hydronium;iron(3+);hexahydroxide;disulfate

Genetic toxicity: in vitro

Administrative data

Endpoint:

in vitro gene mutation study in mammalian cells

Remarks:

Type of genotoxicity: gene mutation

Type of information:

migrated information: read-across based on grouping of substances (category approach)

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Research publication. Well documented meets generally accepted scientific principles, acceptable for assessment.

Justification for type of information:

The Reporting Format for the Chemical Category According to ECHA (2008) Guidance R.6.2.6.2 can be found in the Endpoint Summary of Toxicokinetics, metabolism and distribution.

Cross-reference

Reason / purpose for cross-reference:

read-across: supporting information

Reference

Iron can only be absorbed orally as the ferrous ion. Iron absorption in the rat is higher than humans. The presence of non-complexed iron in the diet rarely results in iron overload conditions. 
There are no reliable acute or repeated dose dermal studies that can be consulted for evidence of absorption via dermal route.
The water soluble inorganic iron salts do not undergo metabolism.
Iron can be inhaled as the ferric or ferrous ion. Ferric and ferrous ions precipitate in lysosomes of alveolar cells. However, it is not clear how fast the clearance of iron particles from pulmonary tissues is. 
Iron is uniformly distributed via blood. Greatest concentrations are in liver, bone marrow and spleen.
If there is an excess of the element within the body, there is no biochemical mechanism for its excretion and this may result in both severe and chronic symptoms if large amounts are ingested. About 1 mg of iron is lost each day through sloughing of cells from skin and mucosal surfaces, including the lining of the gastrointestinal tract. Only 0.01 to 0.02 % of the resorbed iron in humans are excreted daily. The daily losses of iron from the human body correspond to a biological half-time of iron of 10 to 20 years. 
The foetus is protected from the effects of excess iron in the mother.

Bioaccumulation potential:

low bioaccumulation potential

Justification for read-across

This endpoint is covered by the category approach for soluble iron salts (please see below for the category justification/report format).

In addition to the category members, the surrogate material ammonium iron(III) citrate was used for read-across (McCance & Widdowson 1938). This substance also dissociates at physiological pH to the same species produced by the ferric (iron III) salts. As discussed in the category justification section, the equilibrium to iron II species is quickly established. In result iron species burden is assumed to be comparable to equimolar exposure to the iron category member salts. The additional ammonium and citrate are considered not influencing the iron kinetics.

Introduction

Iron is an essential element, and plays an important role in biological processes, and iron homeostasis (biochemical mechanisms maintaining constant concentration in the cell) is under strict control (McCance & Widdowson 1938). Absorption, storage, mobilisation and excretion of iron are all regulated at the surface of cells by a homeostatic mechanism (Hostynek 1993). The counter ions of the soluble inorganic iron salts in question enter the body’s normal homeostatic processes, and are not discussed further.

 

Absorption

Oral

In humans the absorbance and uptake of iron salts from the digestive system is usually rather poor to the extent that treatment of simple anaemia by such means is of limited effectiveness. This is because iron can only be absorbed as the ferrous ion, but the ferrous ion can only exist in an acid medium. Therefore once in the small intestine the ferrous ion cannot exist. Iron absorption in the rat is higher than humans (Mahoney & Hendricks 1984); consequently, rat studies are considered unreliable models for iron toxicology in humans. Uptake is facilitated by the formation of iron chelates such as those with citrate and ascorbate that are present in the diet and in their absence iron absorption by the small intestine is very poor. Additionally, the presence of appreciable amounts of plant tannins may complex iron and further prevents its absorption. The result of this low solubility and low uptake by the human gut means that for healthy individuals, the presence of non-complexed iron in the diet rarely results in iron overload conditions.

There is some evidence that water-soluble iron salts are better absorbed than water-insoluble iron compounds. In both humans and animals, iron absorption from the digestive tract is adjusted to a fine homeostasis with low iron stores resulting in increased absorption and, alternately, sufficient body stores of iron decreasing absorption (Elinder 1986).

Significant differences in iron absorption from salts and food have been noted between rats and humans, with uptake significantly higher from identical meals in rats (Reddy and Cook, 1991), although rats poorly absorb haem (Bjorn-Rassmussen 1974). Dietary enhancers and inhibitors appear to affect non-haem iron absorption in humans to a greater extent than in rats (Reddy & Cook 1991). Growth requirements for iron in the rat are greater, and the dietary intake is about 100 times greater than that of humans, expressed on a body weight basis (WHO 1983).

EFSA (2012) concludes for FeSO4 on rapid absorption (10 % up to 60 % in case of iron deficiency) within 2 to 6 hours.

