Registration Dossier

Data platform availability banner - registered substances factsheets

Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.

The new ECHA CHEM database has been released by ECHA, and it now contains all REACH registration data. There are more details on the transition of ECHA's published data to ECHA CHEM here.

Diss Factsheets

Use of this information is subject to copyright laws and may require the permission of the owner of the information, as described in the ECHA Legal Notice.

REACH

Mill scale (ferrous metal)

EC number: 266-007-8 | CAS number: 65996-74-9 The oxidized surface of steel produced during reheating, conditioning, hot rolling, and hot forming operations. This substance is usually removed by process waters used for descaling, roll and material cooling, and other purposes. It is subsequently recovered by gravity separation techniques. Composed primarily of high-purity iron oxides. May contain varying amounts of other oxides, elements, and trace compounds.

Life Cycle description
Uses advised against

Toxicological information

Endpoint summary

Administrative data

Link to relevant study record(s)

Referenceopen all close all

Reference 1

Endpoint:

basic toxicokinetics in vitro / ex vivo

Remarks:

Bioaccessibility in artificial physiological fluids

Type of information:

experimental study

Adequacy of study:

supporting study

Study period:

Scientific paper published in 2015, experimental work done in the year(s) before.

Reliability:

2 (reliable with restrictions)

Objective of study:

bioaccessibility (or bioavailability)

Principles of method if other than guideline:

The paper describes studies on the solubility of powders of the metals Fe and Mo, the oxides MoO2 and MoO3 and the alloy ferromolybdenum in different artificial physiological fluids. Characteristics of particles and their surfaces (e.g. size, shape, surface area, surface oxidation state) were also investigated and are discussed. The publications itself focusses on the molybdenum compounds, and not all experimental techniques were applied to the iron powder sample. This study record focusses on the data on iron metal powder, and results and methods that were only applied to the other samples are omitted from this study record. Further details in field "Any other information on materials and methods incl. tables" in the technical dossier (IUCLID).

GLP compliance:

Specific details on test material used for the study:

The test material was characterised by the authors.
- Specific surface area (alaysed by XPS, BET (m2/g)): iron metal: 0.032
- Particle size distribution*: approximately 100 µm,
iron metal immersed in PBS: d0.1 (µm): 50.4, d0.5 (µm): 110, d0.9 (µm): 215;
dry iron metal: d0.1 (µm): 50.7, d0.5 (µm): 96.6, d0.9 (µm): 177.

*10, 50, and 90 vol% cut-off values (lm) for each powder when immersed in PBS (relative standard deviation <10%), and at dry conditions (relative standard deviation <21%).

Type:

absorption

Results:

in-vitro bioaccessibility (wt%, corresp. to mg/L as the loading was 100 mg/L): GST 2h: 10.2±0.2, GST 24h: 65±6, ALF 2h: 4.5±1, ALF 24h: 33±8, ASW 2h: ca. 0.05-0.12, ASW 24h: ca. 1, GMB&PBS, 2&24h all < 0.12 (

Sample characteristics:

Relative surface area (BET method): 0.032 m²/g

Particle size distribution, by laser diffraction:
dry powder: D0.1 = 50.4 µm, D0.5 = 110 µm, D0.9 = 215 µm
powder immersed in phosphate buffered saline: D0.1 = 50.7 µm, D0.5 = 96.6 µm, D0.9 = 177 µm

XPS: Binding energiesa (eV) of main peaks of O 1s and Fe 2p3/2 observed for iron powder: 530.7 (O 1s), and 711.9 (2p3/2). The contribution of oxygen related to oxidized carbon constituents in the surface oxide was minor. The observed peaks were assigned to Fe(ox) according to (Fujii et al., 1999; Grosvenor et al., 2004), this showing surface oxidation of the iron powder (not further details reported/discussed on this iron powder).

Bioaccessibility results:

Table: Bioaccessibility of iron from pure iron powder: Fraction of the test item that has dissolved (wt%). Exact numeric values are not reported in text or table in the publication; results read from Figures 4a/b in the publication. Since the loading was 100 mg/L, the same numeric value also stands for the measured dissolved concentration of dissolved iron in mg/L.

	2 hours	24 hours
GST, pH 1.5	10.2 +- 0.2	65 +- 6
ALF, pH 4.5	4.5 +- 1	33 +- 8
ASW, pH 6.5	ca. 0.05-0.12 *	ca. 1
GMB, pH 7.4	< 0.12 (< LOD)	< 0.12 (< LOD)
PBS, pH 7,4	< 0.12 (< LOD)	< 0.12 (< LOD)

*The result for iron in ASW after 2 hours is reported as “similar to the LOD” but the exact LOD for iron in ASW not reported. Instead, the LODs for iron in GST (LOD=0.053 mg/L) and ALF (LOD=0.121 mg/L) are reported as the range of LODs for iron in the five fluids. Thus, the release of iron in ASW at 2h was somewhere in this range of 0.053-0.121 mg/L. that is 0.053-0.121 % at the loading of 100 mg/L.

Reference 2

Endpoint:: basic toxicokinetics in vitro / ex vivo
Remarks:: Bioaccessibility
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: 2018-02-26 to 2018-09-26
Reliability:: 1 (reliable without restriction)
Rationale for reliability incl. deficiencies:: test procedure in accordance with national standard methods
Objective of study:: bioaccessibility (or bioavailability)
Qualifier:: according to guideline
Guideline:: OECD Series on Testing and Assessment No. 29 (23-Jul-2001): Guidance document on transformation/dissolution of metals and metal compounds in aqueous media
Principles of method if other than guideline:: An internationally agreed guideline does not exist for this test (e.g. OECD). However, similar tests have been conducted with several metal compounds in previous risk assessments (completed under Regulation (EEC) No 793/93) and in recent preparation for REACH regulation (EC) No 1907/2006. The test was conducted on the basis of the guidance for OECD-Series on testing and assessment Number 29 and according to the bioaccessibility test protocol provided by the study monitor. The test media were artificial physiological media: gastric fluid (GST), phosphate-buffered saline (PBS), artificial lysosomal fluid (ALF) and Gamble's solution (GMB)
GLP compliance:: yes (incl. QA statement)
Remarks:: signed 2016-05-31
Radiolabelling:: no
Species:: other: in vitro (simulated human body fluids)
Details on test animals or test system and environmental conditions:: Test principle in brief:
- four different artificial physiological media,
- single loading of test substance of ~100 mg/L,
- GST and PBS media: samples taken after 2 and 24 hours agitation (100 rpm) at 37 ± 2 °C
- GMB and ALF media: samples taken after 2, 24 and 168 hours agitation (100 rpm) at 37 ± 2 °C

- two additional method blanks per medium, measurement (ICP-OES) of dissolved Fe concentrations after filtration and centrifugal filtration.
- the study was performed in triplicate

The aim of this test was to assess the dissolution of triiron tetraoxide (Ferroxide Black 86) in four artificial physiological media: Phosphate buffered saline (PBS, pH 7.2-7.4), Artificial gastric fluid (GST, 1.5-1.6), artificial lysosomal fluid (ALF) and Gamble’s solution (GMB). The test media were selected to simulate relevant human-chemical interactions (as far as practical), e.g. a substance entering the human body by ingestion into the gastro-intestinal tract (GST) or via the respiratory system (ALF).
Duration and frequency of treatment / exposure:: Iron concentrations in GST and PBS were determined after 2 and 24 h whereas iron concentrations in GMB and ALF media were assessed after 2, 24 and 168 hours of incubation.
Dose / conc.:: 100 other: mg of test item/L artificial media
Details on study design:: Test setup
Three replicate flasks (500 mL glass flasks) per test medium (PBS, GST) were prepared with a loading of ~ 100 mg/L. The test item was weighed into flasks, adjusted to volume with the respective artificial physiological medium and agitated at 100 rpm at 37°C ± 2°C. Two control blank replicates (same procedure) per test medium were also prepared.
Three replicates containing the test item and two method blanks per artificial medium were tested. All solutions were sampled after 2 and 24 h whereas GMB and ALF media were also sampled after 168h to measure total dissolved Fe concentrations (ICP-OES) after 0.2 µm filtration (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) and centrifugal filtration (i.e. 0.2 μm filtration and 3kDa centrifugal filtration, Sartorius, Göttingen, Germany). In addition, temperature, pH and observations, including the appearance of the solution (including colour, turbidity and particle film on the surface) were recorded.

Sample fortification:
In addition, samples of the artificial physiological media were fortified with a known amount of iron (by standard addition of commercial standards) to determine the standard recovery. For detailed information please refer to "Any other information on materials and methods incl. tables".

Mass balance:
After the test, aqua regia (3 : 1 mixture of concentrated hydrochloric and nitric acid) was added to the vessels containing the test item to reach a final volume of 500 mL, i.e. 120 mL aqua regia were added to approximately 380 mL GST or PBS medium, 180 mL aqua regia were added to approximately 320 mL ALF or GMB medium. From these solutions, 50 mL were taken after 3 - 14 days of “digestion” for mass balance determination.
The filters (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) used for sampling were extensively rinsed with a known volume of aqua regia (ca. 2.5 mL). The added aqua regia was let to drop slowly through the filters and was collected in a clean vial. This procedure was repeated with every syringe and filter used during the study. After collection, the volume was filled up to exactly 10 mL (for the media GST and PBS) and up to 15 mL (for the ALF and GMB media) with aqua regia. Afterwards the concentration of iron in the “filtrated” aqua regia was determined and considered for the determination of the mass balances.

Reagents:
Purified water (resistivity > 18 MΩ·cm, Pure Lab Ultra water purification system from ELGA LabWater, Celle, Germany)
Nitric acid - “Supra” quality (ROTIPURAN® supplied by Roth, Karlsruhe, Germany).
Hydrochloric acid – “Baker-instra-analyzed-plus” quality (J.T. Baker, Griesheim, Germany).
Sodiumhydroxide – pro Analysis quality (Chemsolute, Th. Geyer, Renningen, Germany)
MgCl2 x 6H2O (p.A., Merck, Darmstadt, Germany)
NaCl (p.A., Chemsolute, Renningen, Germany + Merck, Darmstadt, Germany (new GMB medium))
KCl (p.A., Chemsolute, Renningen, Germany + Merck, Darmstadt, Germany (new GMB medium))
Na2HPO4 (p.A. Merck, Darmstadt, Germany)
Na2SO4 (p.A. Merck, Darmstadt, Germany)
CaCl2 x 2H2O (p.A. Merck, Darmstadt, Germany)
NaAcetate (suprapur Merck, Darmstadt, Germany)
NaHCO3 (p.A. Merck, Darmstadt, Germany)
NaOH (p.A., Chemsolute, Renningen, Germany)
Citric acid anhydrous (p.A., Roth, Karlsruhe, Germany)
Glycine (p.A., Merck, Darmstadt, Germany)
Na3Citrate x 2H2O (p.A., Merck, Darmstadt, Germany)
Na2Tartrate x 2H2O (p.A., Merck, Darmstadt, Germany)
NaLactate (98+% Sigma Aldrich, Munich, Germany)
NaPyruvate (p.A., Applichem, Darmstadt, Germany)
KH2PO4 (p.A., Merck, Darmstadt, Germany)
Urea (pure, Applichem, Darmstadt, Germany)
Lactic acid (purum, Fluka, Munich, Germany)
HCl 30% (instra-analyzed, plus J.T. Baker, Griesheim, Germany)

METAL ANALYSIS
- Standards for metal analysis: A commercially available single element standard was used as iron standard (Merck Certipur Iron ICP standard 1000 mg/L lot no. HC68868126; Darmstadt, Germany) to prepare an appropriate stock solution and subsequently calibration solutions for ICP-OES measurements
- Certified reference materials: As quality control standards, certified aqueous reference material TM-DWS.3 (lot no. 0916) and TMDA-70.2 (lot no. 0916 and 0917) obtained from Environment Canada and a multielement standard (Merck Certipur IV ICP standard 1000 mg/L lot no. HC54938555 and HC73962555; Darmstadt, Germany) were analysed for total dissolved iron by ICP-OES.

Instrumental and analytical set-up for the ICP-OES instrument:
Agilent 720, Agilent Technologies, Waldbronn, Germany
Nebulizer: Sea spray nebulizer, from Glass Expansion
Spray chamber: Iso Mist with Twister Helix from Glass Expansion
Plasma stabilization time: at least 30 min before start of the measurements
Plasma gas flow: 15.0 L/min
Additional gas flow: 1.50 L/min
Carrier gas flow: 0.75 L/min
RF power: 1200W
Stabilization time of sample: 15 sec
Repetition time (three internal measurements per sample): 30 sec
Wavelengths: Fe: 238.204 nm, 240.489 nm, 241.052 nm, 258.588 nm and 259.940 nm

- Correlation coefficients (r) for the wavelengths used for evaluation of data were at least >0.999603

The applied LOD/LOQ calculations for the Agilent 720 ICP-OES:
LOD: 3 * standard deviation of calibration blank/slope of the calibration
LOQ: 3 * LOD
The resulting LODs/LOQs are reported in "Any other information on results incl. tables"
Details on dosing and sampling:: Loading:
Detailed loadings of the test vessels are given in "Any other information on materials and methods incl. tables".
Type:: other: Bioaccessibility ALF, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 742 µg Fe/L (dissolved)
Type:: other: Bioaccessibility ALF, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 30824 µg Fe/L (dissolved)
Type:: other: Bioaccessibility ALF, 168h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 75367 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GST, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 296 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GST, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 1186 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GMB, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 0.106 µg Fe/L
Type:: other: Bioaccessibility GMB, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: < LOD (0.19 µg Fe/L)
Type:: other: Bioaccessibility GMB, 168h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 2.16 µg Fe/L (dissolved)
Type:: other: Bioaccessibility PBS, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:
Type:: other: Bioaccessibility PBS, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: < LOD/LOQ (2.28 µg Fe/L) (dissolved)
Bioaccessibility (or Bioavailability) testing results:: Please refer to "any other information on results incl. tables" below.
Conclusions:: The bioaccessibility of Ferroxide Black 86 was determined in vitro by simulating dissolution under physiological conditions considered to mimic artificial body fluids with a loading of 100 mg test item/L. After 2 and 24 h in phosphate buffered saline (PBS, pH 7.2-7.4) solution, dissolved iron concentrations (operationally defined as the dissolved Fe fraction after 0.2 µm filtration and centrifugal filtration (~2.1 nm)) were below the LOD (<0.761 µg/L). With mean concentrations of 0.106 µg Fe/L after 2 h and 2.16 µg Fe/L after 168 h as well as a negative value (after subtraction of background) after 24h, dissolved iron concentrations in Gamble´s solution (GMB, pH 7.4) were also very low. In artificial gastric fluid (GST, pH 1.5-1.6), 296 µg Fe/L and 1.186 µg Fe/L were detected in the dissolved phase after 2 and 24 h, respectively. Mean iron concentrations were highest in artificial lysosomal fluid (ALF, pH 4.5): 742 µg/L, 30,824 µg/L and 75,367 µg/L of iron were found in the dissolved phase after 2, 24 and 168 h. Therefore, with a maximum mean released fraction of ~7.5% after 168 h, the dissolution of diiron trioxide was highest in artificial lysosomal fluid (ALF).

Reference 3

Endpoint:: basic toxicokinetics in vitro / ex vivo
Remarks:: Bioaccessibility
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: 2018-02-26 to 2018-09-26
Reliability:: 1 (reliable without restriction)
Rationale for reliability incl. deficiencies:: test procedure in accordance with national standard methods
Objective of study:: bioaccessibility (or bioavailability)
Qualifier:: according to guideline
Guideline:: OECD Series on Testing and Assessment No. 29 (23-Jul-2001): Guidance document on transformation/dissolution of metals and metal compounds in aqueous media
Principles of method if other than guideline:: An internationally agreed guideline does not exist for this test (e.g. OECD). However, similar tests have been conducted with several metal compounds in previous risk assessments (completed under Regulation (EEC) No 793/93) and in recent preparation for REACH regulation (EC) No 1907/2006. The test was conducted on the basis of the guidance for OECD-Series on testing and assessment Number 29 and according to the bioaccessibility test protocol provided by the study monitor. The test media were artificial physiological media: gastric fluid (GST), phosphate-buffered saline (PBS), artificial lysosomal fluid (ALF) and Gamble's solution (GMB)
GLP compliance:: yes (incl. QA statement)
Remarks:: signed 2016-05-31
Radiolabelling:: no
Species:: other: in vitro (simulated human body fluids)
Details on test animals or test system and environmental conditions:: Test principle in brief:
- four different artificial physiological media,
- single loading of test substance of ~100 mg/L,
- GST and PBS media: samples taken after 2 and 24 hours agitation (100 rpm) at 37 ± 2 °C
- GMB and ALF media: samples taken after 2, 24 and 168 hours agitation (100 rpm) at 37 ± 2 °C

- two additional method blanks per medium, measurement (ICP-OES) of dissolved Fe concentrations after filtration and centrifugal filtration.
- the study was performed in triplicate

The aim of this test was to assess the dissolution of iron hydroxide oxide yellow (Sicovit Yellow 10 E172) in four artificial physiological media: Phosphate buffered saline (PBS, pH 7.2-7.4), Artificial gastric fluid (GST, 1.5-1.6), artificial lysosomal fluid (ALF) and Gamble’s solution (GMB). The test media were selected to simulate relevant human-chemical interactions (as far as practical), e.g. a substance entering the human body by ingestion into the gastro-intestinal tract (GST) or via the respiratory system (ALF).
Duration and frequency of treatment / exposure:: Iron concentrations in GST and PBS were determined after 2 and 24 h whereas iron concentrations in GMB and ALF media were assessed after 2, 24 and 168 hours of incubation.
Dose / conc.:: 100 other: mg of test item/L artificial media
Details on study design:: Test setup
Three replicate flasks (500 mL glass flasks) per test medium (PBS, GST) were prepared with a loading of ~ 100 mg/L. The test item was weighed into flasks, adjusted to volume with the respective artificial physiological medium and agitated at 100 rpm at 37°C ± 2°C. Two control blank replicates (same procedure) per test medium were also prepared.
Three replicates containing the test item and two method blanks per artificial medium were tested. All solutions were sampled after 2 and 24 h whereas GMB and ALF media were also sampled after 168h to measure total dissolved Fe concentrations (ICP-OES) after 0.2 µm filtration (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) and centrifugal filtration (i.e. 0.2 μm filtration and 3kDa centrifugal filtration, Sartorius, Göttingen, Germany). In addition, temperature, pH and observations, including the appearance of the solution (including colour, turbidity and particle film on the surface) were recorded.

Sample fortification:
In addition, samples of the artificial physiological media were fortified with a known amount of iron (by standard addition of commercial standards) to determine the standard recovery. For detailed information please refer to "Any other information on materials and methods incl. tables".

Mass balance:
After the test, aqua regia (3 : 1 mixture of concentrated hydrochloric and nitric acid) was added to the vessels containing the test item to reach a final volume of 500 mL, i.e. 120 mL aqua regia were added to approximately 380 mL GST or PBS medium, 180 mL aqua regia were added to approximately 320 mL ALF or GMB medium. From these solutions, 50 mL were taken after 3 - 14 days of “digestion” for mass balance determination.
The filters (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) used for sampling were extensively rinsed with a known volume of aqua regia (ca. 2.5 mL). The added aqua regia was let to drop slowly through the filters and was collected in a clean vial. This procedure was repeated with every syringe and filter used during the study. After collection, the volume was filled up to exactly 10 mL (for the media GST and PBS) and up to 15 mL (for the ALF and GMB media) with aqua regia. Afterwards the concentration of iron in the “filtrated” aqua regia was determined and considered for the determination of the mass balances.

Reagents:
Purified water (resistivity > 18 MΩ·cm, Pure Lab Ultra water purification system from ELGA LabWater, Celle, Germany)
Nitric acid - “Supra” quality (ROTIPURAN® supplied by Roth, Karlsruhe, Germany).
Hydrochloric acid – “Baker-instra-analyzed-plus” quality (J.T. Baker, Griesheim, Germany).
Sodiumhydroxide – pro Analysis quality (Chemsolute, Th. Geyer, Renningen, Germany)
MgCl2 x 6H2O (p.A., Merck, Darmstadt, Germany)
NaCl (p.A., Chemsolute, Renningen, Germany + Merck, Darmstadt, Germany (new GMB medium))
KCl (p.A., Chemsolute, Renningen, Germany + Merck, Darmstadt, Germany (new GMB medium))
Na2HPO4 (p.A. Merck, Darmstadt, Germany)
Na2SO4 (p.A. Merck, Darmstadt, Germany)
CaCl2 x 2H2O (p.A. Merck, Darmstadt, Germany)
NaAcetate (suprapur Merck, Darmstadt, Germany)
NaHCO3 (p.A. Merck, Darmstadt, Germany)
NaOH (p.A., Chemsolute, Renningen, Germany)
Citric acid anhydrous (p.A., Roth, Karlsruhe, Germany)
Glycine (p.A., Merck, Darmstadt, Germany)
Na3Citrate x 2H2O (p.A., Merck, Darmstadt, Germany)
Na2Tartrate x 2H2O (p.A., Merck, Darmstadt, Germany)
NaLactate (98+% Sigma Aldrich, Munich, Germany)
NaPyruvate (p.A., Applichem, Darmstadt, Germany)
KH2PO4 (p.A., Merck, Darmstadt, Germany)
Urea (pure, Applichem, Darmstadt, Germany)
Lactic acid (purum, Fluka, Munich, Germany)
HCl 30% (instra-analyzed, plus J.T. Baker, Griesheim, Germany)

METAL ANALYSIS
- Standards for metal analysis: A commercially available single element standard was used as iron standard (Merck Certipur Iron ICP standard 1000 mg/L lot no. HC68868126; Darmstadt, Germany) to prepare an appropriate stock solution and subsequently calibration solutions for ICP-OES measurements
- Certified reference materials: As quality control standards, certified aqueous reference material TM-DWS.3 (lot no. 0916) and TMDA-70.2 (lot no. 0916 and 0917) obtained from Environment Canada and a multielement standard (Merck Certipur IV ICP standard 1000 mg/L lot no. HC54938555 and HC73962555; Darmstadt, Germany) were analysed for total dissolved iron by ICP-OES.

Instrumental and analytical set-up for the ICP-OES instrument:
Agilent 720, Agilent Technologies, Waldbronn, Germany
Nebulizer: Sea spray nebulizer, from Glass Expansion
Spray chamber: Iso Mist with Twister Helix from Glass Expansion
Plasma stabilization time: at least 30 min before start of the measurements
Plasma gas flow: 15.0 L/min
Additional gas flow: 1.50 L/min
Carrier gas flow: 0.75 L/min
RF power: 1200W
Stabilization time of sample: 15 sec
Repetition time (three internal measurements per sample): 30 sec
Wavelengths: Fe: 238.204 nm, 240.489 nm, 241.052 nm, 258.588 nm and 259.940 nm

- Correlation coefficients (r) for the wavelengths used for evaluation of data were at least >0.999603

The applied LOD/LOQ calculations for the Agilent 720 ICP-OES:
LOD: 3 * standard deviation of calibration blank/slope of the calibration
LOQ: 3 * LOD
The resulting LODs/LOQs are reported in "Any other information on results incl. tables"
Details on dosing and sampling:: Loading:
Detailed loadings of the test vessels are given in "Any other information on materials and methods incl. tables".
Type:: other: Bioaccessibility ALF, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 6.96 µg Fe/L (dissolved)
Type:: other: Bioaccessibility ALF, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 29 µg Fe/L (dissolved)
Type:: other: Bioaccessibility ALF, 168h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 80.5 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GST, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 2.24 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GST, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 55.9 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GMB, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: negative value after correction
Type:: other: Bioaccessibility GMB, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 0.258 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GMB, 168h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: negative value after correction
Type:: other: Bioaccessibility PBS, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: < LOD/LOQ (1.54 µg Fe/L)
Type:: other: Bioaccessibility PBS, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: < LOD (0.761 µg Fe/L)
Bioaccessibility (or Bioavailability) testing results:: Please refer to "any other information on results incl. tables" below.
Conclusions:: The bioaccessibility of iron hydroxide oxide yellow (Sicovit yellow 10 E172) was determined in vitro by simulating dissolution under physiological conditions considered to mimic artificial body fluids with a loading of 100 mg test item/L. After 2 and 24 h in phosphate buffered saline (PBS, pH 7.2-7.4) solution, dissolved iron concentrations (operationally defined as the dissolved Fe fraction after 0.2 µm filtration and centrifugal filtration (~2.1 nm)) were at least below the LOQ (<1.54 µg/L). With mean concentrations of 0.258 µg Fe/L after 24 h and negative values (after subtraction of background) after 2 and 168 h, dissolved iron concentrations in Gamble´s solution (GMB, pH 7.4) were also very low. In artificial gastric fluid (GST, pH 1.5-1.6), 2.24 µg Fe/L and 55.9 µg Fe/L were detected in the dissolved phase after 2 and 24 h, respectively. Mean iron concentrations were highest in artificial lysosomal fluid (ALF, pH 4.5): 6.96 µg/L, 29.0 µg/L and 80.5 µg/L of iron were found in the dissolved phase after 2, 24 and 168 h. Therefore, with a maximum mean released fraction of <0.1% after 168 h, the dissolution of iron hydroxide oxide yellow (Sicovit Yellow 10 E172) was highest in artificial lysosomal fluid (ALF).

Reference 4

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: 2018-09-11 to 2019-06-21
Reliability:: 1 (reliable without restriction)
Rationale for reliability incl. deficiencies:: guideline study
Objective of study:: toxicokinetics
Qualifier:: according to guideline
Guideline:: OECD Guideline 417 (Toxicokinetics)
Version / remarks:: 2010-07-22
Deviations:: no
GLP compliance:: yes (incl. QA statement)
Remarks:: signed 2017-05-08
Specific details on test material used for the study:: STABILITY AND STORAGE CONDITIONS OF TEST MATERIAL
- Storage condition of test material: stored in tightly-closed, original container, in a cool and dry place.
Radiolabelling:: no
Species:: rat
Strain:: other: Crl:CD(SD)
Details on species / strain selection:: The rat is a commonly used rodent species for toxicity studies.
Sex:: male/female
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Source: Charles River Laboratories Research, Models and Services Germany GmbH, Sandhofer Weg 7, 97633 Sulzfeld, Germany
- Age at study initiation: 59 to 60 days
- Weight at study initiation: males: 262 g to 340 g; females: 200 g to 272 g
- Housing: kept in groups of up to 3 animals (same sex) in MAKROLON cages (type IV) with a basal surface of approximately 55 cm × 33 cm and a height of approximately 20 cm; bedding material: granulated textured wood
- Diet (ad libitum): Commercial diet, ssniff® R/M-H V1534
- Water (ad libitum): drinking water
- Acclimation period: 14 days

ENVIRONMENTAL CONDITIONS
- Temperature: 22°C ± 3°C (maximum range)
- Relative humidity: 55% ± 10% (maximum range)
- Photoperiod (hrs dark / hrs light): 12/12
Route of administration:: other: Ferroxide black 86: oral (gavage); reference item: intravenously injected
Vehicle:: other: Ferroxide black 86: 0.5 % aqueous hydroxypropylmethylcellulose gel; reference item: 0.9 % NaCl solution
Details on exposure:: PREPARATION OF DOSING SOLUTIONS:
1) Ferroxide black 86:
The test items were suspended or dissolved in the vehicle to the appropriate concentration freshly on the administration day and were administered orally by gavage at a constant volume (adminsitration volume: 10 mL/kg bw). The application formulations were continuously agitated by stirring throughout the entire administration procedure.

2) Reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%)
Prior to administration, the reference item and the appropriate vehicle were heated to 70°C and stirred at 50°C for approx. 3 hours until the reference item was completely dissolved. This clear solution was maintained at room temperature until administration. The status as clear solution was monitored and recorded upon administration. Immediately after formulation preparation for the females, the formulations were protected from light by transferring the formulation into brown containers or wrapping in aluminium foil.

The amounts of the test and reference items were adjusted to the animal's current body weight on the administration day.

Administration volume (oral administration / intravenous administration): 10 mL/kg bw/day

Injection speed (intravenous adminsitration): dose per approx. 15 seconds
Duration and frequency of treatment / exposure:: single administration
Dose / conc.:: 1 000 mg/kg bw/day (actual dose received)
No. of animals per sex per dose / concentration:: 5 males / 5 females
Control animals:: yes, concurrent vehicle
Positive control reference chemical:: none
Details on study design:: - Dose selection rationale: the dose levels for this study were selected after consultation with the sponsor based on available toxicity and bioavailability data (as far as available):

1) Reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%):
The oral LD50 value for iron citrate monohydrate was stated as being >2000 mg/kg bw; the oral bioavailability of soluble Fe substances are given in the public domain with 1 to 26% (Fe).

For the test item oral dosing of 1000 mg/kg bw, a very low relative bioavailability was assumed (<1%), considering the very low water solubility and bioacessibility in gastric juice. Since the four iron oxide test items have Fe-contents of approx. 70%, the dose of the reference substance should be adjusted accordingly. Given a test item dose of 1000 mg/kg b.w. (corresponding to 700 mg Fe/kg bw), then 1% of this dose would correspond to 7 mg Fe/kg bw (or 36.8 mg/kg bw iron citrate). Correcting for approx. 20% oral bioavailability of soluble iron substances, this yields a dose for the reference item of 7.4 mg iron citrate/kg bw to be given by intravenous injection.

2) Ferroxide black 86:
The test item oral doses of 1000 mg/kg bw correspond to the limit dose used in a separate 90-day oral toxicity study, which was considered the maximum feasible dose. This dose was also selected in view of the anticipated low bioavailability and the requirements of analytical sensitivity of the analytical method for iron in plasma.

