thesmaller the particle diameter and the higher the numberof charges, the greater the electrical mobility. The in-verse relationship between electrical mobility and parti-cle size allows the determination of aerosol size distri-bution (Flagan 2008). Therefore, for a given voltage,only particles within a narrow range of electrical mobil-ity diameter are transferred into the condensation parti-cle counter (CPC), where the number concentration ofthe particles is measured. The particle number concen-tration that the SMPS can measure is approximately104–109 particles/cm3 (Intra and Tippayawong 2007).The software of these instruments allows to calculate the

surface area, volume and mass of the particles. Themajor limitation of these techniques consists in theassumption that all particles measured are spherical,but some aerosols are composed of NP aggregatesshowing the shape of chains (Shi et al. 2001). Therefore,the electrical mobility analysis tools of this type ofaggregates can lead to an incorrect evaluation of theseparameters.

Another technique developed to measure the aerosolsize distribution is electrical low-pressure impactor(ELPI). It consists of a low-pressure impactor and acorona charger at the inlet of the ELPI to obtain chargedparticles composing the aerosol. The low-pressure cas-cade impactor classifies the particles according to theiraerodynamic diameters. Electrometers measure the cur-rents induced by particles sampled on impaction stagesthat are converted into the particle number concentration(Ouf and Sillon 2009). ELPI allows not only to obtainreal-time measurements of aerosol size distribution, butalso to collect the particles for their morphological andchemical analyses.

To measure total number particle concentration, dif-ferent particle counters have been developed, such asoptical particle counter (OPC) and condensation particlecounter (CPC).

The principle of optical instruments for the measure-ment of total number particle concentration is the detec-tion of light scattered by individual particles as througha focused beam of light (McMurry 2000). Particle num-ber is determined by counting the pulses of scatteredlight reaching the detector. Because the intensity ofscattered light is related to the size of the scatteredparticle, this relationship can be used to determine par-ticle size. In order to measure particles below 100 nm,optical particle counters with high light intensity andincreased detector sensitivity have been developed.However, they are not able to measure particles muchbelow 50 nm (Glantschnig and Chen 1981).

In order to measure number particle concentration ofsmaller NPs, the CPCs, able to count smaller particlesdown to 2 nm, have been developed. In these counters,the particles are grown by condensation to a size of 10–12 μm to allow their detection. The enlargement of sizeof particles in the aerosol by condensation requires thecreation of a region of supersaturation in a condenserwhere nucleation occurs (Hering and Stolzenburg2005). When the aerosol becomes supersaturated, parti-cles above a certain size grow into drops. Then the dropsare focused through a nozzle and are counted by a laser

Page 10 of 41 J Nanopart Res (2020) 22: 235235

beam. Either the properties of the vapour and particles orthe supersaturation achieved determines the minimumparticle size detected by a CPC (Hämeri et al. 2002).Because supersaturation is not the same for all particles,the efficiency of the counter gradually decreases atsmaller sizes (Kulkarni et al. 2012).

Inhalation toxicity of gold nanoparticles

Due to the vast epithelial surface area of the lungs,inhalation is considered the primary route of exposureto NPs, especially at the workplace (Yokel andMacphail 2011). The respiratory tract can be dividedinto three regions: the nasal cavity, the airways (trachea,bronchi and bronchioles) and the alveolar region wherethe actual gas exchange occurs (Fytianos et al. 2016).Organs and structures of the respiratory system areshown in Fig. 4.

The tracheobronchiolar region consists of secretoryand ciliated epithelial cells. The former produce themucus, the latter clear the particles and pathogens thathave been trapped in the mucus and lead to particleexcretion through the mucociliary escalator. This prima-ry defence system is particularly efficient for micron-dimensioned (> 6 μm) particles, while smaller particleshave more chance to escape clearance (El-Sherbinyet al. 2015).

So, inhaled NPs mainly deposit in the alveolar regionand reach the air–blood barrier. The surface of thealveoli is composed of thin type I alveolar epithelialcells, specialised for gas exchange with the underlying

endothelium at the air–blood barrier, and the surfactantsecreting epithelial type II alveolar cells that are able toproliferate and differentiate into type I cells (Fröhlichand Salar-Behzadi 2014). Type II cells are more abun-dant, but type I cells constitute more than 95% of thealveolar surface. Pulmonary surfactant is a lipid-richfilm spreading at the alveolar air–liquid interface. It iscomposed of approximately 80% phospholipids, 10%neutral lipids and 10% proteins. These proteins are thefirst to come in contact with particles in the respiratorytract and so they are probably among the most abundantconstituents of the inhaled particle protein corona, theprotein layer that NPs rapidly acquire upon contact withbiological fluids (Lundqvist et al. 2008; Marchetti et al.2015). Surfactant proteins (SP)-B and SP-C contributeto lower the surface tension together with the phospho-lipids, while SP-A and SP-D are involved in lung de-fence as components of the innate immune system (Judet al. 2013). Also, alveolar macrophages and dendriticcells (DC) belong to the host defence system towardspotentially harmful inhaled particles and microbes. Thefirst can be found on the internal surfaces of the alveoli(the alveolar ducts) and of the bronchioles; the secondhave been described both in the conducting airways andin the distal lung, acting as antigen-presenting cells(Holt et al. 1990).

Inhaled particles deposit in the different regions ac-cording to their size. In particular, NPs with a size rangeof 5–100 nm are deposited mainly in the airways and inthe alveolar region (Oberdörster et al. 2005). From thealveoli, NPs can diffuse to other organs translocatingthrough the extremely thin air–blood barrier. Therefore,

Fig. 4 Schematic representationof the respiratory system

J Nanopart Res Page 11 of 41(2020) 22: 235 235

it is extremely important to determine the adverse healtheffects associated with inhalation of NPs. To this pur-pose, both cellular and animal models have been used.In vitro models of respiratory exposure offer a simpli-fied approach in which it is possible to concentrate on afew parameters at a time and enable researchers tounderstand the interaction of ENMs with cells at themolecular level. On the other hand, in vivo studies areessential to investigate the possible toxic effects ofENMs because they mimic, in a more realistic way,the environmental, occupational or intended exposure(Jud et al. 2013).