Dermal

The water solubility (estimated at 300 g/L) of ferric chloride sulphate suggests that it is unlikely to be absorbed across the lipid-rich stratum corneum. However, there are no reports of percutaneous absorption of iron in non-chelated form to support this prediction. Percutaneous absorption of iron has been reported only for chelated forms administered as ointments in mice (Hostynek 1993). There are no reliable acute or repeated dose dermal studies that can be consulted for evidence of absorption via the dermal route.

Inhalation

In contrast to the wealth of data available on the human toxicology of ingested iron salts, there is only one available study (Johansson et al. 1992) on the potential for adverse health effects via inhalation. In this 2-month repeated dose inhalation study in rabbit, only local pulmonary effects were investigated. Here, macrophages were affected due to ferric chloride (FeCl3). Alveolar macrophages were increased in number in both exposed groups. There were prominent changes in the macrophages such as enlarged Lysosomes containing fibrous-looking structures or iron-rich inclusions leading to accumulation of iron. Since the lungs have a neutral pH of approximately 7.4, it is assumed that ferrous ion following a progressive oxidation in the presence of oxygen will transform to insoluble Fe(OH)3. Since a recovery group in Johansson (1992) study was not investigated, no information is available in regard to reversibility of occurring effects and the clearance of iron particles. Elinder (1986) indicated that yearly lung clearance of iron dust in humans is estimated to be 20 - 40 % of the deposited amount (data obtained from iron welders). This reference could be a hint that clearance of deposited iron in oxidized form from the lungs is relatively slow and possibly not complete.

Distribution

The average adult human stores about 1 to 3 grams of iron in the body. Iron is almost never found in the free ionic state in living cells in appreciable concentrations; it is chaperoned in the form of protein complexes immediately it is absorbed from the diet. In the blood plasma it is transported (as FeIII) by the protein transferrin, which passes it on to dividing cells, particularly the cells in the bone marrow that are the precursors of the red blood cells. This is mediated by the transferrin receptor. Transferrin, which binds iron with high affinity is only 20-35 % saturated, thus the concentration of unbound iron is very low (0.5–1.5 mg/L or 9–27 μmol/L), Tenenbein 2001). Iron is stored principally in the liver in the large proteins haemosiderin and ferretin, although these are also found in all cells and in the blood in lower concentrations. Ferritin exists as hollow spheres of 24 protein subunits and iron is taken up in the FeII state but stored as FeIII. As with transferrin, it is stored in a redox-inactive (and therefore non-toxic) form. Ferritin is also important in recycling iron within the body and is an important biological indicator of iron balance. One consequence of the parsimonious conservation of iron is that if there is an excess of the element within the body, there is no biochemical mechanism for its excretion and this may result in both severe and chronic symptoms if large amounts are ingested.

Foetal exposure

It has been found that extremely elevated maternal serum iron concentrations are not accompanied by corresponding increases in foetal serum iron levels (Curryet et al.1990). This finding suggests that the foetus is protected from the effects of excess iron in the mother.

Metabolism

These water soluble inorganic iron salts do not undergo metabolism per se. As already mentioned iron is bound to transferrin for transport to the bone marrow or contained within storage forms.

Excretion

About 1 mg of iron is lost each day through sloughing of cells from skin and mucosal surfaces, including the lining of the gastrointestinal tract (EVM 2003). Menstruation increases the average daily iron loss to about 2 mg per day in pre-menopausal female adults (Bothwell & Charlton, 1982). No physiological mechanism of iron excretion exists. Consequently, absorption alone regulates body iron stores (McCance and Widdowson 1938).

The daily losses of iron from the human body correspond to a biological half-time of iron of 10 to 20 years. The yearly lung clearance of iron dust is estimated to be 20-40 % of the deposited amount (data obtained from iron welders) (Elinder 1986).