3) Vehicle control group:
In view of the long established circadian variation of plasma iron levels (Lynch et al, 1973)*, a vehicle control group was sampled for blood plasma over a period of 24 hours at identical sampling time points and intervals as the dosed groups.

*Reference:
Lynch et al (1973): Circadian Variation in Plasma Iron Concentration and Reticuloendothelial Iron Release in the Rat, Clinical Science and Molecular Medicine (1973) 45, 331-336.
Details on dosing and sampling:: TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled (delete / add / specify): urine, faeces, blood, plasma, serum or other tissues, cage washes, bile
- Time and frequency of sampling: blood was collected 0 (predose), 0.5, 1, 2, 4, 8, 12, 24, 48 (test item and reference item only), and 72 hours (test item and reference item only) after administration. The whole blood samples were cooled using an IsoTherm-Rack system until centrifugation. Immediately after centrifugation, the isolated plasma was frozen at -20°C ± 10 % and stored at this temperature until analysis.

Pharmacokinetic evaluation of plasma data was performed and a non-compartment model was employed. The following parameters were determined, if possible:
AUC0-∞ = extrapolated area from zero to infinity
AUC0-t last = extrapolated area from time zero to the last quantifiable plasma concentration (i.e. >lower limit of quantification, LLOQ)
Kel = elimination rate constant
t1/2 = elimination half-life

Cmax values were the highest measured plasma concentrations and tmax values were the time points of highest plasma concentrations.

Elimination rate constants (Kel) and plasma elimination half-lives (t½) were calculated by linear regression analysis of the log/linear portion of the individual plasma concentration-time curves (c = concentration, t = time).

Area under the curve (AUC) values were calculated using the linear trapezoidal method and extrapolated to infinite time by dividing the last measurable plasma concentration by the elimination rate constant. Plasma concentrations at time zero were taken to be those at the first blood sampling time.

Furthermore, the AUC0-t last was calculated according to the linear trapezoidal rule. Values below the limit of quantification (LOQ) were excluded from calculation.

In addition, the bioavailability was calculated for the mixture.

For plasma, a pre-treatment by a microwave digestion with HNO3 was necessary to digest the proteins in plasma. Afterwards iron in digested samples was measured by ICP-OES.

OBSERVATIONS
- clinical signs: before and after dosing as well as regularly throughout the working day (7.30 a.m. to 4.30 p.m.) and on Saturdays and Sundays (8.00 a.m. to 12.00 noon; final check at approx. 4.00 p.m).
Special attention was paid to the local tolerance at the injection site(s).
- mortality/morbund: early in the morning and again in the afternoon of each working day as well as on Saturdays and Sundays (final check at approx. 4.00 p.m).
- body weight: at the time of group allocation, before dosing for dose adjustment and on test day 4 before the last blood sampling.

ADMINISTRATION FORMULATION ALANYSIS:
For each test item, that was mixed with the vehicle and the reference substance, tests by appropriate analytical methods were conducted to determine the concentration and stability of the test item in the formulations. For the analysis of the application formulations, one sample of exactly 10 mL from each dosing suspension (test items) or dosing solution (reference item) was taken at the start of the administration (test day 1 of the female animals) and frozen until analysis.

Application solutions of the iron oxide was measured after addition of aqua regia to the samples and after an incubation time for at least four days by ICP-OES. After this measurement the remaining precipitation (only iron oxide application solution) were digested by a microwave procedure and measured by ICP-OES.

ANALYTIC OF REFERENCE ITEM:
The iron content of the reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%) was determined using ICP-OES.
Statistics:: The test item group was compared to the reference group. The following statistical method was used:
- Student's t-test (body weight (at p≤0.05 and p≤0.01; limits used p = 0.05 approx. t = 2.306 p = 0.01 approx. t = 3.355 (for 8 degrees of freedom))
Preliminary studies:: none
Details on absorption:: not specified
Details on distribution in tissues:: not specified
Details on excretion:: not specified
Toxicokinetic parameters:: other: bioavailability
Remarks:: An absolute bioavailability of 0.23%/0.21% (m/f) for Ferroxide Black was calculated for iron following oral administration compared to intravenous administration.
Toxicokinetic parameters:: other:
Remarks:: It should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
Toxicokinetic parameters:: other:
Remarks:: The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels.
Toxicokinetic parameters:: other:
Remarks:: The calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.
Metabolites identified:: not specified
Details on metabolites:: not specified
Bioaccessibility (or Bioavailability) testing results:: An absolute bioavailability of 0.23%/0.21% (m/f) for Ferroxide Black was calculated for iron following oral administration compared to intravenous administration. However, it should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels. Hence, the calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.

Please also refer for results to the field "Attached background information" below.
Conclusions:: An absolute bioavailability of 0.23%/0.21% (m/f) for Ferroxide Black was calculated for iron following oral administration compared to intravenous administration. However, it should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels. Hence, the calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.

Reference 5

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: 2018-09-11 to 2019-06-21
Reliability:: 1 (reliable without restriction)
Rationale for reliability incl. deficiencies:: guideline study
Objective of study:: toxicokinetics
Qualifier:: according to guideline
Guideline:: OECD Guideline 417 (Toxicokinetics)
Version / remarks:: 2010-07-22
Deviations:: no
GLP compliance:: yes (incl. QA statement)
Remarks:: signed 2017-05-08
Specific details on test material used for the study:: STABILITY AND STORAGE CONDITIONS OF TEST MATERIAL
- Storage condition of test material: at room temperature, kept dry and stored in airtight closed containers.
Radiolabelling:: no
Species:: rat
Strain:: other: Crl:CD(SD)
Details on species / strain selection:: The rat is a commonly used rodent species for toxicity studies.
Sex:: male/female
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Source: Charles River Laboratories Research, Models and Services Germany GmbH, Sandhofer Weg 7, 97633 Sulzfeld, Germany
- Age at study initiation: 59 to 60 days
- Weight at study initiation: males: 262 g to 340 g; females: 200 g to 272 g
- Housing: kept in groups of up to 3 animals (same sex) in MAKROLON cages (type IV) with a basal surface of approximately 55 cm × 33 cm and a height of approximately 20 cm; bedding material: granulated textured wood
- Diet (ad libitum): Commercial diet, ssniff® R/M-H V1534
- Water (ad libitum): drinking water
- Acclimation period: 14 days

ENVIRONMENTAL CONDITIONS
- Temperature: 22°C ± 3°C (maximum range)
- Relative humidity: 55% ± 10% (maximum range)
- Photoperiod (hrs dark / hrs light): 12/12
Route of administration:: other: Ferroxide Red 212P: oral (gavage); reference item: intravenously injected
Vehicle:: other: Ferroxide Red 212P: 0.5 % aqueous hydroxypropylmethylcellulose gel; reference item: 0.9 % NaCl solution
Details on exposure:: PREPARATION OF DOSING SOLUTIONS:
1) Ferroxide Red 212P:
The test items were suspended or dissolved in the vehicle to the appropriate concentration freshly on the administration day and were administered orally by gavage at a constant volume (adminsitration volume: 10 mL/kg bw). The application formulations were continuously agitated by stirring throughout the entire administration procedure.

2) Reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%)
Prior to administration, the reference item and the appropriate vehicle were heated to 70°C and stirred at 50°C for approx. 3 hours until the reference item was completely dissolved. This clear solution was maintained at room temperature until administration. The status as clear solution was monitored and recorded upon administration. Immediately after formulation preparation for the females, the formulations were protected from light by transferring the formulation into brown containers or wrapping in aluminium foil.

The amounts of the test and reference items were adjusted to the animal's current body weight on the administration day.

Administration volume (oral administration / intravenous administration): 10 mL/kg bw/day

Injection speed (intravenous adminsitration): dose per approx. 15 seconds
Duration and frequency of treatment / exposure:: single administration
Dose / conc.:: 1 000 mg/kg bw/day (actual dose received)
No. of animals per sex per dose / concentration:: 5 males / 5 females
Control animals:: yes, concurrent vehicle
Positive control reference chemical:: none
Details on study design:: - Dose selection rationale: the dose levels for this study were selected after consultation with the sponsor based on available toxicity and bioavailability data (as far as available):

1) Reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%):
The oral LD50 value for iron citrate monohydrate was stated as being >2000 mg/kg bw; the oral bioavailability of soluble Fe substances are given in the public domain with 1 to 26% (Fe).

For the test item oral dosing of 1000 mg/kg bw, a very low relative bioavailability was assumed (<1%), considering the very low water solubility and bioacessibility in gastric juice. Since the four iron oxide test items have Fe-contents of approx. 70%, the dose of the reference substance should be adjusted accordingly. Given a test item dose of 1000 mg/kg b.w. (corresponding to 700 mg Fe/kg bw), then 1% of this dose would correspond to 7 mg Fe/kg bw (or 36.8 mg/kg bw iron citrate). Correcting for approx. 20% oral bioavailability of soluble iron substances, this yields a dose for the reference item of 7.4 mg iron citrate/kg bw to be given by intravenous injection.

2) Ferroxide Red 212P:
The test item oral doses of 1000 mg/kg bw correspond to the limit dose used in a separate 90-day oral toxicity study, which was considered the maximum feasible dose. This dose was also selected in view of the anticipated low bioavailability and the requirements of analytical sensitivity of the analytical method for iron in plasma.

3) Vehicle control group:
In view of the long established circadian variation of plasma iron levels (Lynch et al, 1973)*, a vehicle control group was sampled for blood plasma over a period of 24 hours at identical sampling time points and intervals as the dosed groups.

*Reference:
Lynch et al (1973): Circadian Variation in Plasma Iron Concentration and Reticuloendothelial Iron Release in the Rat, Clinical Science and Molecular Medicine (1973) 45, 331-336.
Details on dosing and sampling:: TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled: plasma
- Time and frequency of sampling: blood was collected 0 (predose), 0.5, 1, 2, 4, 8, 12, 24, 48 (test item and reference item only), and 72 hours (test item and reference item only) after administration. The whole blood samples were cooled using an IsoTherm-Rack system until centrifugation. Immediately after centrifugation, the isolated plasma was frozen at -20°C ± 10 % and stored at this temperature until analysis.

Pharmacokinetic evaluation of plasma data was performed and a non-compartment model was employed. The following parameters were determined, if possible:
AUC0-∞ = extrapolated area from zero to infinity
AUC0-t last = extrapolated area from time zero to the last quantifiable plasma concentration (i.e. >lower limit of quantification, LLOQ)
Kel = elimination rate constant
t1/2 = elimination half-life

Cmax values were the highest measured plasma concentrations and tmax values were the time points of highest plasma concentrations.

Elimination rate constants (Kel) and plasma elimination half-lives (t½) were calculated by linear regression analysis of the log/linear portion of the individual plasma concentration-time curves (c = concentration, t = time).

Area under the curve (AUC) values were calculated using the linear trapezoidal method and extrapolated to infinite time by dividing the last measurable plasma concentration by the elimination rate constant. Plasma concentrations at time zero were taken to be those at the first blood sampling time.

Furthermore, the AUC0-t last was calculated according to the linear trapezoidal rule. Values below the limit of quantification (LOQ) were excluded from calculation.

In addition, the bioavailability was calculated for the mixture.

For plasma, a pre-treatment by a microwave digestion with HNO3 was necessary to digest the proteins in plasma. Afterwards iron in digested samples was measured by ICP-OES.

OBSERVATIONS
- clinical signs: before and after dosing as well as regularly throughout the working day (7.30 a.m. to 4.30 p.m.) and on Saturdays and Sundays (8.00 a.m. to 12.00 noon; final check at approx. 4.00 p.m).
Special attention was paid to the local tolerance at the injection site(s).
- mortality/morbund: early in the morning and again in the afternoon of each working day as well as on Saturdays and Sundays (final check at approx. 4.00 p.m).
- body weight: at the time of group allocation, before dosing for dose adjustment and on test day 4 before the last blood sampling.

ADMINISTRATION FORMULATION ALANYSIS:
For each test item, that was mixed with the vehicle and the reference substance, tests by appropriate analytical methods were conducted to determine the concentration and stability of the test item in the formulations. For the analysis of the application formulations, one sample of exactly 10 mL from each dosing suspension (test items) or dosing solution (reference item) was taken at the start of the administration (test day 1 of the female animals) and frozen until analysis.

Application solutions of the iron oxide was measured after addition of aqua regia to the samples and after an incubation time for at least four days by ICP-OES. After this measurement the remaining precipitation (only iron oxide application solution) were digested by a microwave procedure and measured by ICP-OES.

ANALYTIC OF REFERENCE ITEM:
The iron content of the reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%) was determined using ICP-OES.
Statistics:: The test item group was compared to the reference group. The following statistical method was used:
- Student's t-test (body weight (at p≤0.05 and p≤0.01; limits used p = 0.05 approx. t = 2.306 p = 0.01 approx. t = 3.355 (for 8 degrees of freedom))
Preliminary studies:: none
Details on absorption:: not specified
Details on distribution in tissues:: not specified
Details on excretion:: not specified
Toxicokinetic parameters:: other: bioavailability
Remarks:: An absolute bioavailability of 0.21%/0.21% (m/f) for Ferroxide Red 212P was calculated for iron following oral administration compared to intravenous administration.
Toxicokinetic parameters:: other:
Remarks:: It should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
Toxicokinetic parameters:: other:
Remarks:: The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels.
Toxicokinetic parameters:: other:
Remarks:: The calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.
Metabolites identified:: not specified
Details on metabolites:: not specified
Bioaccessibility (or Bioavailability) testing results:: An absolute bioavailability of 0.21%/0.21% (m/f) for Ferroxide Red 212P was calculated for iron following oral administration compared to intravenous administration. However, it should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels. Hence, the calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.

Please also refer for results to the field "Attached background information" below.
Conclusions:: An absolute bioavailability of 0.21%/0.21% (m/f) for Ferroxide Red 212P was calculated for iron following oral administration compared to intravenous administration. However, it should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels. Hence, the calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.

Reference 6

Endpoint:: basic toxicokinetics in vitro / ex vivo
Remarks:: Bioaccessibility
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: 2018-02-26 to 2018-09-26
Reliability:: 1 (reliable without restriction)
Rationale for reliability incl. deficiencies:: test procedure in accordance with national standard methods
Objective of study:: bioaccessibility (or bioavailability)
Qualifier:: equivalent or similar to guideline
Guideline:: OECD Series on Testing and Assessment No. 29 (23-Jul-2001): Guidance document on transformation/dissolution of metals and metal compounds in aqueous media
Version / remarks:: test medium, loading and test duration adopted for toxicokinetics assessment
Principles of method if other than guideline:: An internationally agreed guideline does not exist for this test (e.g. OECD). However, similar tests have been conducted with several metal compounds in previous risk assessments (completed under Regulation (EEC) No 793/93) and in recent preparation for REACH regulation (EC) No 1907/2006. The test was conducted on the basis of the guidance for OECD-Series on testing and assessment Number 29 and according to the bioaccessibility test protocol provided by the study monitor. The test media were artificial physiological media: gastric fluid (GST), phosphate-buffered saline (PBS), artificial lysosomal fluid (ALF) and Gamble's solution (GMB)
GLP compliance:: yes (incl. QA statement)
Remarks:: signed 2016-05-31
Radiolabelling:: no
Species:: other: in vitro (simulated human body fluids)
Details on test animals or test system and environmental conditions:: Test principle in brief:
- four different artificial physiological media,
- single loading of test substance of ~100 mg/L,
- GST and PBS media: samples taken after 2 and 24 hours agitation (100 rpm) at 37 ± 2 °C
- GMB and ALF media: samples taken after 2, 24 and 168 hours agitation (100 rpm) at 37 ± 2 °C

- two additional method blanks per medium, measurement (ICP-OES) of dissolved Fe concentrations after filtration and centrifugal filtration.
- the study was performed in triplicate

The aim of this test was to assess the dissolution of nano-sized diiron trioxide in four artificial physiological media: Phosphate buffered saline (PBS, pH 7.2-7.4), Artificial gastric fluid (GST, 1.5-1.6), artificial lysosomal fluid (ALF) and Gamble’s solution (GMB). The test media were selected to simulate relevant human-chemical interactions (as far as practical), e.g. a substance entering the human body by ingestion into the gastro-intestinal tract (GST) or via the respiratory system (ALF).
Duration and frequency of treatment / exposure:: Iron concentrations in GST and PBS were determined after 2 and 24 h whereas iron concentrations in GMB and ALF media were assessed after 2, 24 and 168 hours of incubation.
Dose / conc.:: 100 other: mg of test item/L artificial media
Details on study design:: Test setup
Three replicate flasks (500 mL glass flasks) per test medium (PBS, GST) were prepared with a loading of ~ 100 mg/L. The test item was weighed into flasks, adjusted to volume with the respective artificial physiological medium and agitated at 100 rpm at 37°C ± 2°C. Two control blank replicates (same procedure) per test medium were also prepared.
Three replicates containing the test item and two method blanks per artificial medium were tested. All solutions were sampled after 2 and 24 h whereas GMB and ALF media were also sampled after 168h to measure total dissolved Fe concentrations (ICP-OES) after 0.2 µm filtration (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) and centrifugal filtration (i.e. 0.2 μm filtration and 3kDa centrifugal filtration, Sartorius, Göttingen, Germany). In addition, temperature, pH and observations, including the appearance of the solution (including colour, turbidity and particle film on the surface) were recorded.

Sample fortification:
In addition, samples of the artificial physiological media were fortified with a known amount of iron (by standard addition of commercial standards) to determine the standard recovery. For detailed information please refer to "Any other information on materials and methods incl. tables".

Mass balance:
After the test, aqua regia (3 : 1 mixture of concentrated hydrochloric and nitric acid) was added to the vessels containing the test item to reach a final volume of 500 mL, i.e. 120 mL aqua regia were added to approximately 380 mL GST or PBS medium, 180 mL aqua regia were added to approximately 320 mL ALF or GMB medium. From these solutions, 50 mL were taken after 3 - 14 days of “digestion” for mass balance determination.
The filters (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) used for sampling were extensively rinsed with a known volume of aqua regia (ca. 2.5 mL). The added aqua regia was let to drop slowly through the filters and was collected in a clean vial. This procedure was repeated with every syringe and filter used during the study. After collection, the volume was filled up to exactly 10 mL (for the media GST and PBS) and up to 15 mL (for the ALF and GMB media) with aqua regia. Afterwards the concentration of iron in the “filtrated” aqua regia was determined and considered for the determination of the mass balances.

Reagents:
Purified water (resistivity > 18 MΩ·cm, Pure Lab Ultra water purification system from ELGA LabWater, Celle, Germany)
Nitric acid - “Supra” quality (ROTIPURAN® supplied by Roth, Karlsruhe, Germany).
Hydrochloric acid – “Baker-instra-analyzed-plus” quality (J.T. Baker, Griesheim, Germany).
Sodiumhydroxide – pro Analysis quality (Chemsolute, Th. Geyer, Renningen, Germany)
MgCl2 x 6H2O (p.A., Merck, Darmstadt, Germany)
NaCl (p.A., Chemsolute, Renningen, Germany + Merck, Darmstadt, Germany (new GMB medium))
KCl (p.A., Chemsolute, Renningen, Germany + Merck, Darmstadt, Germany (new GMB medium))
Na2HPO4 (p.A. Merck, Darmstadt, Germany)
Na2SO4 (p.A. Merck, Darmstadt, Germany)
CaCl2 x 2H2O (p.A. Merck, Darmstadt, Germany)
NaAcetate (suprapur Merck, Darmstadt, Germany)
NaHCO3 (p.A. Merck, Darmstadt, Germany)
NaOH (p.A., Chemsolute, Renningen, Germany)
Citric acid anhydrous (p.A., Roth, Karlsruhe, Germany)
Glycine (p.A., Merck, Darmstadt, Germany)
Na3Citrate x 2H2O (p.A., Merck, Darmstadt, Germany)
Na2Tartrate x 2H2O (p.A., Merck, Darmstadt, Germany)
NaLactate (98+% Sigma Aldrich, Munich, Germany)
NaPyruvate (p.A., Applichem, Darmstadt, Germany)
KH2PO4 (p.A., Merck, Darmstadt, Germany)
Urea (pure, Applichem, Darmstadt, Germany)
Lactic acid (purum, Fluka, Munich, Germany)
HCl 30% (instra-analyzed, plus J.T. Baker, Griesheim, Germany)

METAL ANALYSIS
- Standards for metal analysis: A commercially available single element standard was used as iron standard (Merck Certipur Iron ICP standard 1000 mg/L lot no. HC68868126; Darmstadt, Germany) to prepare an appropriate stock solution and subsequently calibration solutions for ICP-OES measurements
- Certified reference materials: As quality control standards, certified aqueous reference material TM-DWS.3 (lot no. 0916) and TMDA-70.2 (lot no. 0916 and 0917) obtained from Environment Canada and a multielement standard (Merck Certipur IV ICP standard 1000 mg/L lot no. HC54938555 and HC73962555; Darmstadt, Germany) were analysed for total dissolved iron by ICP-OES.

Instrumental and analytical set-up for the ICP-OES instrument:
Agilent 720, Agilent Technologies, Waldbronn, Germany
Nebulizer: Sea spray nebulizer, from Glass Expansion
Spray chamber: Iso Mist with Twister Helix from Glass Expansion
Plasma stabilization time: at least 30 min before start of the measurements
Plasma gas flow: 15.0 L/min
Additional gas flow: 1.50 L/min
Carrier gas flow: 0.75 L/min
RF power: 1200W
Stabilization time of sample: 15 sec
Repetition time (three internal measurements per sample): 30 sec
Wavelengths: Fe: 238.204 nm, 240.489 nm, 241.052 nm, 258.588 nm and 259.940 nm

- Correlation coefficients (r) for the wavelengths used for evaluation of data were at least >0.999603

The applied LOD/LOQ calculations for the Agilent 720 ICP-OES:
LOD: 3 * standard deviation of calibration blank/slope of the calibration
LOQ: 3 * LOD
The resulting LODs/LOQs are reported in "Any other information on results incl. tables"
Details on dosing and sampling:: Loading:
Detailed loadings of the test vessels are given in "Any other information on materials and methods incl. tables".
Type:: other: Bioaccessibility ALF, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 39 µg Fe/L (dissolved)
Type:: other: Bioaccessibility ALF, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 90.5 µg Fe/L (dissolved)
Type:: other: Bioaccessibility ALF, 168h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 176 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GST, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 41 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GST, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 134 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GMB, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: negative value after correction
Type:: other: Bioaccessibility GMB, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 1.25 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GMB, 168h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: negative value after correction
Type:: other: Bioaccessibility PBS, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: < LOD (0.514 µg Fe/L)
Type:: other: Bioaccessibility PBS, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: < LOD (0.761 µg Fe/L)
Bioaccessibility (or Bioavailability) testing results:: Please refer to "any other information on results incl. tables" below.
Conclusions:: The bioaccessibility of nano-sized diiron trioxide was determined in vitro by simulating dissolution under physiological conditions considered to mimic artificial body fluids with a loading of 100 mg test item/L. After 2 and 24 h in phosphate buffered saline (PBS, pH 7.2-7.4) solution, dissolved iron concentrations (operationally defined as the dissolved Fe fraction after 0.2 µm filtration and centrifugal filtration (~2.1 nm)) were below the LOD (<0.761 µg/L). With mean concentrations of 1.25 µg Fe/L after 24 h and negative values (after subtraction of background) after 2 and 168 h, dissolved iron concentrations in Gamble´s solution (GMB, pH 7.4) were also very low. In artificial gastric fluid (GST, pH 1.5-1.6), 41 µg Fe/L and 134 µg Fe/L were detected in the dissolved phase after 2 and 24 h, respectively. Mean iron concentrations were highest in artificial lysosomal fluid (ALF, pH 4.5): 39 µg/L, 90.5 µg/L and 176 µg/L of iron were found in the dissolved phase after 2, 24 and 168 h. Therefore, with a maximum mean released fraction of <0.2% after 168 h, the dissolution of diiron trioxide was highest in artificial lysosomal fluid (ALF).

Reference 7

Endpoint:: basic toxicokinetics in vitro / ex vivo
Remarks:: Bioaccessibility
Type of information:: experimental study
Adequacy of study:: key study
Study period:: 2018-02-26 to 2018-09-26
Reliability:: 1 (reliable without restriction)
Rationale for reliability incl. deficiencies:: test procedure in accordance with national standard methods
Objective of study:: bioaccessibility (or bioavailability)
Qualifier:: equivalent or similar to guideline
Guideline:: OECD Series on Testing and Assessment No. 29 (23-Jul-2001): Guidance document on transformation/dissolution of metals and metal compounds in aqueous media
Version / remarks:: test medium, loading and test duration adopted for toxicokinetics assessment
Principles of method if other than guideline:: An internationally agreed guideline does not exist for this test (e.g. OECD). However, similar tests have been conducted with several metal compounds in previous risk assessments (completed under Regulation (EEC) No 793/93) and in recent preparation for REACH regulation (EC) No 1907/2006. The test was conducted on the basis of the guidance for OECD-Series on testing and assessment Number 29 and according to the bioaccessibility test protocol provided by the study monitor. The test media were artificial physiological media: gastric fluid (GST), phosphate-buffered saline (PBS), artificial lysosomal fluid (ALF) and Gamble's solution (GMB)
GLP compliance:: yes (incl. QA statement)
Remarks:: signed 2016-05-31
Radiolabelling:: no
Species:: other: in vitro (simulated human body fluids)
Details on test animals or test system and environmental conditions:: Test principle in brief:
- four different artificial physiological media,
- single loading of test substance of ~100 mg/L,
- GST and PBS media: samples taken after 2 and 24 hours agitation (100 rpm) at 37 ± 2 °C
- GMB and ALF media: samples taken after 2, 24 and 168 hours agitation (100 rpm) at 37 ± 2 °C

- two additional method blanks per medium, measurement (ICP-OES) of dissolved Fe concentrations after filtration and centrifugal filtration.
- the study was performed in triplicate

The aim of this test was to assess the dissolution of nano-size diiron trioxide in four artificial physiological media: Phosphate buffered saline (PBS, pH 7.2-7.4), Artificial gastric fluid (GST, 1.5-1.6), artificial lysosomal fluid (ALF) and Gamble’s solution (GMB). The test media were selected to simulate relevant human-chemical interactions (as far as practical), e.g. a substance entering the human body by ingestion into the gastro-intestinal tract (GST) or via the respiratory system (ALF).
Duration and frequency of treatment / exposure:: Iron concentrations in GST and PBS were determined after 2 and 24 h whereas iron concentrations in GMB and ALF media were assessed after 2, 24 and 168 hours of incubation.
Dose / conc.:: 100 other: mg of test item/L artificial media
Details on study design:: Test setup
Three replicate flasks (500 mL glass flasks) per test medium (PBS, GST) were prepared with a loading of ~ 100 mg/L. The test item was weighed into flasks, adjusted to volume with the respective artificial physiological medium and agitated at 100 rpm at 37°C ± 2°C. Two control blank replicates (same procedure) per test medium were also prepared.
Three replicates containing the test item and two method blanks per artificial medium were tested. All solutions were sampled after 2 and 24 h whereas GMB and ALF media were also sampled after 168h to measure total dissolved Fe concentrations (ICP-OES) after 0.2 µm filtration (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) and centrifugal filtration (i.e. 0.2 μm filtration and 3kDa centrifugal filtration, Sartorius, Göttingen, Germany). In addition, temperature, pH and observations, including the appearance of the solution (including colour, turbidity and particle film on the surface) were recorded.

Sample fortification:
In addition, samples of the artificial physiological media were fortified with a known amount of iron (by standard addition of commercial standards) to determine the standard recovery. For detailed information please refer to "Any other information on materials and methods incl. tables".

Mass balance:
After the test, aqua regia (3 : 1 mixture of concentrated hydrochloric and nitric acid) was added to the vessels containing the test item to reach a final volume of 500 mL, i.e. 120 mL aqua regia were added to approximately 380 mL GST or PBS medium, 180 mL aqua regia were added to approximately 320 mL ALF or GMB medium. From these solutions, 50 mL were taken after 3 - 14 days of “digestion” for mass balance determination.
The filters (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) used for sampling were extensively rinsed with a known volume of aqua regia (ca. 2.5 mL). The added aqua regia was let to drop slowly through the filters and was collected in a clean vial. This procedure was repeated with every syringe and filter used during the study. After collection, the volume was filled up to exactly 10 mL (for the media GST and PBS) and up to 15 mL (for the ALF and GMB media) with aqua regia. Afterwards the concentration of iron in the “filtrated” aqua regia was determined and considered for the determination of the mass balances.

