Registration Dossier

Toxicological information

Endpoint summary

Currently viewing:

Administrative data

Key value for chemical safety assessment

Genetic toxicity in vitro

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (negative)

Genetic toxicity in vivo

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (negative)

Additional information

There are no data available on the genotoxic potential of aluminium oxide in-vitro, while there are few data available on the genotoxic potential of aluminium oxide in-vivo.

 

Information available on the genotoxic potential of aluminium compounds was taken into account for hazard assessment, since the pathways leading to toxic outcomes are considered to be dominated by the chemistry and biochemistry of the aluminium ion (Al3+) (Krewski et al., 2007; ATSDR, 2008).

The publications and reports discussed above have been reviewed by an independent genotoxicity expert (Prof. D J Kirkland), who concurs with the summaries, and the overall weight of evidence for genotoxicity discussed below.

Human Studies

Two human studies, Botta et al. (2006) and Iarmarcovai et al. (2005), became available following the completion of the ATSDR (2008) and Krewski et al. (2007) reviews of aluminium. However, these studies contribute little to the weight of evidence approach for assessing the mutagenic potential of the target substances as the results are confounded by the complex nature of the exposure (a mixture of fume from different welding materials), possible co-exposures, and uncertainties concerning the completeness of the adjustments for age and smoking. 

 

Bacterial test systems

Bacterial mutagenicity assays of simple aluminium compounds have been negative in bacterial reverse mutation tests using several strains of S. typhimurium (e.g.; Marzin and Phi, 1985 (TA102); Ahn and Jeffrey, 1994 (TA98); Gava et al., 1989 (TA92, TA98, TA100, TA104); Blevins and Taylor, 1982 (Spot Test: TA98, TA100, TA1535, TA1537, TA1538); Pan et al., 2010 (TA97a, TA100)) and Escherichia coli WP2 trp uvrA (Pan et al., 2010). Results from the Rec Assay using Bacillus subtilis H17 (Rec+, arg-, tryp-) and B. subtilis M45 (Rec-, arg-, tryp-) have also been negative (Nishioka et al., 1975; Kanematsu et al., 1980).  

As uptake of aluminium depends on the chemical form of the aluminium substance, particularly its solubility and other ligands present, the available mutagenicity assays have tended to use aluminium (III) salts that are more soluble than the target substances.  It should be noted that some strains of bacteria are sensitive to only certain types of genetic changes (Battersby et al., 2007) requiring the use of, for example, a range of S. typhimurium strains to provide suitable sensitivity. Strains TA102 and TA104 are sensitive to oxidising mutagens. Given the possibility of false negatives due to the exclusion of Al3+ ion by cell membranes, and possible insensitivity of assays to mechanisms of metal genotoxicity e.g. induction of large DNA deletions which would lead to cell death rather than mutation (see also Battersby et al., 2007), negative results from bacterial test systems, although contributing to the weight of evidence, are not sufficient to negate further testing.

 

Animal Studies – In-vivo Somatic Cell Tests

The most relevant and methodologically strongest studies are those conducted by Covance (2010a) and by Balasubramanyam et al. (2009a,b). Covance (2010a) investigated the induction of micronuclei in the bone marrow of rats treated with aluminium hydroxide by oral gavage. This study was conducted in accordance with GLP and recognized testing guidelines (Klimisch Score=1). No induction of micronuclei was observed even at the highest dose administered – 2000 mg Al(OH)3/kg bw/day, two administrations 24 hours apart, equivalent to ca. 690 mg Al/kg bw/day.