In this section, we will discuss recent results thataddress the toxicity of gold nanoparticles, obtainedusing in vitro or in vivo respiratory systems.

In vitro studies

For in vitro studies on respiratory exposure, bronchialand alveolar cell lines (such as BEAS-B2, HBEC3-KT,A549, MLE-12) are often preferred to primary cells(NHBE), due to their easy and reproducible use as wellas their purity. MLE-12 mouse cells and, mainly, A549human cells, are widely used as in vitro lung epithelialcell models to assess drugs and airborne nanoparticletoxicity in the lung: these cells have many features ofalveolar epithelial type II (ATII) cells (Foster et al.1998) such as the ability to produce surfactant in certainconditions (Wikenheiser et al. 1993; Lieber et al. 1976).Surfactant protein–AuNP interaction largely affects thebiological properties of NPs (Saptarshi et al. 2013) and,therefore, should be considered in all in vitro studies onNP–lung interaction (Marchetti et al. 2015; Schleh et al.2013b). BEAS-B2, HBEC3-KT and NHBE are humanbronchial epithelial cells. BEAS-2B, an SV40 large T-antigen-immortalised human bronchial epithelial cellline, has been extensively used in many experimentalcontexts, such as the screening of chemical and biolog-ical agents and respiratory injury tests. Advancedin vitro systems that mimic the airway epithelial barrierhave also been successfully utilised to study AuNPtoxicity and will be described. Different endpoints havebeen studied, such as cellular uptake, cytotoxicity, reac-tive oxygen species (ROS) production and genotoxicity.

As far as uptake is concerned, it is particularly im-portant to study the mechanisms used by different cellsto internalise NPs as it can influence their intracellularand extracellular fate. As the results obtained on alimited number of cell types cannot be generalised to

all cellular systems, it is indispensable to take intoconsideration the specific cell type influence on the fateof NPs when NPs are to be used in vivo. Furthermore,genotoxicity is an important parameter to consider sincesmall AuNPs, once internalised, are able to enter thenucleus (Huo et al. 2014) and, interactingwith DNA, arelikely to cause genotoxicity. Finally, it is well knownthat metal NPs can induce oxidative stress (OS)(Limbach et al. 2007). OS is the result of an imbalanceof the cellular redox potential, where reactive oxygenspecies (ROS) are produced at an amount that the cel-lular antioxidant systems are not capable to detoxify(Limbach et al. 2007). As the ROS formation plays akey role in the progression of inflammatory disorders(Mittal et al. 2014), it is very important to measure alsothis parameter.

Gold nanoparticle toxicity in bronchial cell models

Vetten et al. (2013) and Vetten and Gulumian (2019)performed two different in vitro studies on the toxicityof AuNPs on bronchial epithelial cells by using label-free techniques such as cell impedance to analyse bothcell adhesion and viability, and dark field andhyperspectral imaging to investigate uptake. The au-thors characterised AuNP suspensions both in Milli-QH2O and in culture medium. In order to improve thestability of suspensions, they centrifuged the nanoparti-cles suspended in Milli-Q H2O and re-suspended themin culture medium. However, the authors performed thecharacterisation by TEM and DLS, determining primarysize and hydrodynamic diameter of NPs. In the firstresearch (Vetten et al. 2013), the cytotoxicity of AuNPsstabilised with citrate (14 and 20 nm in diameter, 1–5 nM) has been studied in BEAS-2B and in cells ofdifferent origin (the Chinese hamster ovary cell lineCHO, and the human embryonic kidney cell line HEK293) using both conventional and label-free methodol-ogies. In cytotoxicity studies using conventional assays,cells were treated with 14 nm AuNPs (1 or 5 nM),whereas, for the assessment of cytotoxicity using cellimpedance, the cells were treated with 14 nm or 20 nmAuNPs (1 nM, 2 nM and 5 nM). Results obtainedshowed that AuNPs were internalised in a time-dependent manner and that toxicity was not related tothe uptake level, but depended on the cellular systemsand on their specific culture media leading to differentNP agglomeration and protein corona formation which,in turn, may influence the uptake and the toxicity of the

Page 12 of 41 J Nanopart Res (2020) 22: 235235

nanoparticles. In fact, AuNPs were internalised inBEAS-2B cells at high levels with minimal toxicityand CHO cells have proven to be sensitive mainly to20 nm AuNPs.

In the second research (Vetten and Gulumian 2019),the authors investigated the influence of NP surfacegroups on AuNP–cell interactions. Concerning surfacechemistry, literature data show that the functionalisationof AuNPs plays a role on cellular uptake and cytotoxic-ity. The presence of PEG-ligands causes reduction orabsence of uptake, compared to colloidal NPs stabilisedwith citrate (Arnida et al. 2010; Nativo et al. 2008;Rayavarapu et al. 2010; Cruje and Chithrani 2015);however, the additional modification of PEG-ligandscan lead to an increase in AuNP uptake. Based on thishypothesis, Vetten and Gulumian (2019) studied theeffect of conjugation of different functional groups viaPEG on AuNP uptake in BEAS-2B cell line. In partic-ular, hydroxyl-PEG (POH), carboxyl-PEG (PCOOH),biotin-PEG (PBtn), nitrilotriacetic acid-PEG (PNTA)and azide-PEG (PAZ) were utilised. Results obtainedshowed that none of the different AuNPs was cytotoxic,and among the different PEG NPs, only PCOOHAuNPs were internalised. The uptake mechanism ofboth citrate-stabilised and PCOOH AuNPs wascaveolin-mediated endocytosis. Following uptake, theseNPs could bypass endolysosomal degradation, so theabsence of intracellular release of ions can explain theabsence of toxicity. These results demonstrate that themechanism of entry is able to determine the intracellularlocalisation and the subsequent cytotoxicity. Moreover,these studies highlight that label-free techniques can beappropriate and valuable methodologies for the study ofAuNP cytotoxicity and uptake.