Additional references of this section, not entered as studies

• Bjorn-Rassmussen et al. (1974). Food iron absorption in man. Applications of the two-pool extrinsic tag method to measure heme and nonheme iron absorption form the whole diet. J. Clin. Invest. 53:247-55.
• Bothwell and Charlton (1982). A general approach of the problems of iron deficiency and iron overload in the population at large. Seminars in Hematology 19, 54.
• Curryet et al (1990). An ovine model of maternal iron poisoning in pregnancy. Ann. Emerg. Med 19:632-38.
• EFSA European Food Safety Authority (2012). Conclusion on Pesticide Peer Review. Conclusion on the peer review of the pesticide risk assessment of the active substance iron sulphate. Self-published, Parma, Italy. EFSA Journal 10(1):2521. 48 p.
• Elinder (1986) Iron. IN: Friberg L, Nordberg GF, Vouk VB, eds., 1986. Handbook on the toxicology of metals. 2nd ed., the: Elsevier, 277-297 (Vol II).
• EVM Expert Group on Vitamins and Minerals (2003). Safe upper levels for vitamins and minerals. Report of the Expert Group on Vitamins and Minerals. ISBN 1-904026-11-7 Self-published in May by the Food Standards Agency, U.K. 360 p. http://cot.food.gov.uk/pdfs/vitmin2003.pdf
• Hostynek (1993). Metals and the Skin. Critical Reviews in Toxicology 23(2):171-235
• Johansson A, Curstedt T, Rasool O, Jarstrand C, Camner P (1992). Macrophage Reaction in Rabbit Lung following Inhalation of Iron Chloride. PMID 1597169 Environ Res 58(1):66-79.
• Mahoney AW, Hendricks DG (1984). Potential of the rat as a model for predicting iron bioavailability for humans. DOI 10.1016/S0271-5317(84)80067-6 Nutrition Res. 4(5):913-22.
• McCance RA,Widdowson EM (1938). The absorption and excretion of iron following oral and intravenous administration. DOI 10.1113/jphysiol.1938.sp003669 PMID 16995028 J. Phys. 94(1):148 -54. URL http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1393919/pdf/jphysiol01545-0174.pdf
• Reddy MB, Cook JD (1991). Assessment of dietary determinants of nonheme-iron absorption in humans and rats. PMID 1654740 Am. J. Clin. Nutr. 54(4):723-8. URL ajcn.nutrition.org/content/54/4/723.full.pdf
• Tenenbein M (2001). Hepatotoxicity in Acute Iron Poisoning. PMID 11778670 Clin. Toxicol. 39(7):721-6.
• WHO (1983) 571. Iron. Toxicological evaluation of certain food additives and contaminants. WHO Food Additives Series, No. 18, 1983, nos 554-573 on INCHEM URL http://www.inchem.org/documents/jecfa/jecmono/v18je18.htm

Chemical Category Reporting Format according to ECHA Guidance

The following definition complies with the ECHA (2008, chapter R.6) Guidance on QSARs and grouping of chemicals. It should be used in the discussions of the IUCLID 5 section 7 (Toxicology).

Table: Reporting Format for the Chemical Category According to ECHA Guidance R.6.2.6.2