Balasubramanyam et al. (2009a,b) examined the genotoxic effects of aluminium oxide particles in vivo. Single doses of aluminium oxide particulate suspensions were administered to rats by oral gavage. The reporting of these investigations was lacking in some areas but the studies appear to have been conducted according to GLP. The study results were positive for the nano-sized materials with evidence of a dose-response relationship. The relevance of these results to the current hazard identification is unclear as it is not distinghuishable if the observed effects have arisen from the presence of nanoparticles rather than from any solubilized chemical species (“Al3+”) or the chemical substance Al2O3 itself. Low toxicity, poorly soluble substances, such as Al2O3, when in the form of nanoparticles, have produced inflammatory effectsin vitro, possibly due to production of reactive oxygen species (ROS) (Duffin et al., 2007; Dey et al., 2008). Current scientific knowledge does not allow the distinguishing of genotoxic effects due to the physical (in this case “nanoparticle”) nature of the exposure from genotoxic effects due to the chemical characteristics of the substance (Landsiedel et al., 2009; Singh et al., 2009; Gonzalez et al., 2008). However, in the current extensive debate concerning the genotoxic effects of nanoparticles of many different substances, the possibility that nanoparticles stimulate an inflammatory response which leads to oxidative stress and hence to DNA damage has been widely voiced. The genotoxicity levels for 50 to 200μm diameter particles (Al2O3-bulk) were also not statistically significantly different from those for the control. Balasubramanyam et al. (2009a,b) reported tissue aluminium oxide levels elevated in a dose-response manner for the groups treated with nano-sized materials, consistent with transfer of the nano-sized particles across the gastrointestinal mucosa (Florence, 1997; Hagens et al., 2007). A particle size dependence of gastrointestinal absorption was apparent. Aluminium oxide levels in the tissues of animals dosed with the larger 50 to 200μm diameter particles (Al2O3-bulk) were not elevated to a statistically significant level. 

The positive results observed in studies of aluminium sulphate reported by Dhir et al. (1990) and Roy et al. (1992) occurred with non-physiologically-relevant intraperitoneal administration of the test substances and were methodologically weaker (Klimisch Scores of 2-3).

Turkez et al. (2010) studied the clastogenic activity of AlCl3 in hepatocytes of adult male Sprague–Dawley rats (8-weeks old, 5 animals per group) after gavage with 34 mg/kg bw AlCl3 and along with 50 mg/kg bw propolis for 30 days. Isolated hepatocytes were prepared by the collagenase perfusion technique (Wang et al., 2002). Isolated hepatocytes were stained with orange (AO)–40,6-diamidino-2-phenylindole dihydrochloride (DAPI) (AO, 0.5 mg/mL; DAPI, 10 mg/mL) and the numbers of micronucleated hepatocytes (MNHEPs) were counted in 2000 hepatocytes for each animal. MNHEPs were defined as hepatocytes with round or distinct MNs that stained like the nucleus, with a diameter 1/4 or less than that of the nucleus, confirmed by focusing up and down and taking into account hepatocyte thickness. The authors found that repeated gavage with AlCl3 induced a significant increase in the numbers of micronucleated hepatocytes. Simultaneous administration of propolis attenuated the increased numbers MNHEPs induced by AlCl3. In addition, prolonged oral exposure to high doses of AlCl3 caused a significant increase in alkaline phosphatase, transaminases (AST and ALT) and lactate dehydrogenase (LDH) and induced histopathological changes in the liver. The authors suggested the observed clastogenicity after oral AlCl3 may have been mediated, at least in part, by free radicals (Abubakar et al., 2003). It must be noted here that there was no justification for the high oral dose examined and that the increased genetic damage occurred at doses that also induced overall systemic toxicity and cytotoxicity. As the report was written, it could not be excluded that genetic damage was associated with the oxidative stress in the liver as contrast to any direct clastogenic activity of Al. A Klimisch score of 2 was assigned to this study (weight of evidence study). The results of the study have limited utility in Al hazard identification.                     

     

Intraperitoneal injection

Geyikoglu et al. (2012) conducted a liver micronuclei (MN) assay to investigate genotoxic effects in adult male Sprague-Dawley rats (6 animals per group) treated daily with intraperitoneal injections of AlCl3at 5 mg/kg bw per day for 10 weeks. The control group of 6 rats received daily intraperitoneal injections of NaCl (the dose was not reported, but it can be assumed the solution was isotonic NaCl). The liver MN assay was performed in accordance with methods described by Turkez et al. (2010). Daily injections of AlCl3over 10 weeks resulted in a 4-fold increase in the numbers of micronucleated hepatocytes. The adverse effects on the liver and kidney and the histopathological changes in these organs suggested that the clastogenicity might be a consequence of the systemic toxicity and associated cytotoxicity. The reported findings supported previous observations on the genotoxic activity of AlCl3in mice following repeated intraperitoneal injections (Manna and Das, 1972). The Geyikoglu et al. (2012) study has following limitations: lack of detail on test material purity, test solution preparation and volume and pH of the administered solution. In addition there were no descriptions of clinical signs, drinking water or food consumption and body weight gain. No measures of Al levels in diet and drinking water were provided. A Klimisch score of 3 was assigned for this study (weight of evidence study).

