Nanosafety forum for young scientists: a meeting report

Material session

The physico-chemical characteristics of nanoparticles probably represent the most important properties linked to their toxicity, and the production of “safe-by-design” nanoparticles is a strategic objective in nanosafety. In this perspective, advances in the synthesis and characterization of nanoparticles are very relevant.

Four themes were discussed in this session: A synthesis protocol for the production of polyvinylpyrrolidone (PVP) capped metal oxide nanoparticles, and the modifications which these nanoparticles may undergo with time (presented by S. M. Briffa, University of Birmingham), the choice of the best procedures for the dispersion of nanoparticles tested in toxicity studies, (M. Correia, Technical University of Denmark, Copenhagen), the advances in radiolabeling of nanoparticles (M. Dal Miglio, JRC Ispra), and the induction of reactive oxygen species (ROS) in cell free systems (G. Kilic, JRC Ispra).

Briffa’s presentation showed that the protocol for hydrothermal synthesis of PVP capped ceria nanoparticles (2) may be successfully applied to produce PVP capped ZnO and PVP capped Cu₂O nanoparticles. Furthermore, preliminary data suggested that different temperature may influence the size of nanoparticles with time, with no modification of ceria nanoparticles at room temperature up to 60 days, and formation of large aggregates after 21 days at higher temperature.

M. Correia compared two techniques for the dispersion of CuO nanoparticles: ball-milling and probe sonication. Although ball-milling with ZrO₂ beads in dilute acetic acid suspensions resulted in monodispersed nanoparticles with smaller mean hydrodynamic diameters compared with the other technique, it introduced artifacts, represented by clusters/ions release into the solution, so the presenter suggested that dispersion in water by probe sonication should be preferred, in spite of the formation of aggregates. Standardization procedures, such as calibration of the ultrasound probe by calorimetry may lead to similar suspensions across different laboratories (3). M. Correia was selected as the third best presenter.

The third speaker, M. Dal Miglio, presented an overview of the methods developed at the Cyclotron facility of JRC for the labeling of nanoparticles with gamma emitting radionuclides with long enough half-lives for tracing studies over several weeks or months (4, 5). On the basis of this experience, he suggested that intrinsic radiolabel is desirable (e.g. ¹⁰⁵Ag in Ag-nanoparticles); in cases in which this procedure is not technically possible, extrinsic radiolabelling (e.g. ⁵⁶Co in SiO₂ or ⁷Be in multi-walled carbon nanotubes [MWCNTs]) is also a possibility, but radiolabel stability needs to be carefully determined. More work needs to be done to establish radiolabeling protocols and characterise the radiolabeled NPs, for different particle types.

Finally, G. Kilic showed data on the production of excess ROS induced by superparamagnetic iron oxide nanoparticles (SPION). In fact, excessive ROS production is perhaps the most relevant pathophysiologic pathway in nanoparticle toxicity (6). The ROS generating capacity of iron oxide nanoparticles with different surface coatings (silica or oleic acid) was studied in an acellular assay, in which the non-fluorescent dye H2DCFDA becomes green fluorescent when oxidized (7). Ten different doses of nanoparticles, tested in four different acellular systems were mostly unable to produce ROS. Although reassuring, the finding should be checked in experiments including a cellular component.

Exposure session

Six presentations were included in this session.

C. Becker (Universiteit Utrecht and TNO) reported data of the NanoNextNL project dealing with occupational exposure to Nano-Objects, their Agglomerates and Aggregates (NOAA) by means of a multi-metric exposure assessment approach (8). The goal of the project is to evaluate, through different information sources and decision logics, the likelihood and quantification of exposure to NOAA. It was concluded that the decision logic used in this survey allows the overall evaluation of exposure to NOAA and gives semi-quantitative but reliable indication of the level of exposure.

The matter of agglomeration was evaluated in the perspective of exposure by Y. Ding (IST, University of Lausanne and Geneva), who reported data on the patterns of agglomeration/de-agglomeration of aerosolized SiO₂, TiO₂, CeO₂, and ZnO nanoparticles (9). The study showed that the stable dispersion formed with aerosol production may be modified by pressure drop, which seems to break up large agglomerates, thus increasing particle number. Ranking of deagglomeration potential of engineered nanoparticles seems possible.