Reagents:
Purified water (resistivity > 18 MΩ·cm, Pure Lab Ultra water purification system from ELGA LabWater, Celle, Germany)
Nitric acid - “Supra” quality (ROTIPURAN® supplied by Roth, Karlsruhe, Germany).
Hydrochloric acid – “Baker-instra-analyzed-plus” quality (J.T. Baker, Griesheim, Germany).
Sodiumhydroxide – pro Analysis quality (Chemsolute, Th. Geyer, Renningen, Germany)
MgCl2 x 6H2O (p.A., Merck, Darmstadt, Germany)
NaCl (p.A., Chemsolute, Renningen, Germany + Merck, Darmstadt, Germany (new GMB medium))
KCl (p.A., Chemsolute, Renningen, Germany + Merck, Darmstadt, Germany (new GMB medium))
Na2HPO4 (p.A. Merck, Darmstadt, Germany)
Na2SO4 (p.A. Merck, Darmstadt, Germany)
CaCl2 x 2H2O (p.A. Merck, Darmstadt, Germany)
NaAcetate (suprapur Merck, Darmstadt, Germany)
NaHCO3 (p.A. Merck, Darmstadt, Germany)
NaOH (p.A., Chemsolute, Renningen, Germany)
Citric acid anhydrous (p.A., Roth, Karlsruhe, Germany)
Glycine (p.A., Merck, Darmstadt, Germany)
Na3Citrate x 2H2O (p.A., Merck, Darmstadt, Germany)
Na2Tartrate x 2H2O (p.A., Merck, Darmstadt, Germany)
NaLactate (98+% Sigma Aldrich, Munich, Germany)
NaPyruvate (p.A., Applichem, Darmstadt, Germany)
KH2PO4 (p.A., Merck, Darmstadt, Germany)
Urea (pure, Applichem, Darmstadt, Germany)
Lactic acid (purum, Fluka, Munich, Germany)
HCl 30% (instra-analyzed, plus J.T. Baker, Griesheim, Germany)

METAL ANALYSIS
- Standards for metal analysis: A commercially available single element standard was used as iron standard (Merck Certipur Iron ICP standard 1000 mg/L lot no. HC68868126; Darmstadt, Germany) to prepare an appropriate stock solution and subsequently calibration solutions for ICP-OES measurements
- Certified reference materials: As quality control standards, certified aqueous reference material TM-DWS.3 (lot no. 0916) and TMDA-70.2 (lot no. 0916 and 0917) obtained from Environment Canada and a multielement standard (Merck Certipur IV ICP standard 1000 mg/L lot no. HC54938555 and HC73962555; Darmstadt, Germany) were analysed for total dissolved iron by ICP-OES.

Instrumental and analytical set-up for the ICP-OES instrument:
Agilent 720, Agilent Technologies, Waldbronn, Germany
Nebulizer: Sea spray nebulizer, from Glass Expansion
Spray chamber: Iso Mist with Twister Helix from Glass Expansion
Plasma stabilization time: at least 30 min before start of the measurements
Plasma gas flow: 15.0 L/min
Additional gas flow: 1.50 L/min
Carrier gas flow: 0.75 L/min
RF power: 1200W
Stabilization time of sample: 15 sec
Repetition time (three internal measurements per sample): 30 sec
Wavelengths: Fe: 238.204 nm, 240.489 nm, 241.052 nm, 258.588 nm and 259.940 nm

- Correlation coefficients (r) for the wavelengths used for evaluation of data were at least >0.999603

The applied LOD/LOQ calculations for the Agilent 720 ICP-OES:
LOD: 3 * standard deviation of calibration blank/slope of the calibration
LOQ: 3 * LOD
The resulting LODs/LOQs are reported in "Any other information on results incl. tables"
Details on dosing and sampling:: Loading:
Detailed loadings of the test vessels are given in "Any other information on materials and methods incl. tables".
Type:: other: Bioaccessibility ALF, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 16.4 µg Fe/L (dissolved)
Type:: other: Bioaccessibility ALF, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 72.1 µg Fe/L (dissolved)
Type:: other: Bioaccessibility ALF, 168h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 185 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GST, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 9.27 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GST, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 57.5 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GMB, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: 0.03 µg Fe/L (dissolved)
Type:: other: Bioaccessibility GMB, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: below background
Type:: other: Bioaccessibility GMB, 168h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: below background
Type:: other: Bioaccessibility PBS, 2h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: < LOD/LOQ (1.54 µg Fe/L)
Type:: other: Bioaccessibility PBS, 24h @ 37°C (100 mg/L loading, 0.2 µm + 3 kDa filtration for phase separation)
Results:: < LOD (0.761 µg Fe/L)
Bioaccessibility (or Bioavailability) testing results:: Please refer to "any other information on results incl. tables" below.
Conclusions:: The bioaccessibility of nano-size diiron trioxide was determined in vitro by simulating dissolution under physiological conditions considered to mimic artificial body fluids with a loading of 100 mg test item/L. After 2 and 24 h in phosphate buffered saline (PBS, pH 7.2-7.4) solution, dissolved iron concentrations (operationally defined as the dissolved Fe fraction after 0.2 µm filtration and centrifugal filtration (~2.1 nm)) were at least below the LOQ (<1.54 µg/L). With mean concentrations of 0.03 µg Fe/L after 2 h and negative values (after subtraction of background) after 24 and 168 h, dissolved iron concentrations in Gamble´s solution (GMB, pH 7.4) were also very low. In artificial gastric fluid (GST, pH 1.5-1.6), 9.27 µg Fe/L and 57.5 µg Fe/L were detected in the dissolved phase after 2 and 24 h, respectively. Mean iron concentrations were highest in artificial lysosomal fluid (ALF, pH 4.5): 16.4 µg/L, 72.1 µg/L and 185 µg/L of iron were found in the dissolved phase after 2, 24 and 168 h. Therefore, with a maximum mean released fraction of <0.2% after 168 h, the dissolution of diiron trioxide was highest in artificial lysosomal fluid (ALF).

Reference 8

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: disregarded due to major methodological deficiencies
Reliability:: other: not rated according to Klimisch et al.
Rationale for reliability incl. deficiencies:: other:
Remarks:: The references contained in this summary entry represents in vivo experiments with investigations on toxicokinetics with very limited value for risk assessment purposes. The references do not fulfil the criteria for quality, reliability and adequacy of experimental data for the fulfilment of data requirements under REACH and hazard assessment purposes (ECHA guidance R4 in conjunction with regulation (EC) 1907/2006, Annexes VII-X). The information contained therein were included for information purposes only.
Principles of method if other than guideline:: Weissleder, R. et al. (1989):
(A) Relaxation Time and Pathology:
Test substance: superparamagnetic iron oxide (AMI-25), prepared by modification of the method of Whitehead et al. (1985), was synthesized by Advanced Magnetics, Inc., Cambridge, MA; volume median diameter of particles (laser light scattering): 80 nm; relaxivity of preparation: 3 x 10^4/sec*M for 1/T1 and 1 x 10^5 sec*M for 1/T2 in water.
Doses: 18 µmol Fe/kg
No. of animals per sex per dose: 39 for relaxation time measurement and 7 for histology.
Exposure duration: 1 day
Exposure frequency: one single administration
Negative control: without injection
Investigated parameters: (I) MA relaxation times of liver, spleen, lung, kidney, and muscle were determined to measure the effect of AMI-25 on each tissue. Groups (n = 3) of animals were sacrificed at 7.5, 15, 30, 60, 1 20, and 240 min. and at 1, 2, 4, 7, 14, 28, and 91 days after injection. Organs were removed immediately. Relaxation time measurements were performed 1 hr after sacrifice. T2 values were obtained by using an NMA spectrometer. (II) Histologic examinations of liver were performed to document disappearance of stainable iron after administration of AMI-25. Histologic examinations were performed before, at 3 hr, and at 1, 2, 9, 1 6, and 30 days after injection of AMI-25. Samples of livers were stained for iron with Pens’ Prussian blue stain after fixation.

(B) Radiotracer studies:
Test substance: Radioactive AMI-25, synthesized by incorporating 59Fe into the superparamagnetic iron oxide core; specific activity: 8460 µCi/µmoI (313,020 kBq/µmol) Fe; radiolabelled particles were mixed with unlabelled particles to give a specific activity of 36 µCi/µmol (1332 kBq/µmol) Fe.
Doses: 18 µmol Fe/kg (2 µCi/kg [74 kBq/kg])
No. of animals per sex per dose: 4 for whole body clearance, 48 for organ biodistribution, 3 for erythrocyte incorporation.
Exposure duration: 1 day
Exposure frequency: one single administration
Controls: not specified
Investigated parameters: (I) radioactivity were measured in liver, spleen, blood, lung, kidney, and brain before injection and at several points after injection (7.5, 15, and 30 min.; 1, 2, 4, 8, and 16 hr; and 2, 4, 7, 14, 34, 60, and 90 days (n=3 at each time point)). (II) RBC incorporation of iron oxide was determined in blood after 1, 2, 4, 7, 14, 21, 28, 35, 42, 49, and 56 days, and erythrocytes were counted. At day 21, total blood volume was determined by the 51Cr radioisotope dilution technique. The RBC volume was calculated as total blood volume multiplied by haematocrit. (III) For whole-body clearance studies the counts from each rat were immediately obtained in a chamber counter, and then the rats were placed in a metabolic cage. Serial counting was performed at 2 hr and at 1, 2, 4, 7, 14, and 28 days. Separate focal and urine samples were counted at identical time points.
Species:: other: Weissleder, R. et al. (1989 A and B): Sprague-Dawley rat
Sex:: male/female
Route of administration:: other: Weissleder, R. et al. (1989A and B): intravenous injection
Type:: other: Weissleder, R. et al. (1989 A):
Results:: (I) Relaxation time (T2): maximum increase in 1/T2 of liver & spleen within 4 h after administration of AMI-25. Thereafter, 1/T2 decreased consistently. In both liver & spleen, half-time of the 1/T2 effect occurred within 24-48 h.
Type:: other: Weissleder, R. et al. (1989 A):
Results:: (II) Histopathology: iron in Kupffer cells throughout lobules at 3 h but amounts reduced qualitatively thereafter (24 h–9 days) & no stainable iron were seen in liver at 16 & 30 days. No hepatocellular changes in any of study specimens were observed.
Type:: other: Weissleder, R. et al. (1989 B):
Results:: (I) Biodistribution: high levels of 59Fe in liver & spleen (~83 % & 6 % of dose) after 1 h; rel. organ biodistribution: ~6 %/g liver, 11 %/g spleen & minimal amounts in kidney, lung, brain; relaxation times increased in both organs by > 215 % after 1 h.
Type:: other: Weissleder, R. et al. (1989 B):
Results:: (II) Bioavailability: molecular iron of 59Fe-AMI-25 was incorporated into haemoglobin of normal rats in a time-dependent fashion. 1 % of the injected dose was associated with erythrocytes on day 2, 14 % by day 7, and 20 % by day 49.
Type:: other: Weissleder, R. et a. (1989 B):
Results:: (III) Metabolism: max. levels in liver at 2 h & spleen at 4 h (~89 & 19 %); half-life of 59Fe: 3 & 4 days (liver & spleen).
Type:: other: Weissleder, R. et al. (1989 B):
Results:: (III)Excretion: whole body clearance of 59Fe was 20% after 14 days and 35 % after 28 days. Extrapolated half-time of whole-body clearance was 44.9 days (T1/4 & T3/4: 18.7 & 89.9 days). Excretion of 59Fe in urine & faeces was 1.1 & 10.1 % after 2 & 7 days.
Conclusions:: No conclusion can be drawn from the above publications due to lack of quality, reliability and adequacy of the experimental data for the fulfilment of data requirements under REACH.

The references contained in this summary entry represents in vivo toxicokinetics investigations with very limited value for risk assessment purposes. All references do not fulfil the criteria for quality, reliability and adequacy of experimental data for the fulfilment of data requirements under REACH and hazard assessment purposes (ECHA guidance R4 in conjunction with regulation (EC) 1907/2006, Annexes VII-X). The information contained therein were included for information purposes only.

Weissleder, R. et al. (1989):
Publication shows significant methodological deficiencies in the experimental set up and documentation. Test material was insufficiently characterised (no information about the iron oxide form, purity or impurities). There is a lack of information about the rat (age) and the environmental conditions (housing, type of food and drinking water, photoperiod, acclimation, relative humidity). One dose was tested, only. The number of animals per group and the use of control groups was not given in the publication.

Reference 9

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: key study
Study period:: Before April 1978
Reliability:: 2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:: other: Published in a peer-reviewed paper. Sufficient detail for evaluation.
Objective of study:: absorption
Qualifier:: no guideline followed
Principles of method if other than guideline:: Comparison of the absorption and bioavailability of available elemental iron powders in rats.
The bioavailability of several metallic iron powders was compared by feeding rats iron fortified diets and determining the effect on haematocrit in rats.
GLP compliance:: no
Remarks:: Study was conducted before implementation of the GLP system.
Specific details on test material used for the study:: Not specified
Radiolabelling:: no
Species:: rat
Strain:: Wistar
Sex:: male
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Age at study initiation: weanling (3 weeks old)
- Housing: plastic pan cages with stainless steel wire grill tops
- Individual metabolism cages: no. 6/cage
- Diet: ad libitum
- Water: ad libitum
Route of administration:: oral: feed
Vehicle:: unchanged (no vehicle)
Details on exposure:: STABILITY OF TEST MATERIAL
Moisture: the several elemental Fe powders were dusted onto filter papers wet with distilled water and placed on Petri dishes to retain moisture. The dishes were stored in room temperature and examined periodically for a week.
Sodium chloride: the elemental powders were mixed with reagent-grade dry sodium chloride to make 1 % mixtures (gave grey colour); they were stored and examined periodically.
Duration and frequency of treatment / exposure:: 2 weeks
Remarks:: Doses / Concentrations:
described below section 'any other information on materials and methods incl. tables'.
No. of animals per sex per dose / concentration:: described below section 'any other information on materials and methods inl. tables'.
Control animals:: yes, plain diet
Positive control reference chemical:: In a strict sense, there was no positive control. Elemental iron powders were investigated concurrently with FeSO4, which can be regarded as a positive control if the absorption of iron is concerned, due to its known bioavailability. The iron in the salt does not have to be converted to an ionic form to allow absorption. This conversion is a prerequisite for the absorption of the iron in the other elemental iron powders. Moreover, 100-mesh technical grade reduced Fe was included as a low control in anticipation of its low bioavailability.
Details on study design:: The animals were fed an Fe-depleted diet for 4 weeks, so as the hematocrits to fall to a mean value of less than 20%. The rat chow (low Fe-test diet McCall # 170365, supplied by Teklad Mills, P. O. Box 4220) that was given to the rats contained 6 mg of Fe/kg and the average daily food consumption was 10 g feed/100 g bw. After that 4-week period each rat was weighed and individual haematocrits were determined. Successively, the rats were placed on the same Fe-depleted diet supplemented with Fe, either as ferrous sulfate or one of the elemental Fe powders. All powders, but Mallinckrodt #4350 H-reduced, were retested three or four times. At the end of the 2-week period the haematocrits were measured again.
Bioavailability: the efficacy of each Fe powder was calculated as a ratio between the standard dose of ferrous sulfate and the dose of the test substance that gives the same haematocrit response.
Details on dosing and sampling:: not specified
Statistics:: not specified
Preliminary studies:: not specified
Type:: absorption
Details on absorption:: Refer to the field below 'Overall remarks'.
Details on distribution in tissues:: not specified
Details on excretion:: not specified
Metabolites identified:: not specified
Details on metabolites:: not specified
Enzymatic activity measured:: not specified
Bioaccessibility (or Bioavailability) testing results:: Mean bioavailability (FeSO4 = 100):

(A) Carbonyl iron:
(I) Test material (GAF, SF special): 66 %
(II) Test material (GAF, E): 63 %
(III) Test material (GAF, L): 39 %

(B) Electrolytic iron:
(IV) Test material (Glidden, A-131): 48 %

(C) Hydrogen reduced iron:
(V) Test material (Mallinckrodt #4350): 27 %
(VI) Test material (Mallinckrodt #4350): 32 %
(VII) Test material (Glidden, B-131): 35 %

The bioavailability decreases with increasing particle size of the elemental iron powder as can be seen from the carbonyl iron and hydrogen reduced iron powders of different particle sizes. For further information please refer to table 1 (attached document).
Conclusions:: The rate of haemoglobin repletion in Fe-deficient Wistar rats was used to determine the bioavailability of several commercially available elemental Fe powders added to an Fe-free nutritionally balanced chow. Ferrous sulfate was used as a positive control, due to its high bioavailability. Carbonyl Fe powders demonstrated the highest bioavalability among all.

Interpretation of results (migrated information): Carbonyl Fe powders had the highest bioavalability among all. The bioavailability of iron depends not only on solubility but also on particle size. With increasing particle size of the elemental iron powder, the bioavailability decreases. The study provides important information on the bioavailability of Fe particles from different powders.

Reference 10

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: disregarded due to major methodological deficiencies
Reliability:: 3 (not reliable)
Rationale for reliability incl. deficiencies:: significant methodological deficiencies
Remarks:: Publication shows significant methodological deficiencies in the experimental set up and documentation. The self-synthesized test material was insufficiently characterised (no purity or impurities). Only one dose group were tested. The number of animals which were sacrificed at the different time points (after exposure and day 1, 4 and 7 at postexposure) for histopathological examinations varies (n=2 or 3).
Qualifier:: no guideline followed
Principles of method if other than guideline:: Test substance: Fe2O3, self-synthesized; diameter of particles: 0.005 µm; MMAD: 0.15 µm; GSD (ơg): 2.2
Doses: 300 mg/m3
No. of animals per sex per dose: 40 males
Exposure duration: 3 hours
Exposure frequency: single administration
Negative controls: unexposed
Animals were sacrificed immediately after the 3-hour aerosol exposure or at 1, 4 and 7 days postexposure (n=2-3 animals per time point, for total of 10 animals). The trachea and intrapulmonary bronchi were collected and prepared for histopathological examinations. For each section, every cell was identified and scored for its iron content. The number of hemosiderin granules were recorded.
GLP compliance:: not specified
Remarks:: in the publication.
Radiolabelling:: no
Species:: mouse
Strain:: CD-1
Sex:: male
Route of administration:: inhalation: aerosol
Vehicle:: unchanged (no vehicle)
Type:: metabolism
Results:: Iron oxide particles were distributed on the airway surfaces and in pinocytotic vesicles. Intracellularly it was converted to ferritin and hemosiderin (starting within the exposure period).
Conclusions:: No conclusion can be drawn from the above publication due to lack of quality, reliability and adequacy of the experimental data for the fulfilment of data requirements under REACH.
The reference contained in this summary entry represents an in vivo toxicokinetic investigations with very limited value for risk assessment purposes. The reference does not fulfil the criteria for quality, reliability and adequacy of experimental data for the fulfilment of data requirements under REACH and hazard assessment purposes (ECHA guidance R4 in conjunction with regulation (EC) 1907/2006, Annexes VII-X). The information contained therein were included for information purposes only.

Watson, A.Y. et al. (1979):
The publication shows significant methodological deficiencies in the experimental set up and documentation. The self-synthesized test material was insufficiently characterised (no purity or impurities). Only one dose group were tested. The number of animals which were sacrificed at the different time points (after exposure and day 1, 4 and 7 at postexposure) for histopathological examinations varies (n=2 or 3).

Reference 11

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: weight of evidence
Reliability:: 3 (not reliable)
Rationale for reliability incl. deficiencies:: significant methodological deficiencies
Remarks:: The reference contained in this summary entry represents an in vivo experiment with investigations on toxicokinetics with very limited value for risk assessment purposes. The reference does not fulfil the criteria for quality, reliability and adequacy of experimental data for the fulfilment of data requirements under REACH and hazard assessment purposes (ECHA guidance R4 in conjunction with regulation (EC) 1907/2006, Annexes VII-X). The information contained therein were included for information purposes only.
Qualifier:: no guideline followed
Principles of method if other than guideline:: Test substance: radiolabelled 59Fe(II)SO4 (radionuclidic purity of 99.5% and mean specific activity of 21.16 mCi per mg), was purchased from Perkin Elmer Life Sciences (Boston, MA). Radiotracer was added to non-radiolabelled Fe(II)SO4 · 7H2O (Sigma Chemical, St. Louis, MO); MMAD: 2.99 µm; GSD: 1.15.
Doses: 3.70 and 3.99 mg/m3
No. of animals per sex per dose: 32 males
Exposure duration: average of 88.5 min. for the two exposures
Exposure frequency: single exposure
Controls: not specified
Rats were killed immediately or at 1, 2, 4, 8, or 21 days postexposure to the aerosol. Gamma spectrometry was performed to the following tissues at each time point: nasal olfactory mucosa, nasal nonolfactory mucosa, olfactory bulb, olfactory tract plus tubercle, striatum, cerebellum, and rest of brain, lung, liver, kidney and pancreas. Some nasal olfactory mucosa samples were analysed for 59-Fe associated transferrin with HPLC and gamma spectroscopy. Heads were collected and autoradiograms were prepared to visualise the location of 59 Fe from the nose to the brain.
GLP compliance:: not specified
Remarks:: in the publication.
Radiolabelling:: yes
Species:: rat
Strain:: other: Crl:CD (SD)IGSBR rats
Sex:: male
Route of administration:: inhalation: aerosol
Vehicle:: water
Details on exposure:: TYPE OF INHALATION EXPOSURE: nose only
Type:: absorption
Results:: < 4% uptake of 59FeSO4 into brain via olfactory pathway; most of 59Fe remained in nasal regions of olfactory system; 59Fe activity was absent in olfactory regions of brain even 4 days postexposure; it was coeluted in olfactory mucosa with transferrin.
Conclusions:: No conclusion can be drawn from the above publication due to lack of quality, reliability and adequacy of the experimental data for the fulfilment of data requirements under REACH.
The reference contained in this summary entry represents in vivo toxicokinetic investigations with very limited value for risk assessment purposes. The reference do not fulfil the criteria for quality, reliability and adequacy of experimental data for the fulfilment of data requirements under REACH and hazard assessment purposes (ECHA guidance R4 in conjunction with regulation (EC) 1907/2006, Annexes VII-X). The information contained therein were included for information purposes only.

Rao, D.B. et al. (2003) only investigated whether inhaled iron in rats is transported to the brain via the olfactory pathway. Only one dose were tested in two cohorts which were divided in two experimental groups with or without nasal occlusion (1. patent right and left nostrils, 2. only right nostril). According to the authors, iron was not readily transported to the brain via the olfactory tract.
Since only selected parameters were investigated, and the investigated endpoints are not required for building an expert judgement and further assessment of the test substance under REACH (ECHA guidance R4 in conjunction with regulation (EC) 1907/2006, Annexes VII-X).

Reference 12

Endpoint:: basic toxicokinetics in vitro / ex vivo
Type of information:: experimental study
Adequacy of study:: weight of evidence
Reliability:: 3 (not reliable)
Rationale for reliability incl. deficiencies:: significant methodological deficiencies
Remarks:: The experimental design and the results are not described in detail. Test material is insufficiently characterised (no purity or impurities). Furthermore, the test conditions (no. of animals and measurement intervalls) are not always the same for the individual dose groups and the number of animals was too low for some experiments. In addition, the results cannot always be clearly assigned to the individual studies.
Objective of study:: absorption
Qualifier:: equivalent or similar to guideline
Guideline:: OECD Guideline 417 (Toxicokinetics)
Deviations:: yes
Remarks:: Equivalent in the sense that comparable endpoints were studied with comparable reliability and sensitivity.
Principles of method if other than guideline:: The absorption of carbonyl iron as well as its solubilisation in the stomach were determined in rats by means of techniques which will be described below in detail. In several aspects the methods differ from those described in the guidelines.
GLP compliance:: not specified
Remarks:: in publication
Specific details on test material used for the study:: SOURCE OF TEST MATERIAL
- Source of test material: GAF Corp. New York

RADIOLABELLING INFORMATION
- Specific activity: 2 to 8 mCi 59Fe/g
- Locations of the label: part of the iron atoms
- Procedure: prepared by irradiating samples of carbonyl iron in a neutronf lux of 6.8x10^3 neutrons/cm2/sec for 2 to 8 days (Union Carbide Nuclear Reactor Service,
Tuxedo, N.Y.).
Radiolabelling:: yes
Species:: rat
Strain:: Sprague-Dawley
Sex:: male
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Weight at study initiation: 200 - 300 g
- Diet (ad libitum): Purina Rat Chow diet (supplier: Ralston Purina Co., St. Louis)
- Water (ad libitum)
Route of administration:: other: Diets with normal iron, low iron and high iron (1% iron carbonyl). Intra gastric intubation with radiolabelled carbonyl iron.
Vehicle:: not specified
Details on exposure:: Exposure was not systematically described, but has to be reconstructed from fragments of the methods section and the results section. It appears that rats received low iron, normal and high iron feed. The latter was enriched with 1% carbonyl iron. The low iron rats received a diet with 4 to 8 mg iron per kg and were bled as described in one of the references (Huebers et al., J Clin Invest 1978; 62: 944). Apart from feeding the rats with diets containing different amounts of iron, single exposures were applied by means of gavage (intragastric intubation). Carbonyl iron was also injected in the obstructed stomach to study the dissolution in the stomach juice in vivo. In addition iron was injected by needle through the wall of the upper duodenum, with or without ligation of the pylorus.
Duration and frequency of treatment / exposure:: single administration
Dose / conc.:: 0.075 other: mg iron/rat
Remarks:: of carbonyl iron and ferrous ammonium sulfate
Dose / conc.:: 0.2 other: mg iron/rat
Remarks:: of carbonyl iron and ferrous ammonium sulfate
Dose / conc.:: 1 other: mg iron/rat
Remarks:: of carbonyl iron and ferrous ammonium sulfate
Dose / conc.:: 5 other: mg iron/rat
Remarks:: of carbonyl iron and ferrous ammonium sulfate
Dose / conc.:: 20 other: mg iron/rat
Remarks:: of carbonyl iron and ferrous ammonium sulfate
Dose / conc.:: 100 other: mg iron/rat
Remarks:: of carbonyl iron and ferrous ammonium sulfate
Dose / conc.:: 200
Remarks:: mg carbonyl iron
No. of animals per sex per dose / concentration:: 3 males/group (normal & iron-deficient rats) for iron absorption in liver; 3 males/group (normal & iron-deficient rats) and 6 males/group (iron-loaded rats) for iron uptake in different organs
Control animals:: not specified
Positive control reference chemical:: The proportion of absorbed iron in liver at 24 h included the comparison of carbonyl iron with ferrous ammonium sulfate.
Details on study design:: Because of the non-standard nature of the methods the reviewer found it necessary to reproduce them below as they were described in the original publication.

Rats designated as normal were given a Purina Rat Chow diet, those to be made iron deficient were given a diet containing 4 - 8 mg iron/kg and were bled as described elsewhere. Animals were iron loaded by feeding them a 1 % carbonyl iron diet for 4 - 6 months.

Several types of absorption studies were performed. In some instances, the radioactive iron preparation was given by gastric intubation, and absorption determined over ensuing days. Alternately, absorption was measured over a 1- to 2-h period in isolated gut loops with intact blood supply. Iron was also injected by needle through the wall of the upper duodenum with or without ligation al the pylorus. The laparotomy did not appear to affect the amount of iron absorbed.
At the end of the study, animals were exsanguinated under anaesthesia and then perfused with 30 mL saline solution. The gastrointestinal tract was removed in toto. Liver, spleen, femur, and kidneys as well as blood were removed for determination of radioactivity in a gamma scintillation counter (model 5330; Packard Instrument Co.). Total marrow activity was calculated from femur counts x 13 (Hershko et al. (1972)).* Total blood activity was based on an assumed total blood volume of 6% of body weight. By adding activities of these organs, the amount of iron absorbed could be determined. Blood taken at the time of sacrifice was also used for determination of the haematocrit by the micro technique as well as plasma iron and total iron-binding capacity.

In studies of mucosal ferritin, iron was given by gastric intubation; after sacrifice of the animal, the upper third of the small intestine was washed, and the mucosa was than scraped with a glass slide from the absorptive area of the gut. The mucosa was suspended in phosphate saline solution at pH 7.4, homogenised, and sonicated, centrifuged at 27,000 x g for 90 min., and subjected to immunoelectrophoresis with appropriate standards as described elsewhere. The antibody used was against liver ferritin and cross-reacted with mucosal ferritin.

Gastric secretions were collected by placing a ligature at the pylorus and emptying the stomach after 1 or 2 h. Carbonyl iron was either injected into the obstructed stomach so that incubation took place in vivo or was added to gastric juice, removed from the animal, and incubated at 37 °C for 1 h. Solubility of iron was determined, and the mixture of iron in 1 mL gastric juice was also injected into a jejunal loop for determination of absorption. In some studics, the gastric juice was neutralized with 1 N NaOH to a pH of 7.5 to 8.0 before exposure to carbonyl iron.

Solubility of iron solutions was determined in two ways. Centrifugation was carried out at 27,000 x g for 60 min., and the supernatant analysed for radioactivity and by spectrophotometry for soluble iron after the addition of bathophenanthroline. Material was also subjected to Millipore filtration (Millex HA 0.45 µm filter unit), and the filtrate similarly examined for radioactivity and iron content and for its colour reaction when exposed to apotransferrin. In all instances, these methods agreed in the amount of iron solubilized (r =0.9). Solubilized iron was shown to be in the ferrous form by colour development with bathophenanthroline in the absence of reducing agents.