Thus, on weight of evidence, aluminium compounds of normal particle size (i.e. not nanoparticles) do not induce genotoxic effects in somatic cells in vivo when administered by a physiologically relevant route.

 

Animal Studies – In-vivo Germ Cell Tests

Results from animal studies that have employed the dominant lethal assay are inconsistent. Guo et al.(2005) (Klimisch Score = 3) reported a positive response in rats subcutaneously dosed with AlCl3 daily (0, 7 and 13 mg Al/kg bw/day) for a two week period prior to 9 weeks of sequential matings.Two other studies (Dixon et al., 1979; Zelic et al., 1998), both of which were also assigned Klimisch Scores of 3, reported negative results, but the assessments of dominant lethal effects were very limited. Only Guo et al. (2005) reported aluminium levels in the testes indicating that aluminium had reached the target tissue. However, confidence in these results is limited by the high serum Al concentrations in the control group, the time delay in effect, the lack of information on the mating schedule and number of females, and the concurrent effects on fertility (fecundity, libido and male gamete histology).The evidence from animal studies for a mutagenic hazard to germ cells from aluminium ion is inconsistent and no clear conclusions about germ cell mutagenicity can be made based on these studies.

Since there are no known examples of substances inducing genotoxic effects in germ cells that are not genotoxic in somatic cells, it is highly unlikely that aluminium compounds pose any genotoxic risk to germ cells.

 

In-Vitro Gene Mutation Assays

The study by Oberly et al.(1982) (Klimisch Score = 2) was not considered sufficiently robust to meet this information requirement.Two recent gene-mutation studies conducted by Covance, Inc. (Covance, 2010b) using aluminium chloride (tested up to its solubility limit) and aluminium hydroxide (tested up to 10 mM in suspension and subsequently “cleaned” by Percoll density gradient centrifugation), did not find significant mutations at the thymidine kinase (tk) locus of mouse lymphoma L5187Y cells at any of the doses tested. This study type detects both gene mutations and chromosomal damage.These studies were conducted according to guidelines and GLP (Klimisch = 1 for aluminium chloride and 2 for aluminium hydroxide).

Sappino et al. (2011) investigated the potential genotoxicity of AlCl3in cultured MCF-10A cells and in cultured human primary mammary epithelial cells. Aluminium chloride hexahydrate (purity ≥ 99%) was dissolved in water at 1 M and immediately diluted to 10, 100 and 300 µM. The authors suggested that under these conditions, Al precipitation owing to polymerization was expected to be minimal and that no visual evidence of precipitates was observed. Stock solutions were diluted 1:1000 in fresh culture medium twice each week and an equivalent volume of water was used as the control. The addition of AlCl3had only a minor influence on the cell culture medium (control pH = 7.50±0.04; 10 µM: pH = 7.40±0.06; 100 µM: pH = 7.41±0.06; 300 µM: pH = 7.33±0.03). 