1.	Category definition and its members: Dissociating, inorganic and non-toxic iron compounds
1.1.	Category Definition
a.	Category Hypothesis In solid form iron element exists free, or in iron-containing compounds. In aqueous solution, it exists in one or two oxidation states, Fe2+, the ferrous form, and Fe3+, the ferric form. Many of the key biological functions of iron in living systems rely on the high redox potential, enabling rapid conversion between the Fe2+ and Fe3+ forms. The redox potential is relatively harmful in terms of the capacity for oxidative damage to cellular compound such as fatty acids, proteins and nucleic acids. However, iron within the body is normally bound to carrier proteins and/or molecules with antioxidant properties, which minimise the capacity of the free ion to cause oxidative stress (EVM 2003). In the first assumption, due to the neutral pulmonary pH (of about 7.4) the iron ions will precipitate in hydroxide or oxide of ferrous or ferric form and undergo rapid oxidation to ferric form. Therefore iron salts will act as identical material for the both iron oxidation forms. In the second assumption, independently of exposure route, the anions of iron salts are irrelevant for the toxicity: this category covers as well the anions of inorganic ferrous and ferric salts, i.e. chloride, sulphate and their crystalohydrate forms. All of these salts will dissociate immediately in contact with aqueous media to the respective anions and kations, and then be subject to further change of oxidation and speciated state according to the conditions. In the third assumption, the local effects could be slightly different depending on iron oxidation forms, but it should be not relevant in the regulatory context. In the fourth assumption, bioavailability of iron is regulated by homeostasis mechanisms. It is also known that the ferrous ion (Fe2+) has a higher oral bioavailability than the ferric ion (Fe3+) that could impair the category approach. However, this difference is not that significant to be relevant in the regulatory context. Regarding the long-term exposure, a tolerance development could be assumed. Therefore it is not necessary to make any differentiation between category members and read-across between the salts can be used freely for the toxicological property data sets.
b.	Applicability Domain of the category This category comprises five soluble iron salts (ferric and ferrous chloride and ferrous and ferric sulphate and ferric chloride sulphate, including their various hydrated forms).
c.	List of endpoints covered (numbers refer to the IUCLID 5 sections) 7. Toxicological information 7.1 Toxicokinetics, metabolism and distribution 7.2 Acute toxicity 7.3 Irritation / corrosion (limitation: differently as in this category, FeSO4 is irritant for eye) 7.4 Sensitization 7.5 Repeated dose toxicity 7.6 Genetic toxicity 7.7 Carcinogenicity 7.8 Toxicity to reproduction
1.2	Category Members apart from Hydronium jarosite (EC 940-441-4)
	Common name	EC number				CAS number
	Ferric chloride	231-729-4				7705-08-0
	Ferrous chloride	231-843-4				7758-94-3
	Ferric sulphate	233-072-9				10028-22-5
	Ferrous sulphate	231-753-5				7720-78-7
	Ferric chloride sulphate	235-649-0				12410-14-9
	The definition of these salts also covers a number of chemical identities relating to specific hydrates. Under REACH all hydrates are covered by the registration of the anhydrous salt. As the hydrates influence the molecular weight, correction should apply as the hydrate weight is considered in CLP according to ECHA Guidance on the Application of the CLP Criteria Version 2.0 (2012, p 535, Example D). The following hydrates are identified in the public domain though not all are necessarily commercially relevant:
	Chemical name (CAS name)		CAS number		Molecular formula
	Iron chloride (FeCl3), monohydrate (9CI)		60684-13-1		Cl3 Fe . H2O
	Iron chloride (FeCl3), hydrate (2:3) (9CI)		58694-76-1		Cl3 Fe . 3/2 H2O
	Iron chloride (FeCl3), dihydrate (9CI)		54862-84-9		Cl3 Fe . 2 H2O
	Iron chloride (FeCl3), trihydrate (9CI)		58694-75-0		Cl3 Fe . 3 H2O
	Iron chloride (FeCl3), hexahydrate (8CI, 9CI)		10025-77-1		Cl3 Fe . 6 H2O
	Iron chloride (FeCl3), nonahydrate (9CI)		58694-79-4		Cl3 Fe . 9 H2O
	Iron chloride (FeCl3), dodecahydrate (9CI)		58694-80-7		Cl3 Fe . 12 H2O
	Iron chloride (FeCl3), hydrate (8CI, 9CI)		24290-40-2		Cl3 Fe . x H2O
	Iron chloride (FeCl2), hydrate		23838-02-0		Cl2 Fe . x H2O
	Iron chloride (FeCl2), monohydrate		20049-66-5		Cl2 Fe . H2O
	Iron chloride (FeCl2), dihydrate		16399-77-2		Cl2 Fe . 2 H2O
	Iron chloride (FeCl2), tetrahydrate		13478-10-9		Cl2 Fe . 4 H2O
	Iron chloride (FeCl2), hexahydrate		18990-23-3		Cl2 Fe . 6 H2O
	Sulphuric acid, iron(3+) salt (3:2), nonahydrate		13520-56-4		Fe2(SO4)3.9H2O
	Sulphuric acid, iron(3+) salt (3:2), hydrate		15244-10-7		Fe2(SO4)3.xH2O
	Sulphuric acid, iron(3+) salt (3:2), hexahydrate		13761-89-2		Fe . 3/2 H2SO4 . 3 H2O
	Sulphuric acid, iron(3+) salt (3:2), heptahydrate		35139-28-7		Fe . 3/2 H2SO4 . 7/2 H2O
	Sulphuric acid, iron(3+) salt (3:2), tetrahydrate		230310-51-7		Fe . 3/2 H2SO4 . 2 H2O
	Sulphuric acid, iron(2+) salt (1:1), hexahydrate		59261-48-2		Fe . H2SO4 . 6 H2O
	Sulphuric acid, iron(2+) salt (1:1), trihydrate		58694-83-0		Fe . H2SO4 . 3 H2O
	Sulphuric acid, iron(2+) salt (1:1), tetrahydrate		20908-72-9		Fe . H2SO4 . 4 H2O
	Sulphuric acid, iron(2+) salt (1:1), monohydrate		17375-41-6		Fe . H2SO4 . H2O
	Sulphuric acid, iron(2+) salt (1:1), hydrate		13463-43-9		Fe . H2SO4 . x H2O
	Sulphuric acid, iron(2+) salt (1:1), dihydrate		10028-21-4		Fe . H2SO4 . 2 H2O
	Sulphuric acid, iron(2+) salt (1:1), heptahydrate		7782-63-0		Fe . H2SO4 . 7 H2O