*Reference:
- Hershko C, Cook JD, Finch CA. (1972): Storage iron kinetics. II. The uptake of hemoglobin iron by hepatic parenchymal cells. J LAB CLIN MED, 80:624.
Details on dosing and sampling:: Please refer to the field "Details on study design".
Statistics:: Results were presented as means with standard errors.
Preliminary studies:: Not specified
Type:: absorption
Details on absorption:: The relation between absorbed amount of iron and iron applied by means of a single gavage treatment within 5 days is shown in figure 1 in the attachment. The absorption over that period did not differ for iron from carbonyl iron or ferrous ammonium sulfate. Whereas lethality limited the maximum dose applied to 20 mg/animal for the salt, carbonyl iron could be applied up to 200 mg/animal. Absorption was clearly higher in iron depleted animals.
Details on distribution in tissues:: The iron uptake by the rats of the 3 dietary groups in different organs is presented in Table II and Table III of the publication, which are attached. The influence of the diet is clearly shown. The attached figure 3 shows the plasma iron concentrations upon different single doses of carbonyl iron in normal and iron-deplete animals.
Details on excretion:: Not specified
Metabolites identified:: yes
Details on metabolites:: The study clearly reveals that metallic iron particles have to be dissolved in the form of iron II ions in the stomach before absorption is possible.
Enzymatic activity measured:: Not specified
Bioaccessibility (or Bioavailability) testing results:: In vitro solubility studies with stomach juice combined with absorption in vivo from a jejunal gut loop showed unambiguously that carbonyl iron was only significantly absorbed at pH<2 (see attached Table 1). A solubility of >90 % at pH 3 and of 77 % at pH 6 was determined for 50 µg ferrous ammonium sulphate in gastric juice. 50 % of ferric chloride dissolved at pH 3, whereas only a solubility of < 1 % at pH 6. Carbonyl iron showed with < 1 % the lowest solubility. At pH 1.6 and a loading of 20000 µg carbonyl iron, 2560 µg dissolved.
Conclusions:: The mechanism of carbonyl iron absorption has been studied in rats. Solubility by gastric acid was a prerequisite for subsequent absorption.
To examine absorption, SD-rats were given normal diet or diet containing 4 - 8 mg iron/kg. Animals were iron loaded by feeding them a 1 % carbonyl iron diet for 4 - 6 months.
Several types of absorption studies were performed. In some instances, the radioactive iron preparation was given by gastric intubation, and absorption determined over ensuing days. Alternately, absorption was measured over a 1- to 2-h period in isolated gut loops with intact blood supply. Iron was also injected by needle through the wall of the upper duodenum with or without ligation al the pylorus.
At the end of study, the radioactivity of 59Fe in the gastrointestinal tract, liver, spleen, femur, kidneys, and blood was measured after sacrifice for determination of absorption. In addition, haematocrit, plasma iron and total iron-binding capacity were determined in blood samples.
In studies of mucosal ferritin, iron was given by gastric intubation; after sacrifice of animal, the upper third of the small intestine was washed, and the mucosa was scraped from the absorptive area of the gut. The mucosa was subjected to immunoelectrophoresis (antibody against liver ferritin and cross-reacted with mucosal ferritin).
Furthermore, solubility of iron was determined after carbonyl iron was either injected into the obstructed stomach so that incubation took place in vivo or was added to gastric juice, removed from the animal, and incubated at 37 °C for 1 h. The mixture of iron in gastric juice was also injected into a jejunal loop for determination of absorption. Solubility of iron solutions was determined by radioactivity analysis or by spectrophotometry.
The slow rate of solubilisation resulted in a more prolonged absorption, responsible for the low toxicity of carbonyl iron. Large doses of carbonyl iron were held for several days by the gastric mucosa of iron-deficient animals. Once it had been solubilised, the subsequent pathway of absorption by the intestinal mucosa and the amount absorbed was similar to that of ferrous ammonium sulfate. Furthermore, the data of iron absorption in the blood, skeleton, liver and the carcass (minus gut) showed higher absorption in the iron-deficient rats at a period of 4 days in comparison to rats with normal diet after oral administration of a single dose of 200 mg carbonyl iron.
The in vitro solubility of the three iron compounds in gastric juice was dependent on pH. A solubility of >90 % at pH 3 and of 77 % at pH 6 was determined for 50 µg ferrous ammonium sulphate in gastric juice. 50 % of ferric chloride dissolved at pH 3, whereas only a solubility of < 1 % at pH 6. Carbonyl iron showed with < 1 % the lowest solubility. At pH 1.6 and a loading of 20000 µg carbonyl iron, 2560 µg dissolved.

Interpretation of results (migrated information): due to the well-known and extensively studied active regulation of the absorption of iron after oral exposure, this question is not relevant.
The study shows that metallic iron in the form of carbonyl iron particles has to be dissolved in the stomach juice at pH<2 before absorption is possible. Carbonyl iron is only absorbed in the form of iron(II)ions.

The study schows significant methodological deficiencies. The experimental design and the results are not described in detail. Test material is insufficiently characterised (no purity or impurities). In addition, the results cannot always be clearly assigned to the individual studies. Furthermore, the test conditions are not always the same for the individual dose groups. In the study to investigate the uptake of iron in the blood, skeleton, liver and carcass or stomach contents of the various dose groups, a different number of animals was used and the same measurement times were not always chosen, which makes it more difficult to compare the dose groups.

Nevertheless, the study yielded reliable data on absorption in the rat and its dependence of solubilization in the stomach (conversion of the metallic iron to iron ions).

Reference 13

Endpoint:: basic toxicokinetics in vitro / ex vivo
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: Not specified
Reliability:: 3 (not reliable)
Rationale for reliability incl. deficiencies:: significant methodological deficiencies
Remarks:: The description of the methods is insufficient. For the concentration of the test substance (59Fe2O3) in the RPMI medium, only the number of particles in relation to the alveolar macrophages contained was given as 1:1; there is no precise indication of the concentration in [g/L].
Objective of study:: bioaccessibility (or bioavailability)
Principles of method if other than guideline:: Physicochemically uniform radioactively labelled iron(III) oxide (Fe2O3) particles of different sizes were selected and investigated for their extra and intracellular solubility and their ability to activate inflammatory reactions in vivo and in vitro in alveolar macrophages (AM). Particle dissolution rates were measured by determining the content of the radioactive iron (59Fe) label in the filtered media and the lysed cells.
GLP compliance:: not specified
Remarks:: in publication
Specific details on test material used for the study:: SOURCE OF TEST MATERIAL
- Source: self-synthesized, Fe2O3
- Synthesis of test material: 2 batches of monodisperse ferric oxide (Fe2O3) particles were produced from a solution of ferric nitrate (Fe2(NO3)3) dispersed into droplets of the appropriate concentration and uniform size by using a spinning top aerosol generator. Droplets were dried at 140 °C prior to on-line thermal degradation at 800°C in a tube furnace. Particles were prepared with and without labeling by 59Fe.
- Aerodynamic characterisation (by aerodynamic particle sizer & low-angle forward-scattering optical aerosol spectrometry & TEM): particles were spherical, with rough surface area.

Test material A:
- Geometric median diameter: 1.5 µm
- Aerodynamic diameter: 2.8 ± 1.2 µm
- Density (calculated from the median aerodynamic diameter and CMD): 3.8 g/cm3
- Estimated surface area (by TEM & BET): 7.1 m2/g

Tets material B:
- Geometric median diameter: 0.5 µm
- Aerodynamic diameter: 1.3 ± 1.2 µm
- Density (calculated from the median aerodynamic diameter and CMD): 4.1 g/cm3
- Estimated surface area (by TEM & BET): 17 m2/g

RADIOLABELLING INFORMATION
- Locations of the label: iron atom (59Fe) of Fe2O3 and of FeCl2
Radiolabelling:: yes
Species:: rat
Strain:: Wistar Kyoto (WKY)
Sex:: male
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Source: Kyo@Rj; Janvier, France
- Age at study initiation: 12 - 14 weeks old
- Housing: rats were housed in pairs.
- Diet (ad libitum): laboratory animal diet
- Water (ad libitum)

ENVIRONMENTAL CONDITIONS
- Temperature: 22 °C
- Humidity: 55 %
- Photoperiod (hrs dark / hrs light): 12 / 12

Test principle in brief:
- one cell culture medium with alveolar macrophages (AM) obtained from rats
- one single loading of test substance particle to AM ratio of 1:1.
- RPMI medium with AM: samples taken after day 2, 3, 4 and 7 at 37 °C

The aim of this test was to assess the extra- and intracellular solubility of Fe2O3 of two different sizes (1,5 and 0.5 µm) and surface areas (7.1 and 17 m2/g) in a cell culture medium (RPMI medium) with alveolar macrophages: the medium was used for all the in vitro incubations of cells with particles and control cells without particles.
Route of administration:: application in vitro
Vehicle:: not specified
Details on exposure:: IN VITRO APPLICATION
- Concentration of test material and reference chemical: 59Fe2O3 particles were added at a particle to AM ratio of 1:1.
- Cell culture medium: RPMI medium containing penicillin (100 U/mL), streptomycin (100 U/mL), amphotericin (2.5 μg/mL) and 5 % foetal calf serum.
- Incubation temperature: 37 °C
- Cell density in suspension for incubation (nominal and measured): alveolar macrophages covered 5 - 10 % (1*10^5 per well) of the wall botom.
Duration and frequency of treatment / exposure:: 12 days, single administration
No. of animals per sex per dose / concentration:: 4 males
Control animals:: yes, concurrent vehicle
Positive control reference chemical:: Not specified
Details on study design:: TEST SET UP:
(I) Intracellular and extracellular dissolution of 59Fe2O3:
Rat AM (1 × 10^5 cells/well) were incubated with 59Felabeled 0.5 and 1.5 μm Fe2O3 particles in 96 well plates for 12 days at 37°C, and the IPD of 59Fe2O3 particles in AM was determined according to Kreyling et al. (1990).*
Briefly, AM were incubated in full medium using a monolayer technique in 96 well plates. AM were purified by media exchange 2-4 h later. With the new media 59Fe2O3 particles were added at a particle to AM ratio of 1:1. AM covered 5-10% (1*10^5 per well) of the well bottom. These conditions allowed maintenance of fully functional AM through the time of incubation without exchange of media as had been optimized before. As described earlier in detail (Kreyling et al. (1990)) complete phagocytosis of the 59Fe2O3 particles (non-visible at the highest magnification ×600 of inverted microscope) within 24 h was monitored by quantifying phagocytosis of 2 μm fluorescent latex particles (FPSL) using an inverted microscope.* For this test FPSL were added to 2 separate wells at a FPSL to AM ratio of 1:1. Complete phagocytosis was assumed after 24 h when more than 90% of the FPSL were associated with AM as oppose to only 10% after 30 min. The functional state of the AM was monitored during 7 days of incubation by determining the cell concentration and viability in separate wells. IPD measurements were excluded when any of the in vitro functions of AM were markedly impaired.
After 2, 3, 4 and 7 days, intracellular dissolved 59Fe was determined by gamma spectroscopy in filtrates containing the dissolved 59Fe versus particulate 59Fe2O3 on filters of both the medium and the cell lysate. First the cell culture medium was filtered (0.22 μm membrane filter) providing the dissolved 59Fe in the filtrate ADmed(t) and the particulate 59Fe2O3 APmed(t); then the cells were lysed in the wells and subsequently the suspension of cell debris was filtered, too. The filtrates of the cell lysates represented intracellular dissolved iron ADlys(t) which was retained intracellularly while the filters contained the rest of particulate 59Fe2O3 (APlys(t)). Dissolved and particulate iron fractions DFj and PFj were obtained by normalizing activities ADj and APj by the sum of all four samples at a time t:

DFj(t) = ADj(t) / [ADmed(t) + ADlys(t) + APmed(t) and APlys(t)]; j = med, lys (1)
PFj(t) = APj(t) / [ADmed(t) + ADlys(t) + APmed(t) and APlys(t)]; j = med, lys (2)

From the measured dissolution kinetics of time-dependently increasing dissolved 59Fe fractions of the sum of AM medium and lysate, a mean total IPD rate and standard error were calculated. For control, extracellular dissolved 59Fe concentrations DFext(t) were determined in parallel in full media without cells by filtrations at the same time points t in order to determine the dissolved fractions DFext(t) of extracellular particulate fractions PFext(t). From the measured dissolution kinetics of time-dependently increasing dissolved fractions of the sum of AM medium and lysate, a mean total IPD rate and standard error were calculated.

[DFmed(ti) + DFlys(ti) − DFext(ti)] = Ftot *[1 − exp(−ltot t i)]; ti all times t (3)

Dissolved factions DFmed(ti) + DFlys(ti) were already corrected for externally dissolved DFext(ti). The control data of extracellular 59Fe dissolution in full media DFext(t) showed no time dependency but a small immediate leaching effect at the beginning of incubation.

(II) Free extracellular iron uptake of 59FeCl2:
It was determined by cultured AM, additional incubations were done and the cellular uptake of Fe2+ from medium into AM was measured. AM were incubated in 96 well-plates at 1*10^5 per well (0.2 ml) and iron chloride (FeCl2) radio-labeled with 59Fe was added varying between 0.1 - 5.0 μg FeCl2/ml or 1*10^6 AM.

Reagents:
Phosphate buffered saline (PBS) with and without Ca2+/Mg2+ (Biochrome, Berlin, Germany);
RPMI (PAA Laboratories, Linz, Austria);
foetal calf serum (Life Technologies, Eggenstein, Germany);
penicillin (Life Technologies, Eggenstein, Germany);
streptomycin (Life Technologies, Eggenstein, Germany);
amphotericin (Life Technologies, Eggenstein, Germany);
all other chemicals (analytical or HPLC grade, Merck, Darmstadt, Germany)

*Reference:
- Kreyling WG, Godleski JJ, Kariya ST, Rose RM, Brain JD: In vitro dissolution of uniform cobalt oxide particles by human and canine alveolar macrophages. Am J. Respir. Cell Mol. Biol., 2: 413-422.
Details on dosing and sampling:: Preparation of cell culture medium with alveolar macrophages:
- Tissues and body fluids sampled: bronchoalveolar lavage fluid
- Time and frequency of sampling: alveolar macrophages (AM) from healthy rats were isolated by bronchoalveolar lavages (repeated 5 times) using fresh aliquots of Ca2+/Mg2+-free PBS kept at 37°C each time at a volume equivalent to 28 mL/kg total lung capacity. After 20 min centrifugation at 400 g cells were resuspended in RPMI medium containing penicillin (100 U/mL), streptomycin (100 U/mL), amphotericin (2.5 μg/mL) and 5% fetal calf serum. Viability of cells was about 95% as estimated by trypan blue exclusion. After May Grünwald Giemsa staining of cytospin preparations microscopic examination identified about 95-99% of cells as AM.
Statistics:: Statistical significance was determined by analysis of variance and two-sample t-test. Changes with P < 0.05 were considered significant.
Preliminary studies:: Not specified
Type:: other: Bioaccessability in RPMI with alveolar macrophages (AM), 7 days, at 37 °C (1:1 59Fe2O3 particles/AM loading)
Results:: Intracellular dissolved and retained iron (0.5 μm particles): about 3 %
Type:: other: Bioaccessability in RPMI with alveolar macrophages (AM), at 37 °C (1:1 59Fe2O3 particles/AM loading)
Results:: For 0.5 μm particles:
- Extracellular dissolved fraction: 0.00055 ± 0.00022
- Intracellular particle dissolution (IPD) rates: 0.0037 ± 0.0014 1/d
Type:: other: Bioaccessability in RPMI with alveolar macrophages (AM), 7 days, at 37 °C (1:1 59Fe2O3 particles/AM loading)
Results:: Intracellular dissolved and retained iron (1.5 μm particles): about 1 %
Type:: other: Bioaccessability in RPMI with alveolar macrophages (AM), at 37 °C (1:1 59Fe2O3 particles/AM loading)
Results:: For 1.5 μm particles:
- Extracellular dissolved fraction: 0.000534 ± 0.00016
- Intracellular particle dissolution (IPD) rates: 0.0016 ± 0.0012 1/d
Details on absorption:: Not specified
Details on distribution in tissues:: Not specified
Details on excretion:: Not specified
Metabolites identified:: not specified
Details on metabolites:: Not specified
Enzymatic activity measured:: Not specified
Bioaccessibility (or Bioavailability) testing results:: Please refer to "any other information on results incl. tables" below.
Conclusions:: The bioaccessibility of radiolabelled 59Fe2O3 of 2 different particle sizes (0.5 or 1.5 μm) was determined in vitro by simulating intracellular and extracellular dissolution in cell culture medium (RPMI) with alveolar macrophages (AM) with a loading of 1:1 59Fe2O3 particle/AM. The extracellular dissolved fraction of the incubated 0.5 µm particles was 0.00055 ± 0.00022 and for 1.5 particles was 0.00034 ± 0.00016. In contrast, the intracellular dissolved fractions increased with time. The intracellular particle dissolution (IPD) rates from 0.5 and 1.5 μm 59Fe2O3 particles were 0.0037 ± 0.0014 1/d and 0.0016 ± 0.0012 1/d, respectively. From the particle dissolution kinetics data determined from measures of iron in the culture medium and cell lysate (corrected for external dissolved Fe), more than 70% of the dissolved iron remained within AM (likely iron binding protein-associated), whereas only 30% was released out of the cell into the extracellular medium. After 1 week the intracellular dissolved and retained iron progressively increased to about 3% with small and 1% with large 59Fe2O3 particles.
Free extracellular iron uptake was determined by cultured AM, additional incubations were done and the cellular uptake of Fe2+ from medium into AM was measured. Iron chloride (FeCl2) radio-labeled with 59Fe was added varying between 0.1 - 5.0 μg FeCl2/ml or 1*10^6 AM. Extracellular iron Fe2+ at doses between 0.1 - 5 μg 59FeCl2/mL medium (or per 1*10^6 AM) was taken up relatively weakly by AM (24 ± 2.5% of any of the doses provided).

The study by Knopf (2018) also addressed the solubility of Fe2O3 in different artificial physiological media (gastric fluid (GST), phosphate-buffered saline (PBS), artificial lysosomal fluid (ALF) and Gamble's solution (GMB)). Of the four media tested, Gamble's solution comes closest in composition to the test media in the present study.* The values for the solubility of radio labelled Fe2O3 from the in vitro study by Beck-Speier et al. (2009) cannot be really compared with those from the in vitro bioaccessibility study by Knopf (2018) because different media and concentrations of the test substance were used.* In addition, it is unclear in the present study in which concentration the test substance (59Fe2O3) was present in the RPMI medium. It was only stated that the number of particles in relation to the alveolar macrophages was 1:1 contained in the cell medium. A unit of mass for determining the substance concentration in the RPMI medium is not available in the present study. However, the solubility tests by Knopf (2018) with Fe2O3 in Gamble's solution medium (GMB) at a test item concentration of 100 mg/L after 168 h (corresponds to 7 days) also showed no significant solubility (negative value after correction) of the test item.*

*Reference:
- Knopf, B. (2018): Determination of the bioaccessibility of different iron oxides (E172) in artificial media. Report no. EBR-212/5-51.

Reference 14

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: Not specified
Reliability:: 3 (not reliable)
Rationale for reliability incl. deficiencies:: significant methodological deficiencies
Remarks:: Experimental set up was insufficiently described. The test material was insufficiently characterised (no purity). Only males were used. Information about the test animals (age of animals were missing) and environmental conditions (temperature, humidity, air changes) were insufficient.
Objective of study:: other: deposition and retention in the lung of radioactive iron(III) oxide (59Fe).
Qualifier:: no guideline followed
Principles of method if other than guideline:: Rats were exposed to aerosols of iron-59 (III) oxide (59Fe2O3) at a nominal concentration of 20 mg/m3 for 2 h; the effects on alveolar macrophages and the lung radioactivity were examined. The kinetics of particle clearance was measured. The main objective of this study was to determine the effect of a low lung burden of innocuous particles on the alveolar macrophage function. Iron(III) oxide was selected because of its presumed non-toxic character.
GLP compliance:: not specified
Remarks:: in the publication.
Specific details on test material used for the study:: RADIOLABELLING INFORMATION
- Particle size: Exposure I: 1.45 μm with σ(g)=2.9; Exposure II: 1.70 μm with σ(g)=3.0.
- Specific activity: averaged 8.3 x 10^-3 Ci/g Fe, or 4.5 x 10^-3 Ci/g Fe2O3 • 2(H2O), range 1.7 - 3.0 x 10^8 Bq/g.
- Locations of the label: Fe
Radiolabelling:: yes
Remarks:: [59Fe] ferric oxide
Species:: rat
Strain:: Long-Evans
Sex:: male
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Source: Charles Rivers Breeding Laboratories
- Age at study initiation: ≈ 2.5 months old
- Body weight at study: 225 - 250g
- Housing: litter-free cages, SPF facility
Route of administration:: inhalation: aerosol
Vehicle:: unchanged (no vehicle)
Remarks:: anhydrous air
Details on exposure:: TYPE OF INHALATION EXPOSURE: nose-only

GENERATION OF TEST ATMOSPHERE / CHAMBER DESCRIPTION
- Source and rate of air: anhydrous air
- System of generating particulates/aerosols: the ferric oxide aerosols (exposures I and II) were generated from [59Fe]ferric hydroxide hydrosols, prepared as previously described (Gibb and Morrow, 1962; Morrow et al. (1964)), using a DeVilbiss number 42 nebulizer modified for continuous replenishment of the hydrosols.* The aerosols were diluted with anhydrous air and passed at 16 l/min through a 50-mCi 63Ni deionizer maintained at ~70°C with a heating tape.
- Method of particle size determination: particle-size distributions were obtained using a sevenstage cascade impactor (Sandia Research, Albuquerque, N. Mex.).

TEST ATMOSPHERE
- Brief description of analytical method used: aerosol concentrations were measured by drawing the chamber air through glass-fiber filters (type A/E) and assaying the collected radioactivity. Weight measurements involving the drying of the hydrosol preparations at different temperatures indicated the iron oxide as aerosol was in the form of Fe2O3 • 2(H2O).

*Reference:
- Gibb, F. R., and Morrow, P. E. (1962): Alveolar clearance in dogs after inhalation of an iron59 oxide aerosol. J. Appl. Physiol. 17:429-432.
Duration and frequency of treatment / exposure:: 2 h, single administration
Dose / conc.:: 18.2 mg/m³ air
Remarks:: expressed as Fe2O3 * 2(H2O) (Exposure I)
Dose / conc.:: 24.2 mg/m³ air
Remarks:: expressed as Fe2O3 * 2(H2O) (Exposure II)
No. of animals per sex per dose / concentration:: Exposure I: 46 males (n = 8 or 9 per time point);
Exposure II: 49 males (n = 6 or 8 per time point).
Control animals:: yes, sham-exposed
Positive control reference chemical:: Not specified
Details on study design:: Animals from the first exposure series (exposure I) were sacrificed on the day of exposure (day 0), and 1, 3, 7, 14, and 30 d thereafter. Day 0 sacrifices were conducted over a 1- to 4-h period after the aerosol exposures. The second exposure series (exposure II) was designed to confirm and expand on the results obtained in the exposure I studies. Hence, the exposure II rats were sacrificed 1, 4, 11, 20, 50, and 75 d after exposure.
Details on dosing and sampling:: TOXICOKINETIC STUDY
- Tissues and body fluids sampled: lungs, trachea and bronchoalveolar fluid
- Time and frequency of sampling:
Exposure I: on the day of exposure (day 0), and 1, 3, 7, 14, and 30 d thereafter.
Exposure II: on day 1, 4, 11, 20, 50, and 75 after exposure.

ANALYTICAL METHOD
- Each rat was exsanguinated by severance of its carotid arteries, and its trachea was cannulated with an 18-gauge blunt needle secured with a ligature.
- Lungs were lavaged, macrophages isolated, counted and plated. Iron content of lavage fluid, cells, and lungs was recorded as radioactivity.
- The lungs and trachea were excised en bloc and the heart and the esophagus were removed. The trachea-lung preparations were placed vertically in plastic vials and positioned so that the lung bases were in apposition with the vial floors. The vials were placed in a 1 in x VA in Nal(TI) well detector, and, in conjunction with a single-channel analyzer system, lung 59Fe activities were assayed as previously described (Lehnert and Morrow, 1985).
Statistics:: Statistical significance of results was determined using standard
one-way analysis of variance and/or Student's two-tailed t-test with the
assumption of unequal variances in group means. All values expressed in the text and figures represent the mean ± one standard deviation (SD) about the mean; p values <0.05 were considered significant. Least-squares fit analyses of lung clearance data and impactor data were performed.
Preliminary studies:: Rats were exposed, nose-only, to filtered air for 2 h and sacrificed according to the same schedule as exposure I animals. No detectable changes in the phagocytic characteristics of AM (as assessed in the present study) from these animals were found (Lehnert and Morrow, 1984).*

*Reference:
- Lehnert, B. E., and Morrow, P. E. 1984. Functional characteristics of alveolar macrophages following the deposition of iron oxide in the lung. Toxicologist 4:3(abstr.).
Details on absorption:: Not specified
Details on distribution in tissues:: Not specified
Details on excretion:: Not specified
Test no.:: #1
Toxicokinetic parameters:: half-life 1st: 52.5±5.5 d: biological clearance from the lung.
Metabolites identified:: not specified
Conclusions:: Lehnert, B.E. et al. exposed rats to aerosols of iron-59 oxide (MMAD = 1.6 μm, σg = 3.0) at a nominal concentration of 20 mg/m³ for 2 h to determine how a low lung burden (~30 μg) of innocuous particles affects the size of the alveolar macrophage (AM) pool, and the functional status of the AM. The lung radioactivity was also measured.The average biological clearance half-time of 59Fe was calculated to be 52.5 ± 5.5 days.

The study shows some significant methodological deficiencies. Experimental set up was insufficiently described. The test material was insufficiently characterised (no purity). Only males were used. Information about the test animals (age of animals were missing) and environmental conditions (temperature, humidity, air changes) were insufficient.

Reference 15

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: Not specified
Reliability:: 2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:: study well documented, meets generally accepted scientific principles, acceptable for assessment
Remarks:: The experimental set up and results are described in detail. However, the test material was insufficiently characterised (no purity).
Objective of study:: toxicokinetics
Principles of method if other than guideline:: Test substance retention was measured by determination of radioactivity.
Male rats were exposed in a nose only system to 59Fe304 for 2 hrs. at a mean concentration of 15.4 ± 4.5 mg/m3. Following exposure, thoracic 59Fe activity in the live animals was measured. After 120 days, animals were sacrificed, and the lungs were removed for measuerment of 59Fe activities by a Nal (T1) detector.
GLP compliance:: not specified
Specific details on test material used for the study:: RADIOLABELLING INFORMATION
- Name of test material: 59Fe304
- Particle size: 1.5 µm with σ(g)=1.8
- Locations of the label: iron atom
Radiolabelling:: yes
Species:: rat
Strain:: Fischer 344
Sex:: male
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- mean body weight 2 months before study initiation: 110 g
Route of administration:: inhalation: dust
Vehicle:: not specified
Details on exposure:: TYPE OF INHALATION EXPOSURE: nose only

GENERATION OF TEST ATMOSPHERE / CHAMBER DESCRIPTION
- System of generating particulates/aerosols: preparation and generation procedure of the aerosol is described elsewhere (Oberdoerster and Freedman, 1981).*
Duration and frequency of treatment / exposure:: 2 hours
Dose / conc.:: 15.4 mg/m³ air
Remarks:: ± 4.5 mg/m³
No. of animals per sex per dose / concentration:: 80 males
Control animals:: not specified
Positive control reference chemical:: not specified
Details on dosing and sampling:: TOXICOKINETIC (Elimination):
- Tissues sampled: lungs
- Time and frequency of sampling: at study start (day 0) and end of study (day 120)

ANALYTICAL METHOD
- Following exposure, thoracic 59Fe activity in the live animals was measured. Based on these counts, the animals were divided into 8 subgroups of 10 rats, each of which had the same mean thoracic activity assumed to represent the same lung burden.
- The animals of one subgroup of each group were exsanguinated under pentobarbital anesthesia over a period of 120 days, their lungs were excised and air-dried in an inflated state (20 cm H2O). 59Fe activities of the lungs were determined in a 2 x 2 inch Nal (T1) detector. Resulting activity values were normalized for differences in initial lung burdens of 59Fe3O4 due to different exposure concentrations and expressed as relative retention compared to day 0 (=100 %).
Preliminary studies:: not specified
Type:: clearance
Results:: Relative retention of 59Fe3O4: 88 ± 5 % at day 1 and 19 ± 4 % at day 120 of the initially deposited 59Fe3O4 particles were still present in the lungs. Tissue retention was 17.3 ± 6.7 µg Fe3O4 per lung after 2 h exposure.
Details on absorption:: not specified
Details on distribution in tissues:: not specified
Details on excretion:: See toxicokinetic parameters
Toxicokinetic parameters:: half-life 1st: 47,2 ± 5.1 days
Metabolites identified:: not specified
Details on metabolites:: not specified
Enzymatic activity measured:: Not specified
Bioaccessibility (or Bioavailability) testing results:: Not specified
Conclusions:: In the study by Oberdoerster et al. (1984), the test substance retention was measured by determination of radioactivity. Male rats were exposed in a nose only system to 59Fe304 for 2 hrs. at a mean concentration of 15.4 ± 4.5 mg/m3. Following exposure, thoracic 59Fe activity in the live animals was measured. After 120 days, animals were sacrificed, and the lungs were removed for measurement of 59Fe activities by a Nal (T1) detector.
On day 120, 19 ± 4 % of the initially deposited 59Fe3O4 particles were still present in the lungs. Tissue retention was 17.3 ± 6.7 µg Fe3O4 per lung after 2 h exposure. The half-live for 59Fe3O4 was 47 days.

The study is well documented and meets generally accepted scientific principles. The experimental set up and results are described in detail. However, the test material was insufficiently characterised (no purity).