Other studies have been carried out to analyse theinfluence of AuNP surface charge on cytotoxicity bymeasuring cell viability, cell membrane permeabilityand cell proliferation. In the study of Giri et al. (2015)on BEAS-2B cells, pentanethiol-coated AuNPs (corediameters of ~ 2 nm) were functionalised with thioalkyltetra(ethyleneglycol)ated trimethylammonium (TTMA)and tetraethylene glycol (TEGOH) to obtain differentpercentages of positive charge (from 100 to 96, 65, 20and 0%) by varying the TTMA:TEGOH ratio. Theauthors characterised AuNP suspensions both in Milli-Q H2O and in culture medium. In order to improve thestability of the suspensions, they centrifuged the nano-particles suspended in Milli-Q H2O and re-suspendedthem in culture medium. However, the authors

performed the characterisation only by TEM and deter-mined only the primary size of NPs without givinginformation on agglomeration of AuNPs and size distri-bution of AuNP suspensions. Fully positively chargedAuNPs (100% TTMA) were toxic to epithelial cells,while the reduction in the extension of the positivecharge resulted in the absence of cytotoxicity. In partic-ular, at 1 μM, 100% TEGOH particles (fully negativelycharged) showed no cytotoxic effect, while 20% TTMAAuNPs only slightly affected the morphology of cells.Results of experiments in which the effect of 1 μMAuNPs on cell proliferation was studied showed that,after 24 and 48 h of incubation, all positively chargedAuNPs significantly decreased cell proliferation. Takentogether, the results of this study demonstrated thatAuNP surface charge is an essential characteristic thatinfluences their interaction with the cells. Hence, themodulation of the surface charge to a lower positivitycan reduce their cytotoxicity and can be utilised todesign NPs with improved safety features.

The influence of surface charge on toxicity has beenstudied more thoroughly using larger AuNPs and threedifferent lung cell systems (Schlinkert et al. 2015). Inthis study, AuNPs (7 to 10 nm in diameter) were syn-thesised with different coating to give different classesof surface charge ranging from − 50 to + 70 mV: nega-tively charged AuNPs, coated with sodium citrate, andpositively charged NPs, coated with chitosan(Schlinkert et al. 2015). The characterisation was per-formed only after the AuNP synthesis. Data on primarysize by TEM and surface charge by zeta potential areshown. As the diameters of these AuNPs were in a rangeof dimensions in which they can be deposited both in thetracheobronchial tract and in the alveoli (Löndahl et al.2014), the authors selected three different types of hu-man epithelial cells representative of these regions of therespiratory system: the alveolar adenocarcinoma cellline (A549) (Giard et al. 1973), the bronchial epithelialcell line (BEAS-2B) (Albright et al. 1990) and theprimary human bronchial epithelial cells (NHBE)(Lechner and LaVeck 1985). As it will be better de-scribed in the next paragraph, A549 is the cell line mostfrequently used as a model for the evaluation of NP-induced pulmonary cytotoxicity, whereas, concerningbronchial cells, BEAS-2B cells have the advantage ofrapidly forming a tight epithelium and NHBE cells areprimary cells derived from healthy lung tissue. More-over, all these cell types have already been used asmodels for in vitro nano-safety inhalation studies

J Nanopart Res Page 13 of 41(2020) 22: 235 235

(Schlinkert et al. 2015). Two different assays were per-formed to determine the viability of the cells: the mea-sure of membrane integrity and of cell metabolism(proliferation). ROS production was also measured.The results obtained showed that A549 cells were theleast sensitive to AuNPs. The highest sensitivity wasobserved in NHBE cells, in which exposure to AuNPswith a charge greater than + 40 mV induced cellulardamage. Only AuNPs with the highest positive charge(+ 65–75 mV) were toxic for BEAS-2B cells. Positivelycharged AuNPs induced ROS in all cell types, as alreadyobserved in the same cellular systems exposed to titani-um dioxide NPs (Ekstrand-Hammarstrom et al. 2011).ROS production was more evident in A549 cells whereAuNPs with the highest positive charge (+ 65–75 mV)also induced the highest amount of ROS. Taken togeth-er, the results of this study show that AuNPs, generallyconsidered inert, can become toxic if coated withcharged molecules. As a matter of fact, the AuNPsfunctionalised with chitosan have been demonstratedto be toxic for bronchial cell systems. These results arein agreement with those by Giri et al. (2015), whichdemonstrated that AuNPs with a positively chargedsurface were toxic to epithelial cells.

Finally, the effects of AuNPs in bronchial cells wereanalysed in terms of cytotoxicity, assessed by theAlamar Blue assay, and also in terms of genotoxicity,assessed by the comet assay and the micronucleus test.

In particular, the size-dependent genotoxicity ofAuNPs (5 and 50 nm) was studied in HBEC3-KT cells,normal human bronchial epithelial cells immortalisedwith CDK4 and hTERT (Lebedová et al. 2018). AuNPswere characterised by TEM and zeta potential in order todetermine the size and the surface charge. PTA analysishas been performed in culture medium to study NPagglomeration. After cell exposure, only the smallest(5 nm) AuNPs slightly reduced cell viability (tested upto 50 μg/ml). Concerning genotoxicity, following 48 hexposure to AuNPs, only the smallest particles (5 nm)were active showing 3.3, 5.6 and 7.2% DNA in tail, inthe different concentrations tested (1, 10 and 20 μg/ml,respectively), whereas, at the same concentrations,micronuclei formation was not observed. These results,showing a dimension-dependent DNA damage togetherwith the absence of micronuclei induction, are in agree-ment with a previous study on AuNP genotoxicity car-ried out on cells of different origins (Xia et al. 2017;George et al. 2017). The mechanism responsible for thesize-dependent genotoxicity of the AuNPs has not yet

been fully defined (Lebedová et al. 2018). However, ithas been reported that AuNPs smaller than 10 nm, beingable to enter the nucleus through the nuclear pores (Huoet al. 2014), can interact with DNA causinggenotoxicity.