1.3	Purity / Impurities Iron salts are not toxic to the population (cut off category 4 and upwards) and generally only substances with comparable purity or impurity profiles can be assessed together. Technical grade materials may contain relevant quantities of impurities, which exceed the low toxicological effects of the pure compounds comprised in this category. In cases where the sum of impurities triggers the toxicity, assessment has to be based on these impurities considering their concentration and the category assessment is restricted to the iron salt components. The purity of the category members is generally > 80 % w/w. Deviations below this purity value would normally not be acceptable since this would suggest that the substance should not be considered as a mono-constituent substance. Impurities comprise salts of other metals up to ≤ 1 % w/w and free acid up to < 10 % w/w.
2.	Category justification The iron salts category comprises a directly analogous group of iron Fe3+and Fe2+salts, with counter-ions (chloride and sulphate) which are ubiquitous and do not require additional consideration. The formation of a category is justified on the basis of a self-consistent model of the behaviour and properties of these substances. All of these salts will dissociate immediately in aqueous media to the respective anions and kations. The chloride and sulphate anions are of no further interest since they are already ubiquitous in vivo and do not represent health hazard. The ferric and ferrous ions will inter-convert rapidly according to the in vivo conditions that they are found in, and if one ion is introduced into a system, any effects observed may be due to that ion, the other oxidation state, or a mixture of the two. Although each form of iron could be assessed for a particular endpoint and the test substance well-characterised, as a consequence of the inter-conversion it is not always possible to determine the form of the iron responsible for a particular endpoint. In general, under oxygenated conditions, ferrous will be converted to ferric, and in the presence of water ferric hydroxides will precipitate initially. Even where the form of the iron present when the endpoint is reached can be determined there will be uncertainty regarding the species responsible, unless the toxicity can be followed as a function of this species, which means that a dose response-relationship can be established. The majority of toxicological endpoints are covered with experimental tests showing similar mode of action between iron oxidation forms and salts having different counter-ions. Therefore it is logical to consider these five salts together within a single chemical category. Where a data gap may exist for an individual salt, it is considered that relevant data for one or more of the other salts are an acceptable surrogate for the missing data, taking due account of the oxidation state present initially. The iron content of a given salt or solution may be used to convert or compare data.
3.	Data matrix The category approach was evaluated basing on the following experimental data for the acute oral toxicity endpoint. Studies are generally considered reliable or represent the sole data point available.
	Test substance	Identifier of the study		Animal species	LD50 [mg/kg bw] based on test substance	LD50 [mg Fe/kg bw] based on Fe
	FeCl2	Choi 2004a		rat	500 (300 – 2000, toxic classes tested)	220 (132-881, toxic classes tested)
	FeSO4.7H2O	MHLW 2002		rat	> 2000	> 401
	FeSO4	Anon 2000		rat	3200	643
	FeSO4	Parent 2000		rat	3200	1176
	FeSO4	Weaver 1961		rat	2625 (2323-2966, 95% CV)	964 (854-1090, 95% CV)
	FeSO4	Weaver 1961		mouse	1025 (802-1311, 95% CV)	377 (295-482, 95% CV)
	FeSO4	Boccio 1998		mouse	670 (females) 680 (males)	246 (females) 250 (males)
	FeCl3	Hosking 1970		mouse, female	1278 (871-1830, 95% CV)	440 (300-630, 95% CV)
	Fe2(SO4)3	ICI 1991		rat	500-2000 (females) >2000 (males)	140-559 (females) >559 males
	The category approach was evaluated basing on the following experimental data for the skin irritation / corrosion endpoint. Studies are generally considered reliable or represent the sole data point available.
	Test substance	Identifier of the study		Test method & animal species		Result
	FeSO4.7H2O	Clouzeau 1994		Skin irritation / corrosion in vivo, rabbit		Irritating, Category 2
	FeCl2	Park 2004		Skin irritation / corrosion in vivo, rabbit		Not irritating
	FeCl3	BASF 1977		Skin irritation / corrosion in vivo, rabbit		Irritating, Category 2
	The category approach was evaluated based on the following experimental data for eye irritation / corrosion. Studies are generally considered as reliable or represent the sole data point available. In the study of Bayer AG (1992) FeSO4. 7H2O is classified as non-irritating, however, according to Draft Assessment Report for Iron Sulphate (September 2008), FeSO4 is classified as irritating to eyes (R36). Therefore, FeSO4 should be adopted as irritant to eyes.
	Test substance	Identifier of the study		Test method & animal species	Result
	FeCl2	Jeong 2004		Eye irritation / corrosion in vivo, rabbit	Corrosive to rabbit eye, Category 1
	FeSO4.7H2O	Bayer AG 1992		Eye irritation in vivo, rabbit	(Not irritating) Irritating
	FeCl3water free solid	BASF 1977		Eye irritation / corrosion in vivo, rabbit	Corrosive to rabbit eye, Category 1
	The category approach was evaluated basing on the following experimental data for the skin sensitisation.
	Test substance	Identifier of the study		Test method & animal species	Result
	FeSO4	Ikarashi 1992		Skin sensitisation in vivo	negative
	FeSO4	Stitzinger 2010		Skin sensitisation – LLNA, mouse	negative
	FeCl3	Storck 1962		Skin sensitisation in vivo, guinea pigs	Ambiguous, 1 of 2 guinea pigs were positive
	The category approach was evaluated based on the following experimental data for repeated dose toxicity. Studies are generally considered reliable or represent the sole data point available.
	Test substance	Identifier of the study		Test method & animal species	Result based on iron salt	Result based on iron