Reference 16

Endpoint:: basic toxicokinetics, other
Remarks:: in vivo and in vitro
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: not specified
Reliability:: 2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:: study well documented, meets generally accepted scientific principles, acceptable for assessment
Remarks:: At the in vitro solubility study, the dissolution of iron was measured for the various iron compounds over a period of 15 - 150 min, only. The solubility after 24 h was not examined.
Objective of study:: other: determination of bioavailability of six different elemental iron powders and the extent to which their physicochemical properties can predict it.
Qualifier:: no guideline followed
Principles of method if other than guideline:: The study may not directly address the requirements of this endpoint, but it provides some very important information on the bioavailability of Fe powders. Six elemental Fe powders were supplemented in rat diets, in order to determine their bioavailability and the extent to which their physicochemical properties can predict it. Bakery grade ferrous sulfate was used as a reference standard because of its known high bioavailability.
GLP compliance:: not specified
Specific details on test material used for the study:: (I) Carbonyl iron (Ferronyl, U.S.)
- Specific surface area: 362 ± 4 m2/kg
- Microstructure of particles: smaller and spherical with a smooth surface and were more homogeneous than the other iron powders.

(II) Electrolytic iron (A-131, U.S.):
- Specific surface area: 370 ± 3 m2/kg
- Microstructure of particles: highly irregular, coarse, flake-like appearance with apparent fissures.

(III) Electrolytic iron (India):
- Specific surface area: 245 ± 4 m2/kg
- Microstructure of particles: highly irregular, coarse, flake-like appearance with apparent fissures.

(IV) H-reduced iron (AC-325, U.S.):
- Specific surface area: 260 ± 3 m2/kg
- Microstructure of particles: highly irregular, coarse, flake-like appearance with apparent fissures.

(V) Reduced iron (ATOMET 95SP, Canada):
- Specific surface area: 225 ± 6 m2/kg
- Microstructure of particles: highly irregular, coarse, flake-like appearance with apparent fissures.

(VI) CO-reduced iron (RSI-325, Sweden):
- Specific surface area: 90 ± 2 m2/kg
- Microstructure of particles: highly irregular, less coarse with grooved surfaces.

Analysis of test material by the authors:
- Specific surface area of the particles was determined by nitrogen gas absorption (Brunauer et al. (1938)).*
- Microstructure of particles among Fe powders was examined by scanning electron microscopy.

*Reference:
- Brunauer, S., Emmett, P. H. & Teller, E. (1938): Adsorption of gases in multimolecular layers. J. Am. Chem. Soc., 60: 309–319.
Radiolabelling:: no
Species:: rat
Strain:: Sprague-Dawley
Sex:: male
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Source: Charles River/SASCO, Wilmington, MA
- Age at study initiation: weanling
- Weight at study initiation: 100 ± 9 g
- Fasting period before study: 24 d on an Fe deficient diet (depletion period, ~ 1.5 mg Fe/kg AIN-93G diet)
- Housing: wire-bottomed stainless steel mesh cages
- Individual metabolism cages: yes
- Diet: ad libitum
- Water: ad libitum

ENVIRONMENTAL CONDITIONS
- Photoperiod (hrs dark / hrs light): 12
Route of administration:: oral: feed
Vehicle:: other: powder mixture with sucrose
Details on exposure:: DIET PREPARATION
- Mixing appropriate amounts with (Type of food): AIN-93G diet, modified to lower Fe contamination by using vitamin-free casein (Harlan Teklad, Madison, WI), a reagent grade calcium carbonate (J.T. Baker, Phillipsburg, NJ) and a high purity cellulose fiber source, Alphacel (ICN Biomedicals, Irvine, CA). Oil was added when the Fe-sucrose mixture was added in the food, and it was mixed for 30 min.
- Diet composition (without added iron): diet contained 1.6 mg iron/kg diet by analysis and contained (g/kg): 315 cornstarch, 200 vitamin-free casein, 314.5 sucrose, 70 soybean oil, 50 cellulose, 35 iron-free mineral mix (AIN-93G mineral mix omitting ferric citrate), 10 vitamin mix (AIN-93VX; Harlan Teklad, Madison, WI), 3 Lcystine, and 2.5 choline bitartrate.
Duration and frequency of treatment / exposure:: 14 d (after 24 d of Fe depletion diet), daily
Dose / conc.:: 1.5
Remarks:: mg iron/kg diet (control, no added iron)
Dose / conc.:: 6
Remarks:: mg iron/kg diet of bakery-grade ferrous sulfate
Dose / conc.:: 12
Remarks:: mg iron/kg diet of the 6 elemental iron powders and bakery-grade ferrous sulfate.
Dose / conc.:: 18
Remarks:: mg iron/kg diet of bakery-grade ferrous sulfate
Dose / conc.:: 24
Remarks:: mg iron/kg diet of the 6 elemental iron powders and bakery-grade ferrous sulfate.
Dose / conc.:: 32
Remarks:: mg iron/kg diet of the 6 elemental iron powders
No. of animals per sex per dose / concentration:: 9 - 10 males/diet
Control animals:: yes, plain diet
Positive control reference chemical:: no
Details on study design:: (1) IN VIVO BIOAVAILABILITY:
(a) Haemoglobin (HGB) and HGB-Fe were determined colorimetrically. The relative bioavailability was calculated on the basis of absolute Fe intake (µg/d)
(b) diet Fe concentration (mg/kg) and on change in HGB concentration (g/l) as well as change in total body HGB-Fe (mg/rat); blood samples were collected at the beginning (from orbital socket) and at the end (from descending vena cava) of the 14-day period.

(2) IN VITRO BIOACCESSIBILITY:
- Solubility of the Fe powders was measured according to the method of Forbes et al. (1989) with few modifications, in hydrochloric acid solution at pH 1 and 1.7, sampling at 15, 30, 45, 60, 90, 120, 150 min.*

*References:
- Forbes, A. L. et al. (1989): Comparison of in vitro, animal and clinical determinations of iron bioavailability: International Nutritional Anemia Consultative Group Task Force report on iron bioavailability, bioavailability of iron used in fortification and enrichment, electrolytic iron, ferric orthophosphate, ferrous sulfate. Am. J. Clin. Nutr. 49: 225–238.
Details on dosing and sampling:: (1) in vivo BIOAVAILABILITY:
- Body fluids sampled: blood
- Time and frequency of sampling: at the beginning (from orbital socket) and at the end (from descending vena cava) of the 14-day period.
- Analytical method: (a) haemoglobin (HGB) and HGB iron determination: at both times, a modified methaemoglobin method was used for the colorimetric determination of HGB (CELL-DYN 3500 System). HGB-iron was determined on the basis of 3.35 mg iron/g HGB and 0.075 L blood/kg body weight.
(b) analysis of Fe content in diet: each diet (n=6) was ashed by alternating dry heat (500°C) with reflux in concentrated nitric acid, then dissolving in 6 mol/L hydrochloric acid. Samples were analyzed by inductively coupled plasma-atomic emission spectrophotometry (ICP-AES).

(2) in vitro BIOACCESSIBILITY:
- Time and frequency of sampling: Iron in solution was assayed at 15, 30, 45, 60, 90, 120, and 150 min time points.
- Sampling: each time, shaking was stopped and 1 mL of solution was quickly taken and centrifuged at 11,600 x g for 5 min. Immediately after centrifugation, 500 µL of the supernatant was diluted to 5 mL with 0.02 mol/L hydrochloric acid and then analyzed by ICP-AES.
- Analytical method: solubility was measured (n=3) by placing 20 mg iron powder in 250 mL dilute 0.1 and 0.02 mol/L hydrochloric acid (calculated pH of 1.0 and 1.7, respectively) with constant orbital shaking at 150 rpm and 37°C. No glass beads or magnetic stir devices were used.
Statistics:: Solubility and surface area data were analyzed using one-way ANOVA; differences among the means were tested using Tukey’s multiple comparison tests with significance set at P ≤ 0.05 and values expressed as means ± SEM.
Preliminary studies:: no
Details on absorption:: The relative bioavailability did not differ when calculated with the 4 different ways, for each powder. Results can be seen in Table 1 (see attachment 1). Carbonyl Fe powder was the most bioavailable.
Details on distribution in tissues:: not examined
Details on excretion:: not examined
Metabolites identified:: not measured
Details on metabolites:: not examined
Enzymatic activity measured:: not examined
Bioaccessibility (or Bioavailability) testing results:: Physicochemistry: carbonyl Fe was the most soluble and reduced Fe the least soluble (Fig. 3, see attachment 2) at 150 min. Iron solubility was accounted for 36 - 65% (150 min the most predictive time point) and 38 - 82% (30 min the most predictive time point) of the variation in bioavailability for pH 1.7 and 1, respectively (Fig. 4 a, attachment 2). Surface area accounted for 80% of the variation in bioavailability (Fig. 4 b, see attachment). Surface area ranges are shown in Table 2 (see attachment 2).
Conclusions:: The bioavailability of 6 commercial elemental iron powders were determined and examined how physiochemistry influences bioavailability. Relative biological value (RBV) of the iron powders was determined using a haemoglobin repletion/slope ratio method, treating iron-deficient rats with repletion diets fortified with graded quantities of iron powders, bakery-grade ferrous sulfate or no added-iron. Iron powders were assessed physicochemically by measuring iron solubility in hydrochloric acid at pH 1 and 1.7, surface area by nitrogen gas absorption and surface microstructure by electron microscopy.
Bioavailability from the iron powders, based on absolute iron intake, was significantly less than from FeSO4 (100%, p< 0.05) with the following rank order: carbonyl Fe (64%)> electrolytic Fe (54%; USA)> electrolytic Fe (46%;India)> H-reduced Fe (42%)> reduced Fe (24%; Canada)> CO-reduced Fe (21%). Solubility testing resulted in different relative rankings and better RBV predictability with increasing time at pH 1.7. The prediction was improved with less time and lower pH. Surface area ranging from 90 to 370 m2/kg, was also highly predictive of RBV. Bioavailability of Fe powders is less than bakery-grade ferrous sulfate and varies up to 3 times among different commercial forms. Solubility at pH 1 and surface area were predictive of Fe bioavailability in rats.

Interpretation of results (migrated information):
According to this study the solubility of Fe particles at pH 1 and surface area can be predictive of its bioavailability in rats.
The study provides important insight in the bioavailability of Fe particles from different powders, as well as on the factors that determine its extent. The results support its correlation to the solubility, size and surface area of the particles.

The study is well-documented. At the in vitro solubility study, the dissolution of iron was measured for the various iron compounds over a period of 15 - 150 min, only. The solubility after 24 h was not examined.

Reference 17

Endpoint:: basic toxicokinetics, other
Remarks:: in vivo and in vitro
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: Not specified
Reliability:: 2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:: study well documented, meets generally accepted scientific principles, acceptable for assessment
Remarks:: Experimental set up is well-documented, but not all the results are shown in the publication. The 3 doses of test item in diet at the in vivo study were analysed but not reported. At the in vitro solubility study, the dissolution of iron was measured for the various iron compounds over a period of 5 - 90 min, only. The solubility after 24 h was not examined.
Objective of study:: bioaccessibility (or bioavailability)
Qualifier:: no guideline followed
Principles of method if other than guideline:: The bioavailability of Fe from 7 elemental iron powders (produced by electrolysis, reduction with hydrogen (H2), carbon monoxide (CO), and desiccated ammonia (NH3), and carbonyl process) for rats were compared with the in vitro solubility of the powders in 0.2% HCl (w/v, ca 0.05 N) for periods of 5 to 90 min. The relative bioavailability of Fe from the 7 elemental iron powders compared to FeSO4 were evaluated in vivo via the haemoglobin concentrations in blood by the cyanomethaemoglobin procedure.
GLP compliance:: not specified
Remarks:: The study was conducted before the adoption of the GLP principles.
Specific details on test material used for the study:: 1. Electrolytic iron:
- Purity: assumed to be almost 100 % Fe
- Impurities: Na, Mg, Ca, Cl, Al, Cu, Mn, V, Mo, Cr, Sb, Co.
- Particle size (Infrasizer): 7 - 10 µm, about 87 % of particles < 27 µm;
- Specific surface area (Quantimet image analysis): 1368 cm2/g

2. Carbonyl iron A:
- Purity: assumed to be almost 99 % Fe
- Impurities: Na, Mg, Ca, Cl, Al, Cu, Mn, V, Mo, Cr, Sb, Co.
- Particle size (Quantimet image analysis): 0.5 – 11 µm, ca. 92 % of particles < 27 µm and 69 % < 7 µm;
- Specific surface area (Quantimet image analysis): 1720 cm2/g

3. Carbonyl iron B:
- Purity: assumed to be almost 100 % Fe
- Impurities: Na, Mg, Ca, Cl, Al, Cu, Mn, V, Mo, Cr, Sb, Co.
- Particle size (Quantimet image analysis): 1 - 21 µm, ca. 86 % of particles < 27 µm;
- Specific surface area (Quantimet image analysis): 1450 cm2/g

4. CO-reduced iron:
- Purity: assumed to be almost 99 % Fe
- Impurities: Na, Mg, Ca, Cl, Al, Cu, Mn, V, Mo, Cr, Sb, Co.
- Particle size (Infrasizer): 7 - 10 µm;
- Specific surface area (Quantimet image analysis): 1259 cm2/g

5. H2-reduced iron A:
- Purity: assumed to be almost 99 % Fe
- Impurities: Na, Mg, Ca, Cl, Al, Cu, Mn, V, Mo, Cr, Sb, Co.
- Particle size (Infrasizer): 7 - 10 µm;
- Specific surface area (Quantimet image analysis): 1309 cm2/g

6. H2-reduced iron B:
- Purity: assumed to be almost 100 % Fe
- Impurities: Na, Mg, Ca, Cl, Al, Cu, Mn, V, Mo, Cr, Sb, Co.

7. NH3-reduced iron:
- Purity: assumed to be almost 100 % Fe
- Impurities: Na, Mg, Ca, Cl, Al, Cu, Mn, V, Mo, Cr, Sb, Co.
Radiolabelling:: no
Species:: rat
Strain:: Wistar
Sex:: male
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Age at study initiation: 19 days
- Fasting period before study: the animals were made anemic
- Individual metabolism cages: yes
Route of administration:: oral: feed
Vehicle:: unchanged (no vehicle)
Details on exposure:: Diet: the Fe content was determined colorimetrically in the experimental and the basal diet (ca 6-10 mg Fe/kg) before administration
Duration and frequency of treatment / exposure:: 2 weeks, daily
Remarks:: Doses:
3 dose levels for each Fe powder and 3 doses of FeSO4 (positive control) at equal log increments
No. of animals per sex per dose / concentration:: 8 - 10 males
Control animals:: other: positive control
Positive control reference chemical:: yes, fed with FeSO4*7H2O
Details on study design:: The following parameters were investigated:
(1) In vivo study for relative biological activity of Fe:
- 3 levels of ferrous sulfate and 3 levels of each test sample were fed at equal log increments in amounts which gave linear and parallel responses in haemoglobin when plotted against the logarithm of added iron. Blood samples were collected and haemoglobin concentrations were determined by the cyanomethaemoglobin procedure of Crosby et al. (1954).* A colorimetric procedure by Davies et al. (1972) was used to determine the iron content of the basal and experimental diets to verify the adequacy of diet preparation.*

(2) In vitro study for bioaccessibility of elemental iron powders:
- Solubility of the elemental Fe powders were determined in 0.2% HCl (w/v) (ca. 0.05 N) according to the procedure described by Shah et al. (1977) at intervals from 5 to 90 min and measured by the colorimetric procedure by Davies et al. (1972).*

*References:
- Crosby, W.H., Munn, J.I. & Furth, F.W. (1954) U.S. Armed Forces Med. J. 5, 693-703.
- Shah, B.G., Giroux, A. & Belonje, B. (1977) J. Agric Food Chem. 25, 592-594.
- Davies, M.I., Bush, K. and Motzok, I. (1972) J. Assoc. Off. Anal. Chem. 55, 1206-1210.
Details on dosing and sampling:: (1) in vivo study:
- Blood samples were obtained from the tail of the rats.

(2) in vitro study:
- Sampling: a suitable aliquot was removed at 5, 10, 20 and 90 min. Any iron particles in the aliquot were taken out immediately with a Teflon-coated magnetic stirrer and the dissolved iron was measured.
- Procedure: a 100 mg sample of iron was placed in a 500 ml conical flask containing glass beads (5 mm diameter) and 250 ml 0.2% HCl (w/v) (ca. 0.05 N) preheated to 37°C and immediately shaken in a water bath (37°C) with a 1 in.-stroke reciprocating platform set at 150 oscillations/min. Samples were removed at intervals from 5 to 90 min. Any iron particles in the aliquot were taken out and the dissolved iron was measured colorimetrically.
Statistics:: The procedure of Bliss (1952) was used to calculate the RBVs with 95% level of probability.
Preliminary studies:: not performed
Type:: other: bioavailability/absorption
Results:: Relative biological values (versus FeSO4 =100 %):
electrolytic iron: 42 %; carbonyl iron A: 31 %; carbonyl iron B: 27 %; CO-reduced iron: 12 %; H2-reduced iron A: 14 %; H2-reduced iron B: 13 %; NH3-reduced iron: 15 %.
Details on absorption:: (1) in vivo study: elemental Fe powders produced by reduction (either H2 or CO), demonstrated very low bioavailability, in comparison to the electrolytic and both carbonyl Fe powders. The results are presented in Table 1 (see attachment 1).
Details on distribution in tissues:: not measured
Details on excretion:: not measured
Toxicokinetic parameters:: other: Bioaccessibility (in HCl, after 90 min): electrolytic iron: 85 %; carbonyl iron A and B: 89 % and 90 %; CO-reduced iron: 40 %; H2-reduced iron A and B: 47 and 57 %; NH3-reduced iron: 46 %.
Metabolites identified:: not specified
Details on metabolites:: not specified
Enzymatic activity measured:: not specified
Bioaccessibility (or Bioavailability) testing results:: Results on the in vitro solubility of the different iron compounds at various time points are shown in more detail in table 1 of the attachment 1.
Conclusions:: The relative biological value (RBV) for rats of the Fe from 7 elemental powders (produced by electrolysis, reduction with hydrogen, carbon monoxide, and dessicated ammonia, and carbonyl process) were compared with the in vitro solubility of the Fe powders in 0.2 % HCl (w/v, ca 0.05 N) for periods of 5 to 90 min.
Elemental Fe powders produced by reduction (either H2 or CO), demonstrated very low bioavailability, in comparison to the electrolytic and both carbonyl Fe powders. Electrolytic iron, carbonyl iron A and B showed a high solubility in the HCl solution of ≥ 85 % after 90 min. while the H2- and NH3-reduced iron showed a lower solubility of 40 - 57 %.

The methods used for the measurement of particle size distribution have showed contradictory results for carbonyl Fe, and as recognised by the authors their adequacy for measurement of carbonyl Fe has to be further investigated. There was a good agreement among the ratios of RBV/surface area (range 0.0174 -0.0184) for H2-, CO- reduced and carbonyl Fe powders, while this was not the case for elctrolytic Fe (0.0307) (only fractions containing particles from 7 to 10 µm from electrolytic, H2 -, and CO- reduced Fe were used). All Fe powders had a purity of ≥ 99%.

The experimental set up is well-documented, but not all the results are shown in the publication. The 3 doses of test item in diet at the in vivo study were analysed but not reported. At the in vitro solubility study, the dissolution of iron was measured for the various iron compounds over a period of 5 - 90 min, only. The solubility after 24 h was not examined.

Reference 18

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: 2018-09-11 to 2019-06-21
Reliability:: 1 (reliable without restriction)
Rationale for reliability incl. deficiencies:: guideline study
Objective of study:: toxicokinetics
Qualifier:: according to guideline
Guideline:: OECD Guideline 417 (Toxicokinetics)
Version / remarks:: 2010-07-22
Deviations:: no
GLP compliance:: yes (incl. QA statement)
Remarks:: signed 2017-05-08
Specific details on test material used for the study:: STABILITY AND STORAGE CONDITIONS OF TEST MATERIAL
- Storage condition of test material: at room temperature, kept dry and stored in airtight closed containers.
Radiolabelling:: no
Species:: rat
Strain:: other: Crl:CD(SD)
Details on species / strain selection:: The rat is a commonly used rodent species for toxicity studies.
Sex:: male/female
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Source: Charles River Laboratories Research, Models and Services Germany GmbH, Sandhofer Weg 7, 97633 Sulzfeld, Germany
- Age at study initiation: 59 to 60 days
- Weight at study initiation: males: 262 g to 340 g; females: 200 g to 272 g
- Housing: kept in groups of up to 3 animals (same sex) in MAKROLON cages (type IV) with a basal surface of approximately 55 cm × 33 cm and a height of approximately 20 cm; bedding material: granulated textured wood
- Diet (ad libitum): Commercial diet, ssniff® R/M-H V1534
- Water (ad libitum): drinking water
- Acclimation period: 14 days

ENVIRONMENTAL CONDITIONS
- Temperature: 22°C ± 3°C (maximum range)
- Relative humidity: 55% ± 10% (maximum range)
- Photoperiod (hrs dark / hrs light): 12/12
Route of administration:: other: Sicovit Red 30 E172: oral (gavage); reference item: intravenously injected
Vehicle:: other: Sicovit Red 30 E172: 0.5 % aqueous hydroxypropylmethylcellulose gel; reference item: 0.9 % NaCl solution
Details on exposure:: PREPARATION OF DOSING SOLUTIONS:
1) Sicovit Red 30 E172:
The test items were suspended or dissolved in the vehicle to the appropriate concentration freshly on the administration day and were administered orally by gavage at a constant volume (adminsitration volume: 10 mL/kg bw). The application formulations were continuously agitated by stirring throughout the entire administration procedure.

2) Reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%)
Prior to administration, the reference item and the appropriate vehicle were heated to 70°C and stirred at 50°C for approx. 3 hours until the reference item was completely dissolved. This clear solution was maintained at room temperature until administration. The status as clear solution was monitored and recorded upon administration. Immediately after formulation preparation for the females, the formulations were protected from light by transferring the formulation into brown containers or wrapping in aluminium foil.

The amounts of the test and reference items were adjusted to the animal's current body weight on the administration day.

Administration volume (oral administration / intravenous administration): 10 mL/kg bw/day

Injection speed (intravenous adminsitration): dose per approx. 15 seconds
Duration and frequency of treatment / exposure:: single administration
Dose / conc.:: 1 000 mg/kg bw/day (actual dose received)
No. of animals per sex per dose / concentration:: 5 males / 5 females
Control animals:: yes, concurrent vehicle
Positive control reference chemical:: none
Details on study design:: - Dose selection rationale: the dose levels for this study were selected after consultation with the sponsor based on available toxicity and bioavailability data (as far as available):

1) Reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%):
The oral LD50 value for iron citrate monohydrate was stated as being >2000 mg/kg bw; the oral bioavailability of soluble Fe substances are given in the public domain with 1 to 26% (Fe).

For the test item oral dosing of 1000 mg/kg bw, a very low relative bioavailability was assumed (<1%), considering the very low water solubility and bioacessibility in gastric juice. Since the four iron oxide test items have Fe-contents of approx. 70%, the dose of the reference substance should be adjusted accordingly. Given a test item dose of 1000 mg/kg b.w. (corresponding to 700 mg Fe/kg bw), then 1% of this dose would correspond to 7 mg Fe/kg bw (or 36.8 mg/kg bw iron citrate). Correcting for approx. 20% oral bioavailability of soluble iron substances, this yields a dose for the reference item of 7.4 mg iron citrate/kg bw to be given by intravenous injection.

2) Sicovit Red 30 E172:
The test item oral doses of 1000 mg/kg bw correspond to the limit dose used in a separate 90-day oral toxicity study, which was considered the maximum feasible dose. This dose was also selected in view of the anticipated low bioavailability and the requirements of analytical sensitivity of the analytical method for iron in plasma.

3) Vehicle control group:
In view of the long established circadian variation of plasma iron levels (Lynch et al, 1973)*, a vehicle control group was sampled for blood plasma over a period of 24 hours at identical sampling time points and intervals as the dosed groups.

*Reference:
Lynch et al (1973): Circadian Variation in Plasma Iron Concentration and Reticuloendothelial Iron Release in the Rat, Clinical Science and Molecular Medicine (1973) 45, 331-336.
Details on dosing and sampling:: TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled: plasma
- Time and frequency of sampling: blood was collected 0 (predose), 0.5, 1, 2, 4, 8, 12, 24, 48 (test item and reference item only), and 72 hours (test item and reference item only) after administration. The whole blood samples were cooled using an IsoTherm-Rack system until centrifugation. Immediately after centrifugation, the isolated plasma was frozen at -20°C ± 10 % and stored at this temperature until analysis.

Pharmacokinetic evaluation of plasma data was performed and a non-compartment model was employed. The following parameters were determined, if possible:
AUC0-∞ = extrapolated area from zero to infinity
AUC0-t last = extrapolated area from time zero to the last quantifiable plasma concentration (i.e. >lower limit of quantification, LLOQ)
Kel = elimination rate constant
t1/2 = elimination half-life

Cmax values were the highest measured plasma concentrations and tmax values were the time points of highest plasma concentrations.

Elimination rate constants (Kel) and plasma elimination half-lives (t½) were calculated by linear regression analysis of the log/linear portion of the individual plasma concentration-time curves (c = concentration, t = time).

Area under the curve (AUC) values were calculated using the linear trapezoidal method and extrapolated to infinite time by dividing the last measurable plasma concentration by the elimination rate constant. Plasma concentrations at time zero were taken to be those at the first blood sampling time.

Furthermore, the AUC0-t last was calculated according to the linear trapezoidal rule. Values below the limit of quantification (LOQ) were excluded from calculation.

In addition, the bioavailability was calculated for the mixture.

For plasma, a pre-treatment by a microwave digestion with HNO3 was necessary to digest the proteins in plasma. Afterwards iron in digested samples was measured by ICP-OES.

OBSERVATIONS
- clinical signs: before and after dosing as well as regularly throughout the working day (7.30 a.m. to 4.30 p.m.) and on Saturdays and Sundays (8.00 a.m. to 12.00 noon; final check at approx. 4.00 p.m).
Special attention was paid to the local tolerance at the injection site(s).
- mortality/morbund: early in the morning and again in the afternoon of each working day as well as on Saturdays and Sundays (final check at approx. 4.00 p.m).
- body weight: at the time of group allocation, before dosing for dose adjustment and on test day 4 before the last blood sampling.

ADMINISTRATION FORMULATION ALANYSIS:
For each test item, that was mixed with the vehicle and the reference substance, tests by appropriate analytical methods were conducted to determine the concentration and stability of the test item in the formulations. For the analysis of the application formulations, one sample of exactly 10 mL from each dosing suspension (test items) or dosing solution (reference item) was taken at the start of the administration (test day 1 of the female animals) and frozen until analysis.

Application solutions of the iron oxide was measured after addition of aqua regia to the samples and after an incubation time for at least four days by ICP-OES. After this measurement the remaining precipitation (only iron oxide application solution) were digested by a microwave procedure and measured by ICP-OES.

ANALYTIC OF REFERENCE ITEM:
The iron content of the reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%) was determined using ICP-OES.
Statistics:: The test item group was compared to the reference group. The following statistical method was used:
- Student's t-test (body weight (at p≤0.05 and p≤0.01; limits used p = 0.05 approx. t = 2.306 p = 0.01 approx. t = 3.355 (for 8 degrees of freedom))
Preliminary studies:: none
Details on absorption:: not specified
Details on distribution in tissues:: not specified
Details on excretion:: not specified
Toxicokinetic parameters:: other: bioavailability
Remarks:: An absolute bioavailability of 0.22%/0.23% (m/f) for Sicivit Red was calculated for Fe following oral administration compared to intravenous administration.
Toxicokinetic parameters:: other:
Remarks:: It should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
Toxicokinetic parameters:: other:
Remarks:: The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels.
Toxicokinetic parameters:: other:
Remarks:: The calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.
Metabolites identified:: not specified
Details on metabolites:: not specified
Bioaccessibility (or Bioavailability) testing results:: An absolute bioavailability of 0.22%/0.23% (m/f) for Sicivit Red was calculated for iron following oral administration compared to intravenous administration. However, it should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels. Hence, the calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.

Please also refer for results to the field "Attached background information" below.
Conclusions:: An absolute bioavailability of 0.22%/0.23% (m/f) for Sicovit Red was calculated for iron following oral administration compared to intravenous administration. However, it should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels. Hence, the calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.

Reference 19

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: 2018-09-11 to 2019-06-21
Reliability:: 1 (reliable without restriction)
Rationale for reliability incl. deficiencies:: guideline study
Objective of study:: toxicokinetics
Qualifier:: according to guideline
Guideline:: OECD Guideline 417 (Toxicokinetics)
Version / remarks:: 2010-07-22
Deviations:: no
GLP compliance:: yes (incl. QA statement)
Remarks:: signed 2017-05-08
Specific details on test material used for the study:: STABILITY AND STORAGE CONDITIONS OF TEST MATERIAL
- Storage condition of test material: stored in tightly-closed, original container, at ambient temperature, kept dry.
Radiolabelling:: no
Species:: rat
Strain:: other: Crl:CD(SD)
Details on species / strain selection:: The rat is a commonly used rodent species for toxicity studies.
Sex:: male/female
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Source: Charles River Laboratories Research, Models and Services Germany GmbH, Sandhofer Weg 7, 97633 Sulzfeld, Germany
- Age at study initiation: 59 to 60 days
- Weight at study initiation: males: 262 g to 340 g; females: 200 g to 272 g
- Housing: kept in groups of up to 3 animals (same sex) in MAKROLON cages (type IV) with a basal surface of approximately 55 cm × 33 cm and a height of approximately 20 cm; bedding material: granulated textured wood
- Diet (ad libitum): Commercial diet, ssniff® R/M-H V1534
- Water (ad libitum): drinking water
- Acclimation period: 14 days

ENVIRONMENTAL CONDITIONS
- Temperature: 22°C ± 3°C (maximum range)
- Relative humidity: 55% ± 10% (maximum range)
- Photoperiod (hrs dark / hrs light): 12/12
Route of administration:: other: Sicovit Yellow 10 E172: oral (gavage); reference item: intravenously injected
Vehicle:: other: Sicovit Yellow 10 E172: 0.5 % aqueous hydroxypropylmethylcellulose gel; reference item: 0.9 % NaCl solution
Details on exposure:: PREPARATION OF DOSING SOLUTIONS:
1) Sicovit Yellow 10 E172:
The test items were suspended or dissolved in the vehicle to the appropriate concentration freshly on the administration day and were administered orally by gavage at a constant volume (adminsitration volume: 10 mL/kg bw). The application formulations were continuously agitated by stirring throughout the entire administration procedure.

2) Reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%)
Prior to administration, the reference item and the appropriate vehicle were heated to 70°C and stirred at 50°C for approx. 3 hours until the reference item was completely dissolved. This clear solution was maintained at room temperature until administration. The status as clear solution was monitored and recorded upon administration. Immediately after formulation preparation for the females, the formulations were protected from light by transferring the formulation into brown containers or wrapping in aluminium foil.

The amounts of the test and reference items were adjusted to the animal's current body weight on the administration day.

Administration volume (oral administration / intravenous administration): 10 mL/kg bw/day

Injection speed (intravenous adminsitration): dose per approx. 15 seconds
Duration and frequency of treatment / exposure:: single administration
Dose / conc.:: 1 000 mg/kg bw/day (actual dose received)
No. of animals per sex per dose / concentration:: 5 males / 5 females
Control animals:: yes, concurrent vehicle
Positive control reference chemical:: none
Details on study design:: - Dose selection rationale: the dose levels for this study were selected after consultation with the sponsor based on available toxicity and bioavailability data (as far as available):

1) Reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%):
The oral LD50 value for iron citrate monohydrate was stated as being >2000 mg/kg bw; the oral bioavailability of soluble Fe substances are given in the public domain with 1 to 26% (Fe).

For the test item oral dosing of 1000 mg/kg bw, a very low relative bioavailability was assumed (<1%), considering the very low water solubility and bioacessibility in gastric juice. Since the four iron oxide test items have Fe-contents of approx. 70%, the dose of the reference substance should be adjusted accordingly. Given a test item dose of 1000 mg/kg b.w. (corresponding to 700 mg Fe/kg bw), then 1% of this dose would correspond to 7 mg Fe/kg bw (or 36.8 mg/kg bw iron citrate). Correcting for approx. 20% oral bioavailability of soluble iron substances, this yields a dose for the reference item of 7.4 mg iron citrate/kg bw to be given by intravenous injection.

2) Sicovit Yellow 10 E172:
The test item oral doses of 1000 mg/kg bw correspond to the limit dose used in a separate 90-day oral toxicity study, which was considered the maximum feasible dose. This dose was also selected in view of the anticipated low bioavailability and the requirements of analytical sensitivity of the analytical method for iron in plasma.

3) Vehicle control group:
In view of the long established circadian variation of plasma iron levels (Lynch et al, 1973)*, a vehicle control group was sampled for blood plasma over a period of 24 hours at identical sampling time points and intervals as the dosed groups.

*Reference:
Lynch et al (1973): Circadian Variation in Plasma Iron Concentration and Reticuloendothelial Iron Release in the Rat, Clinical Science and Molecular Medicine (1973) 45, 331-336.
Details on dosing and sampling:: TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled: plasma
- Time and frequency of sampling:
blood was collected 0 (predose), 0.5, 1, 2, 4, 8, 12, 24, 48 (test item and reference item only), and 72 hours (test item and reference item only) after administration. The whole blood samples were cooled using an IsoTherm-Rack system until centrifugation. Immediately after centrifugation, the isolated plasma was frozen at -20°C ± 10 % and stored at this temperature until analysis.

Pharmacokinetic evaluation of plasma data was performed and a non-compartment model was employed. The following parameters were determined, if possible:
AUC0-∞ = extrapolated area from zero to infinity
AUC0-t last = extrapolated area from time zero to the last quantifiable plasma concentration (i.e. >lower limit of quantification, LLOQ)
Kel = elimination rate constant
t1/2 = elimination half-life

Cmax values were the highest measured plasma concentrations and tmax values were the time points of highest plasma concentrations.

Elimination rate constants (Kel) and plasma elimination half-lives (t½) were calculated by linear regression analysis of the log/linear portion of the individual plasma concentration-time curves (c = concentration, t = time).

Area under the curve (AUC) values were calculated using the linear trapezoidal method and extrapolated to infinite time by dividing the last measurable plasma concentration by the elimination rate constant. Plasma concentrations at time zero were taken to be those at the first blood sampling time.

Furthermore, the AUC0-t last was calculated according to the linear trapezoidal rule. Values below the limit of quantification (LOQ) were excluded from calculation.

In addition, the bioavailability was calculated for the mixture.

For plasma, a pre-treatment by a microwave digestion with HNO3 was necessary to digest the proteins in plasma. Afterwards iron in digested samples was measured by ICP-OES.

OBSERVATIONS
- clinical signs: before and after dosing as well as regularly throughout the working day (7.30 a.m. to 4.30 p.m.) and on Saturdays and Sundays (8.00 a.m. to 12.00 noon; final check at approx. 4.00 p.m).
Special attention was paid to the local tolerance at the injection site(s).
- mortality/morbund: early in the morning and again in the afternoon of each working day as well as on Saturdays and Sundays (final check at approx. 4.00 p.m).
- body weight: at the time of group allocation, before dosing for dose adjustment and on test day 4 before the last blood sampling.

ADMINISTRATION FORMULATION ANALYSIS:
For each test item, that was mixed with the vehicle and the reference substance, tests by appropriate analytical methods were conducted to determine the concentration and stability of the test item in the formulations. For the analysis of the application formulations, one sample of exactly 10 mL from each dosing suspension (test items) or dosing solution (reference item) was taken at the start of the administration (test day 1 of the female animals) and frozen until analysis.

Application solutions of the iron oxide was measured after addition of aqua regia to the samples and after an incubation time for at least four days by ICP-OES. After this measurement the remaining precipitation (only iron oxide application solution) were digested by a microwave procedure and measured by ICP-OES.

ANALYTIC OF REFERENCE ITEM:
The iron content of the reference item (Iron (III) citrate tribasic monohydrate; Fe content: 21.2%) was determined using ICP-OES.
Statistics:: The test item group was compared to the reference group. The following statistical method was used:
- Student's t-test (body weight (at p≤0.05 and p≤0.01; limits used p = 0.05 approx. t = 2.306 p = 0.01 approx. t = 3.355 (for 8 degrees of freedom))
Preliminary studies:: none
Details on absorption:: not specified
Details on distribution in tissues:: not specified
Details on excretion:: not specified
Toxicokinetic parameters:: other: bioavailability
Remarks:: An absolute bioavailability of 0.26%/0.21% (m/f) for Sicovit Yellow was calculated for Fe following oral administration compared to intravenous administration.
Toxicokinetic parameters:: other:
Remarks:: It should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
Toxicokinetic parameters:: other:
Remarks:: The plasma iron levels of the dosed group falls within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels.
Toxicokinetic parameters:: other:
Remarks:: The calculated absolute bioavailabiliy derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.
Metabolites identified:: not measured
Details on metabolites:: not measured
Bioaccessibility (or Bioavailability) testing results:: An absolute bioavailability of 0.26%/0.21% (m/f) for Sicovit Yellow was calculated for iron following oral administration compared to intravenous administration. However, it should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
The plasma iron levels of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels. Hence, the calculated absolute bioavailabilities derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.

Please also refer for results to the field "Attached background information" below.
Conclusions:: An absolute bioavailability of 0.26%/0.21% (m/f) for Sicovit Yellow was calculated for iron following oral administration compared to intravenous administration. However, it should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control.
The plasma iron levels of the dosed group falls within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels. Hence, the calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.

Reference 20

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: supporting study
Reliability:: 3 (not reliable)
Rationale for reliability incl. deficiencies:: significant methodological deficiencies
Remarks:: The test material was insufficinetly characterised (no purity and particle size). The iron content was determined in only part of the liver, after the organ was obtained after 8, 16 and 24 weeks of age during an open liver biopsy on the living offspring. The values are not representative because the iron content was never measured in the entire liver. In addition, the iron content was not measured before the animals were 8 weeks old (after birth). Furthermore, the amount of feed intake was not recorded, which means that the relative iron absorption in the liver as a function of the amount of iron consumed cannot be calculated.
Objective of study:: other: determination of Fe content in the liver after feeding with carbonyl Fe
Qualifier:: no guideline followed
Principles of method if other than guideline:: 4 pregnant Porton rats were assigned to receive a diet supplemented with 0, 0.5, 1 or 2 % by weight of carbonyl iron. The basal diet contained 66 mg/kg iron. The allocated diets were commenced 2 days after the rats gave birth to initiate iron supplementation to the offspring via breast milk. When the offspring were 3 weeks old, they were separated from the mothers. Male and female offspring were caged separately and continued to receive the assigned supplemented or normal diet until they were 32 weeks of age. Twenty female and 18 male offspring were included in the study. The offspring were weighed weekly. For assessment of hepatic iron loading up to 3 rats from each subgroup had open liver biopsies under anaesthesia at 8, 16 and 24 weeks of age, and liver biopsies were performed on all rats at 32 weeks of age.
GLP compliance:: not specified
Remarks:: in publication.
Radiolabelling:: no
Species:: rat
Strain:: other: Porton
Sex:: male/female
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Age at study initiation: offspring
- Individual metabolism cages: male and female rats were caged seperately
Route of administration:: oral: feed
Vehicle:: unchanged (no vehicle)
Details on exposure:: PREPARATION OF DOSING SOLUTIONS:
Two days after birth of the experimental animals, the mothers were fed with a diet supplemented with 0, 0.5, 1 or 2 % (w/w) carbonyl Fe to initiate Fe supplementation via breastfeeding. The basal diet contained 66 mg Fe/kg feed. After reaching 3 weeks of age, animals were seperated from their mothers and they continued the carbonyl Fe supplemented studies.

DIET PREPARATION:
The supplemented diets were prepared by slightly dampening the rat food pellets (Joint Stock Ration, Ridley Agriproducts, South Australia) with a fine mist of water, adding carbonyl Fe and shaking in a sealed container until the Fe formed a relatively uniform coating over the pellets. The basal diet contained 66 mg/kg iron.
Duration and frequency of treatment / exposure:: via breastfeeding: 19 days; via direct feed: 29 weeks
Dose / conc.:: 0.5
Remarks:: % iron supplementation levels (w/w in diet)
Dose / conc.:: 1
Remarks:: % iron supplementation levels (w/w in diet)
Dose / conc.:: 2
Remarks:: % iron supplementation levels (w/w in diet)
No. of animals per sex per dose / concentration:: males/females: 5/6 (0% w/w carconyl Fe diet), 3/4 (0.5% w/w carconyl Fe diet), 5/4 (1% w/w carconyl Fe diet) and 5/6 (2% w/w carconyl Fe diet)
Control animals:: yes
Positive control reference chemical:: no
Details on dosing and sampling:: TOXICOKINETIC (Absorption)
- Tissues sampled: liver tissue
- Time and frequency of sampling: at 8, 16 and 24 weeks of age (at open liver biopsies) and at 32 weeks of age.
- No. of animals examined: up to 3 rats/group
- weekly body weights of offspring.
- Procedure: each biopsy specimen was divided in two. (a) One half was used for determination of hepatic iron content. This tissue was dried for 24 h at 105°C, weighed, digested in 50 % nitric acid at 70°C for 1 h, and diluted in 0.2 mo/L sodium acetate buffer pH4.5. Iron concentration was determined using a Roche Unimate 5 Iron Kit (Roche Diagnostic Systems), on a Roche Cobas Bio Centrifugal Analyser. (b) The other half of the biopsy specimen was fixed in formalin, processed, embedded in paraffin, and 4 µm sections were cut. These sections were stained with haematoxylin and eosin and Perls' method for iron. The degree of iron loading was assessed histologically in coded sections and graded on a scale from 0 to IV (according to Scheuer et al. (1962)).*

*References:
- Scheuer PJ, Williams R, Muir AR. (1962): Hepatic pathology in relatives of patients with haemochromatosis. J. Pathol. Bacteriol., 84: 53-64.
Statistics:: - Analysis of variance by 2*factor (sex*iron dose): liver Fe content and rat body weight were converted to logarithms for stabilization of variance.
- Polynomial constrants for the factor representing Fe dose level
- Spearman rank correlation coefficient (between Fe content and graded scores)
Preliminary studies:: no
Type:: absorption
Details on absorption:: not examined
Details on distribution in tissues:: The hepatic Fe content is depicted in Fig. 1 & 2 (see attachment).
Details on excretion:: not examined
Metabolites identified:: not specified
Details on metabolites:: not examined
Enzymatic activity measured:: not specified
Bioaccessibility (or Bioavailability) testing results:: - All groups had increased hepatic Fe content, that depended on the dose level of supplemented Fe. At 32 weeks of age, a levelling off of hepatic Fe was observed, when supplemented Fe exceeded 1% (Fig.2, see attachment). Fe content of male rats tended to reach a plateu after 8 -16 weeks, whilst that of female rats continued to rise throughout the experimental period.

- The graded scores were highly correlated with hepatic Fe content.

- The body weights of rats after 32 weeks of Fe supplemented diet are shown in Fig.4 (see attachment). At the highest dose level, the weight of female and male rats was lower by an average of 14% and 19%, respectively, in comparison to the controls.
Conclusions:: 20 female and 18 male Porton rats were exposed to carbonyl Fe, via breastfeeding (19 days) and sucessively via feed (29 weeks). The doses 0, 0.5, 1 and 2 % (w/w) were supplemented to the feed of the mothers. After weaning the offspring continued to receive the assigned diet until the age of 32 weeks.
Liver biopsies examinations revealed that the hepatic Fe content and loading, as well as the depression of growth were depended on the dose of Fe supplemented in the diet.

Interpretation of results (migrated information): other: It is apparent that Fe overload leads to its accumulation in the hepatocytes of the rats, in a dose-dependent manner.
The hepatic Fe content depended on the dose of Fe supplemented in the diet.

This study has methodological deficiencies in the study design. The test material was insufficinetly characterised (no purity and particle size). To determine the iron content, only part of the liver was obtained after 8, 16 and 24 weeks of age during an open liver biopsy on the living offspring. Even at the end of the study (32 weeks of age), again only part of the liver was used for determination of iron content and the other part for histopathological examination. The values are not representative because the iron content was never measured in the entire liver. In addition, the iron content was not measured before the animals were 8 weeks old. Furthermore, the amount of feed intake was not recorded, which means that the relative iron absorption in the liver as a function of the amount of iron consumed cannot be calculated.

Reference 21

Endpoint:: basic toxicokinetics in vivo
Type of information:: experimental study
Adequacy of study:: weight of evidence
Study period:: Not specified
Reliability:: 2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:: study well documented, meets generally accepted scientific principles, acceptable for assessment
Remarks:: However, the test material was insufficiently characterised (no purity). Furthermore, the relative iron absorption in the liver as a function of the amount of iron consumed cannot be calculated, since the food consumption of the rats was not recorded during the study.
Objective of study:: absorption; distribution
Qualifier:: no guideline followed
Principles of method if other than guideline:: The acute oral toxicity of carbonyl iron was examined in male rats at doses of 40 and 50 g/kg bw. In addition, a time-response study was performed of serum and liver non-haem Fe in rats after oral administration of 30 g/kg bw of carbonyl iron. These results represent pertinent information for the endpoint toxicokinetics. Together with carbonyl iron, FeSO4 and NaFeEDTA were investigated.
GLP compliance:: not specified
Remarks:: in the publication.
Specific details on test material used for the study:: Not specified
Radiolabelling:: no
Species:: rat
Strain:: Sprague-Dawley
Sex:: male
Details on test animals or test system and environmental conditions:: TEST ANIMALS
- Source: National Center for Toxicological Research (NCTR)
- Housing: polycarbonate cages
- Diet: ad libitum
- Water: ad libitum
Route of administration:: oral: gavage
Vehicle:: not specified
Details on exposure:: - Administered volume: 2 mL
Duration and frequency of treatment / exposure:: single administration
Dose / conc.:: 30 000 other: mg/kg bw
Dose / conc.:: 40 000 other: mg/kg bw
Dose / conc.:: 50 000 other: mg/kg bw
No. of animals per sex per dose / concentration:: 38 (time-response study, treated with 30 g/kg bw), 16 rats (8/group at acute toxicity study, treated with 40 and 50 g/kg bw), 12 control animals (time-reponse study), and 8 control animals (acute toxicity study).
Control animals:: yes
Positive control reference chemical:: In a strict sense, there was no positive control. However, carbonyl iron was investigated concurrently with FeSO4 and NaFeEDTA. In particular the first compound can be regarded as a positive control if the absorption of iron is concerned. The iron in the salt does not have to be converted to an ionic form to allow absorption. This conversion is a prerequisite for the absorption of the iron in carbonyl iron.
Details on study design:: - Dose selection rationale: the doses used were based on the LD50 levels reported in the literature (Boyd et al. (1963); Hoppe et al. (1955); Shelanski (1950); Weaver et al. (1961)).*

*References:
- Boyd, E. M., and Shanas, M. N. (1963): The acute oral toxicity of reduced iron. Can. Med. Assoc. J. 89, 171–175.
- Hoppe, J. O., Marcelli, G. M. A., and Tainter, M. L. (1955): A review of the toxicity of iron compounds. Prog. Med. Sci. 230, 558–571.
- Shelanski, H. A. (1950): Acute and chronic toxicity tests on carbonyl iron powder. Bull. Natl. Formulary Committee 18, 87–94.
- Weaver, L. C., Gardier, R.W., Robinson, V. B., and Bunde, C. A. (1961): Comparative toxicology of iron compounds. Am. J. Med. Sci. 241, 296–302.
Details on dosing and sampling:: TOXICOKINETIC STUDY (Absorption, distribution)
- Tissues and body fluids sampled: blood serum, liver tissue
- Time and frequency of sampling: prior to treatment for control animals and for test animals at half-hour intervals for 3–4 h after treatment. Additionally, samples were collected at 24 h and one at 48 h. The 40 and 50 g/kg bw doses were used for the acute toxicity test, and determination of serum Fe and liver nonheme Fe levels.

ANALYTICAL METHOD
- Iron concentration in serum:
serum Fe concentrations were measured with an electrochemical technique (Ferrochem II analyzer). Serum samples were analysed in duplicate according to the method of Skikne, B.S. (1987).* The Ferrochem II was calibrated using a 200 µg% copper standard to null the copper response and a 100 µg% iron standard. Serum samples were analysed in duplicate by injecting a 25-µL aliquot into the cell reagent according to the method of Skikne, B.S. (1987).*
- Non-heme iron in liver:
it was determined by the bathophenanthroline reaction and expressed as µg Fe/g liver or total µg of nonheme iron. Approx. 1 g of each previously blotted liver was weighed and placed in a 50-mL polypropylene centrifuge tube, and distilled water was added to bring the volume to 15 mL. The tissue was homogenized for 30 sec with a Polytron. A 3-mL sample of the homogenate was transferred to another 50-mL centrifuge tube, and 10 mL of acid reagent was added (6 M HCl & 1.2 M trichloroacetic acid) (1:1, v/v), mixed and heated in an oven at 65 °C for 20 h, cooled, and centrifuged at 3600 x g for 20 min. Duplicate 0.2-mL samples of the supernatant fraction were pipetted into small glass tubes, and 1.8 mL of freshly prepared colour reagent was added, mixed, and incubated for 10 min at rt. Absorbance was determined spectrophotometrically at 535 nm, and iron concentration (μg Fe/mL) was determined by reference to a standard curve. The bathophenanthroline colour reagent (protected from light) was prepared by dissolving 62.5 mg bathophenanthroline disulfonic acid and 0.25 mL thioglycolic acid in distilled water and diluted to 25 mL. The final colour reagent was a solution of the bathophenanthroline colour reagent, saturated sodium acetate (4.5 M), and distilled, deionized water (1:20:20, v/v).

*Reference:
- Skikne, B. S. (1987): A commercial electrochemical method evaluated for measurement of iron status. Clin. Chem. 33, 1645–1647.
Statistics:: Analysis of variance and Scheffe multiple comparison method, values expressed as means with their standard errors.
Preliminary studies:: not specified
Type:: other: absorption & distribution
Details on absorption:: No mortality was observed. The results obtained regarding the serum levels of Fe are shown in Table 1 (below). The time-response after administration of the 30 g/kg bw is depicted in Fig.3 (see attachment): Fe levels peaked at 0.5 h (787 µg/l) and then gradually decreased up to 4 h. Measurements at 24 and 48 h indicate that the serum Fe level remained constant.
Details on distribution in tissues:: Liver non-haem iron levels peaked at 1.5 h and then fluctuated between 56 and 90 µg/g over the 48h period ( Fig.3, see attachment), after administration of the 30 g/kg bw of carbonyl Fe. Liver non-haem iron levels were increased significantly after administration of 40 and 50 g/kg bw of carbonyl Fe in comparison to the controls
Details on excretion:: not specified
Test no.:: #1
Toxicokinetic parameters:: Tmax: 0.5 h (Fe levels in blood serum after administration of 30 mg/kg bw of carbonyl Fe)
Test no.:: #1
Toxicokinetic parameters:: Cmax: 787 µg/l (Fe levels in blood serum after administration of 30 mg/kg bw of carbonyl Fe)
Test no.:: #1
Toxicokinetic parameters:: Tmax: 1.5 h (non-haem Fe in the liver, after administration of 30 mg/kg bw of carbonyl Fe)
Test no.:: #1
Toxicokinetic parameters:: Cmax: 165 µg/l (non-haem Fe in the liver, after administration of 30 mg/kg bw of carbonyl Fe); estimated from the figure provided
Metabolites identified:: not specified
Details on metabolites:: not specified
Enzymatic activity measured:: not specified
Bioaccessibility (or Bioavailability) testing results:: not specified
Conclusions:: A time-response study was performed of serum and liver nonheme Fe in rats after oral administration of 30 g/kg bw of carbonyl iron. In addition, the acute oral toxicity of carbonyl iron was examined in male rats at doses of 40 and 50 g/kg bw. Together with carbonyl iron, FeSO4 and NaFeEDTA were investigated (tested in different doses). After a dose of 30 g/kg bw, Fe levels peaked at 0.5 h (787 µg/dl) and then gradually decreased up to 4 h. Measurements at 24 and 48 h indicate that the serum Fe level remained constant. Liver nonheme Fe levels (µg/g liver weight) increased significantly in comparison to the controls, which indicates that iron was stored in the liver. Liver non-heme iron reached the maximum at 1.5 h, and then fluctuated between 56 and 90 μg/g over the 48-h period. At the dose of 40 g/kg bw carbonyl iron, the iron level of 406 µg/dl in serum and 448 µg/g in liver were reached, and the rats had an iron level of 358 µg/dl in serum and 407 µg/g in liver at the dose of 50 g/kg bw.

Interpretation of results: Iron overload due to consumption of carbonyl Fe, may result in its accumulation in the liver; however, large Fe particles are not expected to be solubilized, and hence, become systemically available.
The study allows for the comparison of the systemic bioavailability of iron after oral treatment with carbonyl iron, FeSO4 and NaFeEDTA. The lack of acute toxicity observed for carbonyl iron in comparison to the two other compounds could be the result of its toxicokinetic behaviour.

The study is well-documented. However, the test material was insufficiently characterised (no purity). Furthermore, the relative iron absorption in the liver as a function of the amount of iron consumed cannot be calculated, since the food consumption of the rats was not recorded during the study.

Description of key information

No substantial accumulation of iron was observed in blood plasma following oral administration of iron oxides (Fe2O3, FeOOH or Fe3O4).

Key value for chemical safety assessment

Bioaccumulation potential:: no bioaccumulation potential

Additional information

Endpoint summary – Toxicokinetic

1 Introduction

Information taken from EFSA (2004): Opinion of the Scientific Panel on Dietetic Products, Nutrition and Allergies on a request from the Commission related to the Tolerable Upper Intake Level of Iron, The EFSA Journal, 125, 1-34. The references cited in this introduction are not included in the dossier for the sake of brevity.

“Iron is an essential trace element that has important metabolic functions, including oxygen transport and storage and many redox reactions. It is present in biological systems in one of two oxidation states, and redox interconversions of the ferrous (Fe2+) and ferric (Fe3+) forms are central to the biological properties of this mineral. Iron is an essential constituent of oxygen carriers, such as haemoglobin and myoglobin, and the iron contained within haem is essential for the redox reactions of numerous cytochromes.

The adult human body contains 2.2 - 3.8 g iron under iron-adequate conditions (Lynch, 1984). Homeostatic mechanisms have evolved that can alter intestinal iron absorption and supply iron preferentially to functional compartments in response to deficiency or excess.”

1.1 Absorption and regulation of absorption

Tissue concentrations and body stores of iron are controlled at three different levels:

Luminal iron: the extent of uptake of iron by the cells of the gastrointestinal tract affects how much remains unabsorbed and passes to the lower bowel, prior to elimination in faeces;

Mucosal iron: the mucosa is the main site of regulation of iron uptake in relation to liver stores and ferritin levels;

Post-mucosal iron: relates to the impact of iron intake on iron status and body stores.

Luminal iron

Non-haem iron is present in foods largely as salts, which become soluble at acidic pH in the stomach, and absorption from foods depends on its dissolution as ferric salts and subsequent reduction to the ferrous form. Any metallic iron in the diet is probably absorbed as non-haem iron following its dissolution in the acid stomach contents. The absorption of non-haem iron can be increased substantially by the presence of ligands, such as ascorbate, citrate and fumarate, as well as the presence of amino acids (e.g. cysteine) and oligopeptides resulting from meat digestion (Mulvihill et al., 1998). In contrast, very stable complexes, for example with phytates, phosphates and oxalates, impair non-haem iron absorption. Depending on the concentration of supportive or inhibitory ligands in the intestinal lumen, the absorption of non-haem iron can vary by a factor of 10 in single-meal studies, but the effects are less pronounced in more long-term studies (Hallberg and Rossander, 1984; Rossander, 1987; Hunt and Roughead 2000).

Mucosal iron

In addition, homeostatic regulation will influence the extent of non-haem iron absorption. Body iron content is linked to demand by regulated intestinal non-haem iron absorption which, in turn, is regulated to a major extent by the uptake of iron into the cells of the intestinal mucosa (Schümann et al., 1999a and b). This step is mediated by the divalent metal transporter (DMT-1) (Gunshin et al., 1997). The activity of DMT-1 decreases after a period of high oral iron intake (Oates et al., 2000), and such down-regulation of iron uptake may be regarded as protection against iron overload. Transfer of information on hepatic iron stores to the gut may be mediated via an acute phase hepatic protein, hepcidin (Nicolas et al., 2001), or possibly the pro-hormone form pro-hepcidin (Kulaksiz et al., 2004), which influences the extent of iron absorption (Leong and Lonnerdal, 2004) and can result in a mucosal block. Due to the various exogenous and endogenous factors affecting iron absorption, a clear relationship is generally not found between total iron intake and iron status. Total daily iron absorption is about 0.9 mg in males and 0.5 and 0.6 mg higher in menstruating women and blood donors. The mucosal barrier can be overwhelmed by high iron doses, such as occurs in acute iron intoxication (Ellenham and Barceloux, 1988).

Post-mucosal iron

The concentration of free iron in blood is extremely low due to the high affinity binding of iron to transferrin. Transferrin is an 80 kDa plasma protein that binds two Fe3+ ions per molecule with high affinity (binding constant: 10-30). The total iron binding capacity (TIBC) of transferrin in the plasma of healthy adults is approximately 56μmol/L. About 30% and 10% of TIBC is occupied in normal iron status and in iron deficiency respectively. The plasma iron pool is approximately 3 mg, but its daily turnover is more than 30 mg/day (Bothwell et al., 1979). Iron is transferred from the plasma to the tissues via binding of iron-loaded transferrin to transferrin receptors (TfR) at the cell surface, which are subsequently internalised within endocytotic vesicles. High serum TfR values indicate high activity of cellular iron uptake, which in most cases is due to high erythropoietic iron demand. Transferrin releases its iron under acidic conditions in the endocytotic vesicles, and the transferrin and TfR return subsequently to the cell surface. Within cells, iron is preferentially distributed to iron-dependent enzymes and iron-binding proteins. Excessive intracellular iron is sequestered in the storage protein ferritin (500 kDa) or in its degradation product called haemosiderin. One ferritin molecule can store up to 4500 iron atoms as ferric hydroxyphosphate micelles. Such iron can be remobilised and utilised on demand. The serum ferritin concentration is an indication of body iron stores, except when high levels arise due to inflammation (Ponka et al., 1998). Homeostatic mechanisms involving cytosolic iron-regulatory proteins (IRPs) have developed to maintain low concentrations of free iron, in order to provide iron for essential functions and to protect the cells from oxidative damage. IRP-1 is activated by cellular iron-deficiency, but also by reactive oxygen species (ROS). At normal cellular levels of free iron, IRP-1 contains a 4Fe-4S cluster but in iron-deficient cells the 4Fe-4S cluster is lost and the molecule binds to specific base loops of mRNAs for DMT-1, TfR and ferritin, called “iron-response elements”. In deficiency IRP binding to TfR-mRNA increases TfR expression, while the ferritin mRNA is broken down more rapidly, which restores the intracellular free iron concentration by a combination of increased uptake and decreased sequestration.

