Toxicology of metallic nanoparticles

Meeting report by Kate Jones
Health and Safety Laboratory
​​ECG Bulletin January 2012

On 24th June 2011, the RSC Toxicology group held a one‐day meeting on the ‘Role of metals in the toxicity of nanoparticles: Informing the regulation of nanoparticulate safety’.

The aim was to give an overview of the recent research in this area and to collate the toxicological opinions, particularly with reference to regulation. The current concern over the potential toxicity of nanoparticles (environmental, clinical and engineered) is often based on presumptions and uncertainties in their various modes of action. Greater understanding of the specific mechanisms would aid the risk assessment process.

Dr Andy Smith (MRC, Leicester) introduced the meeting giving his view of the current lack of knowledge of toxicological mechanisms for metallic nanoparticles (NPs) and the tendency to presumption of the mode of action by grouping certain species. He posed a number of questions for the day including:

• How might metallic nanoparticles contribute to toxic and carcinogenic processes?
  
• Does the nano state have a specific role to play, other than as a provider of metallic ions?
  

Professor Ken Donaldson (University of Edinburgh) then covered a number of studies that had looked at various aspects of metallic nanoparticulate toxicology and presented data that he had compiled from a number of sources. First, he presented a number of data that demonstrated that smaller particles were better translocated (e.g. gold where a 10‐fold increase in translocation across the lung air/blood barrier was seen for a 10‐fold reduction in particle size) and caused greater inflammation. The inflammation was related to surface area in a linear response, across a number of particle types. He then went on to discuss the role of surface chemistry – different metal oxide NPs with the same surface area showed markedly different free radical activity. For insoluble particles, positive acid zeta potential seems to increase inflammation responses. For soluble particles, the toxicological impact seems directly related to the toxicity of the soluble species, for example Cu²⁺ and Zn²⁺ ions (from metal NPs) are highly toxic and inflammogenic whereas Mg²⁺ ions are not. Professor Donaldson finished his presentation with a discussion of metallic nanofibres and whether they fit the ‘asbestos paradigm’, where the toxicity is related to the high aspect ratio (AR) of the fibre. Metallic nanofibres can exist as nanorods (AR up to 5) and nanowires (AR up to 1000). Inflammation tests have shown that there is very little response to fibres less than 5 μm in length; beyond this ‘trigger point’ there is a dramatic increase in inflammation, with an apparent length related increase up to ~20 μm after which the response plateaus.

Professor Terry Tetley (Imperial College, London) followed with a presentation on the reactivity of NPs in the lung, specifically at the alveolar interface. It is estimated that 50% of inhaled nano‐sized objects will reach the alveoli. The alveolar surface consists of two cell types, type one (AT1) epithelial cells which cover 95% of the alveolar surface and type 2 (AT2) which are progenitors to AT1 cells. Professor Tetley’s research group has undertaken a range of in vitro experiments, looking for AT1 cell responses to NP challenge. Little difference in IL‐8 response was seen on dosing with TiO₂, Ag, tungsten carbide (WC) or ZnO NPs. However, IL‐6 response was increased for all NPs and all doses with responses still elevated after 48 hours recovery (except for ZnO) although Ag and ZnO caused cell death at the higher doses. Polydispersed 244 nm Ag particles caused a significant increase in IL‐8 whereas other smaller Ag particles (5 – 44 nm) with various coatings (sugar, PVP, citrate) didn’t alter IL‐8 release. The use of foetal calf serum (FCS, widely used in tissue culture) can also influence cell death and IL‐6/IL‐8 release – for CuO NPs the TD50 was reduced 3‐fold in the presence of FCS compared to without, whereas for ZnO the TD50 was doubled in the presence of FCS (TiO₂ was unaffected). Experiments looking at the effect of different forms of TiO₂ on AT1 cell mediator releases showed that nanopowder (~600 nm) was no more likely to result in IL‐6, IL‐8 or MCP‐1 release than standard TiO2 (~950 nm), pure anatase (~300 nm) or pure rutile (~150 nm).

Next, Professor Jamie Lead (University of Birmingham) gave an overview of research into silver nanoparticles in the environment. In the environment NPs acquire coatings of organic matter (humus), this can cause aggregates of around 2.5 μm to disaggregate into ~30 nm particles over a 30‐day period. Silver NPs were shown to be highly toxic to Pseudomonas sp. when incubated in media alone but there was no observed effect on growth when incubated in humus‐containing media. By contrast humus had no effect on the high toxicity of silver nitrate. The media causes aggregation so the high toxic response of silver NPs is therefore not explained by dissolution.

