Titanium dioxide

Endpoint summary

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

not specified

Objective of study:

absorption

Qualifier:

no guideline followed

Principles of method if other than guideline:

Groups of 6 male Sprague-Dawley rats were administered a single dose of 4.6 mg/kg of either two different titanium dioxide nanoforms (40 nm and 40-50 nm) each or two micron-sized (120 nm and < 5 µm) titanium dioxide form each. During a post-administration period of 96 hours, rats were housed in metabolic cages for collection of urine and faeces. Furthermore, blood samples were taken during the post-administration period. After sacrifice (96 hours post treatment), tissue samples (liver, brain, heart, kidney, spleen and GI tract) were collected. The samples of urine, faeces, blood and tissues were analysed for titanium levels via ICP-MS.

GLP compliance:

not specified

Specific details on test material used for the study:

TREATMENT OF TEST MATERIAL PRIOR TO TESTING
All of the supplied ‘nanomaterials’ tested contained around ~80 % by weight of larger (typically between 200 and 500 nm) structures. Conversion of mass-based particle size distribution to particle number-based values still showed the majority of the particles to be the nanoscale.
High-powered sonication, followed by centrifugation, separated the nano-sized fraction by sedimenting larger-sized clusters. The nano-dispersions obtained by this method were stable in aqueous media over a few hours. The nanoscale fractions isolated as a supernatant from this procedure were either used straightaway as such, or freeze-dried and resuspended before use. Average nano-fraction obtained by this method was around 17 % (n = 3) of the original as measured by weighing after freeze-drying. Similar results were obtained by DLS method, although the hydrodynamic sizes measured by DLS are generally larger than those measured by other methods. The results of DLS measurements showed that between 10 and 17 % of the particles had a size in the nanoscale, whilst the remaining particles were in larger size range (600–700 nm). It was found that particle agglomeration had again taken place in freeze-dried nanofractions on resuspension in aqueous media. The materials and sample preparation methods were therefore adapted appropriately to obtain a uniform nano-fraction for use in in vivo experiment.

1) Name of test material (as cited in study report): NanoAmor 5430 MR (40 nm)
Dispersed in water and ovalbumin solution, sonicated, nano-fraction separated by centrifugation and used straightaway in experiments

2) Name of test material (as cited in study report): Sigma cat # 637252 (40 - 50 nm)
Dispersed in water and ovalbumin solution, sonicated, nano-fraction separated by centrifugation and used straightaway in experiments

3) Name of test material (as cited in study report): NanoAmor 5430 MR (120 nm)
Dispersed in water and ovalbumin solution. Nanofraction separated by centrifugation and freeze dried, which upon resuspension showed reagglomeration of particles to clusters of ~120 nm

4) Name of test material (as cited in study report): Sigma cat # 224227 (up to 5 µm)
Dispersed in water, sonicated and used straightaway in experiments as the non-nano control (< 5 µm)

Radiolabelling:

not specified

Species:

rat

Strain:

Sprague-Dawley

Details on species / strain selection:

not specified

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Harlan Labs
- Age at study initiation: 8 weeks
- Weight at study initiation: 200 to 429 g (mean: 347.5 ± 46.6 g)
- Housing: a specially designed cage system was used to house individual rats to collect the faeces and urine samples during the experiment (McKenzie et al. 2010)*

*Reference:
- McKenzie J, Charlton A, Donarski J, MacNicoll A, Wilson J (2010) Peak fitting in 2D 1 H–13 C HSQC NMRspectra for metabolomic studies. Metabolomics 6(4):574–582.

Route of administration:

oral: gavage

Vehicle:

other: deionised water or 5 % ovalbumin

Details on exposure:

PREPARATION OF DOSING SOLUTIONS:
Titanium dioxide materials were suspended in either 10 mg/mLdeionised water or 5 % ovalbumin solution.
The required amount of titanium dioxide dose was calculated for each rat and the rats were anaesthetised for test item administration.

Duration and frequency of treatment / exposure:

one administration only

Dose / conc.:

4.6 mg/kg bw/day (actual dose received)

No. of animals per sex per dose / concentration:

6 male rats

Control animals:

yes, concurrent vehicle

Positive control reference chemical:

not specified

Details on dosing and sampling:

TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled: urine, faeces, blood, and tissue samples (liver, brain, heart, kidney, spleen, large intestine and small intestine)

- Time and frequency of sampling: samples of blood, urine and faeces were collected immediately after oral gavage, and then at different time intervals over 4 days (blood: 0-2 , 24, 48, 72, and 96 hours; urine and faeces: 0 - 24, 24 - 48, 48 - 72, and 72 - 96 hours). Blood samples (0.3–0.4 mL) were collected and stored in lithium heparin tubes. Analysis of the sample showed no detectable titanium above background levels in the urine samples, which ruled out possible leaching of titanium dioxide from the faecal pellets into urine. After 96 hours postadministration of titanium dioxide, rats were killed and tissue samples of liver, brain, heart, kidney, spleen, GI tract (large and small intestine) were collected. The tissue samples were kept frozen and the urine, blood and faecal samples were kept in a cold room until analysed for titanium.
For titanium analysis, the samples were digested in acid/hydrogen peroxide and titanium dioxide quantified (as Ti ion) by inductively coupled plasma mass spectrometry (ICP-MS). For this, the TiO2 particles were solubilised by alkali or acid digestion, appropriately diluted, and measured using a ThermoFisher ‘Axiom’ ICP mass spectrometer tuned within 3,000–5,000 resolution to isolate the titanium signal from potential polyatomic interferences. Alkali digestion was used to solubilise TiO2 particles in aliquots of blood and urine and acid digestion was used for tissue samples. The digestion of faecal samples posed a particular challenge and the acid digestion was adapted to ensure complete digestion of TiO2 particles.

Statistics:

The measurement results were analysed in a linear mixed model for main effects only. Fixed Effects: nominal size + pre-treatment + vehicle, with rat, day and analytical run as crossed random effects. This model and method was chosen in response to the way that the experiment had been undertaken, rather than as a reflection of an underlying model for the effects that the various kinds of treatment might have. A value of p<0.05 was used for determining whether a main effect in the model was statistically significant.

Preliminary studies:

A preliminary in vivo study was carried out to measure the background levels of titanium in rats before the main study. Three rats were used for this study, one of which was treated with the vehicle only to provide samples for determination of background exposure to titanium dioxide from diet and the environment, and the other two were orally administered with NanoAmor TiO2 (nominal particle size ~15 nm, measured size 40 nm) at 2 and 5 mg/kg bw. Samples (200 µL) of blood were collected at appropriate time intervals after titanium dioxide administration. Samples of urine and faeces were also collected and analysed for titanium content by ICP-MS.
Results:
The results of analysis of blood and urine indicated that very little titanium dioxide was absorbed within 24 hours at both 2 and 5 mg/kg bw treatment levels. However, the background levels of titanium were significantly above the limit of detection. The authors first thought that diet was the source of high background titanium levels, but an unexpected source was found to be the cardboard ‘‘toys’’ that were provided in the cages as ‘environmental enrichers’ which contained a considerable amount of titanium (~4.5 µg/g). As the rats with high background titanium levels were not suitable for this study, a fresh breeding nucleus of rats was established by feeding on a ‘titanium-free’ diet, and not providing them with the cardboard toys. The parental generation was maintained on this regime for 3–4 weeks before mating. Titanium levels in the tissues and faeces of the 50–60-day-old offspring of these rats were monitored to establish low background levels before use in the main in vivo study.

Type:

absorption

Results:

The statistical analysis showed (with the exception of faecal excretion) that there no significant differences between any of the TiO2 treatments.

Results:

Tissue levels were very low (ng/g): liver (1.43-3.74), kidney (

Results:

Tissue titanium levels (in ng/g) of the control animals were in liver (2.46), in kidneys (

Details on absorption:

The results show that oral administration of TiO2 nano- (or larger) particles did not lead to translocation of titanium to blood in rat at any of the time intervals studied during the 96 hour post-treatment.
The result of statistical analysis showed that there were no significant effects between titanium dioxide treatments following oral exposure to titanium dioxide particles; and the variation in measurement results within each organ was consistent with that caused by analytical measurement variation.
The results show that oral administration of titanium dioxide nano- (or larger) particles did not lead to translocation of titanium to distribution to various organs in rat at any of the time intervals studied during the 96 h post-treatment. With the possible exception of one particle size, the mean concentration of titanium in the GI tract was not significantly affected by treatment.
Trace amounts of titanium were detected in the GI tract samples.
The results show that oral administration of titanium dioxide nano- (or larger) particles did not lead to translocation of titanium to urine in rat at any of the time intervals studied during the 96 hour post-treatment.
The orally administered titanium dioxide materials in all cases were found to be excreted in the faeces, although this happened at different elimination rates in different test animals.
The recoveries of titanium dioxide in faecal samples for the two Sigma materials (<100 and <5,000 nm) were much lower than the samples from the animals treated with other materials. Since titanium dioxide was not absorbed/translocated from the GI tract in any of the materials tested, this discrepancy indicates a possible analytical error due to incomplete solubilisation of the Sigma materials in faecal samples.
In all instances, the titanium dioxide materials were not completely cleared from the faeces at the end of the 96 hour sampling period.

Details on distribution in tissues:

not specified

Details on excretion:

not specified

Metabolites identified:

not specified

Bioaccessibility (or Bioavailability) testing results:

not specified

Conclusions:

The following experimental result can be summarised here briefly:
- there was no detectable titanium in any of the urine samples. Interestingly, they observed a high dietary background in normal laboratory diet, which is why the main study was run with a low (0.44 ng/g) titanium diet. It was however later found that the source of the high background were cardboard “environment enricher” toys place in the rat cages (the abolishment of which led to low titanium background levels).
- the statistical analysis showed (with the exception of faecal excretion) that there no significant differences between any of the TiO2 treatments.
- in view of the dose given (5 mg/kg bw), the tissue levels were generally very low (in ng/g): liver (1.43-3.74), kidney (- the tissue titanium levels (in ng/g) of the control animals (n=6) were in liver (2.46), in kidneys (

Reference 2

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:

documentation insufficient for assessment

Objective of study:

absorption

Qualifier:

no guideline followed

Principles of method if other than guideline:

Groups of four rats were administered with two different titanium dioxide nanomaterials (TiO2 NP) (one anatase and one rutile form) and titanium dioxide E171 food pigment in distilled water. The test items were orally adminstered at dose levels of 1 and 100 mg/kg by gavage daily for 28 consecutive days. A control group was run concurrently. After the exposure period, the animals were sacrificed and a sample of the liver was taken for titanium measurements.

GLP compliance:

not specified

Specific details on test material used for the study:

not applicable

Radiolabelling:

not specified

Species:

rat

Strain:

Wistar

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Stolbovaya Breeding Center
- Diet: balanced casein-based semisynthetic ration

Route of administration:

oral: gavage

Vehicle:

other: distilled water

Details on exposure:

PREPARATION OF DOSING SOLUTIONS:
The test item wsa dispersed in the vehicle.

Duration and frequency of treatment / exposure:

Duration: 28 days
Frequency: daily

Dose / conc.:

1 other: mg/kg (actual received)

Remarks:

Titanium dioxide nanoparticles (anatase)

Dose / conc.:

100 other: mg/kg (actual received)

Remarks:

Titanium dioxide nanoparticles (anatase)

Dose / conc.:

1 other: mg/kg (actually received)

Remarks:

Titanium dioxide nanoparticles (rutile)

Dose / conc.:

100 other: mg/kg (actually received)

Remarks:

Titanium dioxide nanoparticles (rutile)

Dose / conc.:

1 other: mg/kg (actually received)

Remarks:

Titanium dioxide particles (E171)

Dose / conc.:

100 other: mg/kg (actually received)

Remarks:

Titanium dioxide particles (E171)

No. of animals per sex per dose / concentration:

4 rats

Control animals:

yes

Positive control reference chemical:

not specified

Details on study design:

not specified

Details on dosing and sampling:

TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled: liver
- Time and frequency of sampling: at the end of the exposure period, animals were sacrificed and a liver fragment was resected. Titanium in the sample was measured by mass spectrometry with inductive-bound plasma on.

Statistics:

Nonparametric Mann-Whitney test

Type:

absorption

Results:

Anatase TiO2 NP & micron-sized TiO2 particles administered at dose levels of 1 & 100 mg/kg for 28 d induced no significant increase of Ti conc. in livers compared to the control. A similar result was observed in animals treated with 1 mg/kg rutile NP.

Results:

Ti conc. in the liver increased significantly (p<0.05) in rats receiving rutile NP water dispersion at the 100 mg/kg dose level.

Details on absorption:

Administration of water dispersions of titanium dioxide nanoparticles (TiO2 NP) with anatase structure and of titanium dioxide with micron-sized particles in low (1 mg/kg) and high (100 mg/kg) doses for 28 days induced no significant increase of titanium concentration in livers compared to the control group. A similar result was observed in animals treated with rutile nanoparticles in the low dose (1 mg/kg). However, titanium concentration in the liver increased significantly (p<0.05) in rats receiving rutile nanoparticles water dispersion in the high dose (100 mg/kg).
Please also refer to table 1 in the field "Any other information on results incl. tables" below.

Details on distribution in tissues:

not specified

Details on excretion:

not specified

Metabolites identified:

not specified

Bioaccessibility (or Bioavailability) testing results:

not specified

Table 1. Mean Levels of TiO₂in the Liver of Rats Receiving Water Dispersions of Various TiO₂Forms (M±m)

Experiment conditions		Titanium content, mg/kg organ weight
Control		0.456 ± 0.004
Anatase NP	1 mg/kg	0.409 ± 0.040
	100 mg/kg	0.379 ± 0.050
Rutile NP	1 mg/kg	0.500 ± 0.036
	100 mg/kg	0.937 ± 0.267*
Common TiO2	1 mg/kg	0.693 ± 0.170
	100 mg/kg	0.530 ± 0.140

Note. Each group consisted of 4 animals. *p<0.05 in comparison with the control (nonparametric Mann-Whitney test).

Conclusions:

Liver titanium levels measured by ICP-MS in the E171 dosed group were not statistically different from non-dosed controls. Liver titanium levels in nanoparticle-dosed animals were contradictory: whereas anatase-treated animals had lower titanium levels in liver than the controls (1 and 100 mg/kg dose levels), the rutile-treated animals were reported to have a significant increase in liver titanium (100 mg/kg only).

The paper is considered of limited relevance for an assessment of human health risk, since the test material is insufficiently documented.

Reference 3

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

not specified

Objective of study:

absorption

Qualifier:

equivalent or similar to guideline

Guideline:

other: OECD Guideline 408 (Repeated Dose 90-day Oral Toxicity in Rodents)

Version / remarks:

not specified

Principles of method if other than guideline:

Titanium dioxide nanomaterial particles (80:20 anatase:rutile, 26 nm) were administered orally by gavage (vehicle: distilled water) for 13 weeks (7days/week) consecutively to male and female Sprague-Dawley rats (11 rats/group) at doses of 0 (vehicle), 260.4, 520.8 and 1041.5 mg/kg bw/day. The following tissues and body fluids were sampled: urine, faeces, blood, as well as tissue samples of liver, spleen, kidney and brain. The level of titanium in each sample was measured using inductively coupled plasma-mass spectrometry.

GLP compliance:

yes

Specific details on test material used for the study:

DURABILITY OF TEST ITEM IN BIOLOGICAL CONDITIONS
TiO2 nanoparticles were incubated with acidic gastric fluid (pH 1.5–2.0) or under basic conditions (pH 7.4). Acidic gastric fluid was prepared (Wurster et al., 1988)*. Briefly, 2.0 g NaCl and 3.2 g pepsin were dissolved in 1 L of distilled water and the pH was adjusted to 1.5 using 2 N HCl. Distilled water at pH 7.4 was used for the basic solution. Nanoparticles at 5 mg/mL were incubated in solution for up to 24 hours. The degree of ionization was evaluated at 24 hours postincubation using inductively coupled plasma-mass spectrometry (ICP-MS). Briefly, at 24 hours after incubation in solution, nanoparticle-free supernatants were collected by three rounds of centrifugation for 30 minutes. The concentrations of titanium were measured using ICP-MS (Cho, et al., 2010)*. Gross images were taken using a digital camera.

Results:
TiO2 nanoparticles showed minimal dissolution in both acidic gastric fluid and pH 7.4 conditions.

*References:
- Wurster DE, Burke GM, Berg MJ, Veng-Pedersen P, Schottelius DD: Phenobarbital adsorption from simulated intestinal fluid, U.S.P., and simulated gastric fluid, U.S.P., by two activated charcoals. Pharm Res 1988, 5(3):183–186.
- Cho WS, Cho M, Jeong J, Choi M, Han BS, Shin HS, Hong J, Chung BH, Cho MH: Size-dependent tissue kinetics of PEG-coated gold nanoparticles. Toxicol Appl Pharmacol 2010, 245(1):116–123.

Radiolabelling:

not specified

Species:

rat

Strain:

Sprague-Dawley

Details on species / strain selection:

not specified

Sex:

male/female

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: OrientBio Ltd. (Seongnam, Korea)
- Age at after acclimatisation period: 7-weeks old
- Diet (ad libitum): gamma-irradiated rodent diets (LabDiet 5002, PMI nutrition, Richmond, USA); titanium concentration in diet: unknown
- Water (ad libitum): autoclaved water; titanium concentration in water: unknown
- Acclimation period: seven days

ENVIRONMENTAL CONDITIONS
- Temperature: 21 – 24 ºC
- Humidity: 45 – 60%
- Photoperiod (hrs dark / hrs light): 12/12

Route of administration:

oral: gavage

Vehicle:

other: distilled water

Details on exposure:

PREPARATION OF DOSING SOLUTIONS:
Nanoparticles were suspended in the vehicle at each working concentration and orally administered at 10 mL/kg bw.
Administration volume was adjusted based on body weight measured each week.

Duration and frequency of treatment / exposure:

Durtation: 13 weeks
Frequency: 7 days/week

Dose / conc.:

260.4 mg/kg bw/day (actual dose received)

Dose / conc.:

520.8 mg/kg bw/day (actual dose received)

Dose / conc.:

1 041.5 mg/kg bw/day (actual dose received)

No. of animals per sex per dose / concentration:

11 rats (not specified how many of the 11 rats were males or females)

Control animals:

yes, concurrent vehicle

Positive control reference chemical:

not specified

Details on study design:

- Dose selection rationale: a single-dose acute toxicity study and a 14-day repeated toxicity study of the oral route were performed. For the 14-day study, TiO2 was administered orally at 520.8, 1041.5, and 2083 mg/kg/day. Survival, clinical signs, body weight change, haematological change, serum biochemical change, organ weight, and necropsy findings were evaluated for MTD endpoints. There were no treatment-related findings in the TiO2 study; the highest dose over 13 weeks was selected as 1041.5 mg/kg/day for studying long-term exposure.

Details on dosing and sampling:

TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled: urine, faeces, blood, as well as tissue samples of liver, spleen, kidney, and brain
Blood was taken at necropsy. Tissue samples were obtained and weighed. To collect urine and faeces, five animals/group were assigned to a metabolic cage immediately after gavage, and urine and faeces samples were collected for 24 hours. All samples were analyzed for elemental titanium to represent TiO2 concentrations. ICP-MS was used to analyze titanium concentrations in samples. The detection limit of ICP-MS was 0.1–1 ng/L for Ti.

OBSERVATIONS:
- body weights: weekly
- clinical signs/mortality: daily

Statistics:

ICP-MS data were analyzed using one-way ANOVA. Animal data were analyzed using TDMS ver 4.0. When statistically significant differences were indicated (p < 0.05), a Dunnett’s t-test was employed for comparisons between the control and treatment groups.

Type:

absorption

Results:

Titanium blood levels were approx. 0.5 µg/g at all dose levels including vehicle control, indicating low to negligible systemic uptake.

Results:

Titanium levels in all tissue samples (liver, spleen, kidney, and brain) showed no significant increase at any dose level.

Results:

Titanium levels in the urinary excretion were not significantly different from controls in any of the dose groups, with correspondingly high concentration in faeces.

Details on absorption:

The concentrations of titanium in the male TiO2-treatment groups showed a dose-response relationship. Female rats treated with TiO2 showed no dose-response relationship for blood concentration.
The distribution patterns of titanium in the TiO2-treatment groups in the liver, spleen, kidney, and brain showed no significant increase in the sampled organs compared with the vehicle control group.
The urine concentration of titanium in the TiO2-treatment groups showed no significant differences compared with the control group. Titanium concentrations in the faeces were very high compared to concentrations in the urine or tissues, with clear dose responses.

Details on distribution in tissues:

not specified

Details on excretion:

not specified

Metabolites identified:

not specified

Bioaccessibility (or Bioavailability) testing results:

not specified

Body weight changes:

- TiO2 nanoparticle-treatment groups showed no significant changes in body weight compared with the vehicle-control group.

Conclusions:

Titanium blood levels were approx. 0.5 ug/g at all dose levels including vehicle control, indicating low to negligible systemic uptake. Titanium levels in all tissue samples showed no significant increase at any dose level, and the urinary excretion likewise was not significantly different from controls in any of the dose groups, with correspondingly high concentration in faeces.

Reference 4

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

not specified

Objective of study:

absorption

Qualifier:

no guideline followed

Principles of method if other than guideline:

In this study four different TiO2 nanomaterials (NM-101, NM-102, NM-103, and NM-104) were administered to male (n=3/group, all four test materials) and female (n=3/group, only NM-101) Wistar rats by gavage on five consecutive days at a dose of 2.3 mg/animal. Upon sacrifice, samples from liver, spleen and mesenteric lymph nodes were taken for titanium analysis by ICP-MS.

GLP compliance:

not specified

Radiolabelling:

not specified

Species:

rat

Strain:

Wistar

Details on species / strain selection:

not specified

Sex:

male/female

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Charles River, Germany
- Age at study initiation (after acclimation period): 9 weeks
- Housing: housed in polycarbonate cages with a bottom area of 905 cm² and a height of 21.5 cm, with bedding
- Acclimation period: 2 weeks

Route of administration:

oral: gavage

Vehicle:

other: ethanol/rat serum albumin/phosphate buffer

Details on exposure:

PREPARATION OF DOSING SOLUTIONS:
Particle suspensions were made fresh every day according a standardized protocol. A 2.56 mg/mL stock suspension was prepared by pre-wetting the titanium dioxide powder in 96% ethanol (final concentration: 0.5 vol% ethanol) followed by dispersion in 0.05 wt% Rat Serum Albumin in ultrapure water. Probe sonication of the suspensions was performed on ice (Branson Sonifier S-450D equipped with a disruptor horn) for 16 minutes. The stock suspensions were diluted (9:1 v/v) with 10x concentrated phosphate buffer pH 7.4 (702 mg NaH2PO4 x 2H2O, 4155 mg Na2HPO4 x 7H2O, dissolved in 1 L) resulting in an exposure suspension with a concentration of 2.304 mg/mL TiO2. Vehicle control suspensions were also sonicated and diluted according to above description. All suspensions were used immediately. The TiO2 nanomaterials were evaluated for the presence of endotoxin. All preparations investigated showed endotoxin levels below 20 IU. Absence of LPS (endotoxin) was further evaluated and confirmed by fatty acid analysis.

Duration and frequency of treatment / exposure:

- duration: five consecutive days
- frequency: once a day

Dose / conc.:

2.3 other: mg/animal

Remarks:

dose range: 6.8 - 8.5 mg/kg bw/day (males) or 10.9 - 12.0 mg/kg bw/day (females)

No. of animals per sex per dose / concentration:

3 males/group; 3 females/group (NM-101 only)

Control animals:

yes, concurrent vehicle

Positive control reference chemical:

not specified

Details on study design:

- Dose selection rationale: the doses applied were the maxium concentration that resulted in a stable TiO2 nanomaterial dispersion. The doses are also considered to be in the realistic human exposure range (Weir et al. (2012); Lomer et al. (2000))*.

*References:
- Weir A, Westerhoff P, Fabricius L, Hristovski K, von Goetz N: Titanium dioxide nanoparticles in food and personal care products. Environ Sci Technol 2012, 46:2242–2250.
- Lomer MC, Thompson RP, Commisso J, Keen CL, Powell JJ: Determination of titanium dioxide in foods using inductively coupled plasma optical emission spectrometry. Analyst 2000, 125:2339–2343.

Details on dosing and sampling:

TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled (delete / add / specify): blood and tissues (spleen, liver and mesenteric lymph nodes)
- Time and frequency of sampling: rats were sacrificed and tissue sampling was done 24 hours after the last exposure (Day 6). The following tissues were collected: spleen, liver and mesenteria with mesenteric lymph nodes. Full wet weight of these tissues was noted. All tissues were cooled on dry ice and stored at −20°C.

The tissue distribution of the TiO2 nanomaterials administered to the animals was evaluated by determining the Ti content in tissues after administration by High-Resolution Inductively Coupled Plasma Mass Spectrometry (HR-ICP-MS; ELEMENT XR). Depending on the weight either the whole tissue or subsamples were used for the Ti analysis.

The Ti measurements were performed with a HR-ICP-MS using a set-up with on-line addition of an internal standard. Two Ti isotopes were measured in the medium resolution (MR) mode: while 47Ti(MR) was used for the evaluation, 49Ti(MR) was used for control. The accurate masses and abundances are: 47Ti (46.95176 atm): 7.3%, respectively 49Ti (48.94787 atm): 5.5%. The instrumental operating conditions were as follows: RF power 1225 W, cooling gas flow 16 L/min argon, auxiliary gas flow 0.9 L/min argon, and the sample gas flow 0.99 L/min argon.
The solutions were measured against an external calibration with internal standard correction using 115In. A commercial reference material was analysed according to the same procedure (L-2 LOT 1003129 contained 18 ± 5 μg/L Ti, and L-3 LOT 1112691 contained 12.8 ± 0.4 μg/L Ti). The limit of detection (LOD; as 3x STD, n=20) is estimated from the results obtained for tissue material from control animals of this study after applying the complete procedure of digestion and measurement with HR-ICPMS. For the tissue series evaluated for Ti content after oral administration (for five consecutive days) the LOD was 0.03 μg Ti/g tissue, and the LOQ 0.09 μg Ti/g tissue.