Under physiological conditions iron status is almost exclusively regulated by adaptation of intestinal iron absorption according to the demand. In healthy iron-replete humans with iron stores of 800-1200 mg, non-haem iron absorption increases when iron stores are depleted and vice versa (Finch, 1994). This mode of regulation is highly predictable in normal subjects, and iron stores as determined by plasma ferritin values have been used to predict iron absorption (Magnusson et al., 1981). Duodenal non-haem iron absorption is linked to body iron status via the supply of serum iron to intestinal crypt cells. At adequate supply levels, the activity of IRPs in these cells is low, whereas in iron-deficiency IRP activity in the enterocytes is increased and intestinal iron absorption is high. Iron absorption adapts to changes in plasma iron concentration with a lag time of 48h, which corresponds to the time required for young enterocytes to express enzymes and transport proteins, such as DMT-1 and iron regulated gene (IREG), and to migrate up the villi to the site of absorption (Schümann et al., 1999b).

In anaemia, an undefined and unidentified “erythrocyte regulator” can increase iron absorption to 20-40 mg Fe/day from oral iron preparations (Finch, 1994). In addition, iron absorption is increased by an unknown mechanism to up to 66% in pregnant women and returns to normal at 16-24 weeks after delivery (Barrett et al., 1994). Iron absorption is also increased during lactation and growth.

Regulation of non-haem iron absorption protects from iron overload at normal and moderately increased iron-intake levels. A group of 12 Swedish male blood donors and 19 non-donors received standard meals with radioactively labelled non-haem iron (12 mg Fe/day) and haem iron (2 mg/day). Total iron absorption increased when serum ferritin concentrations were less than 60 μg/L, but decreased when serum ferritin exceeded this level, and the homeostatic equilibrium point of the system was estimated to be 60 μg ferritin/L (Hallberg et al., 1997). This study did not investigate the maximum regulatory capacity by the administration of high levels of additional iron, but such data are available from fortification studies. Fortification of curry powder with sodium-iron-EDTA giving additional intakes of 7.5 mg Fe/day for 2 years in iron-replete male subjects (Ballot et al., 1989) did not increase serum ferritin levels. There were no changes in iron stores as estimated by serum ferritin following the addition of 10 mg Fe/day as ferrous sulphate to the food of a healthy male subject for 500 days, or as determined via serial phlebotomies after the end of the iron substitution period (Sayers et al., 1994). In these two studies serum ferritin levels did not change despite long-term challenge with 7.5-10 mg Fe/day in addition to normal dietary iron intake of approx. 10 mg Fe/day. Iron status in the elderly, as indicated by elevated serum ferritin levels, was increased after intake of an additional 30 mg Fe/day as supplements (Fleming et al., 2002), in a study in which subjects with abnormal results for blood leukocytes, C-reactive protein (CRP), and 3 liver enzymes, as indicators of inflammation and liver diseases, were excluded. Theoretical calculations of the accumulation of iron in a fertile woman given different daily intakes (Borch-Iohnsen and Petersson Grawe, 1995) indicated that a daily intake of 60 mg for 5 years would lead to a serum ferritin value close to that seen in iron overload.

1.2 Elimination

Iron excretion via the kidneys is very low, and body iron is highly conserved. Renal elimination is not controlled as part of iron homeostasis or the control of excess body stores. Normally, only about 0.1 mg is lost daily in urine. The sloughing of mucosal enterocytes results in elimination of absorbed iron before it reaches the systemic circulation and accounts for the loss of 0.6mg per day into the intestinal lumen. About 0.2-0.3 mg is lost daily from the skin. The total daily loss is equivalent to about 0.05 % of body iron content (Green et al., 1968). Menstrual losses are variable and may be almost as high as the total loss in non-menstruating women (FNB, 2001).(EFSA, 2004)

1.3 Requirements and recommended intakes

The recommended daily intakes for different groups of the population are based on the amount of ingested iron necessary for absorption of the estimated average amounts of iron lost each day. During the first year of life the body requires approximately 260 mg of iron for metabolism and growth, i.e. 0.6-0.8 mg Fe/day, which corresponds to a dietary intake of 6-8 mg Fe/day, assuming 10% absorption. These data are the rationale for the recommendation of 1 mg Fe/kg body weight per day for children between the 4th month and the 3rd year of life (Oski, 1993). The Scientific Committee on Food recommended daily intakes of 6 mg and 4 mg for infants aged 0.5-1 year and 1-3 years respectively, assuming 15% absorption of the daily intake (SCF, 1993). An adult male loses approx. 1 mg Fe/day, mostly from the intestine (Green et al., 1968), and a daily intake of approx. 10 mg Fe is needed to replace these basal losses, assuming 10% absorption, and the recommended dietary iron intake has been estimated as between 8 and 10 mg Fe/day (SCF, 1993; FNB, 2001; Arbeitsgruppe “Referenzwerte für Nährstoffzufuhr, 2000). Menstrual iron losses are below 1.6 mg Fe/day in 95% of women, which leads to an average total loss of approx. 2.5 mg Fe/day (Baynes and Bothwell, 1990). Assuming 10-20% absorption in iron deficiency, an intake of 15-20 mg Fe/day is recommended for women of reproductive age (SCF, 1993; FNB, 2001; Arbeitsgruppe “Referenzwerte für Nährstoffzufuhr, 2000). During pregnancy, 450 mg Fe is needed to allow increased erythropoiesis, while 270-300 mg and 50-90 mg are transferred to the foetus and placenta, which gives a total extra demand of 770-840 mg. This demand corresponds to approx. 3 mg Fe/day and will be provided by an intake of 30 mg Fe/day and is the rationale for the recommended higher iron intake in pregnancy (FNB, 2001; Arbeitsgruppe “Referenzwerte für Nährstoffzufuhr, 2000). (EFSA, 2004)

The following information is taken into account for any hazard / risk assessment:

Bioaccumulation potential is not relevant since iron is an essential element in human nutrition.

Theoretical calculations of the accumulation of iron in a fertile woman given different daily intakes (Borch-Iohnsen and Petersson Grawe, 1995) indicated that a daily intake of 60 mg for 5 years would lead to a serum ferritin value close to that seen in iron overload. (EFSA, 2004)

2 Experimental data

2.1 Water solubility

In the study by Blust et al. (2010), the water solubility of iron based on Fe(III) was determined mathematically by the MINTEQ chemical equilibrium model. Two standard, unbuffed freshwater toxicity testing media were used as the matrices for the studies. Metal titrations were performed with each model. Open atmosphere (CO2 exchange, or CO2 and O2 exchange in case of PHREEQC) simulations were carried out at pH 6, 7, and 8 at 25 °C. A water solubility of 0.015, 0.0024 and 0.00001 mg/L was calculated at a pH of 6, 7 and 8. It emphasise the high insolubility of iron (III). Solubility decreases dramatically when going from neutral (pH 6) to more basic (pH 8) waters. If the modelled solubility limits are exceeded rapid precipitation would be expected.

The water solubility on pure iron powder (< 250 µm) was investigated by Skeaff et al. (2006) in transformation/dissolution tests in duplicate over 7 days in an aqueous medium with a loading of 100 mg/L at pH 6. In the 100 mg/L loadings of atomised iron, the onset of iron hydrolysis occurred about 2 hours into each T/D test. The orange ferric hydroxide precipitate formed, and the solution exhibited a slight orange hue thereafter. The values of Fe(aq) present a classic pulse formation-decay behaviour comprising the initial increase in Fe(aq) to maxima at 24 h, followed by a declined over the subsequent 144 h. For the two tests the average 24 h maximum was at 1840 µg/L. By 168 h the value of Fe(aq) had declined to an overall average of 19.4 µg/L. These data are typical of the previously reported rapid oxidation of metallic iron to form Fe(II), the subsequent slower oxidation of Fe(II) to Fe(III), and finally the precipitation of Fe(OH)3.

A water solubility of less than 0.1 mg Fe/L for elemental iron was also cited in the literature (SOLLAC (1994), Weast (1976), GESTIS (2006) and Prager (1998)). Iron does not dissolve in water or dilute in aqueous solutions of electrolytes, it is oxidized by (aerated) water but forms mainly insoluble oxides. Iron powder is insoluble in both cold and hot water (temperature and pH of aqueous medium are unknown), and also in alkali medium, but it is soluble in acids (Weast (1979) and Prager (1998)).

Summary

Iron powder is insoluble in water at around neutral pH (< 0.1 mg/L). Metallic iron oxidizes to form Fe(II), then Fe(II) slowly oxidizes to Fe(III) which finally precipitates in Fe(OH)3 (i.e. rust, red iron oxide).

Chemical equilibrium modelling was carried out to calculate a solubility limit in water for elemental iron. Solubility limits of 0.015, 0.0024 and 0.00001 mg/L were calculated at pH 6, 7, and 8, respectively.

A 7-day transformation/dissolution test was run on iron powder at a mass loading of 100 mg/L and a pH of 6.0. Under the conditions of this test, the dissolution of Fe was 19.4 µg/L after 7 days. This is in agreement with the modelled solubility of 0.015 mg/L.

2.2 In vitro bioaccessibility in artificial body fluids

In the in vitro bioaccessibility study by Knopf (2018) generated at Fraunhofer (Schmallenberg, Germany), the iron dissolution from four different iron compounds was measured when exposed to artificial gastric fluid (GST), phosphate-buffered saline (PBS), artificial lysosomal fluid (ALF) and Gamble's solution (GMB) media. The experiments were performed using the shaking flask method (100 rpm, 37 °C) at a loading of 100 mg/L, three replicates per test material and test medium were used. Nano-sized diiron trioxide only dissolve in quantities above the limit of detection (LOD) in GST (at pH 1.5) and ALF (at pH 4.6), while triiron tetraoxide shows low solubility in GST (at pH 1.5), ALF (at pH 4.6) and GMB (at pH 7.4). Also, nano-sized iron hydroxide oxide showed a very low solubility in GST and ALF media.

Table 1: In vitro bioaccessibility results for iron substance compounds in artificial gastric fluid (GST), phosphate-buffered saline (PBS), artificial lysosomal fluid (ALF) and Gamble's solution (GMB).

Fe in [µg/L]	Fe2O3 (Sikovit Red: 85 nm)	Fe2O3 (Ferroxide Red 212P: 82 nm)	Fe3O4 (122 nm)	FeOOH (50 nm)
Test medium
GST 2 h	41	9.27	296	2.24
GST 24 h	134	57.5	1186	55.9
PBS 2 h	< LOD (0.514)	< LOD/LOQ (1.54)	< LOD (0.5)	< LOD/LOQ (1.54)
PBS 24 h	< LOD (0.761)	< LOD (0.761)	< LOD/LOQ (2.28)	< LOD (0.761)
ALF 2 h	39	16.4	742	6.96
ALF 24 h	90.5	72.1	30824	29
ALF 168 h (7 days)	176	185	75367	80.5
GMB 2 h	Negative value after correction	0.03	0.106	Negative value after correction
GMB 24 h	1.25	Below background	< LOD (0.19)	0.258
GMB 168 h (7 days)	Negative value after correction	Below background	2.16	Negative value after correction

Published in vitro bioaccessibility study by Mörsdorf et al. (2015) with micro sized (100 µm) iron metal powder showed a high solubility of approximately 65 % and 33 % in artificial GST (at pH 1.5) and ALF (at pH 4.5) media with a loading of 100 mg/L after 24 h. Only approx. 1 mg/L was dissolved in artificial sweat (at pH 6.5) medium, and iron metal was insoluble (below the LOD) in GMB (at pH 7.4) or PBS media (at pH 7.3) after 24 h.

Furthermore, Swain et al. (2003) analysed the solubility of six different iron compounds (carbonyl iron, electrolytic and reduced iron forms) in hydrochloric acid solution (at pH 1 and 1.7). The experiments were performed using a shaking flask method (150 rpm, 37 °C) at a loading of 20 mg/250 mL, three replicates per test material were used. Carbonyl Fe was the most soluble and reduced Fe the least soluble at 150 min. The carbonyl iron (> 98 % elemental iron, < 47 µm) showed a solubility of > 95 % in HCl (at pH 1.0 and 1.7) after 150 min. More than 90 % of each of the two electrolytic iron compounds dissolved in the HCl medium at pH 1 after 150 min but vary between 60 – 90 % at pH 1.7. The solubility of the reduced iron (reduced, H-/CO-reduced) compounds in the HCl media after 150 min were 58 – 85 % at pH 1 and 50 – 71 % at pH 1.7.

Also, in another study from Motzok et al. (1978) the in vitro solubility of 7 different micro sized iron compounds (carbonyl iron (assumed to be almost 100 % Fe), electrolytic and H2-/CO-/NH3 reduced iron (assumed to be almost 99 – 100 % Fe) were investigated in HCl solution (0.2 %). Each experiment was carried out with the respective iron compound in a flask with a loading of 100 mg/250 mL HCl solution with stirring at 37 °C. Electrolytic iron and both carbonyl iron compounds showed a high solubility in the HCl solution of ≥ 85 % after 90 min, while the H2- and NH3-reduced iron showed a lower solubility of 40 - 57 %.

In addition, Huebers et al. (1986) investigated the in vitro solubility of ferrous ammonium sulphate, ferric chloride and radioactive carbonyl iron (3 – 4 µm). The respective test substance was incubated for 30 min. at 37 °C in 1 ml normal or neutralised rat gastric juice. The in vitro solubility of the three iron compounds in gastric juice was dependent on pH. A solubility of >90 % at pH 3 and of 77 % at pH 6 was determined for 50 µg ferrous ammonium sulphate in gastric juice. 50 % of ferric chloride dissolved at pH 3, whereas only a solubility of < 1 % at pH 6. Carbonyl iron showed with < 1 % the lowest solubility. At pH 1.6 and a loading of 20000 µg carbonyl iron, 2560 µg dissolved.

Summary:

The iron compounds with the highest iron solubility are ferrous ammonium sulphate and ferric chloride. Ferrous ammonium sulphate (50 µg) is up to 90 % soluble in gastric juice from rats after 30 min. at pH 3 and 77 % at pH 6. Ferric chloride is 50 % soluble in gastric juice at pH 3 and the solubility drops to < 1 % at pH 6.

Elemental iron as carbonyl iron and iron metal shows the highest solubility in artificial body fluids. Carbonyl iron (0.5 – 21 µm) has a solubility of approximately 360 mg/L in hydrochloric acid solution (0.2 %) after 90 min, and iron metal (100 µm) of 65 mg/L and 33 mg/L in artificial GST (at pH 1.5) and ALF media (at pH 4.5) after 24 h. The solubility of iron metal decreases with increasing pH value. It dissolves at pH 4.5 at approx. 1 mg/L in artificial sweat medium and is insoluble at pH 7.3 in GMB or PBS media after 24 h. Also, 2560 µg out of 20000 µg of carbonyl iron dissolves in vitro in gastric juice from the rat at pH 1.6 after 30 min. In addition, electrolytic iron and H2-/CO-reduced iron compounds also have a high solubility in a hydrochloric acid solution (0.2 %) after 90 min. Up to 252 and 288 mg/L of H2- and CO-reduced iron dissolves in the acidic medium, while electrolytic iron is soluble up to 348 mg/L after 90 min.

The iron oxide compounds (Fe2O3 and F3O4) show a very low solubility in artificial body fluids. Nano sized Fe2O3 only dissolve in GST (at pH 1.5) and ALF (at pH 4.6) media with 57 - 134 µg/L and approx. 180 µg/L after 24 h, while Fe3O4 shows low solubility of approx. 1.19 mg Fe/L in GST (after 24 h at pH 1.5), 75.4 mg Fe/L in ALF (after 7 days at pH 4.6) and 2.16 µg Fe/L in GMB medium (after 7 days at pH 8.6) at a loading of 100 mg/L. Also, nano sized iron hydroxide oxide shows a very low solubility in GST (55.9 µg Fe/L; after 24 h at pH 1.5) and ALF (80.5 µg Fe/L; after 7 days at pH 4.6) media. Also, the solubility of iron oxide compounds decreases with increasing pH, eventually becoming insoluble at neutral pH.

Overall, the data available shows that the solubility of iron depends on the iron compound and the receiving medium. From the iron compounds, ferrous ammonium sulphate and ferric chloride have the highest solubility. Elemental iron in the form of carbonyl iron and iron metal is soluble in gastric juice at pH < 2, while the iron oxides have very low solubility in various body fluids (phosphate-buffered saline (PBS), artificial lysosomal fluid (ALF) and Gamble's solution (GMB) medium). In addition, electrolytic iron and H2-/CO-reduced iron compounds have a high solubility in the acidic medium. With increasing pH, the solubility of both elemental iron and iron oxides decreases until they are no longer soluble at neutral pH.

3 Toxicokinetic aspects

3.1 Investigations of cellular uptake of iron compounds in-vitro

Beck-Speier et al. (2009) investigated the bioaccessibility of radiolabelled 59Fe2O3 of 2 different particle sizes (0.5 or 1.5 μm) was determined in vitro by simulating intracellular and extracellular dissolution in cell culture medium (RPMI) with alveolar macrophages (AM) with a loading of 1:1 59Fe2O3 particle/AM. The extracellular dissolved fraction of the incubated 0.5 µm particles was 0.00055 ± 0.00022 and for 1.5 particles was 0.00034 ± 0.00016. In contrast, the intracellular dissolved fractions increased with time. The intracellular particle dissolution (IPD) rates from 0.5 and 1.5 μm 59Fe2O3 particles were 0.0037 ± 0.0014 1/d and 0.0016 ± 0.0012 1/d, respectively. From the particle dissolution kinetics data determined from measures of iron in the culture medium and cell lysate (corrected for external dissolved Fe), more than 70% of the dissolved iron remained within AM (likely iron binding protein-associated), whereas only 30% was released out of the cell into the extracellular medium. After 1 week the intracellular dissolved and retained iron progressively increased to about 3% with small and 1% with large 59Fe2O3 particles. Free extracellular iron uptake was determined by cultured AM, additional incubations were done and the cellular uptake of Fe2+ from medium into AM was measured. Iron chloride (FeCl2) radio-labeled with 59Fe was added varying between 0.1 - 5.0 μg FeCl2/ml or 1*10^6 AM. Extracellular iron Fe2+ at doses between 0.1 - 5 μg 59FeCl2/mL medium (or per 1*10^6 AM) was taken up relatively weakly by AM (24 ± 2.5% of any of the doses provided). The study by Knopf (2018) also addressed the solubility of Fe2O3 in different artificial physiological media (gastric fluid (GST), phosphate-buffered saline (PBS), artificial lysosomal fluid (ALF) and Gamble's solution (GMB)). Of the four media tested, Gamble's solution comes closest in composition to the test cell culture media in the present study. However, due to the phagosomal uptake of the particles the values for the low solubility of radio labelled Fe2O3 from the in vitro study by Beck-Speier et al. (2009) could be compared best with the low in vitro bioaccessibility in (ALF) in the study by Knopf (2018). It is unclear in the present study in which concentration the test substance (59Fe2O3) was present in the RPMI medium. It was only stated that the number of particles in relation to the alveolar macrophages was 1:1 contained in the cell medium. A unit of mass for determining the substance concentration in the RPMI medium is not available in the present study. However, the solubility tests by Knopf (2018) with Fe2O3 in Gamble's solution medium (GMB) at a test item concentration of 100 mg/L after 168 h (corresponds to 7 days) also showed no significant solubility (negative value after correction) of the test item.

3.2 In vivo uptake mechanisms and supporting studies

The mechanism of carbonyl iron (3 - 4 µm) absorption has been studied in rats by Huebers et al. (1986). Solubility by gastric acid was a prerequisite for subsequent absorption. To examine absorption, rats were given normal diet or diet containing 4 - 8 mg iron/kg. Animals were iron loaded by feeding them a 1 % carbonyl iron diet for 4 - 6 months. Several types of absorption studies were performed. In some instances, the radioactive iron preparation was given by gastric intubation, and absorption determined over ensuing days. Alternately, absorption was measured over a 1- to 2-h period in isolated gut loops with intact blood supply. Iron was also injected by needle through the wall of the upper duodenum with or without ligation at the pylorus. At the end of study, the radioactivity of 59Fe in the gastrointestinal tract, liver, spleen, femur, kidneys, and blood was measured after sacrifice for determination of absorption. In addition, haematocrit, plasma iron and total iron-binding capacity were determined in blood samples. In studies of mucosal ferritin, iron was given by gastric intubation; after sacrifice of animal, the upper third of the small intestine was washed, and the mucosa was scraped from the absorptive area of the gut. The mucosa was subjected to immunoelectrophoresis (antibody against liver ferritin and cross-reacted with mucosal ferritin). Furthermore, Solubility of iron was determined after carbonyl iron was either injected into the obstructed stomach so that incubation took place in vivo or was added to gastric juice, removed from the animal, and incubated at 37 °C for 1 h. The mixture of iron in 1 mL gastric juice was also injected into a jejunal loop for determination of absorption. Solubility of iron solutions was determined by radioactivity analysis or by spectrophotometry.

In the study in which the iron mixture in gastric juice was injected into the gut loop of individual animals, the iron absorption of ferrous sulphate, ferric chloride and carbonyl iron from an in vivo jejunal intestinal loop was dependent on solubility and the pH. Ferrous sulphate showed an absorption of 57 % (> 90 % solubility of 50 µg Fe) from a jejunal gut loop of iron-deficient rats after 1 hour at pH 3, and iron absorption was 44 % (77 % solubility) at pH 6. Ferric chloride showed a lower solubility and absorption. At pH 3, an absorption of 28 % (50 % solubility) was observed, whereas at pH 6 absorption was 1.9 % (< 1 % solubility). Carbonyl iron had the lowest solubility (< 1 %) and iron absorption (< 1 %) at pH 3 and 6. At pH 1.6 and a loading of 20000 µg carbonyl iron, 306 µg were absorbed.

In the in vivo studies in which the carbonyl iron was orally administered to rats, the slow rate of solubilisation resulted in a more prolonged absorption, responsible for the low toxicity of carbonyl iron. Large doses of carbonyl iron were held for several days by the gastric mucosa of iron-deficient animals. Once it had been solubilised, the subsequent pathway of absorption by the intestinal mucosa and the amount absorbed was similar to that of ferrous ammonium sulphate. Furthermore, the data of iron absorption in the blood, skeleton, liver and the carcass (minus gut) showed higher absorption in the iron-deficient rats over a period of 4 days in comparison to rats with normal diet after oral administration of a single dose of 200 mg carbonyl iron. The study shows that metallic iron in the form of carbonyl iron particles has to be dissolved in the stomach juice at pH<2 before absorption is possible. Carbonyl iron is only absorbed in the form of iron(II)ions.

3.3 Tissue and blood levels of iron

Huebers et al. (1986) investigated the absorption of ferrous ammonium sulphate and radioactive carbonyl iron in male Sprague-Dawley rats (weighing 200 – 300 mg). To determine the iron absorption via the iron plasma levels, groups of 4 rats orally administered 20 mg or 200 mg of carbonyl iron. With an increase in the oral dose from 20 to 200 mg carbonyl iron, the plasma iron levels also increased, but the iron absorption is seven times higher in iron-deficient rats than in normal rats. In iron-deficient rats, the plasma iron levels after oral administration of 200 mg were 0.95 mg/dL and decreased to about 0.7 mg/dL after 3 days, while the plasma iron levels in normal rats were initially about 0.35 mg/dL, and after 3 days were about 0.1 mg/dL.

In another absorption study by Huebers et a. (1986), the iron levels in liver were examined after oral uptake of carbonyl iron or ferrous ammonium sulphate. Groups of 3 rats (weighing 200 – 300 mg) orally administered different iron doses of 0.075 - 100 mg. The absorption of orally administered carbonyl iron or ferrous ammonium sulphate was compared in rats and found to be similar up to a dosage of 20 mg iron. Hepatic iron uptake was < 30 % lower in the iron-deficient animals over the entire dosage range studied.

3.4 Toxicokinetics – in vivo data on oral absorption

3.4.1 Elemental iron

Human data

Three references describing human data on absorption/bioavailability could be identified. After a thorough reliability screening, these references were considered of limited relevance for hazard assessment purposes. The criteria for quality, reliability and adequacy of experimental data under REACH for hazard assessment purposes (ECHA guidance R4 in conjunction with regulation (EC) 1907/2006, Annexes VII-X) are not fulfilled. These references were used as supplementary information.

Hoppe et al. (2006) determined the relative bioavailability (RBV) of seven elemental iron powders (carbonyl iron, electrolytic and reduced iron) in a double-blinded randomized crossover study with male blood donors. Three groups of subjects (n = 16/group; 35 – 61 years old (mean 50 years old) were served bread rolls fortified with 100 mg of the respective elemental iron powder or 272 mg ferrous sulphate monohydrate (FeSO4·H2O) nine weeks apart. RBV was obtained by comparing the increase in serum iron concentration within the 6 hours after oral intake of iron. All elemental iron powders studied were significantly less well absorbed compared to FeSO4·H2O. The electrolytic iron given with 50 mg AA was as well absorbed as FeSO4 (molar ratio = 1:6, AA:Fe). The following mean RBVs were obtained for the iron powders: electrolytic (A-131, RBV = 0.65); electrolytic (Electrolytic, RBV = 0.59); carbonyl (Ferronyl, RBV = 0.58); H-reduced (AC- 325, RBV = 0.56); H-reduced (Hi-Sol, RBV = 0.50); carbonyl (CF, RBV = 0.37); reduced (Atomet 95SP, RBV = 0.36). The reduced iron was distinguished by having significantly lower RBV (0.36) although no significant overall ranking was possible. The authors concluded that the reduced iron powder was absorbed to a lower extent compared to the other iron powders and only 36 % compared to FeSO4·H2O. Ascorbic acid seems to improve the bioavailability of elemental iron even though a rather low molar ratio is used. The major weakness of this epidemiological study is insufficient characterisation of the test materials (no purity, impurities, or particle size).

Gordeuk et al. (1987) investigated in a randomized double-blind trial involving 36 females (18 – 40 years old; menstruating and non-pregnant) blood donors with mild iron deficiency anaemia if high doses of oral iron could shorten the duration of therapy necessary to treat iron deficiency anaemia. The females orally administered a high-dose of 600 mg carbonyl iron or 300 mg ferrous sulphate (equivalent to 60 mg iron) as standard three times per day for three weeks. In the 16-week postexposure period, the iron balance was examined by blood and faeces samples. The 10-fold increase in dose of iron resulted in a 1.4 - 1.6-fold increase in the positive iron balance in the same period of time. Both regimens corrected anaemia but neither replenished storage iron. The major weakness of this epidemiological study is insufficient characterisation of the test materials (no source, purity, impurities or particle size).

Roughead et al. (2000) examined in a randomized, placebo-controlled adaptation trial the haem- and non-haem iron absorption in 27 healthy men and 30 women before and after 12 weeks of supplementation with iron. Humans were given iron supplements (50 mg Fe/d) or a placebo for 12 weeks and studied for the absorption of iron isotope (55Fe in rabbit haemoglobin as a source of haem iron and 59Fe in ferrous sulphate as a source of non-haem iron, both added to a hamburger). Serum and faecal ferritin were measured during supplementation and for 6 months thereafter. Healthy individuals, even those with low iron stores, had reduced non-haem iron absorption from food in response to iron supplementation. Despite this partial adaptation, iron stores were greater after iron supplementation than after placebo and this difference was sustained, except in individuals with the lowest iron stores. Potential confounding factors like smoking, alcohol use, social and history, and daily iron intake by food were statistically not considered. In addition, the test material was insufficiently characterised (no source, purity, impurities or particle size).

Summary entry – oral absorption

Another two references were identified during a literature search, representing investigations on oral absorption. These investigations were conducted with occupationally exposed workers to iron oxides. The study designs are not in accordance with accepted guidelines and are therefore of limited relevance for chemicals hazard assessment. The references usually lack significance due to, e.g., insufficient description of method, uncertainty in number workers analysed and/or confounding factors cannot be excluded. It is therefore concluded that all references do not fulfil the criteria for quality, reliability and adequacy of experimental data for the fulfilment of data requirements under REACH and hazard assessment purposes (ECHA guidance R4 in conjunction with regulation (EC) 1907/2006, Annexes VII-X). The studies given below were included in the IUCLID for information purposes only:

Hallberg, L. et al. (1986): Confounding factors, like smoking, personal and medical history, daily iron intake by food, were not considered. Only few information on subjects is available. A detailed description of the test material is missing. The method of statistical analysis is not specified.

Cook, J.D. et al. (1973): Confounding factors like smoking, personal and medical history, daily iron intake by food, were not considered. The test material was self-synthesised. The method for statistical analysis is missing.

Animal data

In the study by Whittaker et al. (2002), the absorption of carbonyl iron (5 - 6 µm) was examined in male SD-rats at single doses of 40 and 50 g/kg bw. In addition, a time-response study was performed of serum and liver non-haem iron in rats after oral administration of a single dose of 30 g/kg bw carbonyl iron. Together with carbonyl iron, FeSO4 and NaFeEDTA were investigated (tested in different doses). After a dose of 30 g/kg bw carbonyl iron, serum iron levels peaked at 0.5 h (787 µg/dl) and then gradually decreased to a level of approximately 490 µg/dl up to 4 h. Measurements at 24 and 48 h indicate that the serum iron level remained constant. Liver non-haem iron levels increased significantly in comparison to the controls, which indicates that iron was stored in the liver. Liver non-haem iron reached the maximum at 1.5 h, and then fluctuated between 56 and 90 µg/g over the 48-h period. At the dose of 40 g/kg bw carbonyl iron, the iron level of 406 µg/dl in serum and 448 µg/g in liver were reached, and the rats had an iron level of 358 µg/dl in serum and 407 µg/g in liver at the dose of 50 g/kg bw. Iron overload due to consumption of carbonyl iron, may result in its accumulation in the liver; however, large iron particles are not expected to be solubilized, and hence, do not become systemically available. The test material was insufficiently characterised (no purity). Furthermore, the relative iron absorption in the liver as a function of the amount of iron consumed cannot be calculated, since the food consumption of the rats was not recorded during the study.