Results of these studies are summarised in Table 2.

Gold nanoparticle toxicity in alveolar cell models

The first study on alveolar cells showed that non-functionalised 33 nm AuNPs were internalised and in-duced apoptosis in A549 cells but not in other cell linesof non-respiratory origin, showing specificity for lungcells (Patra et al. 2007). The authors detected AuNPclusters near the nucleus by fluorescence and confocalmicroscopy, probably exploiting surface-enhanced Ra-man effect. Since no TEM observation of cells treatedwithAuNPs was performed, it was not possible to assessif the NPs were localised in vesicles or free in thecytoplasm. Similar results were obtained with citrate-stabilised 17-nm AuNPs, which reduced viability in twohuman lung cell lines (A549 and NCI-H1975), inducingboth intrinsic and extrinsic apoptotic pathways, but werefound in membrane-bound vesicles inside the cells byTEM (Choi et al. 2012). The authors observed an in-crease in the particle hydrodynamic diameter (from 36to 45 nm) after incubation in RPMI with 10% foetalbovine serum, presuming that AuNPs adsorbed the se-rum proteins present in the cell culture medium, a phe-nomenon known as protein corona formation(Lundqvist et al. 2008), but they did not explain if theobserved toxicity might be due to this adsorption. Otherauthors, analysing Au nanorod internalisation into A549cells by TEM, observed that NPs were primarily local-ised in lysosomes and membranous vesicles but werealso found scattered in the cytosol as single particles(Tang et al. 2015). Similar results were obtained by usanalysing the entry process of spherical AuNPs in A549cells (unpublished results). In fact, as shown in Fig. 5,also in our experimental conditions, AuNPs areinternalised by the cells and do not enter the nucleusbut are observed as cluster confined to the cytoplasm,free or inside membranous vesicles. Taken together,these observations seem to indicate that AuNPsinternalised into alveolar cells do not reach the nucleusand are confined into the cytoplasm, independently fromtheir shape and size.

Regarding the effect of size, Kirkpatrick’s researchgroup reported a comparable cytotoxicity and

Page 14 of 41 J Nanopart Res (2020) 22: 235235

Tab

le2

AuN

Ptoxicity:invitrostudieson

bronchialcells

Invitrosystem

Size

(nm)

Doserange

Coatin

gMainresults

Conclusions

Reference

BEAS-2B

14 201–5nM

Citrate,anionic

NPsareinternalised

inatim

e-dependent

mannerandarenottoxic.

NPtoxicity

dependson

theirsize

and

onthecellsystem

used.

Vettenetal.(2013)

BEAS-2B

20.01–1

μM

Pentanethiol-

TTMA/TEGOH

Cationic(TTMAcoated)NPs

werefound

tobe

toxic,whileanionicNPs

arenot.

NPsurfacecharge

influences

their

interactionwith

thecells.

Girietal.(2015)

BEAS-2B

NHBE

7 106.25

×10

11–5

×10

12

Citrateor

chito

san

Onlycatio

nic(chitosancoated)NPs

inducedROSandwerefoundto

betoxic.

NPsurfacecharge

influences

their

interactionwith

thecells.

The

type

oflung

epith

elialcellsused

greatly

affectstheresults.

Schlinkertetal.(2015)

HBEC3-KT

5 501–20

μg/ml

Citrate

Only5nm

NPs

(the

smallestones)

werefoundto

begenotoxicandslightly

toxic.

NPtoxicity

dependson

theirsize.

Lebedováetal.(2018)

BEAS-2B

141–5nM

Citrate,PC

OOH

NPs

areinternalised

throughan

activ

eprocessandarenottoxic.

The

entrymechanism

candeterm

ine

intracellularlocalisation.Toxicity

does

notcorrelatewith

thelevel

ofuptake.

VettenandGulum

ian(2019)

POH,P

Btn,

PNTA,P

AZ

Neither

uptake

norcytotoxicity

was

observed.

Studies

arepresentedin

chronologicalo

rder

inwhich

they

werepublished

TTMA,thioalkyltetra(ethyleneglycol)atedtrim

ethylammonium;TE

GOH,tetraethyleneglycol;ROS,

reactiv

eoxygen

species;

ND,notdeterm

ined;PEG,polyethylene

glycol;POH,

hydroxyl-PEG;P

COOH,carboxyl-PE

G;P

Btn,biotin

-PEG;P

NTA

,nitrilo

triacetic

acid-PEG;P

AZ,

azide-PEG

J Nanopart Res Page 15 of 41(2020) 22: 235 235

internalisation by active endocytosis of 9, 11 and 25 nmAuNPs in the human alveolar type II cell lines A549 andNCIH441 (Uboldi et al. 2009; Freese et al. 2012). Theuptake was independent on the presence of a citratecoating that, instead, could affect both cell viabilityand proliferation. Conversely, Choudhury et al. (2013)reported that 40 nm AuNPs were more toxic to A549cells than 20 and 60 nm. This study is particularlyimportant because it suggests for the first time that themechanism of AuNP-induced toxicity in A549 cellsinvolves microtubule aggregation followed by cell cyclearrest and apoptosis (Choudhury et al. 2013).