Statistics:

Pharmacokinetic analysis was performed on sparse data by a non-compartmental approach. The analysis focused on the organ kinetic parameters. Kinetic parameters such as AUC and apparent elimination rate (β) were determined. The AUC represents the ‘area under the curve’ of the organ Ti concentration – time curve, and was calculated using the linear trapezoidal rule. Apparent elimination rate (β) was estimated on the terminal phase of the curve by linear regression of at least 3 points (r2 > 0.95). The apparent elimination half-life (T1/2el) was calculated.
Standards errors were only estimated for Cmax and AUC.
Statistical analysis was performed using the nonparametric Friedman test followed by a pairwise comparison using Wilcoxon test to locate potential differences between groups (i.e. nanomaterials and tissues). The significance level was 0.05.
Data of male and female animals were evaluated separately, but data within the male animal groups or within the female animal groups were analysed together and the titanium levels were not corrected for differences in body weight and thus in dose within these groups.

Type:

absorption

Results:

Liver and spleen tissue samples contained very low titanium levels, mostly below the limit of detection of 0.03 μg/g.

Results:

Twenty-eight measurements were below the LOD, one was at the LOD and only one was above the LOD; it is worthy of note that even in control animals two measurements at the LOD were reported.

Results:

Whereas all samples of mesenteric lymph node tissue (MLN) contained amounts above the LOD, the authors noted that levels in MLN of TiO2 exposed animals were except for one material similar to control animals.

Results:

The authors overall conclude that “the results indicate that after oral administration absorption of TiO2 is very low.”

Details on absorption:

All liver and spleen tissue samples from rats exposed five times contained very low titanium levels, with levels mostly below the limit of detection of 0.03 μg/g. In exposed animals, twenty-eight measurements were below the LOD (0.03 μg/g tissue), 1 was at the LOD (NM-102 liver) and 1 was above the LOD (male liver NM-103). In control animals twice a measurement at the LOD was observed, while 10 times the level was below the LOD.
All samples of mesenteric lymph node (MLN) tissue contained amounts above the limit of detection. Lowest detected concentration was 0.07 μg Ti/g MLN or 0.11 μg Ti for the whole MLN, well above the LOD. Titanium levels in MLN of TiO2 exposed animals were, except for NM-104, similar to control animals. It appears
that female rats (control and NM-101) have a slightly higher Ti concentration in MLN determined per g/MLN, probably due to a lower weight of MLN and higher relative dosing of TiO2. MLN of females and males weighed on average 1.08 g and 1.56 g, respectively.

Details on distribution in tissues:

not specified

Details on excretion:

not specified

Metabolites identified:

not specified

Details on metabolites:

not specified

Bioaccessibility (or Bioavailability) testing results:

not specified

Conclusions:

The authors administered four different TiO2 nanomaterials to male (n=3/group, all four test materials) and female (n=3/group, only one test material) Wistar rats by gavage on five consecutive days at a dose of 2.3 mg/animal. Upon sacrifice, samples from liver, spleen and mesenteric lymph nodes (MLN) were taken for titanium analysis by ICP-MS. Liver and spleen tissue samples contained very low titanium levels, mostly below the limit of detection of 0.03 μg/g. Twenty-eight measurements were below the LOD, one was at the LOD and only one was above the LOD; it is worthy of note that even in control animals two measurements at the LOD were reported. Whereas all samples of MLN tissue contained amounts above the LOD, the authors themselves note that levels in MLN of TiO2 exposed animals were except for one material similar to control animals, and overall conclude that “the results indicate that after oral administration absorption of TiO2 is very low.”

Reference 5

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

not specified

Objective of study:

absorption

Qualifier:

no guideline followed

Principles of method if other than guideline:

A group of six male Sprague-Dawley rats were administered a single dose of 100 mg/kg of titanium dioxide nanoparticles (average size range 18 ± 8 nm, synthesized by flame spray pyrolysis) in MQ water containing sodium citrate at 2 mM. A vehicle control group was run concurrently. 24 hours after administration the rats were sacrificed and liver, spleen, mesenteric lymph nodes, Peyer’s patches, small intestine and caecum were samples. Tissues were analysed for titanium levels via ICP-MS and particles in tissues were visualised with TEM.

GLP compliance:

not specified

Specific details on test material used for the study:

not applicable

Radiolabelling:

not specified

Species:

rat

Strain:

Sprague-Dawley

Details on species / strain selection:

not specified

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Weight: 180 – 230 g
- Diet (ad libitum): Protein Rodent Maintenance Diet 2914C (Harlan)
- Water (ad libitum): autoclaved water

ENVIRONMENTAL CONDITIONS
- Temperature: 20 ± 2 °C
- Humidity: 55 ± 30 %
- Photoperiod (hrs dark / hrs light): 12/12

Route of administration:

oral: gavage

Vehicle:

other: MQ water containing sodium citrate at 2 mM

Details on exposure:

PREPARATION OF DOSING SOLUTIONS:
Titanium dioxide nanoparticles were dispersed at a concentration of 10 mg/mL in the vehicle. Dispersion was achieved by horn sonication.

DOSING VOLUME: 10 mL/kg

Duration and frequency of treatment / exposure:

One administration only

Dose / conc.:

100 other: mg/kg (actually received)

No. of animals per sex per dose / concentration:

6 male rats

Control animals:

yes, concurrent vehicle

Positive control reference chemical:

not specified

Details on study design:

- Dose selection rationale: the dose was selected on the basis of previous toxicological information from literature, so that it could be tolerated while high enough to enable analytical procedures.

Details on dosing and sampling:

TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled: spleen, liver, small and large intestines, and mesenteric lymph nodes
- Time and frequency of sampling: at termination

The intestines were washed with phosphate buffer (pH 7.4) to remove their content. Peyer’s patches were excised and separated from the small intestine. The caecum was separated from the large intestine. One of the Peyer’s patches and a piece of smooth small intestine were immediately preserved in buffered (pH 7.4) glutaraldehyde (2.5%) paraformaldehyde (2%) for later TEM analyses. The remaining samples were kept for chemical analyses.

Tissue samples were acid digested and analyzed by ICP-MS. Acidic digestion (hydrofluoric acid:nitric acid; 1:20) with an analytical microwave was performed to completely dissolve the nanoparticles into metal ions.

Peyer’s patch and smooth small intestine samples preserved in buffered glutaraldehyde-paraformaldehyde were fixed with OsO4.This was followed by serial dehydration through a series of acetone increasing concentrations, and embedding in Epon resin. After resin curing, 50 nm ultrafine cuts were obtained and and placed on Formvar carbon-coated Cu grids. These grids were further contrasted with uranyl acetate and lead citrate. All electron micrographs were obtained with a TEM. Images were obtained with AnalySIS on a Megaview III CCDcamera.

OBSERVATIONS:
- Body weight and clinical signs: on day of administration and on the following day, before sacrifice

Statistics:

T-test analyses (p < 0.05) were performed to evaluate statistical significance of differences between the control group and the treated group.

Type:

absorption

Results:

There was no statistically significant increase in titanium levels in any of the tissues sampled (liver, spleen, mesenteric lymph nodes, Peyer’s patches, small intestine and caecum) compared to controls.

Results:

No TiO2 NPs could be detected in smooth sections of the small intestine, but there were some TiO2 particles in at least one cell in a Peyer’s patch.

Details on absorption:

There was no statistically significant increase in titanium levels in any of the tissues (liver, spleen, mesenteric lymph nodes, Peyer’s patches, small intestine and caecum) evaluated 24 hours after the test item administration.
No titanium dioxide nanoparticles were found when smooth sections of the small intestine were examined by TEM. In contrast, at least one cell containing considerable amounts of titanium dioxide nanoparticles aggregates in a Peyer’s patch section were observed. In this cell, the titanium dioxide nanoparticles were not surrounded by membranes, but they were freely distributed in the cytoplasm. No nanoparticles were observed inside the mitochondria or the nucleus.

Details on distribution in tissues:

not specified

Details on excretion:

not specified

Metabolites identified:

not specified

Bioaccessibility (or Bioavailability) testing results:

not specified

Conclusions:

24 hours post administration, there was no statistically significant increase in titanium levels in any of the tissues sampled (liver, spleen, mesenteric lymph nodes, Peyer’s patches, small intestine and caecum) compared to vehicle controls. With TEM, no titanium dioxide nanoparticles could be detected in smooth sections of the small intestine, but there were some titanium dioxide particles in at least one cell in a Peyer’s patch.

Reference 6

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

not specified

Objective of study:

absorption

Qualifier:

no guideline followed

Principles of method if other than guideline:

Groups of seven young (3 weeks old) or seven adult (8 weeks old) male Sprague-Dawley rats were exposed to titanium dioxide nanoparticles (75 nm) in ultrapure water by gavage at dose levels of 0, 10, 50, 200 mg/kg bw/day once daily for 30 consecutive days. Control groups for each age group were run concurrently. Blood and serum as well as tissues and organs (such as liver, kidney, spleen, testis, lung, heart, brain, stomach and small intestine) were sampled. The following parameters were recorded/examined: clinical signs, mortality, body weight, organ mass, histology, blood biochemistry, biomarker assay, and biodistribution of nanoparticles.

GLP compliance:

not specified

Specific details on test material used for the study:

not applicable

Radiolabelling:

not specified

Species:

rat

Strain:

Sprague-Dawley

Details on species / strain selection:

not specified

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Department of Laboratory Animal Science, Peking University Health Science Center
- Age: 3 weeks old (young rats) or 8 weeks old (adult rats)
- Housing: kept in plastic cages
- Diet (ad libitum): commercial pellet diet
- Water (ad libitum): deionized water
- Acclimation period: one week

ENVIRONMENTAL CONDITIONS
- Temperature: 20 ± 2 °C
- Relative humidity: 50 – 70%
- Photoperiod (hrs dark / hrs light): 12/12

Route of administration:

oral: gavage

Vehicle:

other: ultrapure water

Details on exposure:

PREPARATION OF DOSING SOLUTIONS:
The titanium dioxide nanoparticles were dispersed in ultrapure water and ultrasonic vibrated for 15 minutes. In order to obtain homogenized suspension, the particle dispersion solution was stirred on vortex agitator before every use.

Duration and frequency of treatment / exposure:

Duration: 30 consecutive days
Frequency: once a day

Dose / conc.:

10 mg/kg bw/day (actual dose received)

Dose / conc.:

50 mg/kg bw/day (actual dose received)

Dose / conc.:

200 mg/kg bw/day (actual dose received)

No. of animals per sex per dose / concentration:

Young male rats: 7/group
Adult male rats: 7/group

Control animals:

yes

Positive control reference chemical:

not specified

Details on study design:

- Dose selection rationale: the intragastric doses were selected based on the intake of dietary titanium dioxide particles in the UK, which has been estimated to be about 5 mg/person/day (Powell et al., 2010)* equivalent to approx. 0.1 mg/kg bw/day. In this study, the 100 times dose of the potential human exposure (10 mg/kg bw) was used as the low-dose titanium dioxide nanoparticles exposure in rats.

*Reference:
- J. J. Powell , N. Faria , E. Thomas-McKay , L. C. Pele , J. Autoimmun. 2010, 34, J226.

Details on dosing and sampling:

TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled: blood and serum as well as tissues and organs (such as liver, kidney, spleen, testis, lung, heart, brain, stomach and small intestine)
- Time and frequency of sampling: after 30 days, all animals were sacrificed. Blood samples were collected from the femoral artery. Serum was harvested by centrifuging blood at 3000 rpm (1500 g) for 10 minutes. The tissues and organs were excised and weighed.

Elemental Content Analysis: about 0.1 – 0.3 g of each tissue (liver, kidney, spleen) were weighed and 1 mL of whole blood were taken accurately. All tissues and blood samples were digested and analyzed for elemental content. Briefly, prior to elemental analysis, the tissues and blood samples were digested in nitric acid (ultrapure grade) overnight. After adding H2O2, the mixed solutions were heated (~ 160 ° C) until the samples were completely digested. Then, the solutions were heated at 120 °C to remove the remaining nitric acid. At last, the remaining solutions were diluted with 2% nitric acid. Inductively coupled plasma mass spectrometry (ICP-MS) and inductively coupled plasma optical emission spectrometer (ICP-OES) was used to analyze the titanium, molybdenum, cobalt, strontium, manganese, rubidium, iron, copper, zinc, calcium, magnesium, sodium, potassium and phosphorus concentration in the samples.

TEM observation: liver and the mucosa of the stomach and small intestine were cut up into pieces (1 mm³) and fixed in 2.5% glutaraldehyde (pH 7.4) overnight. Then the samples were treated according to the general protocols for TEM study. The ultra-thin sections (70–100 nm) were stained with lead citrate and uranyl acetate followed by examination using electron microscopy.

OBSERVATIONS/EXAMINATIONS:
The following parameters were observed and recorded:
- clinical symptoms: every day for 30 days
- mortality: every day for 30 days
- body weight: every 4 - 6 days as well as at sacrifice

- coefficients of organs: after weighing the body and tissues, the coefficients of the liver, kidney, spleen, testis, lung, heart, brain to the body weight were calculated as the ratio of tissues
to body weight.

- blood biochemical assay: the subsequent endpoints were investigated by measuring the following parameters in the serum.
1) liver function: total protein, albumin, globulin, albumin/globulin ratio, alanine aminotransferase, aspartate aminotransferase, alanine aminotransferase/aspartate aminotransferase ratio, and total bilirubin.
2) nephrotoxicity: blood urea nitrogen and creatinine.
3) glycolipid metabolism: fasting blood glucose, total cholesterol, triglycerides, high density lipoprotein cholesterol, and low density lipoprotein cholesterol.
4) cardiac damage: enyzmes of creatine kinase, lactate dehydrogenase and alpha-hydroxybutyrate dehydrogenase.

- histological examination: all histopathological examinations were performed using standard laboratory procedures. The tissues were embedded in paraffin blocks, sliced (5 μm), placed onto glass slides, and stained (hematoxylin–eosin staining). Slides were observed.
The mast cells in the stomach and small intestine tissues were identified histochemically by using a toluidine blue staining method. Intact or partially degranulated mast cells were counted. The mast cell density was calculated (mast cell numbers/square millimeter of tissue).

- biomarker assay: concentrations of histamine and IgE in serum were measured by enzyme-linked immunosorbent assay kits.
Plasma was harvested via centrifugation of heparin-anticoagulated whole blood at 3000 rpm (1500 g) for 10 minuntes. The levels of D-lactate and the activity of diamine oxidase in plasma were detected to evaluate the permeability and intergrity of intestinal barrier. The D-lactate levels was measured using the colorimetric method and the activity of diamine oxidase was determined by the reaction of cadaverine dihydrochloride.
For redox state assay, the levels of reduced glutathione and oxidized glutathione were measured using o-phthalaldehyde as fluorescent reagent according to the method of Hissin and Hilf. [ 40 ] The GSH/GSSG ratio were calculated from the GSH and GSSG concentrations. The activity of glutathione peroxidase were tested.

Statistics:

Data were expressed as means ± SD. Independent-samples T test was used to assess the significant difference between two experimental groups. One-way variance (ANOVA) with LSD or Games-Howell tests was applied to evaluate the statistical significance of differences between the experimental groups and the controls. A p value less than 0.05 was considered to be statistically significant.

Type:

absorption

Results:

There were no statistically siginifant differences in levels of titanium either in blood or in livers, kidneys or in spleen after administration of titanium dioxide nanoparticles to young and adult rats.

Details on absorption:

The TEM images showed that some particles with a diameter of approx. 60–200 nm were adhered to the intestinal villi and absorbed in the stomach and small intestine tissue. No particles were found in the liver of young and adult rats. The contents of titanium in the blood, liver, kidney and spleen of young and adult rats in the titanium dioxide nanoparticles-treated groups were not significantly different to the control group after oral exposure to 200 mg/kg bw of the test item for 30 days.

Details on distribution in tissues:

not specified

Details on excretion:

not specified

Metabolites identified:

not specified

Bioaccessibility (or Bioavailability) testing results:

not specified

OBSERVATIONS/EXAMINATIONS:

- clinical signs: daily behaviors such as eating, drinking and activity in test item treated groups were as normal as the control group.

- mortality: death was not observed.

- body weight: During administration, the body weight from each animal was increased: no significant differences in the body weight of the exposed young or adult rats were found compared to the control groups.

- coefficients of organs: significant differences were not observed in the coefficients of the liver, kidney, spleen, testis, lung, heart, and brain of the adult rats. For the young rats, the coefficients of heart in the 50 mg/kg bw titanium dioxide nanoparticles-treated group is significantly lower than the control group.

- histological examination: in young rats, liver oedema was evident in the 50 and 200 mg/kg bw titanium dioxide nanoparticles treated groups. Histology showed hepatic cord disarray, perilobular cell swelling, vacuolization, or hydropic degeneration. In adult rats no liver oedema were observed. Only inflammatory cells infiltration was observed in adult rat liver in the 10 and 50 mg/kg bw titanium dioxide nanoparticles treated groups. No obvious pathological changes were found in the kidney, stomach, small intestine, spleen, testis, lung and heart of the young and adult rats.

Titanium dioxide nanoparticles exposure induced an increase in the number of mast cells in stomach tissues, but no signifi cant changes were found in the small intestine. Especially in young rats, the number of mast cells in stomach tissues showed a dramatic enhancement in the 200 mg/kg bw dose group compared with the control, but the levels of histamine and IgE in serum were not elevated after test item exposure.

- blood biochemical assay: no obvious changes for serum biochemical parameters in adult rats after titanium dioxide nanoparticles exposure were observed except in the serum total bilirubin (significant decreased in 200 mg/kg bw dose group) and blood urea nitrogen levels (significant increased in 50 and 200 mg/kg bw dose groups).

In the 50 and 200 mg/kg bw dose groups of test item treated groups of young rats, the levels of blood glucose, low density lipoprotein cholesterol, the ratio of alanine aminotransferase to aspartate aminotransferase and total bilirubin were significantly higher than those of the control group, and the activities of AST, alpha-hydroxybutyrate dehydrogenase and creatine kinase were significantly reduced. There were no significant changes for blood urea nitrogen and creatinine levels in serum after oral administration of test item in young rats.

- biomarker assay: decreased levels of D-lactate and lowered activity of diamine oxidase in plasma were observed in test item treated groups, especially in the adult rats.

Titanium dioxide nanoparticles exposure significantly increased plasma reduced glutathione/ oxidized glutathione ratios in both young and adult rats. Young rats had higher levels of reduced glutathione and adult rats had reduced oxidized glutathione. Furthermore, the decreased activities of glutathione peroxidase in test item-exposed group were responsible for the low oxidized glutathione levels and high reduced glutathione/ oxidized glutathione ratios in adult rats. On the contrary, the activities of glutathione peroxidase in test item exposed group of young rats were increased with the high reduced glutathione contents and reduced glutathione/ oxidized glutathione ratios. Elevated levels of glucose were found in test item-exposed group of young rats. The higher reduced glutathione/ oxidized glutathione ratios in young rats were associated with higher glucose levels.

- element analysis: results exhibited no obvious differences in elemental content except a significant decrease in the content of molybdenum in adult rats

Conclusions:

There were no statistically siginifant differences in levels of titanium either in blood, livers, kidneys or in spleen after the administration of titanium dioxide nanoparticles (75 nm) to young and adult male Sprague-Dawley rats.

Reference 7

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

Objective of study:

distribution

Qualifier:

no guideline followed

Principles of method if other than guideline:

Mice were orally exposed to micro- and nano-sized TiO2 particles by a syringe via gastrointestinal tract, respectively. Body and organ weights were assessed and tissue distribution analysed.

GLP compliance:

not specified

Specific details on test material used for the study:

Before application, TiO2 suspension was stirred.

Radiolabelling:

no

Species:

mouse

Strain:

CD-1

Sex:

male/female

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Beijing Vitalriver Experimental Animal Technology Co. Ltd.
- Weight at study initiation: 19 ± 2 g
- Housing: animals were housed in stainless steel cages by sex in a ventilated animal room
- Diet ad libitum
- Water ad libitum: distilled water
- Acclimation period: 5 days prior dosing
- Fasting period before study: over night

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 20 ± 2
- Humidity (%): 60 ± 10
- Photoperiod (hrs dark / hrs light): 12/12

Route of administration:

oral: gavage

Vehicle:

other: 0.5 % Hydroxypropylmethylcellulose (HPMC)

Details on exposure:

PREPARATION OF DOSING SOLUTIONS:
- Stirring before administration of dosing suspension
- Suspension was administered by a syringe via gastrointestinal tract in a minute
- Food and water were provided 2 hr after adiministration

Duration and frequency of treatment / exposure:

single application

Dose / conc.:

5 000 mg/kg bw (total dose)

No. of animals per sex per dose / concentration:

10 animals/sex per group (groups: control, 25 nm, 80 nm, 155 nm)

Control animals:

yes, concurrent vehicle

Details on dosing and sampling:

TOXICOKINETIC STUDY (distribution)
- Tissues and body fluids sampled for tissue distribution: liver, kidneys, spleen, lung, brain, red blood cells (females only)
- Time and frequency of sampling for tissue distribution: once, 14 days post exposure
- Tissue distribution was analysed by measurement of titanium content via inductively coupled plasma-mass spectrometry (ICP-MS), detection limit was 0.074 ng/mL

- Tissues and body fluids sampled for assessment of coefficients: liver, kidneys, spleen (males and females)
- Time and frequency of sampling for assessment of coefficients: once, 14 days post exposure

Statistics:

Results were expressed as mean ± standard deviation (S.D.). Multigroup comparisons of the means were carried out by one-way analysis of variance (ANOVA) test. Dunnett’s test was used to compare the differences between the experimental groups and the control group. Student’s t-test was used to compare the means of each nano-group and the corresponding fine group. The statistical significance for all tests was set at p < 0.05.

Type:

other: coefficient (wet weight of tissues/g (body weight)

Results:

Coefficient of liver significant increased in 25 and 80 nm group (females only)

Type:

distribution

Results:

accumulation of titanium (25 nm sized) increased significantly in spleen, kidneys and brain

Type:

distribution

Results:

accumulation of titanium (80 nm sized) incresed significantly in liver, spleen, kidneys, lung and brain

Type:

distribution

Results:

accumulation of titanium (155 nm sized) increased significantly in spleen and brain

Details on absorption:

not applicable

Details on distribution in tissues:

- Accumulation of titanium (25 nm sized) increased significantly in spleen, kidneys and brain
- Accumulation of titanium (80 nm sized) incresed significantly in liver, spleen, kidneys, lung and brain
- Accumulation of titanium (155 nm sized) increased significantly in spleen and brain

Details on excretion:

not applicable

Metabolites identified:

no

Details on metabolites:

Titanium dioxide as an inorganic substance, which is not metabolised.

Bioaccessibility (or Bioavailability) testing results:

not applicable

Conclusions:

Bioaccumulation potential cannot be judged based on study results.
No data can be derived for the assessment of an oral absorption factor for use in HH risk assessment.

Reference 8

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

2013-06-18 to 2014-03-03

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Remarks:

Screening study only

Objective of study:

toxicokinetics

Qualifier:

according to guideline

Guideline:

OECD Guideline 417 (Toxicokinetics)

Version / remarks:

adopted 2010-07-22

Deviations:

yes

Remarks:

pilot study: use of male rats only, one dose level, and 3 rats per the test substance-treated group

GLP compliance:

yes

Specific details on test material used for the study:

STABILITY AND STORAGE CONDITIONS OF TEST MATERIAL
- Stability under test conditions: the test substance appeared to be stable under the conditions of the study; no evidence of instability was observed.

Radiolabelling:

no

Species:

rat

Strain:

other: Crl:CD(SD)

Details on species / strain selection:

The Crl:CD(SD) rat has been selected based on extensive experience with this strain at the performing laboratory, its suitability for toxicokinetic studies, and comparison with companion toxicity studies.

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Charles River Laboratories International, Inc. (Raleigh, North Carolina, U.S.A.)
- Age at study initiation: approximately 11 weeks old
- Weight at study initiation: 287.7 to 305.3 g (mean ± SD =297 ± 8.9 g)(single vehicle control rat weighed 312.0 g)
- Fasting period before study: animals were fasted for approximately 16 (± 1) hour before dosing with test substance. Food was returned approximately 2 hours post-dose.
- Housing: upon arrival at the laboratory animals were housed individually in solid-bottom caging with Bed-o'Cobs® bedding and Nestlets as enrichment. Animals were moved to glass metabolism units for the in-life phase of the study.
- Metabolism cages: yes, animals were moved to glass metabolism units for the in-life phase of the study.
- Diet (ad libitum): PMI® Nutrition International, LLC Certified Rodent LabDiet® 5002
- Water (ad libitum): tap water
- Quarantine period: quarantined for 3 full days, within a 5 day period of initial receipt and pre-dose administration.

ENVIRONMENTAL CONDITIONS
- Temperature: 20-26ºC
- Relative humidity: 30-70%
- Photoperiod (hrs dark / hrs light): approximate 12 hour light/dark cycle

Route of administration:

oral: gavage

Vehicle:

other: sterile water for injection

Duration and frequency of treatment / exposure:

one administration

Dose / conc.:

500 other: mg/kg bw

No. of animals per sex per dose / concentration:

Vehicle control group: one male rat
Test substance group: three male rats

Control animals:

yes, concurrent vehicle

Positive control reference chemical:

none

Details on study design:

One male rat with a jugular vein cannula was administered the vehicle (3.125 g) by oral gavage at the same dose vehicle volume as the rats dosed with the test substance. The dose vehicle volume (10 mL/kg bw) matched that used for the micronucleus test (please refer to the endpoint study record "NANO_k_Myhre_2014" in Section 7.6.2 Genetic toxicity in vivo).