The accidental release of TiO₂ in a simulated indoor environment was reported by G. Kylafis. In the experiment, nanoparticles were nebulized for 70 min in a closed chamber under controlled conditions. The evolution of the particle number concentration over time was characterized by a rapid elevation (in less than 1 minute) and a significant decay within 1 hour. During this period, a balance of coagulation and deposition had occurred and led to smaller shifts in modal diameters.

M. Levin and B. Liguori (Department of Micro- and Nanotechnology, and Department of Environmental Engineering, Technical University of Denmark, respectively) treated the issue of occupational exposure to nanoparticles by comparing the performance of different instruments (10) and control band tools, respectively (11). Levin reported large differences between fast response particle sizing instruments for workplace-like aerosols, whereas a good agreement between scanning mobility particle sizer (SMPS), fast mobility particle sizer (FMPS) and electrical low pressure impactor (ELPI) (secondary instruments) was detected up until 200 nm; Liguori performed an overview of available tools and found a wide range of variability in intended use, approach for the emission potential, use of the same input parameter. The conclusion of the two talks were that secondary instruments should, if possible, be used to confirm data obtained with fast response instruments (Levin) and that several limitations still exists for currently available control banding tools (Liguori).

The last presenter of this session, B. Faure (CEA, France), focused on the relative relevance of two metrics (mass vs number concentration) which may be used for assessing exposure to nanoparticles in the case of personal sampling (12). By checking data obtained with mono- and polydispersed ZnO aerosols, he showed that whereas the mass of zinc measured for the personal nanosampling device is in good agreement with the mass converted from the aerosol counters and granulometers, the same is not true for number concentration.

Human toxicology session

This session was characterized by the highest number of presentations and was therefore divided in two sub-sessions: human toxicology in vivo (7 presentations) and in vitro (6 presentations).

In the in vivo subsession, C. Pavan (Department of Chemisty University of Torino) confirmed that the hemolytic activity of silica nanoparticles may be related to their physico-chemical properties, and showed data still unpublished at that time on a strict correlation between the level of induced hemolysis and the serum concentration of IL-1 β, a marker of inflammation (13). The challenge is to identify specific physico-chemical features linked both to hemolysis and inflammation.

The correlation between specific physico-chemical properties of nanoparticles and toxic response was also the focus of the presentation given by P. Tassone (Department of Occupational and Environmental Medicine, Epidemiology and Hygiene, INAIL, Italy) who investigated the cyto-genotoxic and oxidative effect induced by 24 h exposure to two commercial TiO₂ nanoparticles in human bronchial normal cell line (BEAS-2B).

The two nanoparticles (one of which was mainly anatase, and the other one mainly rutile) had different shape and different tendency to agglomeration and showed different responses (which were more marked for genotoxicity). Also the data of this presentation have been recently published (14).

M. Leppanen (University of Eastern Finland, Kuopio) studied the respiratory pattern in mice after inhalation of nanosized silicon carbide and detected only a limited capability of this nanoparticle to induce irritation or airflow limitation, differently from what has been reported for other silica containing nanoparticles (15).

A lack of toxicity was reported also by L. Vecchione (Department of Biomedicine and Prevention, University of Rome Tor Vergata), who, in contrast with previous data of acute exposure to nanoparticles during pregnancy (16, 17), did not observe any effect on the reproductive ability of female mice after chronic exposure before pregnancy to low dose ZnO and TiO2 nanoparticles.

Another reassuring report was released by M. Mauro (Unit of Occupational Medicine, University of Trieste) showing no penetration of TiO₂ nanoparticles through intact or damaged skin, a finding consistent with the vast majority of studies currently available on this topic (18).

The potential usefulness of digital holographic microscopy and flow cytometry in assessing the shape-dependent toxicity of Ag nanoparticles was highlighted by S. Mues (BioMed Tech, Munster), who showed that these label-free techniques are very useful in giving insights into mechanistic pathways underlying the toxicity of this nanoparticle assessed with traditional techniques.

The talk given by L. A. Santiago (Universitè Paris Diderot) regarded the oxidative potential of several metal oxide nanoparticles on different cell lines. The presenter showed that soluble nanomaterials induced higher cytotoxicity on lung cells as a consequence of their higher reactivity.