The green synthesis of ENMs, in which plant extractscan play the role of a reducing agent to synthesisemetallic NPs, is becoming more and more attractive,the main advantage being the substitution of noxiouschemicals with eco-friendly components. However,there are not many studies comparing the toxicity of

biologically synthesised AuNPs with those chemicallysynthesised on respiratory cells. Sathishkumar et al.(2015) obtained 20–50 nm AuNPs by incubatingchloroauric acid with star anise extract (SA-AuNPs)and compared their toxicity to chemically synthesisedAuNPs in A549 cells. The authors showed that SA-AuNPs, probably because of the protective capping bypolyphenols present in the extract, were less effectiveapoptosis inducers and therefore less cytotoxic thanchemically synthesised AuNPs. Hence, they suggestthat this green method is a good candidate for thesynthesis of AuNPs, provided that batch-to-batch varia-tions are kept to a minimum. It is though important topoint out that chemically and biologically synthesisedAuNPs used in this study had different sizes and so theyshould not be compared. Moreover, SA-AuNPs werenot homogeneous regarding characteristics that can in-fluence their behaviour in a cell system such as size (20–50 nm) and shape, as shown by the presence oftriangular- and hexagonal-shaped platelets in additionto particles in TEM micrographs of the NP suspension.As a general rule, it is not useful to add too manyvariables when comparing the effect of ENMs, sincethis can lead to confusing results.

Concerning the effect of mode of synthesis on me-tallic NP cytotoxicity in respiratory cells, Park et al.(2017) described the use of mangosteen (Garciniamangostana) pericarp waste extract to synthesise gold(GM-AuNPs) and silver (GM-AgNPs) nanoparticlesand compared the cytotoxicity of the different biologi-cally synthetised metallic NPs. Spherical GM-AuNPswith an average size of 15.37 ± 3.99 to 44.20 ±16.99 nm were observed in high-resolution TEM (HR-TEM) images. The cytotoxicity and apoptosis inductionin A549 and mouse fibroblast cells were analysed usinga water-soluble tetrazolium assay and the Annexin V/propidium iodide staining, respectively. The results ob-tained showed that GM-AgNPs were toxic and inducedapoptotic cell death, whereas GM-AuNPs showed nosignificant cytotoxicity, suggesting that low AuNP tox-icity may be related to NP composition.

Colorimetric assays are widely used to test in vitrocytotoxicity of chemicals, but when used to test ENMs,interference phenomena may arise. In this view, Chuanget al. (2013) performed a study to validate a dye-freesystem to monitor cell growth measuring electric im-pedance of the monolayer that was already described forbronchial cells (Vetten et al. 2013; Vetten and Gulumian2019). To test the toxicity of Au nanorods, the system

Fig. 5 Ultrastructural analysis of A549 cells exposed for 24 h toAuNPs (40 nm size) at 20 μg/ml. Micrographs show that AuNPsare internalised by the cells and do not enter the nucleus but areconfined to the cytoplasm either free (a) or inside membranousvesicles (b)

Page 16 of 41 J Nanopart Res (2020) 22: 235235

was compared to classical methods (trypan blue, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS), colony for-mation assay) in different mammalian cell lines, includ-ing lung cells like A549 and MRC5 (human normallung fibroblast) and non-respiratory cells like humangastric adenocarcinoma cells (AGS). Results showedthat ROS production and apoptosis makes a major con-tribution to Au nanorod-induced inhibition of cellgrowth in AGS cells. On the contrary, neither ROSproduction nor apoptosis, but rather a cell cycle delay,was induced by Au nanorods in A549 cells. The gradualand steady slowing of the cell growth rate was measuredwith all tested methods, suggesting that the cell-impedance measurement system is reliable for the eval-uation of NP toxicity. Moreover, impedance resultswere similar in MRC5 cells, which do not efficientlytake up dyes, providing an alternative method when it isnot possible to use colorimetric tests. This study con-firms in other respiratory cell lines what was suggestedby Vetten et al. (2013) for bronchial cells. Other authorshad similar results regarding the toxicity of bigger sizeAu nanorods in A549 cells using traditional assays, butthey obtained completely different results on ROS pro-duction, demonstrating that the oxidative stress inducedby the nanorodsmay be the main cause of toxicity (Tanget al. 2015). A different conclusion was reached in thestudy of Aueviriyavit et al. (2014) in which AuNPs hadno effect on ROS generation in intestinal cells (Caco-2).Interestingly, the results of Tang et al. (2015) are inagreement with what was reported by Schlinkert et al.(2015). As already described, these authors observedthat positively charged AuNPs induced ROS in respira-tory cells and, in particular, that ROS production wasmost prominent in A549 cells. These results are partic-ularly interesting in that ROS formation has been linkedto inflammation and apoptosis (Schlinkert et al. 2015).

In the attempt to recreate in vitro the microenviron-ment of the lung alveoli, Breitner et al. (2015) set up anexperimental model consisting of A549 cells and artifi-cial alveolar fluid (AAF) containing salts supplementedwith phosphatidylcholine, exposed to a dynamic flow ataccurate physiological rates. Data obtained from theanalysis of tannic acid–coated 60 nm AuNP spectralprofiles and DLS measures showed that incubation withAAF resulted in loss of stability and enhanced particleagglomeration and sedimentation. Consequently, whenAuNPs in AAF were added to the cells in static condi-tions, NP deposition was more than double compared to

AuNP suspended in medium, due to the presence oflarge agglomerates. Introducing the variable of dynamicflow did not cause any change because the flow rate, setto low to mimic physiological conditions, was notstrong enough to affect NP deposition. By TEM analy-sis, AAF-suspended AuNPs in static conditions wereobserved as aggregates both bound to the cell surfaceand within intracellular vacuoles, while in dynamicconditions, all AuNPs remained associated to the outersurface of the plasma membrane and no internalisedparticles were identified. This might be due to the mor-phological changes caused in the cells by the experi-mental microenvironment, i.e. cell elongation and mem-brane alterations induced by AAF and dynamic flow,respectively. This means that both fluid environmentand flow status can influence NP deposition andinternalisation, confirming the importance of incorpo-rating as many physiological variables as possible inin vitro models. These results suggest that in physiolog-ical conditions, inhaled NPs are not efficientlyinternalised into lung cells. However, further studiesare required to confirm this hypothesis. For example,the contribution of surfactant proteins, not considered inthis study, should be taken into account because theymay deeply affect the surface chemistry of the NPs and,consequently, their stability, aggregation status and in-teractions with cells (Lundqvist et al. 2008; Marchettiet al. 2015).