Three male rats, each with a jugular vein cannula, were administered the test substance formulated with dose vehicle at an expected dose of 500 mg/kg body weight (491, 486, and 508 mg/kg bw, respectively; mean ± SD = 495 ± 11.1 mg/kg bw). The dose volume was 10 mL/kg bw. The dose administration was performed using a portion of the lowest (50 mg/mL, nominal) dose used in the micronucleus test (please refer to the endpoint study record "NANO_k_Myhre_2014" in Section 7.6.2 Genetic toxicity in vivo).

Details on dosing and sampling:

PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
Vehicle control group and test substance group:
- Tissues and body fluids sampled: urine, faeces, blood, liver
- Time and frequency of sampling: urine and faeces were collected prior to dosing (pre-dose) and at 0-6, 6-12, 12-24, and 24-48 hour intervals over dry ice. Blood (~0.2 mL) was collected at 0, 0.5, 1, 2, 4, 6, 8, 12, 24, and 48 hours via the jugular vein cannula.

Vehicle control group:
- at 48 hours, the rat was sacrificed. The liver was collected. The liver and 48-hour blood sample were submitted for Ti analysis.

Test substance group:
- rats were sacrificed at 48 hours after dose administration. The livers were collected. The blood (from the jugular vein cannulas), liver, urine, and faeces samples were frozen and submitted for Ti analysis. At sacrifice, blood was collected by exsanguination from the vena cava or cardiac puncture.

TITANIUM ANALYSIS
- faeces: samples were prepared for analysis by first ashing the entire sample in a muffle furnace. The ashed sample was then transferred to a microwave vessel and prepared in a chemical microwave using nitric and hydrofluoric acid and then analysed by Inductively Coupled Plasma-Atomic Emission Spectroscpoy (ICP-AES).
- liver: samples were digested in a chemical microwave using nitric acid and hydrogen peroxide and then analysed by Inductively Coupled Plasma-Mass Spectroscopy (ICP-MS).
- blood: samples were ashed in a muffle furnace, adding nitric acid and hydrogen peroxide at the conclusion of the ashing and then analysed by ICP-MS.
- urine: samples were diluted in a dilute nitric acid solution and then analysed by ICP-MS.

Statistics:

Group data were represented as Mean ± SD. Individual samples less than the limit of quantitation (LOQ) were excluded from the mean. Results for the faeces were expressed as percent of the administered dose.

Type:

absorption

Results:

TiO2 concentration (blood): prior to dosing: 0.704 to 1.36 µg/g blood (mean ± SD: 0.978 ± 0.341 µg/g; in treated rats the blood concentrations ranged from 0.088 to 2.22 μg/g across all post-dose time points.

Type:

distribution

Results:

TiO2 concentration (liver): a point estimate of the liver concentration (0.329 μg/g) in the control rat was essentially the same as the mean concentration for treated rats (0.253 ± 0.024 μg/g) at 48 hours after dose administration.

Type:

excretion

Results:

TiO2 concentration (faeces): prior to dosing: <16.7 to 35.4 μg/g faeces; after dosing: 289 to 30,700 μg/g faeces; peak concentrations occurred in the 6-12 or 12-24 hour samples; overall mean recovery was 97.1 ± 9.1% of the administered dose amount.

Type:

excretion

Results:

TiO2 concentration (urine): prior to dosing: 0.045 to 0.159 μg/g urine; after dosing: 0.052 to 0.128 μg/g urine; concentrations of TiO2 in urine were very low and showed overlap between pre-dose (1.48 ± 0.478 μg) and post-dose samples ((1.46 ± 0.435 μg).

Details on absorption:

- blood: individual animal pre-dose blood samples had TiO2 concentrations that ranged from 0.704 to 1.36 μg/g blood with a mean and SD of 0.978 ± 0.341 μg/g. A single measurement at 48 hours in a control animal was 0.637 μg/g blood. In treated rats the blood concentrations ranged from 0.088 to 2.22 μg/g across all post-dose time points. The appearance of minimal increase and similarity of the concentrations across time relative the control (0 hour) sample indicates very low potential for systemic exposure from orally administered TiO2 uf-1. Please also refer to table 3 in the field "Any other information on results incl. tables" below.

Details on distribution in tissues:

- liver: a point estimate of the liver concentration (0.329 μg/g) in the control rat was essentially the same as the mean concentration for treated rats (0.253 ± 0.024 μg/g) at 48 hours after dose administration. Please also refer to table 3 in the field "Any other information on results incl. tables" below.

Details on excretion:

- faeces: individual animal samples collected prior to dose administration had TiO2 concentrations that ranged from <16.7 to 35.4 μg/g faeces. The concentrations after dosing ranged from 289 to 30,700 μg/g faeces. The peak concentrations occurred in the 6-12 or 12-24 hour samples. The overall mean recovery was 97.1 ± 9.1% of the administered dose amount. Please also refer to table 1 in the field "Any other information on results incl. tables" below.

- urine: individual animal samples collected prior to dose administration had TiO2 concentrations that ranged from 0.045 to 0.159 μg/g urine. The concentrations after dosing ranged from 0.052 to 0.128 μg/g urine. Adjusting for the amount of urine produced gave a mean TiO2 amount of 1.48 ± 0.478 μg for the pre-dose group. This value was essentially the same as the amount recovered from 0-24 hours after dose administration (1.46 ± 0.435 μg). Please also refer to table 2 in the field "Any other information on results incl. tables" below.

Metabolites identified:

not measured

Bioaccessibility (or Bioavailability) testing results:

no data

Table 1: TiO2 uf-1 analysis in faeces

Parameter by subject
Parameter	Time	Animal #1	Animal #2	Animal #3	Mean	SD
Concentration (µg/g faeces)	Predose	<16.7	<16.7	35.4	35.4	NA
	0 – 6 h	NA	NA	NA	NA	NA
	6 – 12 h	30700	9060	28900	22900	12000
	12 – 24 h	5120	12000	1490	6200	5340
	24 – 48 h	522	959	289	590	340

 

Table 2: TiO2 uf-1 analysis in urine

Parameter by subject
Parameter	Time	Animal #1	Animal #2	Animal #3	Mean	SD
Concentration (µg/g urine)	Predose*	0.045	0.159	0.105	0.103	0.057
	6	0.117	0.105	0.052	0.091	0.035
	12	0.058	0.122	0.068	0.083	0.034
	24	0.065	0.128	0.093	0.095	0.032
	48	2.39**	0.095	0.087	0.091	0.004

* Predose collection = approximately 16 hours based on start of fasting to time of dosing

** Excluded from mean and SD. Reason for high value likely due to contamination of urine with faeces.

Table 3: TiO2 uf-1 analysis in blood and liver

Concentration (µg/g)
Tissue	Hours	Animal #1	Animal #2	Animal #3	Mean	SD
Treated Blood	0*	0.704	0.871	1.36	0.978	0.341
	0.5	0.786	0.442	0.526	0.585	0.179
	1	1.60	0.088	0.369	0.686	0.804
	2	0.482	1.56	0.539	0.860	0.607
	4	1.10	0.619	1.05	0.923	0.264
	6	0.364	0.893	0.445	0.567	0.285
	8	1.74	0.240	0.467	0.816	0.809
	12	0.526	2.09	0.509	1.04	0.908
	24	0.425	2.22	1.10	1.25	0.907
	48	0.427	0.315	1.24	0.658**	0.447
Liver	48	0.247	0.280	0.233	0.253	0.024
Control
Blood	48	0.637
Liver	48	0.314	-	-	0.329***	NA

* Collected prior to dose administration

** Mean and SD include 48 hour repeat analysis replicates

*** Mean of two replicates

Conclusions:

A screening study was conducted with one control and three treated rats, the latter of which were administered a mean TiO2 uf-1 dose of 495 (±11) mg/kg bw. The majority of dose was recovered in faeces (97.1 ± 9.1%) by 48 hours after administration with peak amounts occurring at 6-12 and 12-24 hours. Concentrations of TiO2 in urine were very low and showed overlap between pre-dose and post-dose samples. Like urine, blood concentrations of TiO2 were also very low and overlapped between pre-dose (0.704 to 1.36 μg/g) and post-dose (0.088 to 2.22 μg/g) time points.
Liver concentrations were similar between the single control (0.329 μg/g) and three treated (0.253 ± 0.024 μg/g) rats 48 hours after dose administration. Collectively these results demonstrate very low potential for absorption of the test material and nearly quantitative recovery of the dose in faeces.

Reference 9

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

not specified

Reliability:

4 (not assignable)

Rationale for reliability incl. deficiencies:

abstract

Objective of study:

absorption

Qualifier:

no guideline followed

Principles of method if other than guideline:

Rats were orally treated with titanium oxide nanoparticles (21 nm; P-25; Aeroxide®) (dose levels: 10 and 100 mg/kg) for five consecutive days. General toxicity, blood chemistry, and serum biochemical analysis were analysed. Titanium concentration in liver, kidney, lung, urine and faeces were measured and histopathology was performed in these organs as well as organ weights were determined.

GLP compliance:

not specified

Specific details on test material used for the study:

not applicable

Radiolabelling:

not specified

Species:

rat

Strain:

Sprague-Dawley

Details on species / strain selection:

not specified

Sex:

not specified

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Weight (mean value):
control: 225.0 g
10 mg/kg dose level: 223.75 g
100 mg/kg dose level: 222.50 g

ENVIRONMENTAL CONDITIONS
- Temperature: 23 ± 3 °C
- Humidity: 50 ± 10 %
- Photoperiod (hrs dark / hrs light): 12/12

Route of administration:

oral: unspecified

Vehicle:

not specified

Details on exposure:

not specified

Duration and frequency of treatment / exposure:

five consecutive days

Dose / conc.:

10 other: mg/kg

Dose / conc.:

100 other: mg/kg

No. of animals per sex per dose / concentration:

4 rats

Control animals:

yes

Positive control reference chemical:

not specified

Details on study design:

not specified

Details on dosing and sampling:

TOXICOKINETIC / PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
- Tissues and body fluids sampled: urine, faeces, blood chemistry (white blood cell, red blood cell, haemoglobin, haematocrit, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, platelet count, lymphocytes, monocytes, eosinophils, basophils, and reticulocytes) serum biochemical analysis (Total protein, Albumin, Albumin/Globulin ratio, Blood urea nitrogen, Creatinine, Glucose, Aspartate aminotransferase, alanine aminotransferase, alkaline phosphate, total cholesterol, triglyceride, calcium, sodium, potassium, and chloride as well as tissue samples (liver, kidney, and lung)
Faeces or urine was collected every day for 24 hours from each rat in metabolic cages.
Accumulation of titanium was measured in the tissue samples, urine, faeces and blood.
Animals were sacrificed 24 hours after last treatment.

OBSERVATIONS:
- general toxicity was observed
- body weight changes were determined
- histopathology was conducted in liver, kidney, and lung
- organ weights of liver, kidney and lung were determined

Statistics:

Mean ± SD
one-way ANOVA test
Dunnett

Type:

absorption

Results:

There was no statistically significant increase in titanium levels in any of the organs sampled (liver, kidney, and lung).

Results:

Excretion was almost exclusively via faeces, and negligible via urine.

Details on absorption:

There was no statistically significant increase in titanium levels in any of the organs sampled (liver, kidney, and lung).
Excretion was almost exclusively via faeces, and negligible via urine (note: results presented only graphically, no tabulated data).

Details on distribution in tissues:

not specified

Details on excretion:

not specified

Metabolites identified:

not specified

Bioaccessibility (or Bioavailability) testing results:

not specified

Induction of toxicological parameters was not observed.

Histopathology:

- no abnormal findings were observed in either livers or kidneys of all the groups.

- no test material-related lesions were found in the lungs

Conclusions:

There was no statistically significant increase of titanium levels in liver, kidney, and lung. Excretion was almost exclusively via faeces, and negligible via urine (results presented only graphically, no tabulated data).

Reference 10

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

Objective of study:

other: pulmonary clearance

Qualifier:

no guideline followed

Principles of method if other than guideline:

Rats were exposed to two grades of titanium dioxide via whole body inhalation for 12 weeks. Influence of lung burden on pulmonary clearance was analysed. Pulmonary clearance was observed using radiolabeled tracer particles (85Sr-polystyrene microspheres (diameter: 3.3 µm) administered either via intratracheal inhalation or intratracheal instillation.

GLP compliance:

not specified

Radiolabelling:

no

Species:

rat

Strain:

Fischer 344

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Weight at study initiation: 220 - 250 g

Route of administration:

inhalation: aerosol

Vehicle:

unchanged (no vehicle)

Details on exposure:

TYPE OF INHALATION EXPOSURE: whole body

GENERATION OF TEST ATMOSPHERE / CHAMBER DESCRIPTION
- Exposure apparatus: whole body exposure chamber

Duration and frequency of treatment / exposure:

6 hr/day, 5 days/week for 12 weeks

Dose / conc.:

22.3 mg/m³ air

Remarks:

pigment grade particles (SD 4.2)

Dose / conc.:

23.5 mg/m³ air

Remarks:

ultrafine particles (SD 2.9)

No. of animals per sex per dose / concentration:

12 animals/dose: 4 animals/dose for lung burden analysis, 4 animals/dose for pulmonary clearance after intratracheal inhalation of tracer particles or instillation, respectively.

Control animals:

yes, sham-exposed

Details on study design:

Different treatments (pigment-grade and ultrafine TiO2) were selected to cause significant impairment of alveolar machrophage-mediated particle clearance by i) lung particle overload due to a high dose of low-toxicity particles of low in vivo solubility (pigment-grade TiO2) and ii) a particle of low-toxicity material but with high specific surface area which is known to increase pulmonary toxicity (ultrafine TiO2)

Details on dosing and sampling:

TOXICOKINETIC
- Tissues sampled: lungs of 4 rats for measurement of lung burden
- Time and frequency of sampling: after 12 weeks of exposure

- Determination of deposited dose of tracer particles (administered after TiO2 exposure for 12 weeks): radioactivity was directly count after administration of tracer particles
- Determination of thoracic activity was counted daily (days 1-8), then twice a week for 5 weeks, then once a week up t day 180 post tracer particle exposure

Statistics:

- One-way ANOVA (comparison of 4 treatments for short and long half-times, seperately for tracer particle inhalation and installation)
- Two-way ANOVA (comparison of 4 treatmentsacross two administration modes, seperately for short and long half-times)
- Tukey method of multiple comparisons at the 5 % level of signficance (comparison of individual groups)

Type:

other: lung burden

Results:

for pigment-grade and ultrafine TiO2 lung burdens were 6.6 ± 1.2 mg and 5.2 ± 0.8 mg, respectively.

Type:

other: pulmonary tracer clearance (85Sr polystyrene microspheres)

Results:

Pigment-grade TiO2: Retention half-times in fast and slow clearing compartment were 1.66/10.9 and 117/99 days, respectively (intratracheal inhalation method/ intratracheal instillation method).

Type:

other: pulmonary tracer clearance (85Sr polystyrene microspheres)

Results:

Ultrafine TiO2: Retention half-times in fast and slow clearing compartment were 1.49/4.7 and 541/606 days, respectively (intratracheal inhalation method/ intratracheal instillation method).

Details on absorption:

not applicable

Details on distribution in tissues:

not applicable

Details on excretion:

not applicable

Metabolites identified:

no

Details on metabolites:

Titanium dioxide as an inorganic substance, which is not metabolised.

Bioaccessibility (or Bioavailability) testing results:

not applicable

Conclusions:

Administration methods resulted in any relevant differences for long-term clearance (alveolar macrophage-mediated clearance), but short-term clearance (mucociliary clearance) showed high variability within the groups when instillation method was used. Thus intratracheal instillation is not recommended for analysis of short-term clearance parameters.
Retention half-times of tracer particles in slow clearing compartement (alveolar macrophage-mediated clearance) differed significantly, with 99 days and 606 days for pigment-grade and ultrafine TiO2 , respectively.

Reference 11

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:

purity and source of test material lacking

Objective of study:

other: lung clearance

Qualifier:

no guideline followed

Principles of method if other than guideline:

The deposition, retention, and clearance of titanium dioxide particles in male Long-Evans rats under varying conditions were determined.

GLP compliance:

not specified

Specific details on test material used for the study:

not applicable

Radiolabelling:

not specified

Species:

rat

Strain:

Long-Evans

Details on species / strain selection:

not specified

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Weight: ~ 200 g

Route of administration:

inhalation: dust

Vehicle:

other: air

Details on exposure:

NOTE: three different experiments were conducted with different experimental conditions.

GENERATION OF TEST ATMOSPHERE / CHAMBER DESCRIPTION
- Exposure apparatus: animals inhaled the test item in an exposure chamber (described by Leach et al., 1959)*.

*Reference:
- Leach, L. F. et al. A multiple chamber exposure unit designed for chronic inhalation studies. Amer. Ind. Hyg. J. 20: 13 (1959).

Duration and frequency of treatment / exposure:

Experiments 1 and 3:
Duration: 7 hours
Frequency: once

Experiment 2:
Duration: different time period (not clearly stated)
Frequency: once

Remarks:

7.5 to 50 mg/m³ (Experiment 1)

Dose / conc.:

54 mg/m³ air

Remarks:

Experiment 2

Dose / conc.:

15 mg/m³ air

Remarks:

Experiment 3

No. of animals per sex per dose / concentration:

about 10 male rats

Control animals:

not specified

Positive control reference chemical:

not specified

Details on study design:

not specified

Details on dosing and sampling:

- Tissues and body fluids sampled: lungs
The titanium dioxide retained in the lungs was determined by a photometric method using 4,4-diantipyrymethane monohydrate.
For histology, the lungs were fixed with a 10 % buffered formalin solution. The formalin was injected intratracheally into the lungs in situ before opening the thorax. With the trachea ligated, the lungs were subsequently removed from the animals and immersed in formalin.

Statistics:

Lung deposition was calculated.

Preliminary studies:

not specified

Type:

other: lung clearance

Results:

The results showed that within the limitation of animal variabillity, experimental conditions, and the titanium dioxide particle parameters used, a deposition within 10 to 20 % can be predicted.

Results:

The retention half-time after exposure is 14 days for approx. the first 8 days and 88 days thereafter.

Details on absorption:

not specified

Details on distribution in tissues:

not specified

Details on excretion:

Experiment 1 (exposure duration: 7 hours; concentrations: 7.5 to 50 mg/m³):
- correlation between deposition and titanium dioxide concentration is very close.

Experiment 2 (exposure duration:different time periods; concentration: 54 mg/m³):
- correlation between exposure time and deposition is very close.

Experiment 3 (exposure duration: 7 hours; concentration: 15 mg/m³):
- results show a very close correlation between wet lung weight and the mass of titanium dioxide deposited.

The values of the retained titanium dioxide for up to approx. 140 days postexposure were given. Three experiments were performed with average initial lung burdens of 60, 139, 247 µg of titanium dioxide. Retention is expressed in percent of these lung burdens; the difference between 100 % and the actual retention value is therefore the amount eliminated. The biphasic retention half-times in these three experiments were about the same, with 14 days applying to the first 8 days, and 88 days, thereafter.

Metabolites identified:

not specified

Bioaccessibility (or Bioavailability) testing results:

not specified

Conclusions:

According to the authors, the results showed that within the limitation of animal variabillity, experimental conditions, and the titanium dioxide particle parameters used, a deposition within 10 to 20 % can be predicted. Furthermore, they stated that the retention half-time after exposure is 14 days for approx. the first 8 days and 88 days thereafter.

Reference 12

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:

Purity and source of test material lacking

Objective of study:

other: lung clearance

Qualifier:

no guideline followed

Principles of method if other than guideline:

Male Long-Evans and Fisher rats were exposed to a titanic dioxide (anatase; TiO2) concentration of 14.9 mg/m³ air via whole body inhalation. The administration was conducted once for a duration of 7 hours. Particle clearance was assessed by killing groups of 10 animals/species/dose/time of sacrifice at days 1, 8, 25, and 130 postexposure and analysing for titanium dioxide content via photometry. Furthermore, the following data were collected: body weight, lung weight, TiO2 distribution in lung lobes, background TiO2 analysis in unexposed rats, and examination of the lung structure.

GLP compliance:

not specified

Specific details on test material used for the study:

not applicable

Radiolabelling:

not specified

Species:

rat

Strain:

other: Long-Evans & Fisher 344

Details on species / strain selection:

Both strains are commonly used experimental rats. The Long_evans rats are an outbred strain considered by many as more resistant to respiratory infections.

Sex:

male

Details on test animals or test system and environmental conditions:

NOTE: information applies to both rat strains.

TEST ANIMALS
- Source: Charles River Breeding Laboratories
- Age: 63 days old
- Housing: housed in a room with overpressurized filtered air

Route of administration:

inhalation: aerosol

Vehicle:

other: air

Details on exposure:

TYPE OF INHALATION EXPOSURE: whole body

GENERATION OF TEST ATMOSPHERE / CHAMBER DESCRIPTION
- Exposure apparatus: 2-m³ Rochester-type chamber
- Method of particle size determination: particle size was determined by means of a cascade impactor:
MMAD: 1.0 µm
GSD: 2.3

TEST ATMOSPHERE
- Brief description of analytical method used: concentration was analysed once per hour

Duration and frequency of treatment / exposure:

Duration: 7 hours
Frequency: once

Dose / conc.:

14.9 mg/m³ air (analytical)

Remarks:

Standard deviation: ± 0.9 mg/m³

No. of animals per sex per dose / concentration:

10 male rats/dose/time point of sacrifice (total: 40 rats)

Control animals:

yes, concurrent no treatment

Positive control reference chemical:

not specified

Details on study design:

not specified

Details on dosing and sampling:

- Tissues and body fluids sampled: lungs
- Time and frequency of sampling: particle clearance was assessed by serially killing groups of 10 animals at days 1, 8, 5, and 130 postexposure and analysing the lung for titanium dioxide content. Titanium dioxide was determined photometrically using 4,4-diantipyryl methane monohydrate.

In addition to lung-clearance assessment, the following data were collected: body weight, lung weight, titanium dioxide distribution in the lung lobes, and background titanium analysis in unexposed rats. In each group, the lung structure of two rats was examined using light microscopy.

Statistics:

Mean ± standard deviation
Regression coefficient
Correlation coefficient

Preliminary studies:

not specified

Type:

other: lung clearance

Results:

Day 1 retention is higher in the Long-Evans rats than in the Fisher 344 rats (157 µg TiO2/lung vs.101 µg TiO2/lung, respectively).

Results:

Clearance at day 8, 25, & 130 for Long-Evans rats is 39%, 48%, & 82%, resp. & for Fisher rats 1%, 24%, and 71%, resp.. Difference at days 8, 25, & 130 are statistically significant.

Details on absorption:

not specified

Details on distribution in tissues:

not specified

Details on excretion:

- titanium dioxide lung content of unexposed rats was 1.79 ± 0.31 µg/Long-Evans rat lung and 1.26 ± 0.35 µg/Fisher 344 rat lung.

1) Titanium dioxide analyses of individual lung lobes/rat at day 1 and day 25 postexposure (sum of the five lobes analyses gives the total lung burden):
- retained titanium dioxide is similarly distributed through the lobes in both strains at each time tested.
- plots of test item content of individual lobes and their wet weight indicated a close correlation between these two parameters. The slopes of the regression lines are almost identical in both strains at day 1 postexposure, they differ at day 25 postexposure.
- titanium dioxide lung content of the Long-Evans rats (155 µg/lung) was higher at day 1 postexposure than that of the Fisher 344 rats (119 µg/lung)(consistent with their different size.)
- at day 25, the retained titanium dioxide is in average 85 µg/Long-Evans rat lung and 83 µg/Fisher 344 rat lung.
- in 25 days, the Long-Evans rats cleared 45 % of the initial retention, the Fisher 344 rats only 30 %.
Please also refer to table 1 in the field "Any other information on results incl. tables" below.

2) Titanium analysis in lung of each rat in toto without dividing the lung into lobes:
- day 1 retention of titanium dioxide is higher in Long-Evans rats than in Fisher 344 rats (157 µg TiO2/lung versus 101 µg TiO2/lung, respectively).
- clearance expressed as percent of initial retention is lower in Fisher 344 rats at each of three time points when retention was determined.
- clearance at day 8, 25, and 130 for Long-Evans rats is 39 %, 48 %, and 82 %, respectively and for Fisher rats 1 %, 24 %, and 71 %, respectivley. The difference at days 8, 25, and 130 are statistically significant (P< 0.01).
Please also refer to table 2 in the field "Any other information on results incl. tables" below.

Metabolites identified:

not specified

Bioaccessibility (or Bioavailability) testing results:

not specified

Body weight and lung weights

- body weight of the Long-Evans rats is higher than that of the Fisher 344 rats throughout the experiment. The corresponding lung weights are also higher in the Long-Evans rats than in the Fisher 344 rats.

- no significant difference were observed between the lung weights of rats exposed or not exposed to titanium dioxide at the postexposure day 1.

- no difference were observed between the lung weight-body weight ratio at the early postexposure days within the two strains.

Pathology/Histopathology

At death, the lungs of all rats looked healthy without any sign of inflammation or pathological change. Light microscopy did not reveal any difference between the two strains.

Table 1: TiO2 retention in lung lobes after a 7 hr exposure (14.9 ± 0.9 mg/m³).

Day post-exposure	Rats			Retention (µg TiO2/lung)	Distribution in lobes (%)
	No.	Strain	%		Left	R1	R2	R3	Medial
1	8	LE	100	155	33	14	16	25	12
1	8	F	100	119	37	13	15	24	11
25	8	LE	55	85	35	13	13	27	12
25	8	F	70	83	33	13	17	25	12

LE = Long-Evans; F = Fisher

Table 2: Lung retention of TiO2 after a 7 -hr exposure (14.9 ± 0.9 mg/m³). Mean ± standard deviation (N = 10).

