4. Does depleted uranium pose a radiation hazard?
The SCHER Opinion states
Uranium is the heaviest naturally occurring element and all isotopes of U are radioactive (Table 1). In order to produce fuel for nuclear reactors and material for nuclear weapons, U has to be "enriched" in the U-235 isotope , which is responsible for nuclear fission. During the enrichment process, the fraction of U-235 is increased from 0.72 % present in natural U to a content of U-235 between 2% and 94%. After removal of the enriched U, the remaining U has significantly reduced concentrations of U-235 and U-234, which is called DU (Table 1). Therefore, only the composition of isotopes is changed in DU as compared to U with the natural isotope composition. DU is defined as U with a percentage fraction by weight of U-235 of less then 0.711%. Typical concentrations of U- 235 in DU are 0.2 to 0.3 weight-%, which represents approximately 30 - 40% of its concentration in natural U (Table 1). The specific activities of natural U (after removal of highly radioactive decay products) and DU (0.2 %) are compared in table 1.
Table 1. Relative isotopic abundance and radioactivity of chemically purified natural U and DU (0.2 %) (Benedict et al., 1981; Bleise et al., 2003; Glastone and Sesonske, 1981; Larsen, 2000). The specific activities (in Bq/mg) of uranium isotopes are 12.44 (U- 238), 80 (U-235 ), and 2.31x105(U-234)
|Abundance||Radioactivity/mg (Bq)||Abundance||Radioactivity/mg (Bq)|
|U-238||99.28 %||12.40||99.8 %||12.40|
|U-235||0.72 %||0.57||0.2 %||0.16|
|U-234||0.0057 %||12.40||0.001 %||2.26|
The radioactivity of freshly prepared DU is only about 60% the radioactivity of natural uranium as DU has less of the more radioactive isotopes U-234 and U-235 per mass unit than natural U (Table 2) (Bleise et al., 2003).
All natural U isotopes emit alpha particles (table 2), i.e. positively charged ions composed of two protons and two neutrons. Both beta (high-energy electrons) and gamma (very high energy photons) activity of relevant U isotopes are low. Due to their relatively large size and charge, alpha particles have little penetrating power. The penetration range of a 5 MeV alpha particle is approximately 4 cm in air and 50 micrometers in soft tissue. Therefore, alpha particles do not penetrate the keratin layer of intact human skin. As a result, U represents a radiation hazard only after inhalation or ingestion.
|Isotope||Average energy per transformation (MeV/Bq)|
DU penetrators collected in Kosovo contained traces of U-236, Pu-239 and Pu-240 (IAEA, 2003; UNEP, 2001). Trace amounts of Am, Np, and 99Tc were also detected (DAF-OO- ALC, 1997; Diehl, 2001). The traces of U-236 (<0.003%) may result from cross- contamination due to the use of the same equipment for handling both non-irradiated and irradiated U (TACOM, 2000). However, the increase in radiation dose due to the trace amounts of these elements and isotopes is less than 1% (WHO, 2001).
The body defences have evolved in such a way that they can deal with any conceivable kind of foreign material that enters the biological system, including bacteria, viruses and particles. The foreign surface is coated with host molecules present at portals of entry. These host molecules act in several ways. Some are opsonins and following binding to particle surfaces, they are recognised by defence cells, which possess receptors for them, resulting in phagocytosis and clearance e.g. scavenger receptors. The scavenger receptor MARCO is required for lung defense against pneumococcal pneumonia and inhaled particles (Arredouani et al. 2004). Proteins of the complement system bind to foreign surfaces. This results in a cascade of effects including production of opsonins and induction of inflammation. The complement system can also be involved in the response to some dusts (Warheit et al. 1991). Immunoglobulins are present in serous fluids at portals of entry. IgG is an opsosin that was found to modify the reponse to some respirable fibres but not others (Donaldson et al. 1995). In addition, a number of host proteins that are specific for certain portals of entry (e.g. Surfactant protein A) and non- specific proteins bind to the particle surface (Kendall 2007). Fine airborne urban particles (PM2.5) sequester lung surfactant and amino acids from human lung lavage. The actual role that these biomolecules play in the subsequent response is not known. However, it is prudent to consider that the outcome of the interaction of a particle with a biological system might depend on the coating that it receives and this has implications for in vitro work and any situation where a particle is delivered into a biolgical system under non- physiological conditions.