To investigate the role of Al in cell transformation, MCF-10A cells were cultured for 6 weeks in the presence of AlCl3  at 100 µM or in the presence of an equivalent volume of water. The authors suggest that “at the expected pH of the cell culture medium (pH ≈7.2), Al chloride and Al2Cl(OH)5 yield the same dissociation product, aluminum hydroxide”. Under these experimental conditions, AlCl3 induced a loss of contact inhibition and increased anchorage-independent growth of cultured MCF-10A cells. No effects of AlCl3on anchorage-independent growth were observed in HaCaT keratinocytes or on C26Ci human colonic fibroblasts cultured for 17 weeks in the presence of 300 µM AlCl3 compared to the same volume of water in controls. In a 7 day cell proliferation assay, AlCl3 at 100 or 300 µM reduced the numbers of MCF-10A cells at the density of 5000 cells per well in triplicate. Apoptosis measured using Annexin V staining revealed no differences between controls and MCF-10A cultures treated with up to 300 µM AlCl3 for 4 days. Exposure to 10, 100 or 300 µM AlCl3 increased the percentage of senescence-associated ß-galactosidase-positive cells in proliferating MCF-10A cultures after 7 days; at the same time, exposure to 100 or 300 µM AlCl3 increased the expression of p16/INK4a, a cyclin-dependent kinase inhibitor and tumor suppressor that enforces growth arrest and is known to increase with ageing in rodent and human tissues (Baker et al., 2011), in proliferating primary human mammary epithelial cells. The addition of AlCl3 increased DNA double strand breaks (DSBs) in a dose- and time-dependent manner in proliferating MCF-10A cells exposed to 10, 100 and 300 µM for 1 and 16 hours. Treatment with 100 and 300 µM AlCl3 increased DNA DSB in proliferating primary human mammary epithelial cells (p < 0.0001, two-sided t-test) but it had little or no effect on proliferating HaCaT keratinocytes (p = 0.21, two-sided t-test).Following X-ray irradiation there was no influence on the repair process of spontaneous DSBs in MCF-10A cells incubated for 16 hr in the presence of 300 µM AlCl3 compared to cells treated with the same volume of water. The cultured MCF-10A cells treated for 10 weeks with 10, 100 or 300 µM AlCl3 responded with upregulation of the p53/p21 pathway that mediates general and premature cell senescence. Based on the results of long-term culture of MCF-10A cells with high concentrations of AlCl3,Sappino et al.(2011) suggested that AlCl3 (up to 300 µM or 60 µM as Al) induced proliferation in MCF-10A cells, increased DSBs and accelerated senescence. According to the authors, the results indicated that induction of DSBs by AlCl3-treatment occurred slowly, suggesting that this effect was indirect and possibly cell-type specific.

There are a number of observations that can be made with regard to the Sappino et al. (2011) report. Other than comparisons to the Al concentrations found in commercial antiperspirants, there were neither justifications for the Al concentrations examined in cell cultures nor for the suggested correlations between Al2Cl(OH)5 and AlCl3 exposure. The Al concentrations examined in cell culture were far greater than the median 0.07 - 0.38 µM (< 10 µg/L) Al concentrations seen in the serum or plasma of healthy people (reviewed in Krewski et al., 2007) and do not reflect levels that could be achieved under normal circumstances.In the Sappino et al. (2011) study, no positive control groups were included, limited details were provided regarding the pH of the cell culture medium and it is not clear if the reported pH values were measured in the fresh or long-term culture medium. Weakly acidic conditions (pH 6.6-6.8) are generally mutagenic and clastogenic for cultured cells and these effects (even at non-cytotoxic concentrations) are the result of oxidative stress (e.g., increased free radicals and reactive oxygen species associated with lipid peroxidation). These changes often decline as cytotoxicity and cell death increase after exposure to higher levels of HCl and other acids (Morita et al., 1992). Sappino et al. (2011) pointed to positive results from other studies of Al genotoxic activity without consideration of the limitations of those studies in that many of the older studies used high concentrations of soluble Al compounds and the pH of the culture medium was not always controlled. Exposure of cultured cells to acidic media in many of the older studies could not be excluded (reviewed in Krewski et al., 2007) and abundant evidence exists to support the fact that acidic conditions in cultured human (Morita et al., 1992; Güngör et al., 2010) and rodent (Cifone et al., 1987; Morita et al., 1989; 1992) cells can increase the numbers of chromosomal aberrations (e.g., chromatid breaks and gaps). It has been established that AlCl3at neutral pH transforms to Al hydroxides including Al trihydroxide and Al oxidehydroxide and these hydroxides can precipitate from solution (Mayeux et al., 2012). The authors stated that they did not observe visually-evident precipitates (detection method was not reported); however, microscopic Al precipitates may have existed in the culture media (particularly at the highest concentration) and in that case, it might be that exposure of cells in culture to Al(OH)3 particulates occurred as compared to conditions in the control cultures.A Klimisch score 3 was assigned.