Day post-exposure	Long Evans rats		Fisher 344 rats
	µg TiO2/lung	%	µg TiO2/lung	%
1	157 ± 24	100	101 ± 9	100
8	95 ± 10	61	100 ± 13	99
25	81 ± 17	52	77 ± 12	76
130	28 ± 6	18	29 ± 7	26

Conclusions:

According to the results presented by the authors, Day 1 retention of titanium dioxide is higher in Long-Evans rats than in Fisher 344 rats (157 µg TiO2/lung vs.101 µg TiO2/lung, respectively). Clearance of titanium dioxide at day 8, 25, and 130 for Long-Evans rats is 39%, 48%, & 82%, respectively. and for Fisher rats 1%, 24%, and 71%, respectively. The difference at days 8, 25, and 130 are statistically significant.

Reference 13

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

Objective of study:

other: lung clearance

Qualifier:

no guideline followed

Principles of method if other than guideline:

Male Long-Evans rats (8 - 10 rats/dose/time point of sacrifice) were administered either 16.5 mg/m³ air of titanium dioxide (anatase) or 19.3 mg/m³ air of titanium dioxide (rutile) via whole body inhalation. The administration was conducted once for a duration of 7 hours. The titanium dioxide retention in the lung was determined immediately after exposure and at days 1, 8, 27 and 132 after the exposure by colourimetry.

GLP compliance:

not specified

Specific details on test material used for the study:

not applicable

Radiolabelling:

not specified

Species:

rat

Strain:

Long-Evans

Details on species / strain selection:

not specified

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Charles River Breeding Laboratories
- Weight: ~ 300 g

Route of administration:

inhalation: aerosol

Vehicle:

other: air

Details on exposure:

TYPE OF INHALATION EXPOSURE: whole body

GENERATION OF TEST ATMOSPHERE / CHAMBER DESCRIPTION
- Exposure apparatus: Rochester-type exposure chamber (2 m³)
- System of generating particulates/aerosols: aerosol was generated by a Wright Dust Feed (Wright, B.M. (1950)*.
- Air flow: about 1.4 m³/min
- Method of particle size determination: particle size information was acquired by using the cascade impactor technique.
MMAD (anatase aerosol): 1.0 µm
GSD (anatase aerosol): 2.3
MMAD (rutile aerosol): 0.83 µm
GSD (rutil aerosol): 2.02

TEST ATMOSPHERE
- Brief description of analytical method used: concentration information was acquired by using the membrane filter.

* Reference:
Wright, B.M.: A new dust feed mechanism. J. Sci. Inst. 27: 12 - 15 (1950).

Duration and frequency of treatment / exposure:

Duration: 7 hours
Frequency: once

Dose / conc.:

16.5 mg/m³ air (analytical)

Remarks:

anatase aerosol
standard deviation: 1.7 mg/m³

Dose / conc.:

19.3 mg/m³ air (analytical)

Remarks:

rutile aerosol
standard deviation: 3.1 mg/m³

No. of animals per sex per dose / concentration:

Titanium dioxide (anatase): 8 - 10 male rats/dose/time point of sacrifice (total: 48 rats)
Titanium dioxide (rutile): 8 - 10 male rats/dose/time point of dacrifice (total: 47 rats)

Control animals:

not specified

Positive control reference chemical:

not specified

Details on study design:

not specified

Details on dosing and sampling:

- Tissues and body fluids sampled: lungs
- Time and frequency of sampling: immediately after exposure and at days 1, 8, 27 and 132 after the exposure, groups of 8 - 10 rats were killed and the lungs were analysed for titanium dioxide content. In addition six rat lungs were examined by light microscopy. Titanium dioxide was determined in whole lungs after drying and fusing the lungs and analysing for titanium dioxide colorimetrically.

Statistics:

The results are given as mean ± standard deviations (SD). Analysis of variance was used and the individual 95 percent confidence intervals for each group mean were calculated. Differences with a probaibility level of P < 0.05 are indicated as statistically significant. The regression equation for anatase or rutile clearance from day 1 to day 132 was computed by fitting a regression line through the data, using the method of least square, on a semi-logarithmic plot, after ln transformation of the titanium dioxide lung contents.

Preliminary studies:

not specified

Type:

other: lung clearance

Results:

Initial deposition at day 0 immediately after the 7-hour exposure was 136 ± 14 µg anatase/lung and 151 ± 30 µg rutile/lung (difference between groups not statistically signficant).

Results:

Amount of titanium dioxide retained in the lungs at days 1, 8, 27 and 132 was similar in both the anatase and the rutile groups (difference between groups not statistically significant).

Results:

Regression of clearance on time from day 1 to day 132 can be expressed by a monoexponential fit: correlation coefficients were r = 0.92 (TiO2 (anatase)) & r = 0.93 (TiO2 rutile)).

Results:

Clearance T1/2 were as follows (difference between groups not significant): TiO2 (anatase): 51 days & TiO2 (rutile): 53 days.

Results:

Histologically a phagocytic macrophage response was observed with the extent and appearance similar in both the anatase and rutile groups. Otherwise the lung tissue was normal.

Details on absorption:

not specified

Details on distribution in tissues:

not specified

Details on excretion:

- initial deposition at day 0 immediately after the 7-hour exposure was 136 ± 14 µg anatase/lung and 151 ± 30 µg rutile/lung (difference between groups not statistically signficant).
- amount of titanium dioxide retained in the lungs at days 1, 8, 27 and 132 was similar in both the anatase and the rutile groups (difference between groups not statistically significant).
- regression of clearance on time, from day 1 to day 132 can be expressed by a monoexponential fit. The correlation coefficients were as follows:
titanium dioxide (anatase): r = 0.92
titanium dioxide (rutile): r = 0.93
- clearance T1/2 were as follows (difference between groups not significant):
titanium dioxide (anatase): 51 days
titanium dioxide (rutile): 53 days
- histologically a phagocytic macrophage response was observed with the extent and appearance similar in both the anatase and rutile groups. Otherwise the lung tissue was normal.
Please also refer to table 1 in the field "Any other information on results incl. tables" below.

Metabolites identified:

not specified

Bioaccessibility (or Bioavailability) testing results:

niot specified

Table 1: Lung retention

Day post exposure	0	1	8	27	132
Anatase	136 ± 14 (10)	129 ± 23 (10)	107 ± 22 (8)	79 ± 9 (10)	23 ± 11 (10)
Rutile	151 ± 30 (10)	130 ± 22 (10)	102 ± 15 (8)	72 ± 14 (10)	23 ± 9 (9)

Mean ± S.D. µg TiO2/lung, (n) = number of animals. No significant difference at P< 0.05.

Conclusions:

According to the authors, particle clearance from the lung, calculated from the retention data, was similar in both the anatase and the rutile titanium dioxide groups with T1/2 of 51 or 53 days, respectivley.

Reference 14

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

Objective of study:

other: pulmonary clearance

Qualifier:

no guideline followed

Principles of method if other than guideline:

Mice were administered one dose via intratracheal instillation. Bronchopulmonary lavage was performed on three animals after 15 min, 1, 5, and 20 hr, and 3 and 7 days postexposure. Lavage samples were analysed for cell count, viability and a differential count after staining with Wright's stain. Tissue samples from lungs and trachea (distal to the larynx) were excised. Lungs were blotted dry and weighed. Titanium content was measured using inductively coupled argon plasma (ICAP) spectroscopy with identically treated standards for evaluation of lung clearance.

GLP compliance:

not specified

Specific details on test material used for the study:

- Treatment of test material prior to testing:
- sample was ground and settled to yield a mass median diameter (MMD) of 1.57 µm and a geometric standard deviation of 2.3
- approproate size fractions were obtained by settling in an Andreason pipet
- dosages were prepared within 2 hr of use the sonicated for 10 min

Radiolabelling:

no

Species:

mouse

Strain:

other: BALB/c BYJ

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Jackson Laboratories, Bar Harbor, ME
- Age at study initiation: 7 and 8 weeks
- Weight after acclimation period: 25.7 - 27.7 g
- Housing: mmice were multiply housed in polycarbonate cages
- Diet (ad libitum): Purina rodent chow
- Water (ad libitum)
- Acclimation period: 2 weeks
- Mice were immunised against Sendai virus by intraperitoneal injection of killed virus (before acclimation period)

Route of administration:

intratracheal

Vehicle:

other: phosphate buffered saline (PBS)

Details on exposure:

Mice were anesthesised using Metafone (Methoxyflurane, Pitman-Moore, Inc., Washington Crossing, NJ) until the onset of slow, deep breathing, then were held on an upright support with fibreoptics illuminating the buccal cavity. Particle suspensions were vortexed and the dose was withdrawn using a micropipet with 1-inch sterile 23-gauge blunt needle slightly blent to deliver the dose at the first bifurcation distal to the vocal cords. Mice were held upright for approx. 30 sec until coughing ceased and placed in a recovery cage for 5 min.

Duration and frequency of treatment / exposure:

single exposure

Dose / conc.:

0.5 mg/kg bw (total dose)

Remarks:

approx. (11.8 µg TiO2 delivered in 20 µL PBS)

No. of animals per sex per dose / concentration:

not specified

Control animals:

yes, concurrent no treatment

yes, concurrent vehicle

Details on dosing and sampling:

- Bronchopulmonary lavage was performed at 15 min, 1, 5, and 20 hr, and 3 and 7 days postexposure
- Tissue samples were obtained at 15 min, 1, 5, and 20 hr, and 3 and 7 days postexposure

Statistics:

Welch approximation (for unkown, unequal variances) to the student's t test was used to assess differences between sample means.
Mean Titanium values and standard deviations from each timepoint were computer fitted using a biexponential fitting program

Type:

other: clearance half-lifes (calculated from lung burden)

Results:

0.5 and 462 hr (biphasic)

Details on absorption:

not applicable

Details on distribution in tissues:

not applicable

Details on excretion:

not applicable

Metabolites identified:

not measured

Details on metabolites:

Titanium dioxide as an inorganic substance is not metabolised.

Bioaccessibility (or Bioavailability) testing results:

not applicable

Conclusions:

After an initial rapid clearance of a significant portion of the dose, later clearance of TiO2 was slow with a half-life of 462 hours.

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

Objective of study:

other: intrapulmonary distribution

Qualifier:

no guideline followed

Principles of method if other than guideline:

Rats were exposed to ultrafine titanium dioxide aerosol for 1 hr via endotracheal tube. Lungs were fixed either 1 hr or 24 hr after the aerosol inhalation by sequential intravascular perfusion. Particle localisation was analysed with a transmission electron microscope.

GLP compliance:

not specified

Specific details on test material used for the study:

TREATMENT OF TEST MATERIAL PRIOR TO TESTING
- Treatment of test material prior to testing: aerosols were generated with a Palas spark generator (Palas GmbH, Germany) in a pure argon plus 0.1 % oxygen stream. The aerosol was quasineutralised by a radioactive 85Kr source, diluted and conditioned prior inhalation. Particles generated were agglomerates of smaller primary structures formed after spark ignition and condensation.
- Count median diameter (CMD): 22 nm (geometric SD of 1.7)
- Mean number concentration: 7.3 x 10^6 particles/cm³ (SD 0.5 x 10^6 particles/cm³)
- Mass concentration: 0.11 mg/m³

Radiolabelling:

no

Species:

rat

Strain:

other: WKY/NCrl BR

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Charles River, Germany
- Weight at study initiation: 250 ± 10 g
- Diet: ad libitum
- Water: ad libitum

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

Route of administration:

intratracheal

Vehicle:

unchanged (no vehicle)

Details on exposure:

TYPE OF INHALATION EXPOSURE: via endotracheal tube
- Method of holding animals in test chamber: anesthesy by intramuscular injection of a mixture of medetomidine, midazolam and fentanyl
- System of generating particulates/aerosols: via spark ignition with a Palas spark generator in a pure argon plus 0.1 % oxygen stream
- Negative-pressure ventilation: -1.5 Pa
- Breathing frequency: 45/min
- Minute volume: approx. 200 cm³/min
- Method of particle size determination: differential electrical mobility particle sizer and particle counter

Duration and frequency of treatment / exposure:

1 hr

Dose / conc.:

0.11 mg/m³ air (nominal)

No. of animals per sex per dose / concentration:

10 animals

Control animals:

no

Details on study design:

not applicable

Details on dosing and sampling:

Lung tissues were sampled 1 hr and 24 hr after inhalation

Statistics:

not specified

Type:

other: pulmonary distribution

Results:

The following amount of particles were found on luminal side of airways and alveoli, within epithelial or endothelial cells, within connective tissue, or within capillaries: 79.3 ± 7.6 %, 4.6 ± 2.5 %, 4.8 ± 4.5 %, 11.3 ± 3.9 %, respectively.

Details on absorption:

not applicable

Details on distribution in tissues:

not applicable

Details on excretion:

not applicable

Metabolites identified:

no

Details on metabolites:

Titanium dioxide as an inorganic substance, which is not metabolised.

Bioaccessibility (or Bioavailability) testing results:

not applicable

Conclusions:

According to the authors, the following amount of particles were found on luminal side of airways and alveoli, within epithelial or endothelial cells, within connective tissue, or within capillaries: 79.3 ± 7.6 %, 4.6 ± 2.5 %, 4.8 ± 4.5 %, 11.3 ± 3.9 %, respectively.

Reference 16

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

2013-07-16 to 2014-03-03

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Remarks:

Screening study only

Objective of study:

toxicokinetics

Qualifier:

according to guideline

Guideline:

OECD Guideline 417 (Toxicokinetics)

Version / remarks:

adopted 2010-07-22

Deviations:

yes

Remarks:

pilot study: use of male rats only, one dose level, and 3 rats per the test substance-treated group

GLP compliance:

yes

Specific details on test material used for the study:

STABILITY AND STORAGE CONDITIONS OF TEST MATERIAL
- Stability under test conditions: the test substance appeared to be stable under the conditions of the study; no evidence of instability was observed.

Radiolabelling:

no

Species:

rat

Strain:

other: Crl:CD(SD)

Details on species / strain selection:

The Charles River Cesarean Derived Sprague-Dawley, Crl:CD(SD), rat has been selected based on extensive experience with this strain at the performing laboratory, its suitability for toxicokinetic studies, and comparison with companion toxicity studies.

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Charles River Laboratories International, Inc. (Raleigh, North Carolina, U.S.A.)
- Age at study initiation: approximately 11 weeks old
- Weight at study initiation: 280.5 to 285.9 g (mean ± SD =283 ± 2.9 g)(single vehicle control rat weighed 295.7 g)
- Fasting period before study: animals were fasted for approximately 16 (± 1) hour before dosing with test substance. Food was returned approximately 2 hours post-dose.
- Housing: upon arrival at the laboratory animals were housed individually in solid-bottom caging with Bed-o'Cobs® bedding and Nestlets as enrichment. Animals were moved to glass metabolism units for the in-life phase of the study.
- Metabolism cages: yes, animals were moved to glass metabolism units for the in-life phase of the study.
- Diet (ad libitum): PMI® Nutrition International, LLC Certified Rodent LabDiet® 5002
- Water (ad libitum): tap water
- Quarantine period: quarantined for 3 full days, within a 5 day period of initial receipt and pre-dose administration.

ENVIRONMENTAL CONDITIONS
- Temperature: 20-26ºC
- Relative humidity: 30-70%
- Photoperiod (hrs dark / hrs light): approximate 12 hour light/dark cycle

Route of administration:

oral: gavage

Vehicle:

other: sterile water for injection

Duration and frequency of treatment / exposure:

one administration

Dose / conc.:

500 other: mg/kg bw

No. of animals per sex per dose / concentration:

Vehicle control group: one male rat
Test substance group: three male rats

Control animals:

yes, concurrent vehicle

Positive control reference chemical:

none

Details on study design:

One male rat with a jugular vein cannula was administered the vehicle(3.004 g) by oral gavage at the same dose vehicle volume as the rats dosed with the test substance. The dose vehicle volume (10 mL/kg bw) matched that used for the micronucleus test (please refer to the endpoint study record "w_Myhre_2014_MN" in Section 7.6.2 Genetic toxicity in vivo).

Three male rats, each with a jugular vein cannula, were administered the test substance formulated with dose vehicle at an expected dose of 500 mg/kg body weight (526, 542, and 512 mg/kg bw, respectively; mean ± SD = 527 ± 14.9 mg/kg bw). The dose volume was 10 mL/kg bw. The dose administration was performed using a portion of the lowest (50 mg/mL, nominal) dose used in the micronucleus test (please refer to the endpoint study record "w_Myhre_2014_MN" in Section 7.6.2 Genetic toxicity in vivo).

Details on dosing and sampling:

PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
Vehicle control group and test substance group:
- Tissues and body fluids sampled: urine, faeces, blood, liver
- Time and frequency of sampling: urine and faeces were collected prior to dosing (pre-dose) and at 0-6, 6-12, 12-24, and 24-48 hour intervals over dry ice. Blood (~0.2 mL) was collected at 0, 0.5, 1, 2, 4, 6, 8, 12, 24, and 48 hours via the jugular vein cannula (test substance group: except for one rat, which had blood collected from tail vein at 12, 24, and 48 hour because the cannula lost patency).

Vehicle control group:
- at 48 hours, the rat was sacrificed. The liver was collected. The liver and 48-hour blood sample were submitted for Ti analysis.

Test substance group:
- rats were sacrificed at 48 hours after dose administration. The livers were collected. The blood (from the jugular vein cannulas), liver, urine, and faeces samples were frozen and submitted for Ti analysis. At sacrifice, blood was collected by exsanguination from the vena cava.

TITANIUM ANALYSIS
- faeces: samples were prepared for analysis by first ashing the entire sample in a muffle furnace. The ashed sample was then transferred to a microwave vessel and prepared in a chemical microwave using nitric and hydrofluoric acid and then analysed by Inductively Coupled Plasma-Atomic Emission Spectroscpoy (ICP-AES).
- liver: samples were digested in a chemical microwave using nitric acid and hydrogen peroxide and then analysed by Inductively Coupled Plasma-Mass Spectroscopy (ICP-MS).
- blood: samples were ashed in a muffle furnace, adding nitric acid and hydrogen peroxide at the conclusion of the ashing and then analysed by ICP-MS.
- urine: samples were diluted in a dilute nitric acid solution and then analysed by ICP-MS.

Statistics:

Group data were represented as Mean ± SD. Individual samples less than the limit of quantitation (LOQ) were excluded from the mean. Results for the faeces were expressed as percent of the administered dose.

Type:

absorption

Results:

TiO2 concentration (blood): prior to dosing: 0.347 to 0.861 μg/g blood (mean ± SD: 0.520 ± 0.295 µg/g; in treated rats the blood concentrations ranged from 0.102 to 1.74 μg/g across all post-dose time points.

Type:

distribution

Results:

TiO2 concentration (liver): a point estimate of the liver concentration (0.348 μg/g) in the control rat was essentially the same as the mean concentration for treated rats (0.256 ± 0.039 μg/g) at 48 hours after dose administration.

Type:

excretion

Results:

TiO2 concentration (faeces): prior to dosing: 45.2 to 57.9 μg/g faeces; after dosing: 684 to 17200 μg/g faeces; peak concentrations occurred in the 6-12 or 12-24 hour samples; overall mean recovery was 93.0 ± 1.3% of the administered dose amount.

Type:

excretion

Results:

TiO2 concentration (urine): prior to dosing: 0.035 to 0.118 μg/g urine; after dosing: 0.020 to 0.247 μg/g urine; conc. of TiO2 in urine were very low and showed overlap between pre-dose (1.34 ± 0.663 μg) & post-dose samples (1.34 ± 0.537 μg; 0-24 hours).

Details on absorption:

- blood: individual animal pre-dose blood samples had TiO2 concentrations that ranged from 0.347 to 0.861 μg/g blood with a mean and SD of 0.520 ± 0.295 μg/g. A single measurement at 48 hours in a control animal was 0.330 μg/g blood. In treated rats the blood concentrations ranged from 0.102 to 1.74 μg/g across all post-dose time points. The appearance of minimal increase and similarity of the concentrations across time relative the control (0 hour) sample indicates very low potential for systemic exposure from orally administered TiO2 pg-1. Please also refer to table 3 in the field "Any other information on results incl. tables" below.

Details on distribution in tissues:

- liver: a point estimate of the liver concentration (0.348 μg/g) in the control rat was essentially the same as the mean concentration for treated rats (0.256 ± 0.039 μg/g) at 48 hours after dose administration. Please also refer to table 3 in the field "Any other information on results incl. tables" below.

Details on excretion:

- faeces: individual animal samples collected prior to dose administration had TiO2 concentrations that ranged from 45.2 to 57.9 μg/g faeces. The concentrations after dosing ranged from 684 to 17,200 μg/g faeces. The peak concentrations occurred in the 6-12 or 12-24 hour samples. The overall mean recovery was 93.0 ± 1.3% of the administered dose amount. Please also refer to table 1 in the field "Any other information on results incl. tables" below.

- urine: individual animal pre-dose urine sample concentrations of TiO2 ranged from 0.035 to 0.118 μg/g urine. The concentrations after dose administration ranged from 0.020 to 0.247 μg/g urine. Adjusting for the amount of urine produced gave a mean TiO2 amount of 1.34 ± 0.663 μg for the pre-dose group. This value was essentially the same as the amount recovered from 0-24 hours after dose administration (1.34 ± 0.537 μg). An additional 0.894 ± 0.243 μg was measured in the 24-48 hour sample. Please also refer to table 2 in the field "Any other information on results incl. tables" below.

Metabolites identified:

not measured

Bioaccessibility (or Bioavailability) testing results:

no data

Table 1: TiO2 uf-1 analysis in faeces

Parameter by subject
Parameter	Time	Animal #1	Animal #2	Animal #3	Mean	SD
Concentration (µg/g faeces)	Predose	52.7	57.9	45.2	51.9	6.38
	0 – 6 h	NA	NA	NA	NA	NA
	6 – 12 h	12400	NA	17200	14800	3390
	12 – 24 h	12400	16100	4220	10900	6080
	24 – 48 h	684	1180	912	925	248

Table 2: TiO2 uf-1 analysis in urine

Parameter by subject
Parameter	Time	Animal #1	Animal #2	Animal #3	Mean	SD
Concentration (µg/g urine)	Predose*	0.118	0.083	0.035	0.079	0.042
	6	0.097	0.065	0.247	0.136	0.097
	12	0.232	0.112	0.072	0.139	0.068
	24	0.047	0.045	0.020	0.037	0.015
	48	0.035	0.050	0.030	0.038	0.008

* Predose collection = approximately 16 hours based on start of fasting to time of dosing

Table 3: TiO2 uf-1 analysis in blood and liver

Concentration (µg/g)
Tissue	Hours	Animal #1	Animal #2	Animal #3	Mean	SD
Treated Blood	0*	0.861	0.347	0.352	0.520	0.295
	0.5	0.631	0.210	0.177	0.339	0.253
	1	1.74	0.280	0.400	0.807	0.811
	2	0.252	0.546	0.329	0.376	0.152
	4	1.18	0.264	0.352	0.598	0.504
	6	1.038	0.224	0.524	0.595	0.412
	8	0.172	0.242	0.169	0.194	0.041
	12	1.45	0.384	0.404	0.746	0.610
	24	0.918	0.162	0.217	0.432	0.421
	48	0.419	0.11	0.125	0.215	0.157
Liver	48	0.279	0.300	0.219	0.256**	0.039
Control
Blood	48	0.330
Liver	48	0.348

* Collected prior to dose administration

** Mean and SD include 48 hour repeat analysis replicates

Conclusions:

A screening study was conducted with one control and three treated rats, the latter of which were administered a mean TiO2 pf-1 dose of 527 (±15) mg/kg bw. The majority of dose was recovered in faeces (93.0 ± 1.3%) by 48 hours after administration with peak amounts occurring at 6-12 and 12-24 hours. Concentrations of TiO2 in urine were very low and showed overlap between pre-dose and post-dose samples. Like urine, blood concentrations of TiO2 were also very low and overlapped between pre-dose (0.347 to 0.861 μg/g) and post-dose (0.102 to 1.74 μg/g) time points. Liver concentrations were similar between the single control (0.348 μg/g) and three treated (0.256 ± 0.039μg/g) rats 48 hours after dose administration. Collectively these results demonstrate very low potential for absorption of the test material and nearly quantitative recovery of the dose in faeces.

Reference 17

Endpoint:

basic toxicokinetics, other

Remarks:

Bioaccessibility

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

2015-10-12 to 2015-10-22

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

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:

2001

Deviations:

yes

Principles of method if other than guideline:

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) and phosphate-buffered saline (PBS).

GLP compliance:

yes (incl. QA statement)

Remarks:

signed 2011-02-07

Species:

other: in vitro (simulated human body fluids)

Details on test animals or test system and environmental conditions:

Test principle in brief:
- two different artificial physiological media,
- single loading of test substance of ~100 mg/L,
- samples taken after 2 and 24 hours agitation (100 rpm) at 37 ± 2 °C,
- two additional method blanks per medium, measurement (by ICP-OES) of dissolved titanium concentrations after filtration and centrifugal filtration.
- the study was performed in triplicate

The aim of this test was to assess the dissolution of TiO2-E171-A in two artificial physiological media: Phosphate buffered saline (PBS, pH 7.2-7.4) and Artificial gastric fluid (GST, 1.5-1.6). The test media were selected to simulate relevant human-chemical interactions (as far as practical), i.e. a substance entering the human body by ingestion into the gastro-intestinal tract.

Details on study design:

PROCEDURE
Three replicate flasks (500 mL glass flasks) per test medium (PBS, GST) were prepared with a loading of ~ 100 mg/L. Two control blank replicates (same procedure) per test medium were also prepared. Thus, three replicates containing the test item and two method blanks per artificial medium were tested; solutions were sampled after 2 and 24 h, to measure total dissolved titanium concentrations (by ICP-OES) after filtration (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) and centrifugal filtration (3kDa centrifugal filter, Sartorius, Göttingen, Germany), temperature and pH. During the study, observations, including the appearance of the solution (including colour, turbidity and particle film on the surface) were recorded.
After the test, 80 mL aqua regia (3 : 1 mixture of concentrated hydrochloric and nitric acid) were added to the vessels containing the undissolved test item. The filters used for sampling were extensively rinsed with 5 mL aqua regia. Solutions were sampled after at least 24 h, titanium concentration were measured in at least one mass balance sample (vessel and filter/syringes) for each medium by ICP-OES, and the mass balance was calculated.