Inflammatory reaction is a key event that may occur following exposure to any solid material, including nanomaterials. For several nanomaterials in vitro induction of inflammatory cytokines was demonstrated (Carlson et al. 2008, Kim et al. 2003, Kocbach et al. 2008, Zhang et al. 2008). Such inflammatory cytokines can also bind to nanomaterials (Kim et al. 2003). This may have implications when in vitro assays are used for the evaluation of the inflammatory properties of nanomaterials.
Important factors for the potential translocation/absorption of nanomaterial may be the protein-nanomaterial interactions both in the lungs and in the gut. Upon contact with body fluids, some nanomaterials have been found to interact with proteins and biomolecules immediately (Linse et al. 2007, Lynch and Dawson 2008). This contact with biological matrices (including food) may determine the behaviour of the nanomaterials, in addition to the nature of the surface (charge, chemistry) itself, although this will of course influence the binding of the biomolecules (Colvin 2003, Lundqvist et al. 2004, Nel et al. 2006).
A recent systematic study of interaction of polystyrene nanoparticles with no modification (plain) or modification with positive (amine) or negative (carboxylic) charges indicates that the surface and the curvature (particle size) both influence the details of the adsorbed proteins, although in all cases, a significant fraction of the proteins bound were common across all particles (Lundqvist et al. 2008). Due to the curvature of the nanoparticle surface this may affect the tertiary structure of the binding protein resulting in malfunctioning (Lynch et al. 2006). These nanoparticle protein interactions may not be static but change during time (Cedervall et al. 2007). Such protein coatings on nanoparticle may enhance membrane crossing and cellular penetration (John et al. 2001, John et al. 2003, Panté and Kann 2002). This may even include crossing the nuclear membrane as demonstrated for gold nanoparticle up to a size of 39 nm bound to the nuclear core complex protein (Panté and Kann 2002).
A recent review has summarised much of the current state-of-the-art in protein- nanoparticle interactions (Lynch and Dawson 2008). However, there are several complicating factors, such as the fact that the biomolecules surrounding the nanomaterial, sometimes referred to as "corona", are not fixed, but are in a dynamic state. The corona equilibrates with the surroundings, with high abundance proteins binding initially, but being replaced gradually by lower abundance, higher affinity proteins. This complicates the measurement of such a protein corona. A considerable portion of the true biologically relevant biomolecules (proteins) will be associated with the nanoparticles for a sufficiently long time that they are not affected by the measurement processes – the so-called "hard-corona” (Lundqvist et al. 2008). Additionally, changes in the biomolecule environment, such as uptake or biodistribution will be reflected as changes in the corona. One may speculate whether the protein determines the fate or distribution of the absorbed nanoparticles or the environment with its specific proteins present. Indeed the coating of 500 nm polystyrene nanoparticles with a bioadhesive tomato lectin molecule did increase considerably the uptake of the particles after oral administration (Hussain et al. 1997). The uptake in the GI tract of various polystyrene latex particles (ranging ion size from 50 nm to 3 microns) could be increased or decreased by modification of the particle surface (Florence et al. 1995).
The significance of this for nano-safety and nano-risk assessment is clear, as it implies that detailed characterisation of the nanoparticles in the relevant biological milieu is vital. Evidence is emerging in the scientific literature that coating of nanoparticles with specific proteins can direct them to specific locations – apolipoprotein E for example has been associated with transport of nanoparticles to the brain (Kreuter et al. 2002). Serum albumin has been shown to induce uptake and anti-inflammatory responses in macrophages, which were not present when the particles were pre-coated with surfactant to prevent albumin binding (Dutta et al. 2007). In addition, coating with polyethylene glycol (PEG) prevents the cellular uptake of nanomaterials and increases their half-life in blood (Niidome et al. 2006).