	FeCl2	Beom 2004		Repeated dose toxicity: oral – OECD 422, rat	NOAEL: 125 mg/kg bw in males, 250 mg/kg bw in females LOAEL: 250 mg/kg bw in males 500 mg/kg bw in females	NOAEL: 55.1 mg/kg bw in males, 110.1 mg/kg bw in females LOAEL: 110.1 mg/kg bw in males 220.5 mg/kg bw in females
	FeSO4.7H2O	Furuhashi 2002		Repeated dose toxicity: oral – OECD 422, rat	NOAEL: 100 mg/kg bw, LOAEL: 300 mg/kg bw for the test item; NOAEL: 54.6 mg/kg bw, LOAEL: 163.9 mg/kg bw for anhydrous FeSO4	NOAEL: 20.1 mg/kg bw LOAEL: 60.3 mg/kg bw
	FeCl3	Sato 1992		Repeated dose toxicity: oral, 90 d rat	NOAEL: 277 mg/kg bw in males, 341 mg/kg bw in females	NOAEL: 95.4 mg/kg bw in males, 117.4 mg/kg bw in females
	The category approach was evaluated based on the following experimental data for the genetic toxicity. Studies are generally considered reliable or represent the sole data point available.
	Test substance	Identifier of the study		Test method & animal species	Result
	FeCl2	Kim 2004		Bacterial mutagenicity – Ames test	negative with and without metabolic activation
	FeCl2	Ji Yoon 2004		In vivo cytogenicity – micronucleus, mouse	negative; 2, 5, 10, 20, 50, 100 and 200 mg/ml in the dose range-finder; 1.25, 2.5 and 5 mg/mL in the micronucleus experiment
	FeSO4	Bianchini 1988		In vivo micronuclei induction in GI tract, oral, mouse	negative
	FeCl3.6H2O	Dunkel 1999		Bacterial mutagenicity – Ames test	negative; not a bacterial mutagen, tested up to 10’000 µg/plate (equivalent to 6001 µg/plate anhydrous FeCl3)
	FeCl3	Dunkel 1999		Mammalian gene mutation – mouse lymphoma assay	negative up to 1030 µg Fe/mL -S9, up to 1.236 µg Fe/mL +S9; cytotoxicity at the highest tested concentrations
	FeCl3	Schulz 2009		In vitro chromosome aberration – micronucleus assay	negative with and without metabolic activation
	FeCl3	Bianchini 1988		In vivo micronuclei induction in GI-tract, oral, mouse	negative; dose related toxic effects were seen in colons of feeding animals and colon and stomach of fasting animals
	The category approach was evaluated based on the following experimental data for the carcinogenicity. Studies are generally considered reliable or represent the sole data point available.
	Test substance	Identifier of the study		Test method & species	Result
	FeCl3	Sato 1992		Similar to OECD 451, rat	no carcinogenic potential
	Iron supplementation	Ullen 1997		Epidemiological, human	Protective effect of iron
	The category approach was evaluated based on the following experimental data for the reproduction toxicity. Studies are generally considered reliable or represent the sole data point available.
	Test substance	Identifier of the study		Test method & animal species	Result based on iron salt	Result based on iron
	FeCl2	Beom 2004		Screening, oral – OECD 422, rat	NOAEL: ≥ 500 mg/kg bw/day	NOAEL: ≥ 220.5 mg/kg bw/day
	FeSO4.7H2O	Furuhashi 2002		Screening, oral – OECD 422, rat	NOAEL: ≥ 1000 mg/kg bw/day	NOAEL: ≥ 200.9 mg/kg bw/day
4.	Conclusions per endpoint for C&L, PBT/vPvB and dose descriptor Considering the acute oral toxicity, the LD50 values show marked variability (220 – 964 mg Fe/kg bw) without clear difference between Fe2+and Fe3+salts. Nevertheless all these iron salt forms clearly have LD50 values > 300 mg/kg bw, that according to CLP support a classification from category 4 to not categorized range and it could be one of the reason to handle the corresponding iron salts as one chemical category. Considering skin irritation, all here mentioned irons salts according their mode of action and following CLP, could be classified as skin irritating – category 2. Considering eye irritation / corrosion, iron salts could be classified as causing irreversible effects on the eye – category 1. An exception may be made for FeSO4, which is already listed in Annex I of the European Plant Protection Products Directive and classified into category 2 (eye irritant). Considering sensitization, in general iron salts are deemed to have a no potential to cause sensitisation that is relevant for classification and could be regarded as one chemical category. Considering repeated dose toxicity, in all reliable oral studies the LOAEL of iron salts is above 100 mg/kg bw (not categorized range according to CLP) and no clear difference between Fe2+and Fe3+is observed. The apparent effect of slightly lower or equal toxicity of FeCl3 in 90-day study is in line with the expectation of bioregulation, i.e. tolerance development. Therefore iron salts could be regarded as one chemical category. In the case of repeated inhalation toxicity, the comparability of the Fe2+and Fe3+oxidation states depends particularly on assumption 1.1.a. of the Category Hypothesis. The scientific evidence of this assumption should be relying on the brother basis of data then they become available. Considering genetic toxicity, iron salts show no genotoxic potential and could be regarded as one chemical category. Considering carcinogenicity, human and animal data together are in agreement that iron is not carcinogenic and accordingly no classification is necessary for the iron salts of this category. Considering toxicity to reproduction, only for FeCl2 and FeSO4 reliable studies (conducted according to OECD TG 422) are available. As the bioavailability of ferrous iron is assumed to be initially higher than that of ferric iron, a read across from these two substances to the ferric salts of this category is deemed valid and conservative. For both substances no adverse effects were seen at the highest tested dose levels where moderate to strong parental toxicity was already present. Based on the available data it can therefore be assumed that the iron salts of this category are not reproductive toxicants. Iron is a bioessential element and uptake of iron is highly regulated by organism, therefore no concerns exist regarding bioaccumulation.