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).

METAL ANALYSIS
- Standards for metal analysis: titanium standard (Merck Certipur Titanium ICP standard 1000 mg/L lot no. HC383640; Darmstadt, Germany; Merck)
- Certified reference materials: certified aqueous reference material TM-25.4 (lot no. 0914) obtained from Environment Canada and a multielement standard (CPI 19 Elements, lot no.15C267)

Instrumental and analytical set-up for the ICP-OES instrument:
- Agilent 720, Agilent Technologies, Waldbronn, Germany
- Nebulizer: Sea spray nebulizer from Agilent
- Spray chamber: Glass cyclonic spray chamber from Agilent
- Carrier gas flow: 0.75 L/min
- RF power: 1200W
- Wavelengths Ti: 308.804 nm, 334.188 nm, 334.941 nm, 368.520 nm and 376.132 nm
- Calibration: blank, 0.5 μg/L, 0.75 μg/L, 1 μg/L, 2.5 μg/L, 5 μg/L, 7.5 μg/L, 10 μg/L, 20 μg/L, 30 μg/L, 40 μg/L and 50 μg/L
- Correlation coefficients (r) for the wavelengths used for evaluation of data were at least 0.999904

The applied LOD/LOQ calculations for the Agilent 720 ICP-OES are (according to DIN 32645):
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" in the validation summary.

From selected artificial physiological media, samples were fortified with a known amount of titanium (by standard addition of commercial standards) to determine the standard recovery of titanium. Detailed tables were stated in "Any other information on materials and methods incl. tables". For fortified test samples, recoveries were 102 % for Ti in GST test samples, 105 – 112 % for Ti in PBS test samples and 95.9 – 113 % for Ti in mass balance samples.

Details on dosing and sampling:

Loading:
Detailed loadings of the test vessels are given in "Any other information on materials and methods incl. tables".

Sampling:
Aqueous samples of approx. 20 mL taken for Ti analysis were transferred into disposable scintillation vials (20 mL scintillation tubes, Sarstedt, Nuembrecht, Germany). These 20 mL were used for filtration and centrifugal filtration, 12 mL of these treated samples were acidified with HNO3 and used for titanium analysis. The remaining 8 mL of filtrated samples were stored at 4 °C without acidification for single particle analysis. Single particle analysis, however, was not performed since results of the pre-study indicated that solutions are free of particles.

Toxicokinetic parameters:

other: bioaccessibility

Remarks:

GST 2 h: total Ti ± SD in samples (background corrected) [μg/L]: 7.67 ± 2.34; PBS 24 h: total Ti ± SD in samples (background corrected) [μg/L]: below LOD

Bioaccessibility (or Bioavailability) testing results:

Under conditions of this test (flasks with different artificial physiological media; TiO2-E171-A loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.06 (in two samples below LOQ*; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 7.88 ± 2.34
- total Ti ± SD in samples (background corrected) [μg/L]: 7.67 ± 2.34
* Concentrations < LOQ and > LOD were estimated with LOD + ½* (LOQ – LOD)

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.05 (in two samples below LOQ; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 32.42 ± 2.16
- total Ti ± SD in samples (background corrected) [μg/L]: 32.21 ± 2.16

PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L]: in three samples below LOD; in one sample below LOQ
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

Method validation summary ICP-OES

Validation parameter	Results	Comment
Selectivity	Similar data with different wavelengths for ICP-OES method	-
Linearity	Applied calibration functions were linear	correlation coefficient at least 0.999904
Limit of detection	Ti: 0.03 – 0.19 μg/L
Limit of quantification	Ti: 0.09 – 0.56 μg/L
Method blanks	GST: - two method blanks after 2h: below LOQ, - two method blanks after 2h: above LOQ but on average 37.4 fold lower than concentrations in test samples after 2h, - two method blanks after 24h: below LOQ, - two method blanks after 24h: above LOQ but on average 155 fold lower than concentrations in test samples after 24h PBS: All method blanks: below LOD/LOQ
Accuracy measurement / Reproducibility - mass balance samples	Mean recovery for CRM TM- 25.4 (dilution factor 8, 3.15 μg Ti/L): Ti: 102 ± 2.0 % (n = 4)	Low concentration range mass balance
Accuracy measurement/ Reproducibility - GST test samples	Mean recovery for CRM TM- 25.4 (dilution factor 3, 8.40 μg Ti/L): Ti: 99.9 ± 1.1 % (n = 5)	Mid concentration range GST
Accuracy measurement / Reproducibility – PBS test samples and remeasurement of GST 3b 2h test sample	Mean recovery for CRM TM- 25.4 (dilution factor 10, 2.52 μg Ti/L): Ti: 99.8 ± 6.4 % (n = 7)	Mid concentration range PBS
Trueness measurement - GST test and mass balance samples	Mean recovery for quality control standard (CPI 2.50 μg Ti/L): Ti: 101 ± 1.5 % (n = 9)	Low concentration range
Trueness measurement - PBS test samples	Mean recovery for quality control standard (CPI 3 μg Ti/L): Ti: 96.3 ± 1.2 % (n = 5)	High concentration range PBS
Trueness measurement - PBS test samples and remeasurement of GST 3b 2h test sample	Mean recovery for quality control standard (CPI 5 μg Ti/L): Ti: 95.6 ± 0.5 % (n = 6)	High concentration range PBS
Trueness measurement - GST test and mass balance samples	Mean recovery for quality control standard (CPI 25 μg Ti/L): Ti: 97.9 ± 1.4 % (n = 9)	High concentration range
Trueness measurement - PBS test samples and remeasurement of GST 3b 2h test sample	Mean recovery for recalibration standard (1 μg Ti/L): Ti: 101 ± 3.4 % (n = 6)	Low concentration range PBS
Trueness measurement - GST test and mass balance samples	Mean recovery for recalibration standard (10 μg Ti/L): Ti: 101 ± 3.4 % (n = 9)	Mid concentration range
Trueness test samples	Fortification of samples: GST- Ti: 102 % PBS- Ti: 105 – 112 %
Trueness mass balance	Fortification of samples: Ti: 95.9 – 113 %

Solution pH values

After preparation of the artificial physiological media, solution pH was adjusted to their respective target pH. Solution pH values prior to the test and after 2 h and 24 h in the different media were measured using a multi 9430 digital IDS sensor equipped with a SenTix 940-3 electrode from WTW (Weilheim, Germany) directly in the test vessel. The target pH of all media before addition of the test substance was in the nominal range. During the study, the pH of GST and PBS solutions remained stable (± 0.02).

Temperature control

The test was performed on an incubated laboratory shaker (Shaking incubation cabinet Minitron, INFORS AG, Bottmingen, Switzerland) at 100 rpm. The temperature was adjusted to 37.5 °C in the thermostatically controlled shaking cabinet to reach a temperature of 37 ± 2 °C in the media. The temperature in the cabinet was monitored by a certified thermometer and a thermo data logger.

Mass balance calculation

Measured dissolved titanium concentrations (filtered through a 0.2 μm filter) in test solutions and in the rinse of the filters indicate an incomplete dissolution of the test material TiO2- E171-A after addition of aqua regia to the physiological media in sample vessels. An incomplete dissolution was confirmed by the presence of particles in the media. Hence, recoveries for the mass balance are low. The results and the recovery of the test item are compiled in following table

Amount of titanium in mass balance samples (vessels, filter/syringes and taken samples) for each media

Media	Measured dissolved Ti after addition of aqua regia + filter/syringe rinse + samples [mg] *	Nominal amount [mg] #	Recovery [%]
GST vessel 1	0.044	30.27	0.145
GST vessel 2	0.036	30.20	0.120
GST vessel 3	0.046	30.17	0.151
PBS vessel 1	0.058	30.10	0.191
PBS vessel 2	0.055	30.10	0.184
PBS vessel 3	0.054	30.01	0.181

* Measured dissolved Ti includes the amount of titanium in test solutions after addition of aqua regia, the amount of titanium rinsed of used filters and syringe and the amount of titanium removed with samples taken during the study.

Example: 0.032 mg (amount Ti in vessels) + 0.010 mg (amount Ti in filters/syringes) + 0.002 mg (amount Ti in samples taken) = 0.044 mg

# nominal amount TiO2 = 99.3% in test item (according to CoA) -> 59.9343 % Ti in TiO2 -> 29.8 mg Ti in 50 mg test item -> nominal amount 29.8 * initial weight / 50mg

Example: 29.8 mg Ti * 50.861 mg TiO2 / 50 mg TiO2 = 30.3 mg Ti

Conclusions:

Under conditions of this test (flasks with different artificial physiological media; TiO2-E171-A loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.06 (in two samples below LOQ*; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 7.88 ± 2.34
- total Ti ± SD in samples (background corrected) [μg/L]: 7.67 ± 2.34
* Concentrations < LOQ and > LOD were estimated with LOD + ½* (LOQ – LOD)

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.05 (in two samples below LOQ; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 32.42 ± 2.16
- total Ti ± SD in samples (background corrected) [μg/L]: 32.21 ± 2.16

PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L]: in three samples below LOD; in one sample below LOQ
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

Reference 18

Endpoint:

basic toxicokinetics, other

Remarks:

Bioaccessibility

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

2015-10-12 to 2015-10-22

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

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:

2001

Deviations:

yes

Principles of method if other than guideline:

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) and phosphate-buffered saline (PBS).

GLP compliance:

yes (incl. QA statement)

Remarks:

signed 2011-02-07

Species:

other: in vitro (simulated human body fluids)

Details on test animals or test system and environmental conditions:

Test principle in brief:
- two different artificial physiological media,
- single loading of test substance of ~100 mg/L,
- samples taken after 2 and 24 hours agitation (100 rpm) at 37 ± 2 °C,
- two additional method blanks per medium, measurement (by ICP-OES) of dissolved titanium concentrations after filtration and centrifugal filtration.
- the study was performed in triplicate

The aim of this test was to assess the dissolution of TiO2-E171-B in two artificial physiological media: Phosphate buffered saline (PBS, pH 7.2-7.4) and Artificial gastric fluid (GST, 1.5-1.6). The test media were selected to simulate relevant human-chemical interactions (as far as practical), i.e. a substance entering the human body by ingestion into the gastro-intestinal tract.

Details on study design:

PROCEDURE
Three replicate flasks (500 mL glass flasks) per test medium (PBS, GST) were prepared with a loading of ~ 100 mg/L. Two control blank replicates (same procedure) per test medium were also prepared. Thus, three replicates containing the test item and two method blanks per artificial medium were tested; solutions were sampled after 2 and 24 h, to measure total dissolved titanium concentrations (by ICP-OES) after filtration (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) and centrifugal filtration (3kDa centrifugal filter, Sartorius, Göttingen, Germany), temperature and pH. During the study, observations, including the appearance of the solution (including colour, turbidity and particle film on the surface) were recorded.
After the test, 80 mL aqua regia (3 : 1 mixture of concentrated hydrochloric and nitric acid) were added to the vessels containing the undissolved test item. The filters used for sampling were extensively rinsed with 5 mL aqua regia. Solutions were sampled after at least 24 h, titanium concentration were measured in at least one mass balance sample (vessel and filter/syringes) for each medium by ICP-OES, and the mass balance was calculated.

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).

METAL ANALYSIS
- Standards for metal analysis: titanium standard (Merck Certipur Titanium ICP standard 1000 mg/L lot no. HC383640; Darmstadt, Germany; Merck)
- Certified reference materials: certified aqueous reference material TM-25.4 (lot no. 0914) obtained from Environment Canada and a multielement standard (CPI 19 Elements, lot no. 15C267)

Instrumental and analytical set-up for the ICP-OES instrument:
- Agilent 720, Agilent Technologies, Waldbronn, Germany
- Nebulizer: Sea spray nebulizer from Agilent
- Spray chamber: Glass cyclonic spray chamber from Agilent
- Carrier gas flow: 0.75 L/min
- RF power: 1200W
- Wavelengths Ti: 308.804 nm, 334.188 nm, 334.941 nm, 368.520 nm and 376.132 nm
- Calibration: blank, 0.5 μg/L, 0.75 μg/L, 1 μg/L, 2.5 μg/L, 5 μg/L, 7.5 μg/L, 10 μg/L, 20 μg/L, 30 μg/L, 40 μg/L and 50 μg/L
- Correlation coefficients (r) for the wavelengths used for evaluation of data were at least 0.999939

The applied LOD/LOQ calculations for the Agilent 720 ICP-OES are (according to DIN 32645):
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" in the validation summary.

From selected artificial physiological media, samples were fortified with a known amount of titanium (by standard addition of commercial standards) to determine the standard recovery of titanium. Detailed tables were stated in "Any other information on materials and methods incl. tables". For fortified test samples, recoveries were in the range of 99.7 – 101 % for Ti in GST test samples, 101 – 104 % for Ti in PBS test samples and 98.2 - 114 % for Ti in mass balance samples.

Details on dosing and sampling:

Loading:
Detailed loadings of the test vessels are given in "Any other information on materials and methods incl. tables".

Sampling:
Aqueous samples of approx. 20 mL taken for Ti analysis were transferred into disposable scintillation vials (20 mL scintillation tubes, Sarstedt, Nuembrecht, Germany). These 20 mL were used for filtration and centrifugal filtration, 12 mL of these treated samples were acidified with HNO3 and used for titanium analysis. The remaining 8 mL of filtrated samples were stored at 4 °C without acidification for single particle analysis. Single particle analysis, however, was not performed since results of the pre-study indicated that solutions are free of particles.

Toxicokinetic parameters:

other: bioaccessibility

Remarks:

GST 2h: - total Ti ± SD in samples (background corrected) [μg/L]: 7.22 ± 0.95; PBS 24 h: - total Ti ± SD in samples (background corrected) [μg/L]: below LOD

Bioaccessibility (or Bioavailability) testing results:

Under conditions of this test (flasks with different artificial physiological media; TiO2-E171-B loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.06 (in two samples below LOQ*; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 7.43 ± 0.95
- total Ti ± SD in samples (background corrected) [μg/L]: 7.22 ± 0.95
* Concentrations < LOQ and > LOD were estimated with LOD + ½* (LOQ – LOD)

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.05 (in two samples below LOQ; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 30.30 ± 1.84
- total Ti ± SD in samples (background corrected) [μg/L]: 30.09 ± 1.84

PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L] :in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L] :in three samples below LOD; in one sample below LOQ
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

Method validation summary ICP-OES

Validation parameter	Results	Comment
Selectivity	Similar data with different wavelengths for ICP-OES method	-
Linearity	Applied calibration functions were linear	correlation coefficient at least 0.999939
Limit of detection	Ti: 0.03 – 0.19 μg/L
Limit of quantification	Ti: 0.09 – 0.56 μg/L
Method blanks	GST: - two method blanks after 2h: below LOQ, - two method blanks after 2h: above LOQ but on average 35.3 fold lower than concentrations in test samples after 2h, - two method blanks after 24h: below LOQ, - two method blanks after 24h: above LOQ but on average 144 fold lower than concentrations in test samples after 24h PBS: All method blanks: below LOD/LOQ
Accuracy measurement / Reproducibility - mass balance samples	Mean recovery for CRM TM- 25.4 (dilution factor 8, 3.15 μg Ti/L): Ti: 102 ± 2.0 % (n = 4)	Low concentration range mass balance
Accuracy measurement/ Reproducibility - GST test samples	Mean recovery for CRM TM- 25.4 (dilution factor 3, 8.40 μg Ti/L): Ti: 99.9 ± 1.1 % (n = 5)	Mid concentration range GST
Accuracy measurement / Reproducibility – PBS test samples	Mean recovery for CRM TM- 25.4 (dilution factor 10, 2.52 μg Ti/L): Ti: 97.8 ± 0.3 % (n = 5)	Mid concentration range PBS
Trueness measurement - GST test and mass balance samples	Mean recovery for quality control standard (CPI 2.50 μg Ti/L): Ti: 101 ± 1.5 % (n = 9)	Low concentration range
Trueness measurement - PBS test samples	Mean recovery for quality control standard (CPI 3 μg Ti/L): Ti: 96.3 ± 1.2 % (n = 5)	High concentration range PBS
Trueness measurement - PBS test samples	Mean recovery for quality control standard (CPI 5 μg Ti/L): Ti: 95.5 ± 0.6 % (n = 5)	High concentration range PBS
Trueness measurement - GST test and mass balance samples	Mean recovery for quality control standard (CPI 25 μg Ti/L): Ti: 97.9 ± 1.4 % (n = 9)	High concentration range
Trueness measurement - PBS test samples	Mean recovery for recalibration standard (1 μg Ti/L): Ti: 103 ± 3.0 % (n = 5)	Low concentration range PBS
Trueness measurement - GST test and mass balance samples	Mean recovery for recalibration standard (10 μg Ti/L): Ti: 101 ± 3.4 % (n = 9)	Mid concentration range
Trueness test samples	Fortification of samples: GST- Ti: 99.7 – 101 %. PBS- Ti: 101 – 104 %
Trueness mass balance	Fortification of samples: Ti: 98.2 – 114 %

Solution pH values

After preparation of the artificial physiological media, solution pH was adjusted to their respective target pH. Solution pH values prior to the test and after 2 h and 24 h in the different media were measured using a multi 9430 digital IDS sensor equipped with a SenTix 940-3 electrode from WTW (Weilheim, Germany) directly in the test vessel. The target pH of all media before addition of the test substance was in the nominal range. During the study, the pH of GST and PBS solutions remained stable (± 0.02).

Temperature control

The test was performed on an incubated laboratory shaker (Shaking incubation cabinet Minitron, INFORS AG, Bottmingen, Switzerland) at 100 rpm. The temperature was adjusted to 37.5 °C in the thermostatically controlled shaking cabinet to reach a temperature of 37 ± 2 °C in the media. The temperature in the cabinet was monitored by a certified thermometer and a thermo data logger.

Mass balance calculation

Measured dissolved titanium concentrations (filtered through a 0.2 μm filter) in test solutions and in the rinse of the filters indicate an incomplete dissolution of the test material TiO2- E171 -B after addition of aqua regia to the physiological media in sample vessels. An incomplete dissolution was confirmed by the presence of particles in the media. Hence, recoveries for the mass balance are low. The results and the recovery of the test item are compiled in following table

Amount of titanium in mass balance samples (vessels, filter/syringes and taken samples) for each media

Media	Measured dissolved Ti after addition of aqua regia + filter/syringe rinse + samples [mg] *	Nominal amount [mg] #	Recovery [%]
GST vessel 1	0.035	29.96	0.116
GST vessel 2	0.036	29.97	0.122
GST vessel 3	0.035	30.13	0.116
PBS vessel 1	0.108	30.08	0.359
PBS vessel 2	0.049	30.05	0.162
PBS vessel 3	0.049	30.14	0.163

* Measured dissolved Ti includes the amount of titanium in test solutions after addition of aqua regia, the amount of titanium rinsed of used filters and syringe and the amount of titanium removed with samples taken during the study.

Example: 0.025 mg (amount Ti in vessels) + 0.01 mg (amount Ti in filters/syringes) + 0.002 mg (amount Ti in samples taken) = 0.035 mg

# nominal amount TiO2 = 99.5% in test item (according to CoA) -> 59.9343 % Ti in TiO2 -> 29.8 mg Ti in 50 mg test item -> nominal amount 29.8 * initial weight / 50mg

Example: 29.8 mg Ti * 50.238 mg TiO2 / 50 mg TiO2 = 30.0 mg Ti

Conclusions:

Under conditions of this test (flasks with different artificial physiological media; TiO2-E171-B loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.06 (in two samples below LOQ*; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 7.43 ± 0.95
- total Ti ± SD in samples (background corrected) [μg/L]: 7.22 ± 0.95
* Concentrations < LOQ and > LOD were estimated with LOD + ½* (LOQ – LOD)

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.05 (in two samples below LOQ; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 30.30 ± 1.84
- total Ti ± SD in samples (background corrected) [μg/L]: 30.09 ± 1.84

PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L] :in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L] :in three samples below LOD; in one sample below LOQ
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

Reference 19

Endpoint:

basic toxicokinetics, other

Remarks:

Bioaccessibility

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

2015-10-12 to 2015-10-22

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

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:

2001

Deviations:

yes

Principles of method if other than guideline:

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) and phosphate-buffered saline (PBS).

GLP compliance:

yes (incl. QA statement)

Remarks:

signed 2011-02-07

Species:

other: in vitro (simulated human body fluids)

Details on test animals or test system and environmental conditions:

Test principle in brief:
- two different artificial physiological media,
- single loading of test substance of ~100 mg/L,
- samples taken after 2 and 24 hours agitation (100 rpm) at 37 ± 2 °C,
- two additional method blanks per medium, measurement (by ICP-OES) of dissolved titanium concentrations after filtration and centrifugal filtration.
- the study was performed in triplicate

The aim of this test was to assess the dissolution of TiO2-E171-C in two artificial physiological media: Phosphate buffered saline (PBS, pH 7.2-7.4) and Artificial gastric fluid (GST, 1.5-1.6). The test media were selected to simulate relevant human-chemical interactions (as far as practical), i.e. a substance entering the human body by ingestion into the gastro-intestinal tract.

Details on study design:

PROCEDURE
Three replicate flasks (500 mL glass flasks) per test medium (PBS, GST) were prepared with a loading of ~ 100 mg/L. Two control blank replicates (same procedure) per test medium were also prepared. Thus, three replicates containing the test item and two method blanks per artificial medium were tested; solutions were sampled after 2 and 24 h, to measure total dissolved titanium concentrations (by ICP-OES) after filtration (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) and centrifugal filtration (3kDa centrifugal filter, Sartorius, Göttingen, Germany), temperature and pH. During the study, observations, including the appearance of the solution (including colour, turbidity and particle film on the surface) were recorded.
After the test, 80 mL aqua regia (3 : 1 mixture of concentrated hydrochloric and nitric acid) were added to the vessels containing the undissolved test item. The filters used for sampling were extensively rinsed with 5 mL aqua regia. Solutions were sampled after at least 24 h, titanium concentration were measured in at least one mass balance sample (vessel and filter/syringes) for each medium by ICP-OES, and the mass balance was calculated.

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).

METAL ANALYSIS
- Standards for metal analysis: titanium standard (Merck Certipur Titanium ICP standard 1000 mg/L lot no. HC383640; Darmstadt, Germany; Merck)
- Certified reference materials: certified aqueous reference material TM-25.4 (lot no. 0914) obtained from Environment Canada and a multielement standard (CPI 19 Elements, lot no. 15C267)

Instrumental and analytical set-up for the ICP-OES instrument:
- Agilent 720, Agilent Technologies, Waldbronn, Germany
- Nebulizer: Sea spray nebulizer from Agilent
- Spray chamber: Glass cyclonic spray chamber from Agilent
- Carrier gas flow: 0.75 L/min
- RF power: 1200W
- Wavelengths Ti: 308.804 nm, 334.188 nm, 334.941 nm, 368.520 nm and 376.132 nm
- Calibration: blank, 0.5 μg/L, 0.75 μg/L, 1 μg/L, 2.5 μg/L, 5 μg/L, 7.5 μg/L, 10 μg/L, 20 μg/L, 30 μg/L, 40 μg/L and 50 μg/L
- Correlation coefficients (r) for the wavelengths used for evaluation of data were at least 0.999939

The applied LOD/LOQ calculations for the Agilent 720 ICP-OES are (according to DIN 32645):
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" in the validation summary.

From selected artificial physiological media, samples were fortified with a known amount of titanium (by standard addition of commercial standards) to determine the standard recovery of titanium. Detailed tables were stated in "Any other information on materials and methods incl. tables". For fortified test samples, recoveries were in the range of 101 – 105 % for Ti in GST test samples, 107 – 108 % for Ti in PBS test samples and 98.4 – 113 % for Ti in mass balance samples.

Details on dosing and sampling:

Loading:
Detailed loadings of the test vessels are given in "Any other information on materials and methods incl. tables".

Sampling:
Aqueous samples of approx. 20 mL taken for Ti analysis were transferred into disposable scintillation vials (20 mL scintillation tubes, Sarstedt, Nuembrecht, Germany). These 20 mL were used for filtration and centrifugal filtration, 12 mL of these treated samples were acidified with HNO3 and used for titanium analysis. The remaining 8 mL of filtrated samples were stored at 4 °C without acidification for single particle analysis. Single particle analysis, however, was not performed since results of the pre-study indicated that solutions are free of particles.