Exposure to aerosol experiments can be performed inrespiratory cells grown in submersed culture or kept inair–liquid interface (ALI) conditions, a more advancedmodel of respiratory exposure. In the ALI system, cellsare cultured on transwell insert, which are placed into aculture well; the medium is supplied from the basal site;and cells are exposed to an aerosol at the apical part. Theexposure apparatus consists of a nebuliser, an exposurechamber and a flow system, that mimics the physiolog-ical conditions of aerosol inhalation in the lungs morerealistically (Lenz et al. 2009). Brandenberger et al.(2010a) using this exposure system on cells grown inALI conditions reported the quantitative analysis ofcitrate- and PEG-coated AuNP uptake in A549. Here,the authors describe how surface coating may changethe internalisation pattern of the particles. For example,citrate-stabilised AuNP uptake, which was the mostefficient, occurred mainly by macropinocytosis, whilePEG-coated AuNPs were internalised mostly by caveo-lin–clathrin-mediated endocytosis. Successively, usingthe same exposure system, Bachler et al. (2015) studied

J Nanopart Res Page 17 of 41(2020) 22: 235 235

AuNP translocation across the lung epithelial tissuebarrier combining in vitro and in silico methods. In thisstudy, the authors confirmed that alveolar epithelialcellular monolayers grown for few days on cell cultureinserts and exposed to the particles at the air–liquidinterface can represent the air–blood barrier to modelAuNP translocation. Moreover, they used a physiolog-ically based pharmacokinetic model to predict the dis-tribution of AuNPs to secondary organs. Results fromthis research showed similar translocation kinetics ofAuNPs through the human A549 and murine MLE-12alveolar type II cell lines, suggesting the absence ofspecie-specific mechanisms. The translocated fractionwas independent of the applied dose (up to 100 ng/cm2) and, as expected, was inversely proportional tothe particle size. These results are encouraging sincethey show the feasibility of improving in vitro modelswith the contribution of in silico models to more realis-tically mimic in vivo exposure.

To improve the predictive value of in vitro experi-ment of exposure to NPs, advanced culture models havebeen developed by several authors of the same researchgroup (Rothen-Rutishauser et al. 2007; Brandenbergeret al. 2010b; Fytianos et al. 2017; Durantie et al. 2017).They set up a triple cell co-culture model of the humanairway wall (also called 3D lung model) composed ofepithelial cells (A549), monocyte-derived macrophages(MDM) and monocyte-derived dendritic cells (MDDC)from human blood monocytes. A549 cells were grownfor a few days on cell culture inserts (polyester mem-brane, 3.0 μm pores) before MDM addition on top ofthe epithelial monolayer and MDDC underneath theinsert. Both the material and the size of the pores arevery important since it is known that pores as small as0.4 μm and polycarbonate membranes are likely tohinder the free passage of NPs (Fröhlich 2018). In anearly study, they found that in this cell model, exposedto bovine serum albumin-coated 25 nm AuNPs as sus-pension in submerged conditions, there was a significantincrease of the pro-inflammatory cytokine tumour ne-crosis factor-α (TNF-α). Moreover, by energy filteringTEM, they showed that AuNPs could be seen in all celltypes as free particles and also in the nucleus (Rothen-Rutishauser et al. 2007). To obtain an even more accu-rate alveolar model, Brandenberger et al. (2010b) usedthe same triple cell co-culture in ALI conditions. Whenthe triple cell culture was ready, the medium was re-moved from the upper transwell chamber and the cellswere kept in air–liquid interface conditions for 24 h

prior to particle exposure. The cultures were then trans-ferred to the exposure chamber where they were incu-bated with aerosolised 15 nm AuNPs for 20 min andkept in air–liquid interface conditions for different post-exposure times. In these conditions, which mimic parti-cle deposition in vivo, not only particle agglomerationwas decreased, but it was possible to obtain a morehomogeneous and efficient particle deposition. In thissystem, no inflammatory response was induced byAuNP treatment, and although the particles were ableto enter the cells and translocate, they were seen inintracellular vesicles (Brandenberger et al. 2010b) andnot free in the cytoplasm as in the previous report(Rothen-Rutishauser et al. 2007). As expected, differentways of exposure elicit different cell response and re-sults obtained in air–liquid interface conditions(Brandenberger et al. 2010b) contradict with what wasreported in submerged conditions (Rothen-Rutishauseret al. 2007). On the other hand, these results on uptakeare in agreement with those obtained with classicalexperiments showing that AuNPs enter the cells byendocytosis and therefore can be found inside the cellssurrounded by lipidic membranes (Choi et al. 2012;Uboldi et al. 2009; Freese et al. 2012; Tang et al. 2015).

Fytianos et al. (2017) used the aerosol exposuresystem to expose a 3D lung model grown in ALI con-ditions to poly(vinyl alcohol) (PVA)-coated AuNPsfunctionalised with an antibody specific for dendriticcells (DC-SIGN), plus a negative (PVA-COOH) or apositive (PVA-NH2) surface charge. They evaluated thepossibility of using nanocarriers to target immune cellswith the aim of enhancing or suppressing the immuneresponse in the lung. As expected, they found a signif-icant increase of particle uptake by dendritic cells. Asregards the non-DC-targeted AuNPs, the positivecharged ones were taken up more easily, especially bymacrophages, while uptake was low for all NPs in A549cells. None of the particles induced any cytotoxic effectthat could be ascribed to the gold itself, nor increasedcytokine secretion. These results are in agreement withthe observation that the positive surface charge canenhance AuNP uptake efficiency (Hauck et al. 2008;Alkilany et al. 2009; Cho et al. 2009; Liang et al. 2010;Liu et al. 2013).