• Anon. 2000. Acute toxicity data submission 96-95. Int J Toxicol 19: No 5. Available from the secondary source OECD (2007, table 145).
• EVM Expert Group on Vitamins and Minerals (2003). Safe upper levels for vitamins and minerals. Report of the Expert Group on Vitamins and Minerals. ISBN 1-904026-11-7 Self-published in May by the Food Standards Agency, U.K. 360 p. http://cot.food.gov.uk/pdfs/vitmin2003.pdf
• ICI (1991). Acute range oral toxicity study to the rat (CT20-126). Unpublished report. Testing laboratory: ICI Central Toxicology Laboratories. Report no.: Internal data Report No CTL/L/4392. Study number: AR5351. Report date: 1991-12-12. Available from the secondary source OECD (2007, table 145).
• OECD Organisation for Economic Co-operation and Development (2007). Chemical Category: Iron Salts. SIDS Initial Assessment Report for SIAM 24, held in Paris, France, 17-20 April. Self-published, Paris, France. 138 p.

Data source

Reference

Reference Type:

publication

Title:

Genotoxicity of iron compounds in Salmonella typhimurium and L5178Y mouse lymphoma cells.

Author:

Dunkel VC, San RHC, Seifried HE, Whittaker P

Year:

1999

Bibliographic source:

DOI 10.1002/(SICI)1098-2280(1999)33:1<28::AID-EM4>3.0.CO;2-S PMID 10037321 Environ Mol Mutagenesis 33(1):28-41.

Materials and methods

Test guideline

Qualifier:

equivalent or similar to guideline

Guideline:

OECD Guideline 476 (In Vitro Mammalian Cell Gene Mutation Test)

Deviations:

not specified

Principles of method if other than guideline:

Method: other: Clive et al/1975 and 1979

GLP compliance:

not specified

Type of assay:

mammalian cell gene mutation assay

Test material

Test material information

Constituent 1

Reference substance name:

7782-63-0

Cas Number:

7782-63-0

IUPAC Name:

7782-63-0

Constituent 2

Reference substance name:

ferrous sulfate (FeSO4 x 7 H2O)

IUPAC Name:

ferrous sulfate (FeSO4 x 7 H2O)

Test material form:

not specified

Details on test material:

Ferrous Sulfate
The substance tested was the heptahydrate (CAS number 7782-63-0), but this does not affect the chemical species available to the test organisms.
Purity: 99%
Source: Sigma