Toxicokinetic parameters:

other: bioaccessiblity

Remarks:

GST 2h: total Ti ± SD in samples (background corrected) [μg/L]: 0.82 ± 0.13; PBS 24 h: total Ti ± SD in samples (background corrected) [μg/L]: below LOD

Bioaccessibility (or Bioavailability) testing results:

Under conditions of this test (flasks with different artificial physiological media; TiO2-E171-C loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.06 (in two samples below LOQ*; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 1.03 ± 0.13
- total Ti ± SD in samples (background corrected) [μg/L]: 0.82 ± 0.13
* Concentrations < LOQ and > LOD were estimated with LOD + ½* (LOQ – LOD)

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.05 (in two samples below LOQ; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 4.76 ± 0.55
- total Ti ± SD in samples (background corrected) [μg/L]: 4.55 ± 0.55
PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L] :in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L] :in three samples below LOD; in one sample below LOQ
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

Method validation summary ICP-OES

Validation parameter	Results	Comment
Selectivity	Similar data with different wavelengths for ICP-OES method	-
Linearity	Applied calibration functions were linear	correlation coefficient at least 0.999939
Limit of detection	Ti: 0.03 – 0.19 μg/L
Limit of quantification	Ti: 0.09 – 0.56 μg/L
Method blanks	GST: - two method blanks after 2h: below LOQ, - two method blanks after 2h: above LOQ but on average 4.89 fold lower than concentrations in test samples after 2h, - two method blanks after 24h: below LOQ, - two method blanks after 24h: above LOQ but on average 22.7 fold lower than concentrations in test samples after 24h PBS: All method blanks: below LOD/LOQ
Accuracy measurement / Reproducibility - mass balance samples	Mean recovery for CRM TM- 25.4 (dilution factor 8, 3.15 μg Ti/L): Ti: 102 ± 2.0 % (n = 4)	Low concentration range mass balance
Accuracy measurement/ Reproducibility - GST test samples	Mean recovery for CRM TM- 25.4 (dilution factor 3, 8.40 μg Ti/L): Ti: 99.9 ± 1.1 % (n = 5)	Mid concentration range GST
Accuracy measurement / Reproducibility – PBS test samples	Mean recovery for CRM TM- 25.4 (dilution factor 10, 2.52 μg Ti/L): Ti: 97.8 ± 0.3 % (n = 5)	Mid concentration range PBS
Trueness measurement - GST test and mass balance samples	Mean recovery for quality control standard (CPI 2.50 μg Ti/L): Ti: 101 ± 1.5 % (n = 9)	Low concentration range
Trueness measurement - PBS test samples	Mean recovery for quality control standard (CPI 3 μg Ti/L): Ti: 96.3 ± 1.2 % (n = 5)	High concentration range PBS
Trueness measurement - PBS test samples	Mean recovery for quality control standard (CPI 5 μg Ti/L): Ti: 95.5 ± 0.6 % (n = 5)	High concentration range PBS
Trueness measurement - GST test and mass balance samples	Mean recovery for quality control standard (CPI 25 μg Ti/L): Ti: 97.9 ± 1.4 % (n = 9)	High concentration range
Trueness measurement - PBS test samples	Mean recovery for recalibration standard (1 μg Ti/L): Ti: 103 ± 3.0 % (n = 5)	Low concentration range PBS
Trueness measurement - GST test and mass balance samples	Mean recovery for recalibration standard (10 μg Ti/L): Ti: 101 ± 3.4 % (n = 9)	Mid concentration range
Trueness test samples	Fortification of samples: GST- Ti: 101 – 105 %. PBS- Ti: 107 – 108 %
Trueness mass balance	Fortification of samples: Ti: 98.4 – 113 %

Solution pH values

After preparation of the artificial physiological media, solution pH was adjusted to their respective target pH. Solution pH values prior to the test and after 2 h and 24 h in the different media were measured using a multi 9430 digital IDS sensor equipped with a SenTix 940-3 electrode from WTW (Weilheim, Germany) directly in the test vessel. The target pH of all media before addition of the test substance was in the nominal range. During the study, the pH of GST and PBS solutions remained stable (± 0.02).

Temperature control

The test was performed on an incubated laboratory shaker (Shaking incubation cabinet Minitron, INFORS AG, Bottmingen, Switzerland) at 100 rpm. The temperature was adjusted to 37.5 °C in the thermostatically controlled shaking cabinet to reach a temperature of 37 ± 2 °C in the media. The temperature in the cabinet was monitored by a certified thermometer and a thermo data logger.

Mass balance calculation

Measured dissolved titanium concentrations (filtered through a 0.2 μm filter) in test solutions and in the rinse of the filters indicate an incomplete dissolution of the test material TiO2- E171 -C after addition of aqua regia to the physiological media in sample vessels. An incomplete dissolution was confirmed by the presence of particles in the media. Hence, recoveries for the mass balance are low. The results and the recovery of the test item are compiled in following table .

Amount of titanium in mass balance samples (vessels, filter/syringes and taken samples) for each media

Media	Measured dissolved Ti after addition of aqua regia + filter/syringe rinse + samples [mg] *	Nominal amount [mg] #	Recovery [%]
GST vessel 1	0.012	29.97	0.039
GST vessel 2	0.012	29.89	0.040
GST vessel 3	0.011	29.90	0.037
PBS vessel 1	0.063	30.04	0.210
PBS vessel 2	0.015	30.08	0.050
PBS vessel 3	0.015	30.05	0.050

* Measured dissolved Ti includes the amount of titanium in test solutions after addition of aqua regia, the amount of titanium rinsed of used filters and syringe and the amount of titanium removed with samples taken during the study.

Example: 0.008 mg (amount Ti in vessels) + 0.003 mg (amount Ti in filters/syringes) + < 0.000 mg (amount Ti in samples taken) = 0.012 mg

# nominal amount TiO2 = 99.2% in test item (according to CoA) -> 59.9343 % Ti in TiO2 -> 29.8 mg Ti in 50 mg test item -> nominal amount 29.7 * initial weight / 50mg

Example: 29.7 mg Ti * 50.405 mg TiO2 / 50 mg TiO2 = 30.0 mg Ti

Conclusions:

Under conditions of this test (flasks with different artificial physiological media; TiO2-E171-C loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.06 (in two samples below LOQ*; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 1.03 ± 0.13
- total Ti ± SD in samples (background corrected) [μg/L]: 0.82 ± 0.13
* Concentrations < LOQ and > LOD were estimated with LOD + ½* (LOQ – LOD)

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.08, LOQ = 0.24
- total Ti ± SD in method blanks [μg/L]: 0.21 ± 0.05 (in two samples below LOQ; in two samples above LOQ)
- total Ti ± SD in samples [μg/L]: 4.76 ± 0.55
- total Ti ± SD in samples (background corrected) [μg/L]: 4.55 ± 0.55
PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L] :in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.19, LOQ = 0.56
- total Ti ± SD in method blanks [μg/L] :in three samples below LOD; in one sample below LOQ
- total Ti ± SD in samples [μg/L]: in all samples below LOD
- total Ti ± SD in samples (background corrected) [μg/L]: -

Reference 20

Endpoint:

basic toxicokinetics, other

Remarks:

Bioaccessibility

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

2015-11-27 to 2016-01-14

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

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:

2001

Deviations:

yes

Principles of method if other than guideline:

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) and phosphate-buffered saline (PBS).

GLP compliance:

yes (incl. QA statement)

Remarks:

signed 2011-02-07

Species:

other: in vitro (simulated human body fluids)

Details on test animals or test system and environmental conditions:

Test principle in brief:
- two different artificial physiological media,
- single loading of test substance of ~100 mg/L,
- samples taken after 2 and 24 hours agitation (100 rpm) at 37 ± 2 °C,
- two additional method blanks per medium, measurement (by ICP-OES) of dissolved titanium concentrations after filtration and centrifugal filtration.
- the study was performed in triplicate

The aim of this test was to assess the dissolution of TiO2-E171-D in two artificial physiological media: Phosphate buffered saline (PBS, pH 7.2-7.4) and Artificial gastric fluid (GST, 1.5-1.6). The test media were selected to simulate relevant human-chemical interactions (as far as practical), i.e. a substance entering the human body by ingestion into the gastro-intestinal tract.

Details on study design:

PROCEDURE
Three replicate flasks (500 mL glass flasks) per test medium (PBS, GST) were prepared with a loading of ~ 100 mg/L. Two control blank replicates (same procedure) per test medium were also prepared. Thus, three replicates containing the test item and two method blanks per artificial medium were tested; solutions were sampled after 2 and 24 h, to measure total dissolved titanium concentrations (by ICP-OES) after filtration (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) and centrifugal filtration (3kDa centrifugal filter, Sartorius, Göttingen, Germany), temperature and pH. During the study, observations, including the appearance of the solution (including colour, turbidity and particle film on the surface) were recorded.
After the test, 80 mL aqua regia (3 : 1 mixture of concentrated hydrochloric and nitric acid) were added to the vessels containing the undissolved test item. The filters used for sampling were extensively rinsed with 5 mL aqua regia. Solutions were sampled after at least 24 h, titanium concentration were measured in at least one mass balance sample (vessel and filter/syringes) for each medium by ICP-OES, and the mass balance was calculated.

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).

METAL ANALYSIS
- Standards for metal analysis: titanium standard (Merck Certipur Titanium ICP standard 1000 mg/L lot no. HC383640; Darmstadt, Germany; Merck)
- Certified reference materials: certified aqueous reference material TM-25.4 (lot no. 0914) obtained from Environment Canada and a multielement standard (CPI 19 Elements, lot no. 15C267)

Instrumental and analytical set-up for the ICP-OES instrument:
- Agilent 720, Agilent Technologies, Waldbronn, Germany
- Nebulizer: Sea spray nebulizer from Agilent
- Spray chamber: Glass cyclonic spray chamber from Agilent
- Carrier gas flow: 0.75 L/min
- RF power: 1200W
- Wavelengths Ti: 307.866 nm, 324.199 nm, 332.945 nm, 334.188 nm, 334.941 nm, 337.280 nm and 376.132 nm
- Calibration: blank, 0.5 μg/L, 0.75 μg/L, 1 μg/L, 2.5 μg/L, 5 μg/L, 7.5 μg/L, 10 μg/L, 20 μg/L, 30 μg/L, 40 μg/L, 50 μg/L, 75 µL, 100 µg/L and 250 µg/L
- Correlation coefficients (r) for the wavelengths used for evaluation of data were at least 0.999890

The applied LOD/LOQ calculations for the Agilent 720 ICP-OES are (according to DIN 32645):
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" in the validation summary.

From selected artificial physiological media, samples were fortified with a known amount of titanium (by standard addition of commercial standards) to determine the standard recovery of titanium. Detailed tables were stated in "Any other information on materials and methods incl. tables". For fortified test samples, recoveries were in the range of 102 - 103 % for Ti in GST test samples, 93.8 – 105 % for Ti in PBS test samples and 90.5 – 110 % for Ti in mass balance samples.

Details on dosing and sampling:

Loading:
Detailed loadings of the test vessels are given in "Any other information on materials and methods incl. tables".

Sampling:
Aqueous samples of approx. 20 mL taken for Ti analysis were transferred into disposable scintillation vials (20 mL scintillation tubes, Sarstedt, Nuembrecht, Germany). These 20 mL were used for filtration and centrifugal filtration, 12 mL of these treated samples were acidified with HNO3 and used for titanium analysis. The remaining 8 mL of filtrated samples were stored at 4 °C without acidification for single particle analysis. Single particle analysis, however, was not performed since results of the pre-study indicated that solutions are free of particles.

Toxicokinetic parameters:

other: bioacessibility

Remarks:

GST 2 h: - total Ti ± SD in samples [μg/L]: 8.24 ± 1.43; PBS 24 h: total Ti ± SD in samples [μg/L]: in all samples below LOD

Bioaccessibility (or Bioavailability) testing results:

Under conditions of this test (flasks with different artificial physiological media; TiO2-E171-D loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In all samples below LOQ
- total Ti ± SD in samples [μg/L]: 8.24 ± 1.43

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In one sample below LOD, in all other samples below LOQ
- total Ti ± SD in samples [μg/L]: 31.1 ± 1.73

PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: In one sample below LOQ, in all other samples below LOD

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]:in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD

Method validation summary ICP-OES

Validation parameter	Results	Comment
Selectivity	Similar data with different wavelengths for ICP-OES method	-
Linearity	Applied calibration functions were linear	correlation coefficient at least 0.999890
Limit of detection	Ti: 0.06 – 0.27 μg/L
Limit of quantification	Ti: 0.18 – 0.81 μg/L
Method blanks	GST: In one sample below LOD, in all other method blank samples below LOQ PBS: In all samples below LOD
Accuracy measurement / Reproducibility - mass balance samples	Mean recovery for CRM TM-25.4 (dilution factor 10, 2.52 μg Ti/L): Ti: 99.4 ± 3.6 % (n = 4)	Low concentration range mass balance
Accuracy measurement/ Reproducibility -   mass balance samples	Mean recovery for CRM TM-25.4 (dilution factor 2, 12.60 μg Ti/L): Ti: 100 ± 1.3 % (n = 4)	Mid concentration range  mass balance
Accuracy measurement/ Reproducibility - GST test samples	Mean recovery for CRM TM-25.4 (dilution factor 3, 8.40 μg Ti/L): Ti: 99.8 ± 1.1 % (n = 5)	Mid concentration range GST
Accuracy measurement / Reproducibility – PBS test samples	Mean recovery for CRM TM-25.4 (dilution factor 10, 2.52 μg Ti/L): Ti: 96.1 ± 1.8 % (n = 5)	Low concentration range PBS
Trueness measurement - mass balance samples	Mean recovery for CPI quality control standard (50 μg Ti/L): Ti: 97.9 ± 1.1 % (n = 3)	Mid concentration range, mass balance
Trueness measurement - mass balance samples	Mean recovery for CPI quality control standard (100 μg Ti/L): Ti: 96.2 ± 1.1 % (n = 3)	High concentration range, mass balance
Trueness measurement - GST test samples	Mean recovery for CPI quality control standard (2.50 μg Ti/L): Ti: 107 ± 1.6 % (n = 5)	Low concentration range, GST
Trueness measurement - GST test samples	Mean recovery for CPI quality control standard (25 μg Ti/L): Ti: 100 ± 0.4 % (n = 5)	High concentration range, GST
Trueness measurement - PBS test samples	Mean recovery for CPI quality control standard (3 μg Ti/L): Ti: 98.5 ± 3.6 % (n = 5)	Mid concentration range, PBS
Trueness measurement - PBS test samples	Mean recovery for CPI quality control standard (5 μg Ti/L): Ti: 96.0 ± 1.4 % (n = 5)	High concentration range, PBS
Trueness measurement - mass balance samples	Mean recovery for recalibration standard (25 μg Ti/L): Ti: 101 ± 2.4 % (n = 4)	Mid concentration range, mass balance
Trueness measurement - mass balance samples	Mean recovery for recalibration standard (75 μg Ti/L): Ti: 100 ± 1.2 % (n = 4)	High concentration range, mass balance
Trueness measurement - GST test samples	Mean recovery for recalibration standard (10 μg Ti/L): Ti: 102 ± 0.5 % (n = 5)	Mid concentration range; GST
Trueness measurement - PBS test samples	Mean recovery for recalibration standard (1 μg Ti/L): Ti: 99.1 ± 4.6 % (n = 5)	Low concentration range, PBS
Trueness test samples	Fortification of samples: GST- Ti: 102 - 103 % PBS- Ti: 93.8 - 105 %
Trueness mass balance	Fortification of samples: Ti: 90.5 – 110 %

Solution pH values

After preparation of the artificial physiological media, solution pH was adjusted to their respective target pH. Solution pH values prior to the test and after 2 h and 24 h in the different media were measured using a multi 9430 digital IDS sensor equipped with a SenTix 940-3 electrode from WTW (Weilheim, Germany) directly in the test vessel. The target pH of all media before addition of the test substance was in the nominal range. During the study, the pH of GST and PBS solutions remained stable (± 0.05).

Temperature control

The test was performed on an incubated laboratory shaker (Shaking incubation cabinet Minitron, INFORS AG, Bottmingen, Switzerland) at 100 rpm. The temperature was adjusted to 37.5 °C in the thermostatically controlled shaking cabinet to reach a temperature of 37 ± 2 °C in the media. The temperature in the cabinet was monitored by a certified thermometer and a thermo data logger.

Mass balance calculation

Measured dissolved titanium concentrations (filtered through a 0.2 μm filter) in test solutions and in the rinse of the filters indicate an incomplete dissolution of the test material TiO2- E171 -D after addition of aqua regia to the physiological media in sample vessels. An incomplete dissolution was confirmed by the presence of particles in the media. Hence, recoveries for the mass balance are low. The results and the recovery of the test item are compiled in following table .

Amount of titanium in mass balance samples (vessels, filter/syringes and taken samples) for each media

Media	Measured dissolved Ti after addition of aqua regia + filter/syringe rinse + samples [mg] *	Nominal amount [mg] #	Recovery [%]
GST vessel 1	0.026	29.761	0.086
GST vessel 2	0.025	29.736	0.085
GST vessel 3	0.026	29.751	0.088
PBS vessel 1	0.034	29.743	0.114
PBS vessel 2	0.034	29.746	0.115
PBS vessel 3	0.034	29.743	0.114

* Measured dissolved Ti includes the amount of titanium in test solutions after addition of aqua regia, the amount of titanium rinsed of used filters and syringe and the amount of titanium removed with samples taken during the study.

Example: 0.024 mg (amount Ti in vessels) + 0.00001 mg (amount Ti in filters/syringes) + 0.001 mg (amount Ti in samples taken) = 0.026 mg

# nominal amount TiO2 = 99.2% in test item (according to CoA) -> 59.9343 % Ti in TiO2 -> 29.7 mg Ti in 50 mg test item -> nominal amount 29.7 * initial weight / 50mg

Example: 29.7 mg Ti * 50.056 mg TiO2 / 50 mg TiO2 = 29.761 mg Ti

Conclusions:

Under conditions of this test (flasks with different artificial physiological media; TiO2-E171-D loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In all samples below LOQ
- total Ti ± SD in samples [μg/L]: 8.24 ± 1.43

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In one sample below LOD, in all other samples below LOQ
- total Ti ± SD in samples [μg/L]: 31.1 ± 1.73

PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: In one sample below LOQ, in all other samples below LOD

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]:in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD

Reference 21

Endpoint:

basic toxicokinetics, other

Remarks:

Bioaccessibility

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

2015-11-27 to 2016-01-14

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

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:

2001

Deviations:

yes

Principles of method if other than guideline:

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) and phosphate-buffered saline (PBS).

GLP compliance:

yes (incl. QA statement)

Remarks:

signed 2011-02-07

Species:

other: in vitro (simulated human body fluids)

Details on test animals or test system and environmental conditions:

Test principle in brief:
- two different artificial physiological media,
- single loading of test substance of ~100 mg/L,
- samples taken after 2 and 24 hours agitation (100 rpm) at 37 ± 2 °C,
- two additional method blanks per medium, measurement (by ICP-OES) of dissolved titanium concentrations after filtration and centrifugal filtration.
- the study was performed in triplicate

The aim of this test was to assess the dissolution of TiO2-UF-E in two artificial physiological media: Phosphate buffered saline (PBS, pH 7.2-7.4) and Artificial gastric fluid (GST, 1.5-1.6). The test media were selected to simulate relevant human-chemical interactions (as far as practical), i.e. a substance entering the human body by ingestion into the gastro-intestinal tract.

Details on study design:

PROCEDURE
Three replicate flasks (500 mL glass flasks) per test medium (PBS, GST) were prepared with a loading of ~ 100 mg/L. Two control blank replicates (same procedure) per test medium were also prepared. Thus, three replicates containing the test item and two method blanks per artificial medium were tested; solutions were sampled after 2 and 24 h, to measure total dissolved titanium concentrations (by ICP-OES) after filtration (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) and centrifugal filtration (3kDa centrifugal filter, Sartorius, Göttingen, Germany), temperature and pH. During the study, observations, including the appearance of the solution (including colour, turbidity and particle film on the surface) were recorded.
After the test, 80 mL aqua regia (3 : 1 mixture of concentrated hydrochloric and nitric acid) were added to the vessels containing the undissolved test item. The filters used for sampling were extensively rinsed with 5 mL aqua regia. Solutions were sampled after at least 24 h, titanium concentration were measured in at least one mass balance sample (vessel and filter/syringes) for each medium by ICP-OES, and the mass balance was calculated.

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).

METAL ANALYSIS
- Standards for metal analysis: titanium standard (Merck Certipur Titanium ICP standard 1000 mg/L lot no. HC383640; Darmstadt, Germany; Merck)
- Certified reference materials: certified aqueous reference material TM-25.4 (lot no. 0914) obtained from Environment Canada and a multielement standard (CPI 19 Elements, lot no. 15C267)

Instrumental and analytical set-up for the ICP-OES instrument:
- Agilent 720, Agilent Technologies, Waldbronn, Germany
- Nebulizer: Sea spray nebulizer from Agilent
- Spray chamber: Glass cyclonic spray chamber from Agilent
- Carrier gas flow: 0.75 L/min
- RF power: 1200W
- Wavelengths Ti: 307.866 nm, 324.199 nm, 332.945 nm, 334.188 nm, 334.941 nm, 337.280 nm and 376.132 nm
- Calibration: blank, 0.5 μg/L, 0.75 μg/L, 1 μg/L, 2.5 μg/L, 5 μg/L, 7.5 μg/L, 10 μg/L, 20 μg/L, 30 μg/L, 40 μg/L, 50 μg/L, 75 µL, 100 µg/L and 250 µg/L
- Correlation coefficients (r) for the wavelengths used for evaluation of data were at least 0.999890

The applied LOD/LOQ calculations for the Agilent 720 ICP-OES are (according to DIN 32645):
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" in the validation summary.

From selected artificial physiological media, samples were fortified with a known amount of titanium (by standard addition of commercial standards) to determine the standard recovery of titanium. Detailed tables were stated in "Any other information on materials and methods incl. tables". For fortified test samples, recoveries were in the range of 104 - 110 % for Ti in GST test samples, 98.2 – 105 % for Ti in PBS test samples and 93.3 – 99.9 % for Ti in mass balance samples.

Details on dosing and sampling:

Loading:
Detailed loadings of the test vessels are given in "Any other information on materials and methods incl. tables".

Sampling:
Aqueous samples of approx. 20 mL taken for Ti analysis were transferred into disposable scintillation vials (20 mL scintillation tubes, Sarstedt, Nuembrecht, Germany). These 20 mL were used for filtration and centrifugal filtration, 12 mL of these treated samples were acidified with HNO3 and used for titanium analysis. The remaining 8 mL of filtrated samples were stored at 4 °C without acidification for single particle analysis. Single particle analysis, however, was not performed since results of the pre-study indicated that solutions are free of particles.

Toxicokinetic parameters:

other: bioaccessibility

Remarks:

GST 2 h: total Ti ± SD in samples [μg/L]: 2.36 ± 0.18; PBS 24 h: total Ti ± SD in samples [μg/L]: In one sample below LOQ, in all other samples below LOD

Bioaccessibility (or Bioavailability) testing results:

Under conditions of this test (flasks with different artificial physiological media; TiO2-UF-E loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In all samples below LOQ
- total Ti ± SD in samples [μg/L]: 2.36 ± 0.18

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In one sample below LOD, in all other samples below LOQ
- total Ti ± SD in samples [μg/L]: 4.72 ± 0.12

PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: In one sample below LOQ, in all other samples below LOD

Method validation summary ICP-OES

Validation parameter	Results	Comment
Selectivity	Similar data with different wavelengths for ICP-OES method	-
Linearity	Applied calibration functions were linear	correlation coefficient at least 0.999890
Limit of detection	Ti: 0.06 – 0.27 μg/L
Limit of quantification	Ti: 0.18 – 0.81 μg/L
Method blanks	GST: In one sample below LOD, in all other method blank samples below LOQ PBS: In all samples below LOD
Accuracy measurement / Reproducibility - mass balance samples	Mean recovery for CRM TM-25.4 (dilution factor 10, 2.52 μg Ti/L): Ti: 99.4 ± 3.6 % (n = 4)	Low concentration range, mass balance
Accuracy measurement/ Reproducibility -   mass balance samples	Mean recovery for CRM TM-25.4 (dilution factor 2, 12.60 μg Ti/L): Ti: 100 ± 1.3 % (n = 4)	Mid concentration range, mass balance
Accuracy measurement/ Reproducibility - GST test samples	Mean recovery for CRM TM-25.4 (dilution factor 3, 8.40 μg Ti/L): Ti: 99.8 ± 1.1 % (n = 5)	Mid concentration range, GST
Accuracy measurement / Reproducibility – PBS test samples	Mean recovery for CRM TM-25.4 (dilution factor 10, 2.52 μg Ti/L): Ti: 96.1 ± 1.8 % (n = 5)	Low concentration range, PBS
Trueness measurement - mass balance samples	Mean recovery for CPI quality control standard (50 μg Ti/L): Ti: 97.9 ± 1.1 % (n = 3)	Mid concentration range, mass balance
Trueness measurement - mass balance samples	Mean recovery for CPI quality control standard (100 μg Ti/L): Ti: 96.2 ± 1.1 % (n = 3)	High concentration range, mass balance
Trueness measurement - GST test samples	Mean recovery for CPI quality control standard (2.50 μg Ti/L): Ti: 107 ± 1.6 % (n = 5)	Low concentration range, GST
Trueness measurement - GST test samples	Mean recovery for CPI quality control standard (25 μg Ti/L): Ti: 100 ± 0.4 % (n = 5)	High concentration range, GST
Trueness measurement - PBS test samples	Mean recovery for CPI quality control standard (3 μg Ti/L): Ti: 98.5 ± 3.6 % (n = 5)	Mid concentration range, PBS
Trueness measurement - PBS test samples	Mean recovery for CPI quality control standard (5 μg Ti/L): Ti: 96.0 ± 1.4 % (n = 5)	High concentration range, PBS
Trueness measurement - mass balance samples	Mean recovery for recalibration standard (25 μg Ti/L): Ti: 101 ± 2.4 % (n = 4)	Mid concentration range, mass balance
Trueness measurement - mass balance samples	Mean recovery for recalibration standard (75 μg Ti/L): Ti: 100 ± 1.2 % (n = 4)	High concentration range, mass balance
Trueness measurement - GST test samples	Mean recovery for recalibration standard (10 μg Ti/L): Ti: 102 ± 0.5 % (n = 5)	Mid concentration range; GST
Trueness measurement - PBS test samples	Mean recovery for recalibration standard (1 μg Ti/L): Ti: 99.1 ± 4.6 % (n = 5)	Low concentration range, PBS
Trueness test samples	Fortification of samples: GST- Ti: 104 - 110 % PBS- Ti: 98.2 – 105 %
Trueness mass balance	Fortification of samples: Ti: 93.3 – 99.9 %

Solution pH values

After preparation of the artificial physiological media, solution pH was adjusted to their respective target pH. Solution pH values prior to the test and after 2 h and 24 h in the different media were measured using a multi 9430 digital IDS sensor equipped with a SenTix 940-3 electrode from WTW (Weilheim, Germany) directly in the test vessel. The target pH of all media before addition of the test substance was in the nominal range. During the study, the pH of GST and PBS solutions remained stable (± 0.05).