In a biodistribution study, Durantie et al. (2017)investigated the effect of NP aggregation on cellularuptake and translocation in the same multicellular lungsystem described above. Citrate-capped 14.5 nm coresize AuNPs were prepared and coated with tiopronin to

Page 18 of 41 J Nanopart Res (2020) 22: 235235

add carboxylic groups on the surface. AuNP aggrega-tion was obtained by protonation of the negativelycharged groups. Both single and aggregated tiopronin-coated AuNP solutions were stabilised with a polymermixture of PVA/PAAm (poly(allylamine)) generatingparticles with final hydrodynamic diameters of 32 nmand 106 nm, respectively. These sizes are consistentwith TEM images showing that each aggregate wascomposed by a mean of 4 single AuNPs. Cells wereexposed to aerosolised AuNPs onto the air–liquid inter-face for 4, 24 and 48 h and no cytotoxic or pro-inflammatory effect was observed. Both single and ag-gregated AuNPs were internalised, and although theuptake of aggregated AuNPs was faster, the final resultwas similar. Also, the translocation rate across the lungbarrier model was not affected by the aggregation statusof AuNPs and was quite low (2–5%), similar to whatwas found in vivo (Lipka et al. 2010; Schleh et al.2013a; Kreyling et al. 2014). Taken together, theseresults suggest that the aggregation status of AuNPsuspensions can affect cellular uptake kinetics earlyduring exposure. Moreover, this study proposes a so-phisticated in vitro model, able to mimic NP behaviourand cell response in a realistic way, as a useful tool forthis kind of studies.

Results of the studies on the different alveolar cellmodels are summarised in Tables 3 and 4.

Results of studies in the different alveolar cell sys-tems showed that AuNP surface charge influences theirinteraction with cells; hence, these NPs, generally as-sumed inert, can become toxic if coated with positivelycharged molecules. Moreover, advanced in vitro lungmodels and exposure systems suggest that the closer weget to physiological conditions, the lower is the NPtoxicity. Concerning the air–liquid interface system toexpose lung cells, it must be taken into considerationthat it requires cell culture inserts so, in these studies,AuNP toxicity has been generally assessed analysingcell morphology, cell layer integrity, release of the cy-tosolic enzyme lactate dehydrogenase and pro-inflammation response. Therefore, the difference inAuNP toxicity could be due to the different parametersexamined.

Overall, the in vitro works highlight a lack of infor-mation on the preparation of AuNP suspensions in theculture medium used in the assays and on dispersionprocedures. Not all authors perform the characterisationof the AuNPs suspended in the culture medium, crucialto understand the changes in their physico-chemical

characteristics, such as agglomeration, surface chargeand the adsorption of serum protein that influence theinteractions with cell systems.

Taken together, results from in vitro studies showthat surface charge influences AuNP interaction withcells and, consequently, their cytotoxicity. NP uptakemechanism can be affected by surface charge, but un-fortunately, this aspect has not been considered by mostauthors. However, these investigations have an impor-tant flaw, as charge and size are not studiedsimultaneously. As a matter of fact, only Lebedováet al. (2018) utilised two NPs of very different sizewithout considering the influence of charge, whereasthe other studies, that considered the influence ofcharge, have the limitation of using NPs in a verynarrow size range.

In vivo studies

As reported above, NPs can be inhaled or ingested whileusing products containing ENMs, and it is well knownthat inhalation represents the most common route ofexposure to NPs in the workplaces (Yah et al. 2012).

NPs can be inhaled in the form of aerosol or powdersuspensions, or artificially administered by instillationinto the respiratory tract for toxicity studies. Once NPshave reached the lungs, they can remain there or enterthe bloodstream and move to other organs or cells, withthe probability of bioaccumulation in the heart, thespleen, the kidneys, the bone marrow and the liver.

As shown in the previous paragraphs, although sev-eral cytotoxicological studies have been conducted onAuNPs, there is no systematic procedure that has givendefinitive results. This is because the different studieshave used not only different biosystems, but also, re-gardless of the size and shape of the particles, varioustypes of coating. These variants, such as size, surfacecharge, etc., are extremely important, since the mole-cules or ions that surround the AuNPs produce firstcontact with the biosystem of interest. Consequently,no definitive conclusions could be drawn from theseinvestigations.

In vitro studies focussing on the toxicological effectsof AuNPs greatly outnumber in vivo studies; the latterare obviously more demanding, but take into accountthe complexity of the organism, hence more in vivotoxicology studies are needed to assess AuNP safety,including long-term studies for workers’ exposure. Al-though inhalation represents the main route of exposure

J Nanopart Res Page 19 of 41(2020) 22: 235 235

Tab

le3

AuN

Ptoxicity:invitrostudieson

alveolar

cells

Invitrosystem

Size

(nm)

Doserange

Coatin

gMainresults

Conclusions

Reference

A549

330.01–0.12μM

Citrate

NPs

wereinternalised

bythecells

and

inducedapoptosis.

Noresponse

inothercelllin

es.

Cellspecificity

Patraetal.(2007)

A549

NCIH

441

9 11 25

50–70μM

None,citrate

NPs

wereinternalised

byactiv

eendocytosis.

Citratecoatingim

paired

cellviability

and

proliferation.

Citrateshould

bereducedto

aminim

umlevelfor

biom

edicalapplications.

Uboldietal.(2009)

Freese

etal.(2012)

A549(A

LI)

150.02

μM

aCitrateor

PEG

Citrate-stabilisedAuN

Puptake

occurred

mainlyby

macropinocytosis.

PEG-coatedAuN

Puptake

occurred

mainly

bycaveolin–clathrin-mediatedendocytosis.

NPsurfacecoatingmodulates

endocytotic

uptake

pathways.

Brandenberger

etal.(2010a)

A549

NCIH

1975

175–35

μg/ml

Citrate

NPs

foundin

mem

brane-boundvesicles

inA549cells.

NPs

affected

cellviability

andinduced

mem

branedamageandapoptosis.

Bothintrinsicandextrinsic

apoptotic

pathwaysare

induced.