Plummer et al. (1997) examined the absorption of iron in the offspring of Porton rats. Rats were exposed to carbonyl iron, via breastfeeding (19 days) and successively via feed (29 weeks). The doses of 0, 0.5, 1 and 2 % (w/w) were supplemented to the feed of the mothers. After weaning the offspring continued to receive the assigned diet until the age of 32 weeks. Liver biopsies examinations revealed that the hepatic iron content and loading, as well as the depression of growth were depended on the dose of iron supplemented in the diet. iron overload leads to its accumulation in the hepatocytes of the rats, in a dose-dependent manner. The hepatic iron content depended on the dose of Fe supplemented in the diet. The main disadvantages of the study are the measurement of iron content in only one part of the liver and the lack of recording of the feed intake of the offspring. The iron content was determined in only part of the liver, after the organ was obtained after 8, 16 and 24 weeks of age during an open liver biopsy on the living offspring. The values are not representative because the iron content was never measured in the entire liver. In addition, the iron content was not measured before the animals were 8 weeks old (after birth). The relative iron absorption in the liver as a function of the amount of iron consumed cannot be calculated in this study, because food consumption of offspring was not recorded.

Motzok evaluated the relative bioavailability of iron from the elemental iron powders in vivo via the haemoglobin levels when plotted against the logarithm of added iron compared to FeSO4 (with a bioavailability of 100 %). The in vivo bioavailability of iron from elemental iron powders (produced by electrolysis, reduction with hydrogen (H2), carbon monoxide (CO), or desiccated ammonia (NH3), and carbonyl process) were compared with the in vitro solubility of the powders in 0.2 % HCl (w/v, ca 0.05 N) for periods of 5 to 90 min. In addition, particle size distributions and specific surface areas of the iron powders were measured. At the bioavailability study, anaemic Wistar rats were fed daily to 3 levels of each test substance for 2 weeks. The elemental Fe powders produced by reduction (either H2 or CO), demonstrated very low relative bioavailability (10 - 17 %) in comparison to the electrolytic (36 - 48 %) and both carbonyl Fe powders (23 - 35 %). Electrolytic iron and carbonyl iron showed a high solubility in the HCl solution of ≥ 85 % after 90 min, while the H2- and NH3-reduced iron showed a lower solubility of 40 - 57 %. All iron powders had a purity of ≥ 99% but the carbonyl iron particles have the lowest particle size (0.5 - 21 µm) with the highest specific surface area (1450 - 1720 cm2/g), while the particles of electrolytic and reduced iron showed a higher particle size of 7 - 10 µm and a lower specific surface area of 1259 - 1368 cm2/g. In this study, the experimental set up is well-documented, but not all the results are shown in the publication. The 3 doses of test item in diet at the in vivo study were analysed but not reported.

Swain et al. (2003) also investigated the bioavailability but as a function of the solubility of elemental iron powders. Different elemental iron compounds (carbonyl iron, electrolytic iron reduced, H- or CO-reduced iron) were supplemented in the diet. SD-Rats received daily doses of 6 - 32 mg iron/kg diet for 14 days. It was shown that carbonyl iron powder was the most bioavailable elemental iron form. Bioavailability from the iron powders, based on absolute iron intake, was significantly less than from FeSO4 (100 %) with the following rank order: carbonyl Fe (64 %)> electrolytic Fe (54 %; USA)> electrolytic Fe (46 %; from India)> H-reduced Fe (42 %)> reduced Fe (24 %; from Canada)> CO-reduced Fe (21 %). Bioavailability of iron powders is less than bakery-grade ferrous sulphate and varies up to 3 times among different commercial forms. Solubility at pH 1 and surface area were predictive of Fe bioavailability in rats. Bioavailability and hence, absorption of iron particles, was correlated with their solubility (in different pH and time points) and surface area. The smaller particles of carbonyl iron with the highest specific surface area showed the highest solubility and bioavailability compared to the larger particles of electrolytic and reduced iron with a smaller surface area.

Sacks et al. (1978) used the rate of haemoglobin repletion in Fe-deficient Wistar rats to determine the bioavailability of several commercially available elemental Fe powders (carbonyl iron, electrolytic and hydrogen reduced iron) added to an iron-free nutritionally balanced chow. Ferrous sulphate was used as a reference substance, due to its high bioavailability. Carbonyl Fe powders demonstrated the highest bioavailability among all. Also in this study, carbonyl iron powders were identified as the compound with the highest bioavailability. The bioavailability of carbonyl iron powders in the rat were 66 % (substance a: 90 % < 5 µm), 63 % (substance b: 90 % < 8 µm) and 39 % (substance c: 80 % < 20 µm) compared to ferrous sulphate. The increase in the bioavailability with decrease particle size of the elemental iron powders was also seen with the hydrogen reduced iron powders.

3.4.2 Iron oxides

The iron absorption and excretion were examined in vivo by Leuschner (2020). SD-rats were administered Fe2O3, FeOOH (< 100 nm) or Fe3O4 (> 100 nm) at a single dose of 1000 mg/kg bw by oral gavage. Iron(III) citrate tribasic monohydrate were used as reference substance and were intravenously injected to the animals. The urine, faeces, cage washes and bile were collected to investigate the iron excretion, and the iron absorption in blood was examined at differential time points. An absolute bioavailability of 0.21 - 0.26 % (both sexes) for Fe2O3, Fe3O4 and FeOOH was calculated for iron following oral administration compared to intravenous administration. However, it should be noted that this evaluation was done on the substance-specific data without consideration of the vehicle control. The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels. Hence, the calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide is similarly minimal to negligible.

3.4.3 Summary (oral absorption):

In vitro bioavailability data in body fluids or artificial body fluids clearly show that the solubility of iron, particularly elemental iron such as carbonyl iron, is greatest at low pH, dropping to a very low solubility at neutral and alkaline pH. Huebers et al. (1986) showed in his different in vitro/in vivo bioaccessibility and bioavailability studies that metallic iron in the form of carbonyl iron particles has to be dissolved in the stomach juice at pH < 2 before absorption is possible. Carbonyl iron is only absorbed in the form of iron(II)ions.

There is also a relationship between the solubility of iron and its bioavailability. In the studies by Matzok et al. (1978) and Swain et al. (2003) on the bioavailability and solubility of iron, it was shown after feeding rats with various elemental iron compounds (carbonyl iron, electrolytic and reduced iron) that the bioavailability relative to FeSO4 is greatest for carbonyl iron and electrolytic iron. The bioavailability of carbonyl iron is 23 - 35 % and 46 - 54 % for electrolytic iron in male Sprague-Dawley rats (Swain et al. (2003)). For male iron-deficient Wistar rats it is 36 - 48 % for electrolytic iron and 64 % for carbonyl iron (Motzok et al. (1978)). In the study by Sacks et al. (1978), a bioavailability of 66 % (A), 63 % (B) and 39 % (C) was determined in iron-deficient male Wistar rats for carbonyl iron of different particle sizes (A: 90 % of particles < 5 µm; B: 90 % < 8 µm and C: 80 % < 20 µm). The bioavailability of hydrogen reduced iron powders also increased with decreasing particle size. When comparing the animal and human data on the bioavailability of the various elemental iron compounds, it is noticeable that the bioavailability of iron differs somewhat between humans and rats. In the double-blinded randomized crossover study by Hoppe et al. (2006), the bioavailability was examined in 48 men after oral administration of 100 mg carbonyl iron, electrolytic or reduced iron compared to 272 mg ferrous sulphate monohydrate. A bioavailability of 59 % or 65 % for electrolytic iron, 37 or 58 % for carbonyl iron, 36 % for reduced iron, and 50 or 56 % for hydrogen-reduced iron were determined. The relative bioavailability of the respective elementary iron compound is higher in humans than in rats, with exception of carbonyl iron. A comparison of the different elemental iron compounds tested in these four studies is not possible, since the iron compounds in the human study were not further characterized. Furthermore, carbonyl iron has the highest solubility in hydrochloric acid solution at pH < 2, closely followed by electrolytic iron. Bioavailability and hence, absorption of iron particles, is correlated with their solubility (in different pH and time points), particle size and surface area. The smaller particles of carbonyl iron with the higher specific surface area show a higher solubility and bioavailability compared to the larger particles of electrolytic and reduced iron with a smaller surface area.

In contrast, the iron oxides in the studies by Leuschner (2020) show only low solubility and bioavailability. To investigate the bioavailability, iron oxides (Fe2O3, FeOOH (< 100 nm) or Fe3O4 (> 100 nm)) were administered to rats by oral gavage at a single dose of 1000 mg/kg bw. Iron(III) citrate tribasic monohydrate were used as reference substance. The plasma iron level of the dosed group falls practically within the boundaries of the vehicle control group which reflects long established daily circadian variation of plasma iron levels. Hence, the calculated absolute bioavailability derived by the pharmacokinetic analysis can therefore be seen as conservative overestimates, thus leading to the conclusion that the bioavailability of iron from the tested oxide (0.21 – 0.26 %) is similarly minimal to negligible.

After a single oral dose of 30 g/kg bw carbonyl iron to male rats, the iron reaches a maximum value in the blood after just 0.5 h (787 µg/dl) and in the liver only after 1.5 h (about 170 µg/g liver) (Whittaker et al. (2002)). The iron levels then fell again and remain constant at about 490 µg/dl in the blood or fluctuate between 50 and 90 µg/g in the liver over the 48-h period. This indicates that orally ingested iron first enters the blood before entering the liver. Iron overload due to consumption of micro sized carbonyl iron (5 - 6 µm), may result in its accumulation in the liver of rats. However, large iron particles are not expected to be solubilized, and hence, become systemically available.

Plummer et al. (1997) also investigated the accumulation of iron in rats, but in the offspring. Rats were exposed to carbonyl iron, via breastfeeding (19 days) and successively via feed (29 weeks). The doses of 0, 0.5, 1 and 2 % (w/w) were supplemented to the feed of the mothers. After weaning the offspring continued to receive the assigned diet until the age of 32 weeks. Liver biopsies examinations revealed that the hepatic iron content and loading, as well as the depression of growth were depended on the dose of iron supplemented in the diet. Iron overload leads to its accumulation in the hepatocytes of the rats, in a dose-dependent manner. The hepatic iron content depended on the dose of iron supplemented in the diet.

Furthermore, the absorption studies from Huebers et al. (1986) showed in male rats that the iron absorption in the blood, skeleton, liver and the carcass are higher in the iron-deficient rats after a single oral intubation of 200 mg carbonyl iron (3 - 4 µm) in comparison to rats with normal diet over a period of 4 days.

A randomized double-blind trial with humans by Gordeuk et al. (1987), there 32 women with anaemia received orally 600 mg carbonyl iron or 300 mg ferrous sulphate (equivalent to 60 mg iron) three times per day for 3 weeks (cumulative daily dose: 1800 mg Fe for carbonyl iron, 180 mg Fe for iron sulfate), a 10-fold increase in iron dose result in a 1.4 - 1.6-fold increase in the positive iron balance in humans in the same time period. Both regimens correct anaemia but neither replenish storage iron.

Roughead et al. (2000) showed in a randomized, placebo-controlled that a supplementation of 50 mg iron/day as ferrous sulphate for 12 weeks reduces the absorption of non-haem iron in food in healthy individuals, even those with low iron stores. In addition, iron supplementation increases iron stores compared to placebo, and this difference persists except in individuals with low iron stores.

3.5 Toxicokinetics – in vivo data on inhalation absorption

3.5.1 Iron oxides

The kinetics of particle clearance in the respiratory tract were examined by Lehnert et al. (1985a & b). The main objective of this study was to determine the effect of a low lung burden of innocuous particles on the alveolar macrophage (AM) function. Lehnert et al. (1985a & b) exposed male Long-Evans rats once nose-only to aerosols of radioactive iron-59 oxide (MMAD = 1.6 μm, σ(g) = 3.0) at concentrations of 18.2 (exposure I: 46 males) or 24.2 mg/m³ (exposure II: 49 males) for 2 h to determine how a low lung burden (~30 μg) affects the size of the AM pool, and the functional status of the AM. The lung radioactivity was also measured. The average biological clearance half-time of 59Fe was calculated to be 52.5 ± 5.5 days. On day 0 postexposure, a mass of 25.7 ± 7.7 μg Fe2O3 was found in the lungs of rats at exposure I, while it was 43.9 ± 20.2 μg in rats at exposure II. The average effective clearance half-time for radioactive iron (59Fe) from the lungs of all animals was 24 ± 2.5 days. Based on the 45-d physical half-life of 59Fe, this translates into an average biological clearance half-time of 52.5 ± 5.5 days, which is within the elimination half time range for alveolar clearance in a non-overloading-state for rats (reported before to be 50 - 65 days); hence, no clearance disturbance is observed. Radioactivity was associated with macrophages (~60 % on the day of exposure and >90 % afterwards). Lung retention decreased from ~30000 counts per minute within 70 days. The study showed significant methodological deficiencies. Experimental set up was insufficiently described. The test material was insufficiently characterised (no purity). Only males were used. Information about the test animals and environmental conditions were insufficient.

Furthermore, Oberdoerster et al. (1984) determined the lung retention of radioactive 59Fe304 in rats. Male Fischer 344 rats were exposed in a nose-only system to 59Fe304 (particle size: 1.5 µm, σ(g): 1.8) for 2 hrs at a single concentration of 15.4 ± 4.5 mg/m3. Following exposure, thoracic 59Fe activity in the live animals was measured by a Nal (T1) detector. After 120 days, animals were sacrificed, and the lungs were removed for measurement of 59Fe activities by a Nal (T1) detector. On day 120, 19 ± 4 % of the initially deposited 59Fe3O4 particles were still present in the lungs. Tissue retention was 17.3 ± 6.7 µg Fe3O4 per lung after 2 h exposure. The half-live for 59Fe3O4 was 47 days.

In another study, the uptake and transport of sub-micrometer insoluble particles was investigated by the airway epithelium (Watson et al. (1979)). Male CD-1 Mice were exposed by inhalation to an aerosol of nano sized Fe2O3 (diameter of particles: 0.005 µm, MMAD: 0.15 µm, GSD: 2.2) at a single concentration of 300 mg/m3 for 3 hours. Participation of the tracheal and bronchial epithelium in the uptake of iron oxide was investigated immediately following the exposure and at 1 day, 4 days and 7 days postexposure. At each time point, animals were sacrificed, and the trachea and intrapulmonary bronchi were histopathologically examined. Iron oxide nanoparticles are pinocytised by the epithelial cells of the airways and subsequently give rise to the formation of ferritin and hemosiderin. Quantification of hemosiderin shows that iron storage increases with time after exposure to iron(III) oxide (Fe2O3). The study provides some evidence of dissolution of the iron oxide after entering the cells. However, the very small particle size has to be taken into account. In this study, the experimental set up and results are described in detail. However, the test material was insufficiently described (no purity).

Whether iron is transported to the rat brain via the olfactory tract following inhalation exposure was investigated by Rao et al. (2003). In this study, rats were exposed to radiolabelled 59Fe(II)SO4 (radionuclidic purity: 99.5 %, MMAD: 2.99 µm, GSD: 1.15) at a single concentration of 3.70 or 3.99 mg/m3 for 88.5 min. Then rats were killed immediately or at 1, 2, 4, 8, or 21 days postexposure. Gamma spectrometry was performed to the following tissues at each time point: nasal olfactory and non-olfactory mucosa, olfactory bulb, olfactory tract plus tubercle, striatum, cerebellum, and rest of brain, lung, liver, kidney and pancreas. Some nasal olfactory mucosa samples were analysed for 59-Fe associated transferrin with HPLC and gamma spectroscopy. Heads were collected and autoradiograms were prepared to visualise the location of 59 Fe from the nose to the brain. Most of the 59Fe remained in the nasal regions of the olfactory system and less than 4 % uptake of 59Fe(II) SO4 into the brain via the olfactory pathway was measured. 59Fe activity was absent in the olfactory regions of brain even 4 days postexposure. Further HPLC-gamma spectroscopy analyses indicated that 59Fe in the olfactory mucosa was coeluted with transferrin. Hence, iron is not readily transported to the brain via the olfactory tract. Rao et al. (2003) only investigated the absorption of iron in rats to the brain via the olfactory pathway after inhalation. Only one dose was tested in two cohorts which were divided in two experimental groups with or without nasal occlusion (1. patent right and left nostrils, 2. only right nostril).

3.5.2 Summary (inhalation absorption):

In the studies by Lehnert et al. (1985a & b), a nose-only inhalation exposure to aerosols of radioactive 59Fe2O3 (MMAD: 1.6 μm, σg: 3.0) at a concentration of 20 mg/m³ for 2 h leads to a low lung burden (~30 μg) in the male rats of innocuous particles. Radioactive Fe2O3 has an average biological clearance half-time of 52.5 ± 5.5 days. On day 0 postexposure, a mass of 25.7 ± 7.7 μg Fe2O3 is found in the lungs of rats after an exposure level of 18.2 mg/m³, and 43.9 ± 20.2 μg after an exposure of 24.2 mg/m³. The average effective clearance half-time of 59Fe from the lungs is 24 ± 2.5 days and it has a biological clearance half-time of 52.5 ± 5.5 days, which is within the elimination half-time range for alveolar clearance in a non-overloading-state for rats, hence, no clearance disturbance is observed. Radioactivity is associated with macrophages (~60 % on the day of exposure and >90 % afterwards). Lung retention decreases from ~30000 cpm within 70 days.

After a single exposure of radiolabelled 59Fe304 (MMAD: 1.5 µm σ(g)=1.8) for 2 h at a concentration of 15.4 ± 4.5 mg/m3, a tissue retention of 17.3 ± 6.7 µg Fe3O4 per lung is found (Oberdoerster et al. (1984)). The half-live for 59Fe3O4 in rats is 47 days. 19 ± 4 % of the initially deposited 59Fe3O4 particles are still present in the lungs on day 120.

In another study from Watson et al. (1979), the uptake and transport of sub-micrometer insoluble particles was investigated by the airway epithelium. In male mice, exposed by inhalation to an aerosol of Fe2O3 nanoparticles (particle diameter 0.005 µm, MMAD: 0.15 µm, GSD: 2.2) at a concentration of 300 mg/m3 for 3 hours, iron oxide nano-particles are pinocytised by the epithelial cells of the airways and subsequently give rise to the formation of ferritin and hemosiderin. Iron storage increases with time after exposure.

Furthermore, Rao et al. (2003) examined the iron absorption to the brain of male (SD)IGSBR rats via the olfactory tract following inhalation. However, the results show that iron is not readily transported to the brain via the olfactory tract. Most of the 59Fe remain in the nasal regions of the olfactory system and less than 4 % uptake of 59Fe(II) SO4 are deposited into the brain via the olfactory pathway after a single nose-only inhalation exposure to radiolabelled 59Fe(II)SO4 (radionuclidic purity: 99.5 %, MMAD: 2.99 µm, GSD: 1.15) at a concentration of 3.70 or 3.99 mg/m3 for 88.5 min. 59Fe activity is absent in the olfactory regions of brain even 4 days postexposure. Further, 59Fe in the olfactory mucosa is coeluted with transferrin.

References:

Arbeitsgruppe: Referenzwerte für die Nährstoffzufuhr (2000). Umschau Braus Verlag, Frankfurt, pp174-179.

Ballot DE, MacPhail AP, Bothwell TH, Gillooly M, Mayet FG (1989). Fortification of curry powder with NaFe(III) EDTA in an iron-deficient population: Report of a controlled iron-fortification trial. Am J Clin Nutr 49: 156-161.

Barrett JFR, Whittaker PG, Eilliams JG, Lind T (1994). Absorption of non-haem iron from food during normal pregnancy. Br Med J 309: 79-82.

Baynes RD and Bothwell TH (1990). Iron deficiency. Ann Rev Nutr 10: 133-148.

Borch-Iohnsen B and Ptersson Grawe K (1995). In Risk evaluation of essential trace elements. Ed. Oskarsson A. Nordic Council of Ministers, Copenhagen, 1995, pp 67-117.

Bothwell TH, Charlton RW, Cook JD, Finch CA (1979). Internal iron kinetics In: Iron metabolism in man. Blackwell Scientific Publishers, Oxford, London, pp 327-353.

EFSA (2005): Opinion of the Scientific Panel on Dietetic Products, Nutrition and Allergies on a request from the Commission related to the Tolerable Upper Intake Level of Iron, The EFSA Journal, 125, 1-34.

Ellenham MJ and Barceloux DG (1988). Iron. In: Medical Toxicology. Elsevier, New York, Amsterdam, London, pp 1023-1030.

Finch C (1994). Regulators of iron balance in humans. Blood 84: 1697-1702.

Fleming DJ, Tucker KL, Jaques PF, Dallal GE, Wilson PWF, Wood RJ (2002). Dietary factors associated with risk of high iron stores in the elderly Framingham Heart Study Cohort. Am J Clin Nutr, Am J Clin Nutr 76: 1375-1384.

FNB (Food and Nutrition Board) (2001). Dietary Reference Intakes: Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodide, Iron, Manganese Molybdenum, Nickel, Silicon, Vanadium and Zinc. Institute of Medicine, National Academy Press, Washington D.C.

Green R, Charlton R, Seftel H, Bothwell TH, Maget F (1968). Body iron excretion in man. A collaborative study. Am J Med 45: 336-353.

Gunshin H, Mackenzie B, Berger UV, Gunshin Y, Romero MF, Boron WF, Nussberger S, Gollan JL, Hediger MA (1997). Cloning and characterization of a mammalian proton-coupled metal-ion transporter. Nature 388: 482-488.

Hallberg L and Rossander L (1984). Improvement of iron nutrition in developing countries: comparison of adding meat, soy protein, ascorbic acid, citric acid, and ferrous sulphate on iron absorption from a simple Latin American-type of meal.Am J Clin Nutr 39: 577-583.

Hallberg L, Hulthen L, Garby L (1997). Iron stores in man in relation to diet and iron requirements. Eur J Clin Nutr 52: 623-631.

Hunt JR and Roughead ZK (2000). Adaptation of iron absorption in men consuming diets with high or low iron bioavailability.Am J Clin Nutr 71: 94-102.

Kulaksiz H, Gehrke SG, Janetzko A, Rost D, Bruckner T, Kallinowski B, Stremmel W. (2004). Pro-hepcidin: expression and cell specific localisation in the liver and its regulation in hereditary haemochromatosis, chronic renal insufficiency, and renal anaemia. Gut 53: 735-743.

Leong WI and Lonnerdal B (2004). Hepcidin, the recently identified peptide that appears to regulate iron absorption. J Nutr 134: 1-4.

Lynch SR (1984). Iron. In: Solomons NW, Rosenberg ICH, eds., Absorption and Malabsorption of Mineral Nutrients. Alan R Liss Inc., New York, pp 89-124.

Magnusson B, Bjorn-Rassmussen E, Hallberg L, Rossander L (1981). Iron absorption in relation to iron status. A model proposed to express results of food absorption measurements. Scand J Haematol 27: 201-208.

Mulvihill B, Kirwan FM, Morrissey PA, Flynn A (1998). Effects of myofibrillar muscle protein on the in vitro bioavailability of non-haem iron. Int J Food Sci Nutr 49: 187-192.

Nicolas G, Bennoun M, Devaux I, Beaumont C, Grandchamp B, Kahn A, Vaulont S (2001). Lack of hepcidin gene expression and severe tissue iron overload in upstream stimulatory factor 2 (USF2) knockout mice. PNAS 98: 8780-8785.

Oates PS, Trinder D, Morgan EH (2000). Gastrointestinal function, divalent metal transporter-1 expression and gastrointestinal iron absorption. Pflügers Arch.-Eur J Physiol 440: 496-502.

Oski FA (1993). Iron deficiency in infancy and childhood. N Engl J Med 329: 190-193.

Ponka P, Beaumont C, Richardson DR (1998). Function and regulation of transferrin and ferritin. Semin Hematol 35: 35-54.

Rossander L (1987). Effect of Fiber on iron absorption in man. Scand J Gastroenterol 22 (Suppl) 129: 68-72.

Sayers MH, English G, Finch CA (1994). The capacity of the store regulator in maintaining iron balance. Am J Hematol 47: 194-197.

SCF (Scientific Committee on Food) (1993). Reports of the Scientific Committee on Food (31st series). Commission of the European Community, Luxembourg, pp. 177-189.

Schümann K, Elsenhans B, Forth W (1999a). Kinetic analysis of 59Fe movement across the intestinal wall in duodenal rat segments ex vivo.Am J Physiol 276: G431-G440.

Information on Registered Substances comes from registration dossiers which have been assigned a registration number. The assignment of a registration number does however not guarantee that the information in the dossier is correct or that the dossier is compliant with Regulation (EC) No 1907/2006 (the REACH Regulation). This information has not been reviewed or verified by the Agency or any other authority. The content is subject to change without prior notice.
Reproduction or further distribution of this information may be subject to copyright protection. Use of the information without obtaining the permission from the owner(s) of the respective information might violate the rights of the owner.

sample name	target pH	pH prior to the test	pH after 2h	pH after 24 h
GSTblankvessel1	1.5–1.6	1.55	1.57	1.59
GSTblankvessel2	1.5–1.6	1.54	1.58	1.58

sample name	target pH	pH prior to the test	pH after 2h		pH after 24h
GSTvessel10	1.5–1.6	1.55		1.61	1.61
GSTvessel11	1.5–1.6	1.55		1.60	1.65
GSTvessel12	1.5–1.6	1.54		1.61	1.62

samplename	targetpH	pHpriortothetest	pHafter2h	pHafter24h
PBSblankvessel1	7.2–7.4	7.30	7.30	7.33
PBSblankvessel2	7.2–7.4	7.30	7.33	7.33

samplename	targetpH	pHpriortothetest	pHafter2h	pHafter24h
PBSvessel10	7.2–7.4	7.30	7.34	7.37
PBSvessel11	7.2–7.4	7.30	7.33	7.35
PBS vessel 12	7.2–7.4	7.29	7.34

samplename	targetpH	pHpriortothetest	pHafter2h	pHafter24h	pHafter7d
GMBblankvessel1	7.4	7.46	8.10	8.68	9.09
GMBblankvessel2	7.4	7.46	8.14	8.74	9.09

sample name	target pH	pH prior to the test	pH after 2h	pH after 24h
PBS vessel 1	7.2–7.4	7.31	7.35	7.36
PBS vessel 2	7.2–7.4	7.29	7.32	7.32
PBS vessel 3	7.2–7.4	7.31	7.36	7.37

	n (deaths)	Dose (g/kg bw)	Body weight	Liver weight	Liver nonheme Fe (µg/g)	Total liver nonhem Fe (µg)	Serum Fe (µg/dl)
Control	8 (0)	0	121± 4	5.87± 0,23*,°	42.9± 3.7*	252± 23*	229± 11*,°
Carbonyl Fe	8 (0)	40	89± 3	3.89±0.18*	448± 50~	1687± 130°,~	406± 30°
Carbonyl Fe	8 (0)	50	92± 5	4.32± 0.45*	407± 58°,~	1582± 97°,~	358± 28*,°

samplename	targetpH	pHpriortothetest	pHafter2h	pHafter24h	pHafter7d
ALFblankvessel1	4.5	4.56	4.57	4.62	4.62
ALFblankvessel2	4.5	4.56	4.59	4.62	4.62

samplename	targetpH	pHpriortothetest	pHafter2h	pHafter24h	pHafter7d
ALFblankvessel1	4.5	4.56	4.57	4.62	4.62
ALFblankvessel2	4.5	4.56	4.59	4.62	4.62

samplename	targetpH	pHpriortothetest	pHafter2h	pHafter24h	pHafter7d
ALFblankvessel1	4.5	4.56	4.57	4.62	4.62
ALFblankvessel2	4.5	4.56	4.59	4.62	4.62

samplename	targetpH	pHpriortothetest	pHafter2h	pHafter24h	pHafter7d
GMBvessel4	7.4	7.45	8.19	8.80	9.24
GMBvessel5	7.4	7.46	8.20	8.71	9.24
GMBvessel6	7.4	7.45	8.22	8.86	9.22

Registration Dossier

Registration Dossier

Data platform availability banner - registered substances factsheets

Diss Factsheets

Mill scale (ferrous metal)

Endpoint summary

Administrative data

Link to relevant study record(s)

Referenceopen allclose all

Description of key information

Key value for chemical safety assessment

Additional information

1 Introduction

1.1 Absorption and regulation of absorption

1.2 Elimination

1.3 Requirements and recommended intakes

2 Experimental data

2.1 Water solubility

2.2 In vitro bioaccessibility in artificial body fluids

3 Toxicokinetic aspects

3.1 Investigations of cellular uptake of iron compounds in-vitro

3.2 In vivo uptake mechanisms and supporting studies

3.3 Tissue and blood levels of iron

3.4 Toxicokinetics – in vivo data on oral absorption

3.4.1 Elemental iron

3.4.2 Iron oxides

3.4.3 Summary (oral absorption):

3.5 Toxicokinetics – in vivo data on inhalation absorption

3.5.1 Iron oxides

3.5.2 Summary (inhalation absorption):

Referenceopen all close all