Method

Target gene:

thymidine kinase (TK) gene

Species / strain

Species / strain / cell type:

mouse lymphoma L5178Y cells

Details on mammalian cell type (if applicable):

- Type and identity of media: Fischer's medium for leukemic cells of mice supplemented with 10% horse serum and 0.02% pluronic F-68
- Properly maintained: yes
- Periodically checked for Mycoplasma contamination: yes
- Periodically checked for karyotype stability: no data
- Periodically "cleansed" against high spontaneous background: no data

Additional strain / cell type characteristics:

not specified

Metabolic activation:

with and without

Metabolic activation system:

Aroclor induced rat liver S9

Test concentrations with justification for top dose:

without S9: 309-1030 µg Fe/mL (i.e. mg Fe/L)
with S9: 0.206-1.236 µg Fe/mL (i.e. mg Fe/L)

Vehicle / solvent:

Vehicle used: Distilled water

Controls

Controls 1

Untreated negative controls:

no

Negative solvent / vehicle controls:

yes

Remarks:

destilled water

True negative controls:

no

Positive controls:

yes

Positive control substance:

ethylmethanesulphonate

Remarks:

without metabolic activation

Controls 2

Untreated negative controls:

no

Negative solvent / vehicle controls:

yes

Remarks:

destilled water

True negative controls:

no

Positive controls:

yes

Positive control substance:

7,12-dimethylbenzanthracene

Remarks:

with metabolic activation

Details on test system and experimental conditions:

METHOD OF APPLICATION: in medium

DURATION
- Preincubation period: no data
- Exposure duration: 4 hours
- Expression time (cells in growth medium): 48 hours
- Selection time (if incubation with a selection agent): 10-12 days
- Fixation time (start of exposure up to fixation or harvest of cells): 12-14 days

SELECTION AGENT (mutation assays): trifluorothymidine (final concentration 3ug/ml) added to cloning mediumfor mutant selection

NUMBER OF REPLICATIONS: duplicate cultures

NUMBER OF CELLS EVALUATED: 1E+06 for mutant selection, 200/plate for mutant count, colony size range 0.2-1.1 mm, counted with an Artek automated colony counter

DETERMINATION OF CYTOTOXICITY
- Method: relative total growth

Doses of test compound selected for mutagenicity assay were within the range yielding approximately 0-90% cytotoxicity

Evaluation criteria:

Doubling of mutant frequency over  concurrent solvent treated control value together with dose relationship.

Statistics:

Colonies larger than 0.1 mm diameter were counted with an Artek automated colony counter. Colony size was also determined.

Results and discussion

Test results

Species / strain:

mouse lymphoma L5178Y cells

Metabolic activation:

with and without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

cytotoxicity

Vehicle controls validity:

valid

Untreated negative controls validity:

not examined

Positive controls validity:

valid

Remarks on result:

other: all strains/cell types tested

Remarks:

Migrated from field 'Test system'. Remarks: L5178Y TK+/- 3.7.C mouse lymphoma cells

Any other information on results incl. tables

Table 1: Number of revertants per plate (mean of 2 plates) 

Concentration µg Fe/ml	Mutant frequency*	Growth**	Concentration µg Fe/ml	Mutant frequency*	Growth**
	-S9	-S9		+S9	+S9
0*	25	100	0	27	100
20.1	25	84.5	0.804	53	61.5
50.25	29	72.5	1.005	67	54.5
100.5	32	51.5	1.206	70	41.5
150.75	46	27.5	1.508	88	5.0
201.0	80	10.5
Positive control	430	46	Positive control	171	52

* per 1E+06 survivors

** as % of control

PRECIPITATION CONCENTRATION: None reported

Applicant's summary and conclusion

Conclusions:

Interpretation of results (migrated information):
negative

In the mouse lymphoma TK+/- assay ferrous sulphate FeSO4 x 7H2O showed a weak positive response in the absence of metabolic activation at cytotoxic concentrations and a dose-related increase in mutant frequency in the presence of metabolic activation, with marked increase of cytotoxicity.

Executive summary:

A L5178Y TK+/- mouse lymphoma cell assay was conducted with FeSO4 x 7H2O. The cytotoxicity of the test substance was determined with and without metabolic activation (rat liver S9 mix) prior to the mutagenicity test in order to determine the appropriate testing concentrations. The mutagenicity assay was performed in duplicates. As indication of a positive effect doubling of the mutant frequency was used. The mouse lymphoma cells showed with and without metablic activation a weak increase in the number of induced mutants at cytotoxic concentrations.

FeSO4 x 7H2O is negative for mutagenicity in absence and presence of metabolic activation under the conditions of this test system.