Choietal.(2012)

A549

5–6

10–20

20–50

0.01–0.2

μM

NaB

H4,

citrate,SA

AuN

Psinducedcytotoxicity,oxidativ

estress

andapoptosis:NaB

H4>citrate>SA

.Biologically

synthesised

AuN

Psareless

cytotoxic.

Sathishkum

aretal.(2015)

A549

MRC5

10×39

10×41

10×45

0.072–0.72

μg/ml

None

Intracellularaccumulation

Slow

ingof

cellgrow

thrate,noapoptosis

NoROSproductio

n(A

549)

Cell-im

pedancemeasurement

system

isreliableforthe

evaluatio

nof

NPtoxicity.

Chuangetal.(2013)

A549

20 40 60

12–25pM

Citrate

40nm

AuN

Psweremoretoxicthan

20and

60nm

.40

nmAuN

Psinduced

microtubuleaggregation

andapoptosis.

Choudhury

etal.(2013)

A549

25×52

2.5–15

μg/ml

None

NPs

foundmainlyin

mem

brane-bound

vesicles.D

ose-dependentcytotoxicity

Oxidativ

estress

isthemain

causeof

toxicity.

Tangetal.(2015)

A549+AAF+

dynamicflow

6025

μg/ml

Tannicacid

AAF-suspendedAuN

Pswereobserved

with

inintracellularvacuoles,w

hilewith

dynamicflow

,noNPs

wereinternalised.

Slight

toxicity

with

dynamicflow

.Absence

ofoxidativestress

Feasibility

ofim

provingin

vitromodels

Microenvironm

entaffectsthe

nano-cellularinterface.

Breitn

eretal.(2015)

A549

MLE-12(A

LI)

2 7 18 46 80

25–200

ng/cm

2Citrate

NPs

foundfree

inA549cytoplasm.

Translocatio

ninverselyproportio

naltoNPsize

Feasibility

ofcombining

invitro

andin

silicomethods

toreplace

anim

alstudies

Bachler

etal.(2015)

A549

7 10100–700ng/cm

2Citrateor

chito

san

Cellv

iabilityandmem

braneintegrity

were

notaffected.Anionic(citratecoated)NPs

didnotinduceROS,w

hilecationic(chitosan

coated)NPs

inducedthehighestamount

ofROS.

Divergent

cytotoxicity

datain

differentcelltypes

bSchlinkertetal.(2015)

A549

15–44

18–75μg/ml

None

Nocytotoxicity

orapoptosisinduction

Green-synthesised

NPs

canbe

used

inbiom

edicalapplications.

Park

etal.(2017)

Studiesarepresentedin

chronologicalo

rder

inwhich

they

werepublished

SA,staranise;AAF,artificialalveolarfluid;

ALI,air–liquidinterface;PEG,polyethyleneglycol;R

OS,reactiv

eoxygen

species

aSu

spension

fordropletg

eneration

bSeeresults

onbronchialcellsreported

inTable2

Page 20 of 41 J Nanopart Res (2020) 22: 235235

Tab

le4

AuN

Ptoxicity:invitrostudieson

3Dlung

models

Invitrosystem

Size

(nm)

Doserange

Coating

Mainresults

Conclusions

Reference

A549

MDM

MDDC

2510

10particles/ml

BSA

AuN

Pscouldbe

seen

inallcelltypes

asfree

particles(translocatio

n).

Inductionof

asignificantincrease

ofthepro-inflam

matorycytokine

TNF-α

NPsize

andmaterial

affecttheirtoxicological

potential.

Rothen-Rutishauser

etal.(2007)

A549

MDM

MDDC(A

LI)

1540–400

μg/mla

Citrate

AuN

Pswereseen

inintracellular

vesicles

inallcelltypes

(translocatio

n).A

bsence

ofpro-inflam

matorycytokine

productio

nor

oxidativestress

induction

Lackof

adverseeffects

might

besize

related.

Brandenberger

etal.(2010b)

A549

MDM

MDDC(A

LI)

66 111

100

175

120μg/mla

PVA-COOH(negativelycharged)

PVA-N

H2(positively

charged)

PVA-COOH+DC-SIG

NPVA-N

H2+DC-SIG

N

Low

uptake

inA549cells

Potentialtargetin

gand

activ

ationof

immune

cells

inthelung

bygold

nanocarriers

Fytianosetal.(2017)

Low

uptake

inA549cells

NPs

weretakenup

moreeasily,

especially

bymacrophages,

andinducedapoptosis.

Low

uptake

inA549cells

Antibodyspecificfordendritic

cells

(DC-SIG

N)significantly

increasesuptake

inMDDC.

A549

MDM

MDDC

(ALI)

3250–500

μg/mla

Citrate

+tio

pronin

+PV

A/PAAm

(single)

Nocytotoxicor

pro-inflam

matory

effectwas

observed.

Bothsingleandaggregated

NPs

wereinternalised.

AggregatedAuN

Puptake

was

faster.

Translocatio

nrateacross

thelung

barrierwas

notaffectedby

the

aggregationstatus

ofAuN

Psandwas

similarto

whatw

asfoundin

vivo.

Durantie

etal.(2017)

106

Citrate+tio

pronin

+PVA/PAAm

(aggregated)

Studiesarepresentedin

chronologicalo

rder

inwhich

they

werepublished

MDM,m

onocyte-derivedmacrophages;M

DDC,m

onocyte-deriveddendritic

cells;A

LI,air–liquidinterface;PVA,poly(vinylalcohol);DC-SIG

N,functionalised

with

anantib

odyspecific

fordendritic

cells;P

AAm,poly(allylamine)

aSu

spension

fordropletg

eneration

J Nanopart Res Page 21 of 41(2020) 22: 235 235

for workers, researchers and consumers, intravenousinjection of AuNPs is often used for in vivo studies.As it is easily understandable, this method of NP ad-ministration does not reflect reality, and results fromthese studies are not predictive of the risk associated toAuNP use and manipulation. To overcome these limi-tati