Turkez et al. (2011) conducted chromosome aberration (CAs) and sister chromatid exchange (SCEs) assays with alum (aluminum sulfate) in human lymphocytes. Experimental studies were conformed to the guidelines of the World Medical Assembly (Declaration of Helsinki). Whole heparinized blood samples were obtained from three healthy non-smoking donors with no history of exposure to any genotoxic agent. The Al2(SO4)3 concentrations tested were 0, 10 and 20 µg/mL (equivalent as 0, 1.57 and 3.15 µg Al/mL).The effects of bismuth subnitrate (BSN) on oxidative status of erythrocytes and cytogenetic changes in human lymphocytes were studied at 0.0625, 0.125, 0.25 and 0.5 µg/mL alone and applied to the cultures together with Al sulfate. Concentrations of Al2(SO4)3 at 10 µg/mL alone did not influence the frequency of SCEs and CAs; however, identical study with 20 µg/mL increased the frequency of SCEs per cell and CAs compared with controls. It must be noted here that no biologically significant increases (> 2-fold) of SCEs/cell were observed (publication Fig.1).Although there was no effect of Al2(SO4)3 at 10 µg/mL on oxidative stress markers in erythrocytes compared with the controls, the highest concentration of Al2(SO4)3 caused significant decrements in the activities of antioxidant enzymes (G-6-PDH, SOD and CAT) and reduced glutathione (GSH) in erythrocytes.. The authors suggested that increased SCEs and CAs resulted from the decreased activity of the antioxidant enzymes seen at 20 µg/mL. Concomitant treatment with BSN (except for 0.5 mg/mL) reduced the increased number of SCEs and CAs and the increased oxidative stress associated with Al2(SO4)3. Limitations to the Turkez et al. (2011) protocol include the fact only one time point was examined, reference mutagens were not included, only thirty well-spread metaphases were scored per sample for CA assay when OECD Test Guideline #473 requires that at least 200 well-spread metaphases be scored. In addition, the number of the second cycle metaphases examined for SCEs was not reported and the authors provided few details on laboratory methods. The highest Al concentration (3.15 µg/ml or 3,157 µg/L) examined was ~300x the Al concentrations (1.9-10.3 µg/L) present in normal human plasma and serum (Krewski et al., 2007). It should be noted that structural CAs can also occur as a result of cytotoxicity (Galloway et al., 2000) and in the presence of > 50% cytotoxicity, CA increases are most all artifactual and can represent false positives (Battersby et al., 2007 ; Kirkland et al., 2007; Galloway, 2000). The following criteria should be considered in selection of the highest concentration of the test substance: cytotoxicity, solubility of the compound in the test system, changes in pH and changes in osmolality (in OECD guideline for the in-vitro chromosome aberration test (OECD Test Guideline #473). Overall, the results of this study are equivocal. A Klimisch Score of 3 was assigned to this study.

 

In-Vitro Mammalian Cell Assays

The in-vitro micronucleus assay results of Migliore et al. (1999) for aluminium sulphate and the chromosome aberration assay of Lima at el. (2007) using aluminium chloride provide evidence that the aluminium ion is an in-vitro clastogen. Treatment of human lymphocytes with aluminum as AlCl3 has also been observed to induce oxidative DNA damage and inhibit repair of DNA damage from exposure to ionizing radiation (Lankoff et al., 2006). Caicedo et al. (2008), however, did not observe double DNA strand breaks at concentrations up to 5000 µM-Al (as AlCl3) in human jurkat T-cells, supporting an oxidative mechanism of action that produces single strand effects only. Available studies provide evidence for an indirect genotoxic mechanism of action for the aluminium ion involving the production of single strand breaks. An oxidative mechanism of action would be expected to exhibit a threshold, which may be expected to be higherin-vivodue to more efficient defence mechanisms than in cultured cells.