In the in vitro subsession, K. Aimonen (FIOH, Finland), discussed the potential genotoxicity of four Nanofibrillar cellulose (NFC) materials. Toxicity was assessed both in vitro (BEAS 2B cells) and in vivo (C57BL/6 mice). Whereas in vitro the tested materials did not induce significant DNA double strand breaks or chromosomal damage as measured by the comet and micronucleus assays, in vivo one of the four tested NFC showed a transient genotoxic effect. This finding suggested that more evoluted in vitro models are needed for a reliable transposition of the data to living organisms.

The role of TiO₂ nanofiber (NF) length on cell viability and barrier competence was the topic of the talk given by M. Allegri (Department of Biomedical, Biotechnological and Translational Sciences University of Parma), who showed that the reduction of TiO₂NF aspect ratio mitigated the effects on cell viability and barrier competence, however it enhanced the pro-inflammogenic activity of the nanoparticle.

The length-toxicity relationship was also treated by L. Di Cristo (Department of Clinical and Experimental Medicine, University of Parma). The presenter showed that long MWCNTs caused more marked toxicity on macrophages and alterations of barrier competence in lung epithelial cells; however short MWCNTs caused M1 macrophage activation, which was not altered by long MWCNTs (19).

H. Libalova (Institute of Experimental Medicine, Poland) investigated the cyto-genotoxic potential of four fractions of different size of particulate matter (PM) and did not detect significant differences between the ultrafine (UF) and non UF components, except for a slightly higher induction of oxidated nucleotide by UF. The talk also highlighted the technical difficulties for the efficient fractionation of different size group PM.

The role of components of the protein corona in modulating toxicity of nanoparticles was the topic presented by A. Marrucco (Department of Chemistry and NIS, University of Torino), who reported about experiment involving the interaction of nano SiO₂, nano TiO₂ and CNTs with fibrinogen, an important component of the protein corona. Only nanoparticles covered by fibrinogen induced alterations in alveolar macrophages, monitored as production of NO and inducible nitric oxide synthase (iNOS) activity.

The last presenter, C. Muoth (EMPA, St. Gallen, Switzerland), selected as the best presenter of the forum, reported about the development of a novel 3 D in vitro model of the human placenta. The presentation was appreciated by the audience, which recognized that this in vitro system may represent a promising tool to study the nanoparticle induced mechanisms of toxicity on the embryo/fetus.

Ecotoxicology session

This session included nine presentations, two of which regarded the possible toxic effects of polystyrene nanoparticles (E. Bergami, Department of Physical, Earth and Environmental Sciences Università di Siena) and of titanium dioxide (A. Pinsino, Istituto di Biomedicina e Immunologia Molecolare, Università di Palermo) on the sea urchin. Carboxylate polystyrene did not induce any toxic effect on the sea urchin embryo, while the aminated form did, probably as a consequence of a better dispersion of the aminated form in seawater, and/or of the different surface charge (20). In the study reported by Pinsino, TiO₂ NOAA were injected into the sea urchins body cavity to expose immune cells. It was found that TiO₂ NOAA elicited a receptor-mediated endocytosis by phagocytes, involving the p38 MAPK signaling pathway. Sea urchin immune cells display functions similar to those of the vertebrates, thus representing a potential good model to investigate molecular pathways of NP phagocytosis (21).

In the presentation of E. Bocci (Department of Physical, Earth and Environmental Sciences Università di Siena) algae, zooplankton and sea urchin were the marine organisms taken as models and mesoporous TiO₂ the studied nanoparticle. It was found that mesoporous TiO₂ per se did not exert any toxic effect, however in the case of co-exposure with Cd, the toxicity of Cd was increased. Since mesoporous TiO₂ has been proposed as polluttant removal (including metals) this finding highlights the possibility of indirect toxic effects of seemingly safe nanoparticles.

A. Bermejo-Nogales (INIA, Madrid) described the topminnow fish (Poecilipsis lucida) hepatoma cell line (PLHC-1) as an in vitro model to elucidate the mechanisms of toxicity underlying the action of nanoparticles in fish cells. Cytotoxic effects were mainly produced by silver and zinc oxide nanoparticles, whereas the remaining tested nanoparticles, such as CeO₂, MWCNTs, SiO₂, and TiO₂ tested atoxic.

The ability of nanoparticles to induce oxidative stress was discussed also in this session by Zhao J (University of Lausanne). Airborne generated Ag nanoparticles, coated with low volatile organic compounds (LVOC) did not show a significant decrease in their ability to induce ROS generation in acellular environment; by contrast, Ag nanoparticles coated with paraffin almost completely lost the ROS generating ability. This finding shows that inert and active coatings have different effects on the modulation of nanoparticle toxicity.