Professor Frank Kelly (King’s College, London) presented results from a study on ambient particulate matter (PM) toxicity. It has been estimated that 29,000 deaths are due to airborne particles. In cities, most of the particulate matter comes from traffic pollution and the problem has been exacerbated by the increase in the use of diesel. Particles consist of a carbonaceous core with components adsorbed onto the surface (organics e.g. PAHs, metals, biological material e.g. endotoxins). Three sites (semi‐rural, urban residential and high urban traffic) were studied for oxidative potential of the PM collected at each site – the high urban traffic area showed greater oxidative potential (both as a fraction of PM and per unit volume air) than the other sites. The RAPTES (Risk of Airborne Particulates: A hybrid Toxicological‐ Epidemiological Study) study aims to assess the metal content and redox activity of ambient PM during human volunteer challenge. Volunteers were exposed to an underground train platform environment as well as a traffic intersection and an urban garden. Ambient iron‐containing PM concentrations were 100 times greater and PM0.18 concentrations were 300 times greater in the underground scenario. Nasal lavage samples showed that post‐underground levels of iron PM deposits were three times greater than for the other scenarios. There was also evidence of increased redox activity of the PM deposited in the nasal cavity and this activity was correlated to the iron content of the nasal lavage fluid.

Continuing the iron theme, Dr Shareen Doak (University of Swansea) is studying the genotoxicity of iron oxide NPs. Ultrafine superparamagnetic iron oxide nanoparticles (USPION) have potential medical applications in MRI imaging, drug delivery and magnetic tumour ablation. There are three different composition types – Fe₃O₄, γ‐Fe₂O₃ and α‐Fe₂O₃ and the supramagnetic properties rely on a particle size of <35 nm. In vitro experiments showed that serum content had an impact on apparent hydrodynamic diameter (at 1% serum average diameter was nine times greater than at 10% serum for the same nominal 10 nm particles). Cellular uptake of dextran‐coated Fe₂O₃ was three times greater in 1% serum than 10% and also three times greater than dextran‐coated Fe₃O₄. However, uncoated Fe₂O₃ was absorbed to the same extent as uncoated Fe3O4 and the amount of serum did not affect uptake. Fe₂O₃ particles were shown to cause dose‐dependent increased oxidative NA adducts and the cellular response was similar to iron overload in hepatocytes.

Dr Patrick Case (University of Bristol) concluded the day’s presentations with a study on cobalt and chromium NPs in vitro and in vivo. Metal‐on‐metal (MoM) replacement joints are becoming increasingly common as younger patients receive implants. Although MoM implants have been used since the 1930s, two new adverse reactions have been observed. Within five years of surgery, 1% of patients suffer ‘pseudotumours’ destroying local tissues, or local ‘hypersensitivity’ immune reactions. Long‐term risks include melanoma (up 43%), kidney (up 22%) and bladder (up 15%) cancers after 10 years. In vitro, CoCr NPs were more cytotoxic and caused more DNA damage than an equivalent micron‐sized particle dose. In vivo, blood and urine levels of Co or Cr are increased post‐operation from six months to at least 2 years. Chromosomal aberrations also increase. In 2010 a medical device alert was issued describing the potential adverse immune reactions. In cases where replacement surgery is required, corrosion of the implant has been observed.
Following the presentations there was a brief discussion of the issues. The general consensus was that nanoparticle toxicology seemed to be very complicated! There seemed no apparent clear case for groupings, although “low toxicity” NPs might be filtered out on surface area/inflammation ratios. It is also perhaps reasonable that nanofibres are considered on their physical properties, such as aspect ratio. It is clear that NPs are not a single entity and surface reactivity may be important in determining toxicology. Routes of entry also seem to be relevant and in vitro experiments may not always reflect what is seen in vivo.

KATE JONES
Principal Scientist,                                                          
Health & Safety Laboratory, UK

Reproduced with permission from the RSC Toxicology Group Newsletter, Autumn/Winter 2011.

Visit the Toxicology Group web pages at http://www.rsc.org/Membership/Networking/InterestGroups/Toxicology/Meetings.asp for the PowerPoint presentations from this meeting.