Thus, there is some evidence that soluble aluminium salts may induce DNA damage, probably by an oxidative mechanism, but these findings were not confirmed in recent GLP studies using the sensitive mouse lymphomatkassay. 

 

Other Relevant Information

In a weight of evidence assessment for a mutagenic effect in humans, the levels at which effects are seen in animal studies and the systemic bioavailability of the target substances need to be considered. The study conducted by Covance (2010b) in non-fasted rats observed no induction of micronuclei in bone marrow at the acute maximum tolerated dose (MTD) for aluminium hydroxide when administered by oral gavage, namely 2000 mg Al(OH)3/kg bw/day. The MTD had been determined previously in a range-finding experiment. Balasubramanyam et al. (2009a) also observed no statistically significant genotoxic effects in rat bone marrow after a single oral gavage of 2000 mg Al2O3/kg bw in the form of particles with a size-range of 50 to 200 µm. Although current toxicokinetic information does not allow the prediction of time profiles of levels of aluminium in target tissues as a function of realistic external exposures, when administered orally or by inhalation the target substances exhibit low bioavailability.

 

Conclusion for genetic toxicity

The available information does not provide indications for significant mutations at the thymidine kinase (tk) locus of mouse lymphoma L5187Y cells treated with aluminium hydroxide and aluminium chloride at any of the doses tested (Covance, 2010b; Oberly et al., 1982).

In vitro studies with human blood lymphocytes showed positive responses to aluminium sulphate for micronuclei formation (Migliore et al., 1999) and to aluminium chloride for the induction of chromosome aberrations (Lima et al., 2007). Aluminium chloride has also been shown to induce oxidative DNA damage in human lymphocytes (Lankoff et al., 2006). However, double DNA strand breaks were not observed at concentrations up to 5000 µM-Al (as aluminium chloride) in human jurkat T-cells (Caicedo et al., 2008). This supports an oxidative mechanism of action leading to single strand effects only. Thus, there is some evidence that soluble aluminium salts may induce DNA damage, probably by an oxidative mechanism.

The most relevant and methodologically strongest in vivo studies are those conducted by Covance (2010a) and by Balasubramnyam et al. (2009a, b).

In the Covance (2010a) study, the induction of micronuclei in the bone marrow was investigated in rats given aluminium hydroxide by oral gavage. No induction of micronuclei was observed up to the highest dose administered (2000 mg aluminium hydroxide/kg bw/day, corresponding to 700 mg Al/kg bw/day).

In the studies by Balasubramanyam et al. (2009a, b), the genotoxic effects of aluminium oxide particles were investigated in vivo. Single doses of aluminium oxide particulate suspensions were administered to rats by oral gavage. The study results were positive for the nano-sized materials with evidence of a dose-response relationship, while the genotoxicity levels for aluminium oxide bulk material (50 to 200 μm diameter particles) were not statistically significantly different from those for the control. The relevance of the results with nanomaterials for hazard assessment is unclear as the observed effects may have been related to the presence of nanoparticles as foreign bodies in the cells rather than to the chemical properties of the test material itself. Low toxicity, poorly soluble substances, such as aluminium oxide, have produced inflammatory effects in vitro, when present as nanoparticles. The proposed mechanism of action is the production of reactive oxygen species (ROS) (Donaldson and Stone, 2003; Nel et al., 2006; Oberdörster et al., 2005, 2007; Duffin et al., 2007; Dey et al., 2008). Current scientific knowledge does not allow differentiation of genotoxic effects due to the physical (nanoparticle) nature from genotoxic effects due to the chemical characteristics of the test substance (Landsiedel et al., 2009; Singh et al., 2009; Gonzalez et al., 2008). However, in the current scientific debate regarding the genotoxic effects of nanoparticles of many different substances, the possibility that nanoparticles stimulate an inflammatory response leading to oxidative stress in the cells and consequently to DNA damage is the most accepted hypothesis. Balasubramanyam et al. (2009a, b) reported tissue aluminium oxide levels elevated in a dose-response manner for the groups treated with nano-sized materials, consistent with transfer of the nano-sized particles across the gastrointestinal mucosa (Florence, 1997; Hagens et al., 2007). A particle size dependence of gastrointestinal absorption was apparent. Aluminium oxide levels in the tissues of animals dosed with the larger 50 to 200 μm diameter particles were not elevated to a statistically significant level, consistent with the notion of a low bioavailability of aluminium compounds (see Toxicokinetics).