The theme of hierarchy in the toxic effects of nanoparticles was the focus of the presentations of M. Heinlaan and of M. Sihtmäe (both from the National Institute of Chemical Physics and Biophysics, Tallin). In the study reported by Heinlaan, bacteria, yeast, protozoa and Daphnia magna were challenged with several nanoparticles, showing the following toxicity order: nAg>nZnO>nCuO>nSiO₂=MWCNTs=nTiO₂. nSiO₂, nTiO₂ and MWCNTs were not toxic in any screening test (EC50>100 mg/L) and nano Ag was toxic to all organisms, especially to Daphnia magna.

The study presented by Sihtmäe regarded bacteria and protozoa. In bacteria, CuO, ZnO, Sb₂O₃, and WO₃ had EC50<100 mg/L. CuO, ZnO, Fe₃O₄, Mn₃O₄ and Co₃O₄ inhibited bacterial colony forming ability at concentrations ≤100 mg/L. In protozoa, the most toxic NPs were ZnO and CuO. These two nanoparticles tested as most toxic in both species, probably due to the release of metal ions.

S. Bohme from the Helmholtz Centre for Environmental Research (Leipzig) was selected as the second best presenter. Bohme presented a new method coupling a Laser Ablation system with an ICP-MS (LA-ICP-MS), allowing to coupling the reliability of the quantitative data obtained by ICP-MS along with the crucial information about the distribution of nanoparticles in tissues. She tested the reliability of the system in zebrafish embryo by investigating the time and concentration dependent nanoparticle uptake (AgNP, AuNP, Al₂O₃-NP) and the accumulation areas inside the organisms. Through these measurements the embryo chorion resulted a biological barrier for the tested NPs. Data on the application of this technique to quantify nanoparticles in human cell lines have been recently published (22).

Finally, G. Pulido-Reyes (Universidad de Alcalà) talked about the effect of colloidal stability and valence states at the surface of cerium oxide nanoparticles on cellular toxicity. The effect of seven different CeNPs were evaluated on a model aquatic microorganism, the green alga Pseudokirchneriella subcapitata; it was found a good positive correlation between the surface percent of Ce³⁺ and toxicity (related with an increase in intracellular oxidative stress). By knocking out the oxygen vacancy of Ce³⁺ the toxicity and oxidative stress was totally prevented. Conversely, colloidal stability did not modulate the cellular toxicity.

Systems biology

The Systems biology session included two presentations given by H. Bouwmeester (Wageningen UR) and M. D. Driessen (German Federal Institute for Risk Assessment, Berlin). H. Bouwmeester reported about experiments in which silver nanoparticles of different size and AgNO₃ (control) were administered to two cell lines (CaCo-2 and MCF-7); the following endpoints were evaluated: viability (WST-1 assay), uptake (inductive coupled plasma-mass spectroscopy [ICP-MS], confocal microscopy), gene expression (Affymetrix arrays), and systems biology (hierarchical clustering and pathway analysis connectivity map, pathway analysis, gene set enrichment analysis). Gene expression profiles were altered in a similar manner by AgNO₃ and by silver nanoparticles, although silver nanoparticles seemed to exert and additive effect.

In the study presented by Driessen, the oxidative potential of 12 nanoparticles (each in different form) was evaluated in one acellular system (electron spin resonance) and in two cell-based systems (dichlorofuorescein [DCF] assay and the analysis of protein carbonylation, respectively). Protein carbonylation showed a good fit with other in vitro tests, furthermore, it was observed that each nanoparticle generated a unique pattern of protein carbonyl.

Databases session

Two presentations, regarding the expertise developed in the context of two EU-FP-7 projects (NanoReg and eNanoMapper) were proposed in this session.

T. Oosterwijk (TNO Amsterdam) showed the interim results of an inventory of physico-chemical parameters of nanoparticles selected in NanoReg with the purpose to link them to their (eco)toxicological properties.

N. Jeliazkova (Idea Consult Ltd., Sofia) made a review of available databases regarding nanoparticles, with particular reference to the eNanoMapper database, emphasizing the need of adequate recording and the possibility of integrating different databases by showing the most used techniques for this purpose.