Temperature control

The test was performed on an incubated laboratory shaker (Shaking incubation cabinet Minitron, INFORS AG, Bottmingen, Switzerland) at 100 rpm. The temperature was adjusted to 37.5 °C in the thermostatically controlled shaking cabinet to reach a temperature of 37 ± 2 °C in the media. The temperature in the cabinet was monitored by a certified thermometer and a thermo data logger.

Mass balance calculation

Measured dissolved titanium concentrations in test solutions (filtered through a 0.2 μm filter) and in the rinse of the filters confirmed an incomplete digestion of TiO2-UF-E by aqua regia. An incomplete dissolution was confirmed by the presence of particles in the media. Hence, recoveries for the mass balance are low. The results and the recovery of the weighted test item (approx. 50 mg/ 500 mL) are compiled in following table.

Amount of titanium in mass balance samples (vessels, filter/syringes and taken samples) for each media

Media	Measured dissolved Ti after addition of aqua regia + filter/syringe rinse + samples [mg] *	Nominal amount [mg] #	Recovery [%]
GST vessel 1	0.017	28.702	0.059
GST vessel 2	0.017	28.692	0.058
GST vessel 3	0.017	28.703	0.060
PBS vessel 1	0.046	28.700	0.161
PBS vessel 2	0.046	28.710	0.162
PBS vessel 3	0.047	28.687	0.165

* Measured dissolved Ti includes the amount of titanium in test solutions after addition of aqua regia, the amount of titanium rinsed of used filters and syringe and the amount of titanium removed with samples during the study.

Example: 0.016 mg (amount Ti in vessels) + 0.0002 mg (amount Ti in filters/syringes) + 0.0003 mg (amount Ti in samples) = 0.017 mg

# nominal amount TiO2 = 95.70% in test item (according to CoA) -> 59.9343 % Ti in TiO2 -> 28.7 mg Ti in 50 mg test item -> nominal amount 28.7 * initial weight / 50mg

Example: 28.7 mg Ti * 50.040 mg TiO2 / 50 mg TiO2 = 28.702 mg Ti

Conclusions:

Under conditions of this test (flasks with different artificial physiological media; TiO2-UF-E loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In all samples below LOQ
- total Ti ± SD in samples [μg/L]: 2.36 ± 0.18

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In one sample below LOD, in all other samples below LOQ
- total Ti ± SD in samples [μg/L]: 4.72 ± 0.12

PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: In one sample below LOQ, in all other samples below LOD

Reference 22

Endpoint:

basic toxicokinetics, other

Remarks:

Bioaccessibility

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

2015-11-28 to 2016-01-14

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

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:

2001

Deviations:

yes

Principles of method if other than guideline:

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) and phosphate-buffered saline (PBS).

GLP compliance:

yes (incl. QA statement)

Remarks:

signed 2011-02-07

Species:

other: in vitro (simulated human body fluids)

Details on test animals or test system and environmental conditions:

Test principle in brief:
- two different artificial physiological media,
- single loading of test substance of ~100 mg/L,
- samples taken after 2 and 24 hours agitation (100 rpm) at 37 ± 2 °C,
- two additional method blanks per medium, measurement (by ICP-OES) of dissolved titanium concentrations after filtration and centrifugal filtration.
- the study was performed in triplicate

The aim of this test was to assess the dissolution of TiO2-UF-F in two artificial physiological media: Phosphate buffered saline (PBS, pH 7.2-7.4) and Artificial gastric fluid (GST, 1.5-1.6). The test media were selected to simulate relevant human-chemical interactions (as far as practical), i.e. a substance entering the human body by ingestion into the gastro-intestinal tract.

Details on study design:

PROCEDURE
Three replicate flasks (500 mL glass flasks) per test medium (PBS, GST) were prepared with a loading of ~ 100 mg/L. Two control blank replicates (same procedure) per test medium were also prepared. Thus, three replicates containing the test item and two method blanks per artificial medium were tested; solutions were sampled after 2 and 24 h, to measure total dissolved titanium concentrations (by ICP-OES) after filtration (Syringe Filter w / 0.2 μm, polyethersulfon membrane, DIA Nielsen, Dueren, Germany) and centrifugal filtration (3kDa centrifugal filter, Sartorius, Göttingen, Germany), temperature and pH. During the study, observations, including the appearance of the solution (including colour, turbidity and particle film on the surface) were recorded.
After the test, 80 mL aqua regia (3 : 1 mixture of concentrated hydrochloric and nitric acid) were added to the vessels containing the undissolved test item. The filters used for sampling were extensively rinsed with 5 mL aqua regia. Solutions were sampled after at least 24 h, titanium concentration were measured in at least one mass balance sample (vessel and filter/syringes) for each medium by ICP-OES, and the mass balance was calculated.

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).

METAL ANALYSIS
- Standards for metal analysis: titanium standard (Merck Certipur Titanium ICP standard 1000 mg/L lot no. HC383640; Darmstadt, Germany; Merck)
- Certified reference materials: certified aqueous reference material TM-25.4 (lot no. 0914) obtained from Environment Canada and a multielement standard (CPI 19 Elements, lot no. 15C267)

Instrumental and analytical set-up for the ICP-OES instrument:
- Agilent 720, Agilent Technologies, Waldbronn, Germany
- Nebulizer: Sea spray nebulizer from Agilent
- Spray chamber: Glass cyclonic spray chamber from Agilent
- Carrier gas flow: 0.75 L/min
- RF power: 1200W
- Wavelengths Ti: 307.866 nm, 324.199 nm, 332.945 nm, 334.188 nm, 334.941 nm, 337.280 nm and 376.132 nm
- Calibration: blank, 0.5 μg/L, 0.75 μg/L, 1 μg/L, 2.5 μg/L, 5 μg/L, 7.5 μg/L, 10 μg/L, 20 μg/L, 30 μg/L, 40 μg/L, 50 μg/L, 75 µL, 100 µg/L and 250 µg/L
- Correlation coefficients (r) for the wavelengths used for evaluation of data were at least 0.999890

The applied LOD/LOQ calculations for the Agilent 720 ICP-OES are (according to DIN 32645):
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" in the validation summary.

From selected artificial physiological media, samples were fortified with a known amount of titanium (by standard addition of commercial standards) to determine the standard recovery of titanium. Detailed tables were stated in "Any other information on materials and methods incl. tables". For fortified test samples, recoveries were in the range of 101 - 103 % for Ti in GST test samples, 93.7 – 102 % for Ti in PBS test samples and 94.5 – 104 % for Ti in mass balance samples.

Details on dosing and sampling:

Loading:
Detailed loadings of the test vessels are given in "Any other information on materials and methods incl. tables".

Sampling:
Aqueous samples of approx. 20 mL taken for Ti analysis were transferred into disposable scintillation vials (20 mL scintillation tubes, Sarstedt, Nuembrecht, Germany). These 20 mL were used for filtration and centrifugal filtration, 12 mL of these treated samples were acidified with HNO3 and used for titanium analysis. The remaining 8 mL of filtrated samples were stored at 4 °C without acidification for single particle analysis. Single particle analysis, however, was not performed since results of the pre-study indicated that solutions are free of particles.

Toxicokinetic parameters:

other: bioaccessibility

Remarks:

GST 2 h: total Ti ± SD in samples [μg/L]: 1.12 ± 0.07; PBS 24 h: total Ti ± SD in samples [μg/L]: in all samples below LOD

Bioaccessibility (or Bioavailability) testing results:

Under conditions of this test (flasks with different artificial physiological media; TiO2-UF-F loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In all samples below LOQ
- total Ti ± SD in samples [μg/L]: 1.12 ± 0.07

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In one sample below LOD, in all other samples below LOQ
- total Ti ± SD in samples [μg/L]: 3.68 ± 0.02

PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: In one sample below LOQ, in all other samples below LOD

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD

Method validation summary ICP-OES

Validation parameter	Results	Comment
Selectivity	Similar data with different wavelengths for ICP-OES method	-
Linearity	Applied calibration functions were linear	correlation coefficient at least 0.999890
Limit of detection	Ti: 0.06 – 0.27 μg/L
Limit of quantification	Ti: 0.18 – 0.81 μg/L
Method blanks	GST: In one sample below LOD, in all other method blank samples below LOQ PBS: In all samples below LOD
Accuracy measurement / Reproducibility - mass balance samples	Mean recovery for CRM TM-25.4 (dilution factor 10, 2.52 μg Ti/L): Ti: 99.4 ± 3.6 % (n = 4)	Low concentration range, mass balance
Accuracy measurement/ Reproducibility -   mass balance samples	Mean recovery for CRM TM-25.4 (dilution factor 2, 12.60 μg Ti/L): Ti: 100 ± 1.3 % (n = 4)	Mid concentration range, mass balance
Accuracy measurement/ Reproducibility - GST test samples	Mean recovery for CRM TM-25.4 (dilution factor 3, 8.40 μg Ti/L): Ti: 99.8 ± 1.1 % (n = 5)	Mid concentration range, GST
Accuracy measurement / Reproducibility – PBS test samples	Mean recovery for CRM TM-25.4 (dilution factor 10, 2.52 μg Ti/L): Ti: 96.1 ± 1.8 % (n = 5)	Low concentration range, PBS
Trueness measurement - mass balance samples	Mean recovery for CPI quality control standard (50 μg Ti/L): Ti: 97.9 ± 1.1 % (n = 3)	Mid concentration range, mass balance
Trueness measurement - mass balance samples	Mean recovery for CPI quality control standard (100 μg Ti/L): Ti: 96.2 ± 1.1 % (n = 3)	High concentration range, mass balance
Trueness measurement - GST test samples	Mean recovery for CPI quality control standard (2.50 μg Ti/L): Ti: 107 ± 1.6 % (n = 5)	Low concentration range, GST
Trueness measurement - GST test samples	Mean recovery for CPI quality control standard (25 μg Ti/L): Ti: 100 ± 0.4 % (n = 5)	High concentration range, GST
Trueness measurement - PBS test samples	Mean recovery for CPI quality control standard (3 μg Ti/L): Ti: 98.5 ± 3.6 % (n = 5)	Mid concentration range, PBS
Trueness measurement - PBS test samples	Mean recovery for CPI quality control standard (5 μg Ti/L): Ti: 96.0 ± 1.4 % (n = 5)	High concentration range, PBS
Trueness measurement - mass balance samples	Mean recovery for recalibration standard (25 μg Ti/L): Ti: 101 ± 2.4 % (n = 4)	Mid concentration range, mass balance
Trueness measurement - mass balance samples	Mean recovery for recalibration standard (75 μg Ti/L): Ti: 100 ± 1.2 % (n = 4)	High concentration range, mass balance
Trueness measurement - GST test samples	Mean recovery for recalibration standard (10 μg Ti/L): Ti: 102 ± 0.5 % (n = 5)	Mid concentration range; GST
Trueness measurement - PBS test samples	Mean recovery for recalibration standard (1 μg Ti/L): Ti: 99.1 ± 4.6 % (n = 5)	Low concentration range, PBS
Trueness test samples	Fortification of samples: GST- Ti: 101 - 103 % PBS- Ti: 93.7 – 102 %
Trueness mass balance	Fortification of samples: Ti: 94.5 – 104 %

Solution pH values

After preparation of the artificial physiological media, solution pH was adjusted to their respective target pH. Solution pH values prior to the test and after 2 h and 24 h in the different media were measured using a multi 9430 digital IDS sensor equipped with a SenTix 940-3 electrode from WTW (Weilheim, Germany) directly in the test vessel. The target pH of all media before addition of the test substance was in the nominal range. During the study, the pH of GST and PBS solutions remained stable (± 0.06).

Temperature control

The test was performed on an incubated laboratory shaker (Shaking incubation cabinet Minitron, INFORS AG, Bottmingen, Switzerland) at 100 rpm. The temperature was adjusted to 37.5 °C in the thermostatically controlled shaking cabinet to reach a temperature of 37 ± 2 °C in the media. The temperature in the cabinet was monitored by a certified thermometer and a thermo data logger.

Mass balance calculation

Measured dissolved titanium concentrations in test solutions (filtered through a 0.2 μm filter) and in the rinse of the filters confirmed an incomplete digestion of TiO2-UF-F by aqua regia. An incomplete dissolution was confirmed by the presence of particles in the media. Hence, recoveries for the mass balance are low. The results and the recovery of the weighted test item (approx. 50 mg/ 500 mL) are compiled in following table.

Amount of titanium in mass balance samples (vessels, filter/syringes and taken samples) for each media

Media	Measured dissolved Ti after addition of aqua regia + filter/syringe rinse + samples [mg] *	Nominal amount [mg] #	Recovery [%]
GST vessel 1	0.028	29.869	0.092
GST vessel 2	0.028	29.853	0.094
GST vessel 3	0.028	29.865	0.093
PBS vessel 1	0.011	29.886	0.038
PBS vessel 2	0.012	29.868	0.039
PBS vessel 3	0.011	29.887	0.036

* Measured dissolved Ti includes the amount of titanium in test solutions after addition of aqua regia, the amount of titanium rinsed of used filters and syringe and the amount of titanium removed with samples during the study.

Example: 0.028 mg (amount Ti in vessels) + 0.00001 mg (amount Ti in filters/syringes) + 0.0002 mg (amount Ti in samples) = 0.02717 mg

# nominal amount TiO2 = 99.60 % in test item (according to CoA) -> 59.9343 % Ti in TiO2 -> 29.8 mg Ti in 50 mg test item -> nominal amount 29.8 * initial weight / 50mg Example: 29.8 mg Ti * 50.036 mg TiO2 / 50 mg TiO2 = 29.869 mg Ti

Conclusions:

Under conditions of this test (flasks with different artificial physiological media; TiO2-UF-F loadings of 50 mg per 500 mL, shaking at 37 °C, sampling after 2 h and 24 h), the following concentrations of dissolved titanium were measured:

GST 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In all samples below LOQ
- total Ti ± SD in samples [μg/L]: 1.12 ± 0.07

GST 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.13, LOQ = 0.40
- total Ti ± SD in method blanks [μg/L]: In one sample below LOD, in all other samples below LOQ
- total Ti ± SD in samples [μg/L]: 3.68 ± 0.02

PBS 2 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: In one sample below LOQ, in all other samples below LOD

PBS 24 h:
- LOD/LOQ of Ti measurement series [μg/L]: LOD = 0.27, LOQ = 0.81
- total Ti ± SD in method blanks [μg/L]: in all samples below LOD
- total Ti ± SD in samples [μg/L]: in all samples below LOD

Reference 23

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

2003-01-22 to 2003-04-03

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

comparable to guideline study

Objective of study:

absorption

distribution

excretion

Qualifier:

according to guideline

Guideline:

OECD Guideline 417 (Toxicokinetics)

Deviations:

yes

Remarks:

Measurements of the excretion were not conducted until 95% of the administered dose had been excreted or for 7 days.

GLP compliance:

yes (incl. QA statement)

Remarks:

signed 2001-04-03

Radiolabelling:

no

Species:

rat

Strain:

Sprague-Dawley

Sex:

male/female

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Charles River Ltd. Margate, UK
- Age at study initiation: 6-10 weeks
- Weight at study initiation: 117 g-154 g
- Housing:
- Animals not required for excretion/balance experiments were housed in groups
- Prior excretion/balance experiments (only for animals required for excretion/balance experiments): animals were housed individually in stainless steel cages with suspended mesh floors
- During excretion/balance experiments: animals were housed individually in glass Metabowls®, equipped with urine/faeces separators
- Individual metabolism cages: yes
- Diet ad libitum: SDS Rat & Mouse No. 1 fine ground (Special Diets Services Ltd., Witham, Essex, England)
- Water ad libitum: Anglian Water mains supply
- Acclimation period: 5 days

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 21 ± 2
- Humidity (%): 55 ± 15
- Air changes (per hr): minimum of 15
- Photoperiod (hrs dark / hrs light): 12/12

Route of administration:

oral: feed

Vehicle:

other: feed

Details on exposure:

DIET PREPARATION
- Rate of preparation of diet: prepared on single occasion
- Mixing appropriate amounts with (Type of food): Standard powdered rat diet was preepared by homogenisation of standard pelleted diet. Test substances were incorporated into standard powdered diet. Control diet was prepared by the same procedure but in the absence of test substance.

The test animals were fed with diets (powdered form) with different forms of TiO2 at approx. 200 ppm (equivalent to ca. 30 mg/kg bw) for 7 days after which the diets were replaced by control diet (pelleted form).
Diets were available to the animals ad libitum. Groups of animals were sacrificed at 1, 24 and 72 hours after withdrawal of the treated diets.

Duration and frequency of treatment / exposure:

7 days

Dose / conc.:

200 ppm (nominal)

Remarks:

according to 30 mg/kg bw/day assuming daily consumption of 30 g diet by a 200 g rat

No. of animals per sex per dose / concentration:

9 + 1 in excess

Control animals:

yes

yes, plain diet

Details on study design:

Dose selection rationale: dose level is lower than the NOAEL in a 103 week dietary study (=2500mg/kg bw/day)

Details on dosing and sampling:

TOXICOKINETIC STUDY (tissue distribution)
- Tissues and body fluids sampled: livers, kidneys, skeletal muscle, blood
- Time and frequency of sampling: 1 hr and 24 hr after withdrawal of treated diet
- From how many animals: 3 animals/sex/group

EXCRETION/BALANCE AND TISSUE DISTRIBUTION STUDIES (absorption, distribution, excretion)
- Tissues and body fluids sampled: urine, faeces, cage wash, livers, kidneys, skeletal muscle, blood
- Time and frequency of sampling: 72 hr after withdrawal of treated diet, additional sampling of urine and faeces at 24, 48 hr
- From how many animals: 3 animals/sex/group
- Method type(s) for identification: inductively coupled plasma atomic emission spectrometer (ICP-AES)
- Limits of detection and quantification: Lower limit of quantification (LLOQ): 2ng/mL

Statistics:

Data for tissues could not be statistically analysed because nearly all values were below the limit of quantification. For faeces, an analysis combining sexes was performed by two-way analysis of variance (ANOVA) with group and sex as factors. As the sex by group interaction was not significant, data from both sexes were pooled and analysed using one-way analysis of variance.

Type:

excretion

Results:

The total amount of titanium excreted in faeces accounted for means of 1.1-2.2 mg for male rats and 1.1-1.3 mg for female rats during 0 – 72 hours after withdrawal of the treated diets.

Type:

excretion

Results:

Titanium concentrations in urine were below the limit of quantification (<0.04 mg/l equivalent to <2% daily dose/l) in all but one sample: 0-24 hour urine from one male rat (0.05 mg/l).

Type:

distribution

Results:

Ti concentrations in liver, kidney and muscle were mainly below the limit of detection (<0.1 - <0.2 mg/kg wet weight) or in the range of 0.1 – 0.3 mg/kg wet weight for all groups including controls.

Type:

absorption

Results:

Whole-blood concentrations of titanium were <0.04 mg/ml for all animals and treatments.

Details on absorption:

not applicable

Details on distribution in tissues:

Whole-blood concentrations of titanium were <0.04 mg/ml for all animals and treatments.
Titanium concentrations in liver, kidney and muscle were mainly below the limit of detection (<0.1 - <0.2 mg/kg wet weight) or in the range of 0.1 – 0.3 mg/kg wet weight for all groups including controls. The limit of detection was expressed as a range of values as a consequence of the variability in weights of individual samples taken for analysis.

Details on excretion:

The total amount of titanium excreted in the faeces accounted for means of 1.4 - 2.2 mg for male rats and 1.1 - 1.3 mg for female rats during 0 - 72 hr after withdrawal of the treated diet. Titanium was only excreted in significant amounts in the faeces accounting for means of 39 - 63 % daily dose during 0 - 72 hr after withdrawal of treatment of the different forms of titanium dioxide. For each sampling interval there was no statistical evidence to suggest that faecal excretion between groups (i.e. different forms of titanium dioxide) was significantly different.
Titanium concentrations in urine were below the limit of quantification (< 0.04 mg/L equivalent to < 2 % daily dose/L) in all but one sample, which was 0.05 mg/L at 24 hr interval.

Metabolites identified:

no

Details on metabolites:

Titanium dioxide as an inorganic substance, which is not metabolised.

Bioaccessibility (or Bioavailability) testing results:

not applicable

Conclusions:

The results suggest that there is no substantial accumulation of titanium in tissues following administration of diets with different forms of titanium dioxide. According to the authors, the absence of titanium in the urine is indicative of low absorption of titanium, although biliary excretion cannot be excluded.

Reference 24

Endpoint:

basic toxicokinetics, other

Remarks:

bioaccessibility

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

comparable to guideline study with acceptable restrictions

Objective of study:

bioaccessibility (or bioavailability)

Qualifier:

no guideline followed

Principles of method if other than guideline:

Solubility of test item in simulated human fluids. Principle of test is similar to Transformation/Dissolution testing according to OECD Series 29 (2001)

GLP compliance:

no

Species:

other: in vitro (simulated human body fluids)

Details on test animals or test system and environmental conditions:

The test item was exposed in four different test media screening a relevant pH window ranging from about 1.5 to 7.5.
The test media were:
• Phosphate-buffered saline (PBS, pH 7.4), as a standard physiological solution that mimics the ion strength of human blood serum. It is generally used within research and medical health care and normally serves as a reference test solution for comparison of data under simulated physiological conditions.
• Gamble’s solution (GMB, pH 7.4) mimics interstitial fluid within the deep lung at normal health conditions.
• Artificial lysosomal fluid (ALF, pH 4.5) simulates interstitial conditions in the lung occurring in conjunction to phagocytosis by cells, which involves relatively aggressive conditions similar to an immunologic reaction of the body.
• Artificial gastric fluid (GST, pH 1.5) mimics the very aggressive digestion milieu of high acidity in the stomach.

The test media where chosen in order to simulate a relevant inhalation scenario where the titanium powder may enter the human body through inhalation and, subsequently by ingestion of inhaled particles that are translocated to the gastro-intestinal tract. It should be stressed though, that the different test media only simulate physiological conditions to a limited extent. The complexity and function of the real body fluids are difficult to simulate, however, in vitro results in synthetic biological media can, in a simple way, provide information that could be relevant for a real situation.

The test solutions were prepared using ultra-pure water and chemicals of analytical grades. The pH of ALF and GMB was adjusted using 50% NaOH and 25% HCl respectively. The pH of PBS was adjusted with 50% NaOH.

Artificial gastric fluid was prepared according to the ASTM standard using 4 g of 25% HCL solution that was diluted with ultra-pure water to 1 L (0.07 N HCl, N = ”Normal solution”, 1 N solution equals 1 g equivalent weight of compound in 1 L of water). This resulted in a pH about 1.5-1.6 (ASTM D5517).

References:
ASTM D5517-03 (2003), ”Standard Test Method for Determining Extractability of Metals from Art Materials”

Details on study design:

Experimental Procedure
Triplicate TiO2 powder samples were prepared for exposure in the different test media for two time periods respectively. In addition, one blank sample containing only the test solution was incubated together with the triplicates for each time period. 5 ± 0.5 mg of titanium dioxide powder was weighed and placed in a TPX Nalgene® jar. 50 mL of the test solution were then added to the powder sample, before incubation in the dark in a Merck mini incubator regulated at 37 ± 2 °C that was gently shaken (bi-linearly) with an intensity of 25 cycles per minute (no adjustment of solution volume to powder mass was made). The testing periods were 2 hours and 24 hours, respectively.

Details on dosing and sampling:

After the testing period, the samples were allowed to cool to ambient room temperature before the final pH of the test solution was measured. The test medium was then separated from the powder particles by centrifugation at 3000 rpm for 10 minutes. In some cases fine particles were floating on the solution surface after centrifugation. These particles were removed before the supernatant solution was decanted into a LDPE storage flask and acidified to pH <2 with 65% suprapur HNO3 prior to solution analysis.

The particle loading 0.1 g/L was selected since it is experimentally feasible even when low concentrations of released metal are expected. It is also used for testing according to the OECD transformation/dissolution protocol for sparingly soluble metals and metal compounds, which facilitates comparison with other data of metal release/dissolution (OECD, 2001).

Type:

other: bioaccessibility

Results:

Bioaccessibility data on titanium released from titanium dioxide were determined when exposed to synthetic biological media of varying pH and composition.

Results:

Only a small fraction of Ti was released/dissolved from the TiO2 powder during exposure to any of the media matrices of varying acidity & composition. A trend with somewhat higher release rates with increasing acidity & exposure period was evident.

Bioaccessibility (or Bioavailability) testing results:

Bioaccessibility data on titanium released from titanium dioxide were determined when exposed to synthetic biological media of varying pH and composition.
Only a small fraction of titanium was released/dissolved from the titanium dioxide powder during exposure to any of the media matrices of varying acidity and composition. A trend with somewhat higher release rates with increasing acidity and exposure period was evident.

Characterization of the test material

The specific surface area of titanium dioxide powder measured by BET-analysis is 8.51 m²/g.

 

Appearance and surface morphology:

Except for adventitious carbon on the surface, no other elements except titanium and oxygen were detected at the outermost surface layer of the unexposed powder. A relatively high content of carbon, 53 at% was observed on the TiO2 powder. (Atomic percent, at%, is the number of atoms of an element per unit volume, divided by the number of atoms per unit volume of the substance containing the element.) Some oxidized carbon was also observed.

XPS core level spectrum for Ti 2p, indicated that for the TiO2-powder, the binding energy is in good agreement with literature values on TiO2(XPS 1999). For TiO2, the relative (0.28) and the theoretical value (0.33) were similar.