Thus, on a weight of evidence approach, aluminium compounds in non-nanoparticle size ranges do not induce genotoxic effects in somatic cells in vivo when administered by a physiologically relevant route.

Taken together, the weight of evidence does not support a systemic mutagenic hazard for soluble and insoluble aluminium compounds. 

With regard to the nanosized material it is inconclusive whether the aluminium oxide shows a gentotoxic potential or not because the studies by Balasubramanyam et al. 2009a and b are not trustworthy due to the following reasons:

These two papers describe studies that have been well-designed and generally comply with the OECD guidelines that were in place at the time. Negative control CA, MN and comet data are normal, and positive control chemicals were effective. The results with the 30 and 40 nm samples of Al2O3 certainly suggest they are genotoxic, However, the data are so perfect they raise concerns. Such clear dose-responses for different endpoints over multiple sampling times are rarely seen following a single administration, and cannot be easily explained. The standard deviations, at least for the MN scores, are consistently to low to be credible taking normal inter-replicate and inter-animal variability into account. In order to confirm or refute the findings published by Balasubramanyam et al, it is recommended that a new combined MN and comet assay be performed, with oral dosing, to a robust OECD protocol. Therefore, these two in vivo studies are necessary for the correct assessment of aluminium oxide nano materials.

 

References not in IUCLID

Dey S, Bakthavatchalu V,et al.(2008). Interactions between SIRT1 and AP-1 reveal a mechanistic insight into the growth promoting properties of alumina (Al2O3) nanoparticles in mouse skin epithelial cells. Carcinogenesis 29(10): 1920-1929.

Duffin R, Tran L, Brown D et al. (2007). Proinflammogenic effects of low-toxicity and metal nanoparticles In Vivo and In Vitro: Highlighting the role of particle surface area and surface reactivity. Inhalation Toxicology 19: 849-856.

Florence AT. (1997). The oral absoprtion of micro- and nanoparticulates: Neither exceptional or unusual. Pharmaceut Res 14(3): 259-266.

Gonzalez L, Lison D, Kirsch-Volders M. (2008).Genotoxicity of engineered nanomaterials: A critical review. Nanotoxicology 2(4): 252-273.

Hagens WI, Oomen AG, de Jong WH et al. (2007). What do we (need to) know about the kinetic properties of nanoparticles in the body? Reg Toxicol Pharmacol 49: 217 229.

Landsiedel R, Kapp MD, Schulz M, et al.(2009).Genotoxicity investigations on nanomaterials: Methods, preparation and characterization of test material, potential artifacts and limitations - Many questions, some answers. Mutat Res 681: 241-258.

Nel A, Xia T, Madler L, Li N (2006). Toxic potential of materials at the nanolevel, Science 311: 622-627

Oberdorster G, Oberdorster E, Oberdorster J (2005).Nanotoxicology: An emerging discipline evolving from studies of ultrafine particles, Environmental Health Perspectives 113: 823-839

Oberdorster G, Stone V, Donaldson K (2007).Toxicology of nanoparticles: A historical perspective, Nanotoxicology 1: 2-25

Singh N, Manshian B, Jenkins GJS et al. (2009). Nanogenotoxicology: The DNA damaging potential of engineered nanomaterials. Biomaterials 3891-3914.


Short description of key information:
Overall, the read-across from aluminium compounds within a weight of evidence approach does not support a systemic mutagenic hazard for aluminium oxide and there are also no clastogenicn effects in valid in vivo studies

Endpoint Conclusion: No adverse effect observed (negative)

Justification for classification or non-classification

Based on the read-across from aluminium compounds within a weight of evidence approach (GLP-guideline studies) for genetic toxicity, no classification is required according to CLP (1272/2008/EC) classification criteria.