Dissemination and standardization session

Also in this session just two young scientists presented their work. The first presentation (A. Porovic, Nanotechnology Industries Association) described the project “Nanodiode”, aimed to developing innovative outreach and dialogue on responsible nanotechnologies in EU civil society, having as starting point the fact that people have concomitantly great expectations and concerns in relation to nanotechnology, and that, therefore, active science participation and not passive engagement should be the target of scientists involved in studying the different aspects of this new technology.

The second speaker (V. Vaananen, Finnish Institute of Occupational Health) illustrated some basic concepts in the processes of standardization and dissemination and described the follow-up activities of the FP7 nanoSTAIR project which ended in February 2014, mainly consisting in establishing a process and a platform to support standardization for nanotechnologies.

Modeling session

The session on modeling included young scientists involved in short-term scientific missions for the COST project MODENA which is devoted to promote the development of reliable modeling for the assessment of toxicity of nanoparticles.

Two of the presenters, F. Messina and T. Tosco, were from the same Institution, (DIATI Politecnico di Torino), and presented a modeling approach regarding the transport of nanoparticles. Messina proposed a new equation offering some potential advantages in comparison to the existing ones (is always smaller than one; has not local maximum values; is valid also if only one or two transport mechanisms are inactive; is valid also for point particles), whereas Tosco described the application of an user friendly software for nanoparticle transport simulation which was successfully applied to transport in porous media of iron nanoparticles for the remediation of contaminated aquifer systems (23).

P. Kamath (Universitat Rovira i Virgili Tarragona) proposed a model based on the relative abundance and (relatively) simple properties of the proteins forming the biocorona to predict the interaction of nanoparticles with living cells. The model was successfully applied to gold nanoparticles with various modifications of their surfaces.

The interaction of nanoparticles with proteins was also the focus of the model proposed by M. Witt (Lab of Environmental Chemistry, University of Gdansk), regarding fullerene and its interaction with proteins of the broncho-alveolar lavage fluid. Six model complexes were built, and the following criteria were applied for selecting the most plausible fullerene binding sites: 1) frequency of the binding site in 1000 repeated docking trials; 2) binding energy of the complex; 3) relevance of a location taken by the ligand for the protein functioning. It was found that fullerene molecules tend to interact with active sites of proteins and interfere with the conformational changes of the proteins required in its activation process. Malfunctioning of proteins crucial in biochemical processes may therefore lead to toxic effects.

Risk assessment session

The topic discussed in this session is one of the most interesting in the field of nanosafety, so it is not surprising that the 4 presentations were followed by an active debate.

The talk of K. Gerloff (JRC) regarded Adverse Outcome Pathway (AOP), a conceptual framework that portrays existing knowledge on the links between a Molecular Initiating Event (MIE) and an Adverse Outcome (AO). As an example of the application of the concept, it was described the possible development of hepatitis after exposure to the following nanoparticles: TiO₂, ZnO, SiO₂, and CeO₂. On the basis of the information taken from different sources, putative AOPs for metal oxide-induced hepatitis were built. These nano-AOP may be useful for regulatory application and to inform smart testing strategies for high throughput screening (HTS).

I. Rio-Echevarria (JRC) highlighted the gaps and challenges in environmental risk assessment of nanoparticles. She concluded that exposure, environmental fate and environmental transport are the key aspects of nanoparticle environmental impact determination. The main challenge is represented by the difficulty to estimate nanoparticle concentrations in the environment because of unknowns in nanoparticle production volume and product life cycles.

V. Subramanian (Ca’ Foscari University of Venice) described a decision support system framework for sustainable nanotechnologies (24). To this end, an user-friendly software is being developed. By means of this tool, it should be possible to determine risk-benefit balance of nanoparticles and nanoproducts by estimating risk and impacts at each lifecycle stage, technical and cost effectiveness of the available technologies and best practices in risk reduction. The prototype should be ready in 2015, and will be tested to relevant case studies (e.g. copper oxide used as an anti-microbial wood coating).

In the last talk of this session, V. Vaananen (Finnish Institute of Occupational Health) presented a detailed guidance for monitoring the exposure to nanoparticles and health at the construction sites. Qualifying points of the guidance are: institution of nanoexposure registers, the type of mandatory information of the exposure register, its recording in excel format, a tiered approach to health monitoring.