 

XPS-data on unexposed powder:

Powder                          Ti 2p [eV]            Ti/(Ti+O) [rel at%]

Reference-Ti met         453,8

Reference-TiO2            458,7                           0,33

Test-item-TiO2               458,8 / 458,7             0,30 / 0,26

 

All values using adventitious C 1s as reference (285 eV)

Reference data: XPS International Inc., 1999

 

Solubility of titanium dioxide in simulated human fluids

Generally, the titanium concentrations were very low, however the titanium released in the more acidic test media (ALF and GST) show some kinetic behavior with increased release rates after 24 hours of exposure. The opposite with decreasing release rate with longer exposure time, is noted for the exposure in the matrix PBS of more neutral acidity.In the case of 2 hours of exposure in ALF matrix the result is based on duplicate samples since results from one of the three parallel samples deviated in orders of magnitude from the other two and therefore was excluded due to possible experimental errors. No titanium could be detected in the GMB test media after exposure for 2 and 24 hours (the detection limit in this media was 1.83μg/L. However, the value of released titanium are close to the limit of detection.

Total concentration of released Ti [µg/L]

Sample Material	PBS pH 7.4	GMB pH 7.4	ALF pH 4.5	GST pH 1.6
TiO22 hours	12.6 ± 3.9	not detected (< 1.83)	13.0 ± 0.4*	21.2 ± 3.8
TiO224 hours	2.8 ± 2.3	not detected (< 1.83)	51.0 ± 2.4	48.3 ± 4.0

 

The results of titanium release from titanium dioxide powder are reported as release rates of titanium per unit surface area and hour of exposure (μg/cm2/hour). In general, the titanium release rate from titanium dioxide powder is lower after 24 hours than after the first two hours of exposure. The highest release rate, 0.0013μg/cm2/hour, is obtained after 2 hours of exposure in the most acidic (aggressive) test media, GST. Calculated titanium release rates from titanium dioxide powder samples exposed in different artificial biological media for 2 and 24 hours are as follows:

 

Release rate of Ti [μg/cm²/hour].

Sample Material	PBS pH 7.4	GMB pH 7.4	ALF pH 4.5	GST pH 1.6
TiO22 hours	0.00070 +/- 0.00023	not detected	0.00077 +/- 0.00003*	0.00129 +/- 0.00025
TiO224 hours	0.00001 +/- 0.00001	not detected	0.00025 +/- 0.00001	0.00024 +/- 0.00002

 *result based on duplicate samples

Based on generated release data on titanium, the amount of titanium from the titanium dioxide powder material that did not dissolve or transform during the exposure to different test media was estimated. Results are presented as the titanium percentage of the initial amount of powder loaded (mass%). In general terms more than 99.9% of the titanium dioxide resisted dissolution in PBS, GMB and GST and ALF. The amount of dissolved/transformed titanium is presented as the maximum percentage determined from the triplicate (in one case duplicate) samples investigated.

 

Amount of Ti not transformed (dissolved) from titanium dioxide powder [mass%]

Sample Material	PBS pH 7.4	GMB pH 7.4	ALF pH 4.5	GST pH 1.6
TiO22 hours	> 99.9%	> 99.9%	> 99.9%*	> 99.9%
TiO224 hours	> 99.9%	> 99.9%	> 99.9%	> 99.9%

 *result based on duplicate samples

Conclusions:

Bioaccessibility data on titanium released from titanium dioxide were determined when exposed to synthetic biological media of varying pH and composition.
Only a small fraction of titanium was released/dissolved from the titanium dioxide powder during exposure to any of the media matrices of varying acidity and composition. A trend with somewhat higher release rates with increasing acidity and exposure period was evident.

Reference 25

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

Objective of study:

other: alveolar retention and clearance

Qualifier:

no guideline followed

Principles of method if other than guideline:

Rats were exposed to titanium dioxide particles in a whole-body exposure chamber for 24 months. Measurements for lung and lung-associated lymph nodes (Experiment 1 only) were conducted and alveolar clearance was determined via short-term nose only exposure to radioactive labeled tracer particles. In Experiment1 female Wistar rats were exposed to TiO2 of anatase type. In Experiment2, male and female F344 rats were exposed to TiO2 of rutile type.

GLP compliance:

not specified

Specific details on test material used for the study:

Experiment1:
Materials were aerolised by a dry dispersion technique.

Experiment2:
not specified

Radiolabelling:

no

Species:

rat

Strain:

other: Experiment1: Wistar, Experiment2: F344

Remarks:

Experiment1: female only

Sex:

male/female

Details on test animals or test system and environmental conditions:

Experiment1:

TEST ANIMALS
- Housing: housed individually in wire mesh cages

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 22 - 24
- Relative humidity (%): 40 - 60
- Air flow: 3.8 m³/min
- Photoperiod (hrs dark / hrs light): 12/12

Experiment2:
not specified

Route of administration:

inhalation: aerosol

Vehicle:

unchanged (no vehicle)

Details on exposure:

Experiment1:
TYPE OF INHALATION EXPOSURE: whole body

GENERATION OF TEST ATMOSPHERE / CHAMBER DESCRIPTION
- Exposure apparatus: horizontal flow type
- System of generating particulates/aerosols: dry dispersion technique
- Air flow rate: 3.8 m³/min

Experiment2:
not specified

Duration and frequency of treatment / exposure:

Experiment1:
18 hr/day, 5 days/week for 24 months

Experiment2:
6 hr/day, 5 days/week for 24 months

Dose / conc.:

7.5 mg/m³ air (nominal)

Remarks:

Experiment1: for months 1-4 of the study

Dose / conc.:

15 mg/m³ air (nominal)

Remarks:

Experiment1: for months 5-8 of the study

Dose / conc.:

10 mg/m³ air (nominal)

Remarks:

Experiment1: for months 9-24 of the study

Dose / conc.:

5 mg/m³ air (nominal)

Remarks:

Experiment2

No. of animals per sex per dose / concentration:

Experiment1: 4-7 investigated animals per sacrifice date (sacrifice dates set at 3, 6, 12, 18, 22, 24 months)

Experiment2: 4-8 investigated animals per sacrifice date (sacrifice dates set at 3, 9, 15, 21, 24, 24 months)

Control animals:

yes

Positive control reference chemical:

Experiment2: Silicium dioxide

Details on study design:

not applicable

Details on dosing and sampling:

TOXICOKINETIC STUDY

Experiment1:
- Tissues sampled: lung (sampled at all time points), lung associated lymph nodes (sampled at 22 months)
- Time and frequency of sampling: 3, 6, 12, 18, 22, 24 months after first exposure

Experiment2:
- Tissues sampled: lung (sampled at all time points)
- Time and frequency of sampling: 3, 9, 15, 21, 24 months after first exposure

Statistics:

Dunnett test

Type:

other: lung weight

Results:

Experiment1: lung weight significantly increased, progessing with study duration. Experiment2: no significant changes in lung weight/body weight (related to controls).

Type:

other: lung retention

Results:

retained mass after 24 months: Experiment1: 39.3 µg/lung (females), Experiment2: 3.20 mg (males) and 2.24 mg (females)

Type:

other: lung associated lymph node retention

Results:

Experiment1: retained mass after 22 months: 5749 µg/animal

Type:

other: pulmonary tracer clearance (59Fe2O3)

Results:

Experiment1and 2: half-time increased during study

Type:

other: pulmonary tracer clearance (85SR-polystyrene)

Results:

Experiment1: half-time decreased during study. Experiment2: half-time increased during study

Details on absorption:

not applicable

Details on distribution in tissues:

not applicable

Details on excretion:

not applicable

Metabolites identified:

no

Details on metabolites:

Titanium dioxide as an inorganic substance, which is not metabolised.

Bioaccessibility (or Bioavailability) testing results:

not applicable

Conclusions:

According to the authors, alveolar clearance rate decreases with increasing lung burden. Particle overloading of lung generates a variety of nonspecific effects, thus simple first-order modelling underestimates pulmonary retention. The authors recommend semiempirical modelling using the first-order clearance rate coefficient as an explicit function of lung burden.

Reference 26

Endpoint:

dermal absorption in vitro / ex vivo

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: well-conducted and reported study

Qualifier:

according to guideline

Guideline:

OECD Guideline 428 (Skin Absorption: In Vitro Method)

GLP compliance:

yes

Radiolabelling:

no

Species:

pig

Strain:

other: Pietrain-Deutsche Landrasse-Hybrid

Sex:

not specified

Details on test animals or test system and environmental conditions:

not applicable

Type of coverage:

other: diffusion cell

Vehicle:

physiological saline

Duration of exposure:

up to 24 hours

Doses:

The test formulations were applied to 1cm² exposed skin at nominal doses of 4 mg/cm² for 24 hours corresponding to nominal doses of about 400 µg/cm² of zinc oxide or titanium dioxide or to nominal doses of 360 and 240 µg/cm² of zinc or titanium, respectively.

No. of animals per group:

The skin absorption studies were performed with three pigs per formulation, and the test formulations were applied to 8 (ZnO) or 3 (TiO2) skin preparations from each pig in a parallel experimental setup

Control animals:

no

Remarks:

in vitro testing

Details on study design:

no

Details on in vitro test system (if applicable):

Full thickness skin samples of visually intact skin from the lateral abdominal region of 5 months old pigs were dermatomed to a thickness of around 500 µm. The samples were mounted in modified Franz static dermal penetration cells consisting of an upper donor compartment and a lower receptor compartment with the external surface of the stratum corneum facing the donor chamber.
The receptor chambers were filled with approx. 4 ml of the receptor fluid and the cells were placed on a magnetic stirrer (32°C). Each skin sample was checked for integrity by measuring its electrical resistance with a LCR bridge prior testing.
Samples of receptor fluid were taken at various time intervals (3, 6, 12 and 24 hours) after application of the test formulation to the skin and retained for analysis. The skin was removed from the diffusion cell and put onto parafilm. Titanium was removed from the skin preparations by washing with sponge pieces dipped into soap solution, and subsequent tape stripping was used to remove titanium together with the superficial layers of the stratum corneum.
The pools of tape strips, the skin remaining after tape stripping, and all retained receptor fluid samples were acid-treated and analysed for titanium by atomic spectroscopy (AS). Depending on the concentration of the samples, inductively coupled plasma-atomic emission spectrometry or - mass spectrometry was applied for Ti analyses.

Signs and symptoms of toxicity:

not examined

Remarks:

in vitro absorption was investigated

Dermal irritation:

not examined

Remarks:

in vitro absorption was investigated

Absorption in different matrices:

Reported results include full mass-balance (recovery 86-100%), tape stripping and epidermal and receptor medium analysis.
Absorbed dose (receptor ): 0%
Potentially absorbable dose (skin): 0.1-0.5 %
(excl. surface tape strippings)

Total recovery:

see table
Mean total recoveries of Ti ranged from 98% to 100% (T-Lite SF-S) and 86% to 93% (T-Lite SF) of the total Ti applied, respectively.

Dose:

10%

Parameter:

percentage

Absorption:

0 %

Remarks on result:

other: after 24 hours

Remarks:

T-Lite SF-S (TiO2) formulation

Dose:

10%

Parameter:

percentage

Absorption:

0 %

Remarks on result:

other: after 24 hours

Remarks:

T-Lite SF (TiO2) formulation

Conclusions:

Results show that the test substance was not able to penetrate porcine stratum corneum.

Description of key information

No substantial accumulation of titanium was observed in tissues following oral administration of titanium dioxide.

Titanium dioxide has been shown not to penetrate human skin to any appreciable degree, so that the dermal absorption of titanium dioxide through human skin is considered negligible.

Key value for chemical safety assessment

Bioaccumulation potential:

no bioaccumulation potential

Additional information

Introduction

 

There is only a limited number of studies available on the toxicokinetics of TiO₂. The inherent biological variation in in-vivo toxicokinetic studies coupled with the poor oral bioavailability and lack of sensitivity of analytical techniques has made precise quantification difficult. This pertains both to pigment grade as well as nanoforms of TiO2.

The poor oral bioavailability has led some researchers (such as Geraets et al., 2014) particularly with respect to nano-TiO2 to resort to intravenous application, assuming the thereby maximized systemic dose would yield reliable information as to the systemic distribution of any TiO2 nanoparticles in the blood stream.

This has been strictly contradicted recently in a series of three publications (Kreyling et al, 2017a-c) who compared toxicokinetics and tissue distributions by three different routes of administration (oral, inhalation and intravenous) and using a sensitive radiotracer technique to enhance the limits of detection. In brief, the authors conclude reliably that intravenous administration is not a suitable surrogate for oral route (for details, see tabular summaries below).

Another important aspect is that because of the very low water solubility (<10 ug/L) of TiO2, researchers in the past have tended to assume “zero” dissolution of TiO2 upon gastric/intestinal passage. However, the possibility that a very small portion of TiO₂from any ingested material may in fact be dissolved to a minor extent during passage of the GI tract, then be available for systemic transport and potentially precipitates under physiological pH conditions in tissues cannot be discounted, as discussed further below.

 

Water solubility /bioaccessibility studies with different grades of nano- and pigmentary TiO2

 

Reliable water solubility data for TiO2 have previously not been available, with text books generally referring to the material as being “insoluble”. Likewise, data on solubility in surrogate physiological media (“bioaccessibility”) are scarce. Therefore, water solubility (cited in phys-chem section) and bioaccessibility of 3 nano and 3 pigmentary TiO₂grades have been generated at Fraunhofer (Schmallenberg, Germany) in a first step.

 

 

In vitro bioaccessibility in artificial bodyfluids

 

In an in vitro bioaccessibility test the titanium release from six different titanium dioxide samples was determined when exposed to synthetic gastric fluid (GST) and phosphate buffered saline (PBS) for 2 and 24 hours (Knopf, B, 2015). The experiments were performed using the shaking flask method (100 rpm, 37°C) at a loading of 100mg/L, three replicates per test material and medium were used. Titanium dioxide rutile and anatase only dissolve in quantities above the limit of detection in GST (at pH 1), no dissolution could be detected in neutral PBS. In GST, anatase shows a slightly higher dissolution compared with rutile, but with no influence of the particle size.

 

Table: In vitro bioaccessibility results for 6 different titanium dioxide samples in synthetic gastric fluid (GST) and phosphate buffered saline (PBS) after 2 and 24 hours. Results given as [µg Ti per L].

 

Ti inμg/L	TiO2-E171-A	TiO2-E171-B	TiO2-E171-C	TiO2-E171-D	TiO2-UF-E	TiO2-UF-F
Mineral form	Anatase	Anatase	Rutile	Anatase	Anatase	Rutile
Grade	Pigment	Pigment	Pigment	Pigment	Nano	Nano
GST 2h	7.67 ± 2.34	7.22 ± 0.95	0.82 ± 0.13	7.98 ± 1.28	2.09 ± 0.17	0.85 ± 0.09
GST 24h	32.21 ± 2.16	30.09 ± 1.84	4.55 ± 0.55	30.89 ± 1.55	4.49 ± 0.11	3.45 ± 0.07
PBS 2h	< 0.19 (LOD)	< 0.19 (LOD)	< 0.19 (LOD)	< 0.32 (LOD)	< 0.32 (LOD)	< 0.32 (LOD)
PBS 24h	< 0.19 (LOD)	< 0.19 (LOD)	< 0.19 (LOD)	< 0.32 (LOD)	< 0.32 (LOD)	< 0.32 (LOD)

 

In another in vitro bioaccessibility study, the dissolution of one titanium dioxide powder sample (purity 99.5%) was measured in four different artificial bodyfluids (phosphate buffered saline PBS, Gamble’s solution GMB, artificial lysosomal fluid ALF and artificial gastric fluid GST) (Midander, K, 2005). Only a small fraction of titanium was released/dissolved from the titanium dioxide powder during exposure to any of the media matrices of varying acidity and composition. A trend with somewhat higher release rates with increasing acidity and exposure period was evident.

 

Published in vitro bioaccessibility data (KTH, 2010) on an unspecified TiO2 sample (Source: Kemira Pigments Oy, FIN) in four different media yielded a 24h solution level of 48 µg/L in gastric juice. A similar study (KTH, 2007) on pigment grade TiO2 (Kronos) in the same four test media yielded a 24h solution level of 116 µg/L in gastric juice.

  

Previously published solubility data

 

Jones et al. (2015) found heavy aggregation in simulated gastric juice for three different TiO2 particle sizes (10nm, 70 nm, 1.8µm), but did not analyse/report dissolved concentrations. In a previous water solubility test (ICI, 2004) on three different pigment grades of TiO2 (no particle size stated), water solubilities of 0.3-72 µg/l at pH=9, 10-24 µg/l at neutral pH, and 2-44 µg/L at pH=4 were reported.

 

Schmidt and Vogelsberger (2006) describe the dissolution kinetics of different types of titanium dioxide nanoparticles (P25 Degussa, 24-30nm; DT51D, Millenium 24nm; G5, Millenium 10nm; TIPO, own synthesis 1-6nm) in water. The authors report an increase of solubility when the medium pH drops below 3. At ph=1.5, the equilibrium saturation concentration of the tested commercial TiO2 NPs is given as approx. 250 nmol/L (20 ug/L), and that of their own synthesis product at 10 umol/L (800 ug/L); in the underlying Dissertation Thesis by Schmidt (1979, University of Jena, D), these data are described in much more detail, and corroborated by cited data from Lencka et al. (1993) and Ziemniak et al (1993).

 

Sugimoto et al (2002) analysed the Ti(IV) ionic species in solution as a function of pH. Soluble Ti(IV) species remained at a plateau solubility of approx. 10^-5.5M/L, rising to as much as 10^-2M/L at pH=1.0. The authors also determined the mole fractions in dependence of pH, as follows:

 

 

The concentrations of Ti(OH)₂²⁺, Ti(OH)₃⁺and Ti(OH)₄complexes (in equilibrium with solid Ti(OH)₄) as measured by Sugimoto et al as a function of pH are presented graphically below:

 

Reference: Sugimoto et al (2002), Synthesis of uniform anatase TiO2 nanoparticles by gel-sol method, J. Coll Interf Sci 252, 339-346

 

 

Summary of solubility data:

 

Pigment grade TiO2 exhibits water solubilities of 0.3-72 µg/l at pH=9, 10-24 µg/l (neutral pH), 2-44 µg/L at pH=4. At and above a pH of 4-5, dissolved titanium originating from NPs is likely to precipitate as Ti(OH)₄; in the pH range characteristic for the gastric milieu, positively charged Ti complexes begin to dominate. This concurs with an increased solubility, rising from approx. 100 µg/L to approx. 100 mg/L. Different sized TiO2 NPs (4.7nm, 6.6nm, 17.8nm, 28.3nm) reported solubility in aqueous NaCl solutions (pH=1.5) at 25 and 37°C ranging between 20-800 µg/L.

 

In vitro bioaccessibility studies with different pigment grades of TiO2 yielded 24h dissolution values in simulated gastric juice ranging between approx. 30-115 µg/L, whereas the solubility in PBS was generally close to or below detection limit.

 

Overall, the data available does not support the paradigm of “insolubility” used in dietary risk assessment of TiO2. Whereas at neutral pH TiO2 has a very low solubility, this is by comparison substantially increased at gastric pH. This may explain to some extent the albeit very low, but not “zero” oral bioavailability.

 

SolNanoTox project

This project was launched in March 2014, with an anticipated duration of 3 years (prolonged in the meantime until 31.12.2017). Among other aspects, this project aims at investigating the solubilisation capacity of NMs as a likely important determinant of nanomaterial uptake. Interestingly, representatives of two different classes of NMs are scheduled to be investigated: titanium dioxide as an example for insoluble species due to its stability in water and aluminium representing the soluble category. It is hypothesized by the project initiators that aluminium nanoparticles form aluminium ions, either before or during the uptake in the intestine, whilst titanium dioxide nanoparticles may cross the intestine as intact nanoparticles. This difference in behaviour is assumed to explain the different target organs and toxicity for the two NMs. The identified collaborators for this project are: Dr. Valérie Fessard (ANSES, France), Prof. Alfonso Lampen, Dr. Peter Laux (BfR, Germany), Dr. Fabienne Gauffre (ISCR, France), Dr. Gérald Casterou (PMRic TEM, France).

 

References

Harvey, S.; Fields, B. (2004). Titanium dioxide pigment – determination of water solubility and effect of pH on water solubility Study AE91. ICI Measurement Sci. Group GLP Facility.

 

Jones, K. et al. (2015). Human in vivo and in vitro studies on gastrointestinal absorption titanium dioxide Q1 nanoparticles. Toxicol. Letters 233, 95-101.

 

Kungliga Tekniska högskolan (KTH), Sweden (2007). Orientating analysis of bioaccessibility of different metals released from various Ti slag materials in synthetic biological media. Royal Institute of Technology (KTH). Unpublished report.

 

Kungliga Tekniska högskolan (KTH), Sweden (2010). Assessment of bioaccessibility by comparative analysis of metal cation release from titanium dioxide particles in synthetic biological media. Royal Institute of Technology (KTH). Commissioned by International Antimony Oxide Industry Association (IAOIA). Unpublished report.

 

Lencka et al. (1993): Thermodynamic modeling of hydrothermal synthesis of ceramic powders. Chem Mater. 5, 61-70.

 

Schmidt, J. and Vogelsberger, W. (2006).Dissolution kinetics of titanium dioxide nanoparticles - the observation of an unusual kinetic size effect, J. Phys Chem B (110), 3955-3963.

 

Sugimoto, T. et al. (2002). Synthesis of uniform anatase TiO2 nanoparticles by gel-sol method, J. Coll Interf Sci 252, 339-346.

 

Ziemniak, S. F., Jones, M. E., and Combs, K. E. S. (1993). J. Sol. Chem. 22, 601.

 

 

5. Toxicokinetic aspects

 

Titanium is one of ten most abundant elements in the earth’s crust, ubiquitously present in soils and sediments and occurs in the environment in its thermodynamically stable tetravalent form. This is also the form in which it enters the body and is distributed and excreted, without any form of metabolic conversion whatsoever. This section is therefore restricted to a summary of available information on absorption, distribution and elimination after ingestion of TiO2.

 

 

5.1 Investigations of cellular uptake of TiO2 NPs in-vitro

 

Undisputedly, TiO2 nanoparticles have been shown to be taken up readily in cell model systems, such as in A549 cells (carcinomic human alveolar basal epithelial cells); on the other hand, the passage though differentiated cell types such as Caco-2 (human epithelial colorectal adenocarcinoma cells) has been shown to be very low (Janer et al., 2014).

 

For the reasons stated above, “side-by-side” investigations comparing in-vitro uptake and in vivo oral gavage studies in laboratory animals with the same TiO2 nanomaterials have a particular value. For example Janer et al. (2014) conducted such experiments with uncoated TiO2 NPs with average size range of 18+8 nm, synthesized by flame spray pyrolysis (for in vivo results, please refer to the text below). Jones et al. (2015) measured < 0.02% uptake through a layer of Caco-2 cells for three different TiO2 particle sizes (10nm, 70 nm, 1.8 µm). McNicoll et al (2015) investigated the uptake of four different nanoforms (15-120nm) and one micron-sized TiO2 form in-vitro in a transwell system with a Caco-2 and M-cell coculture. In brief, it was shown that none of the TiO2 particles crossed the in vitro gut epithelium model barrier. Brun et al. (2014) investigated TiO2 NP uptake with a combination of in-vitro, as well as in-vivo and ex-vivo analysis of intestinal tissues after dosing of mice (strain and sex not stated) with a single gavage of 12.5 mg TiO2-NP/kg bw. In vivo and ex vivo, they report that agglomerates of TiO2-NPs were translocated through both the regular gut epithelium and through the FAE. They attempted to define the level of dissolved titanium by Xray absorption spectroscopy, but the concentrations were too low to be adequately analysed. They also concluded that the amount of TiO2-NPs in gut tissues was so low that they were unable to quantify it by particle-induced X-ray emission (local limit of detection: 20-30 ppm). Koenemann et al. (2010) describe different degrees of transcytosis in monolayers of Caco-2 epithelial cell monolayers, but without placing these into perspective with likely intra-intestinal TiO2 concentrations upon ingestion.

 

The briefly summarised investigations above do not represent an exhaustive review of available data, but are instead intended to document the existence of results, albeit with somewhat contradictory outcomes. Given the artificial nature of these cellular systems, it is considered highly questionable whether results obtained with these systems are of any relevance for in vivo situations.

 

 

5.2 In vivo uptake mechanisms and supporting studies

 

Given the apparent discrepancy between in-vitro model systems and the complex in-vivo situation, a more detailed assessment of potential uptake mechanism for particles in the digestive tract appears warranted:

 

In brief, the small intestine is highly specialised for uptake of macromolecules (duodenum, jejunum, ileum). Intestinal epithelial cells are covered by a mucous layer that NPs must be able to penetrate for any uptake. In the ileum, enterocytes are the most important cell type facilitating nutrient absorption. Beyond this, the ileum contains areas with lymphoid tissue mediating immune response, localised in Peyer’s patches; these Peyer's patches are lymphoid nodules in the lower small intestine (the ileum), their major function being facilitation of the immune response within the mucosa (pathogenic microorganisms, antigens). Their epithelium contains specialized microfold cells (M cells) which take up „antigens“ directly from the lumen and express them to immune cells on their basolateral side, with subsequent transport via the lymphatic system. A detailed overview has been published recently (Bellmann et al. (2015), and is summarised graphically in the figure below:

 

 

Source: Bellmann et al. (2015): Mammalian gastrointestinal tract parameters modulating the integrity, surface properties and absorption of food-relevant nanomaterials, WIREs Nanomed Nanobiotech, Wiley Periodicals

 

 

Janer et al. (2014) conducted in vivo uptake experiments in male Sprague-Dawley rats (in comparison to in-vitro model cell systems) with a single dose of 100 mg/kg bw of TiO2 NPs (18nm) by oral gavage. At 24 h post treatment, the rats were sacrificed and the following tissues were analysed for Ti via ICP-MS: spleen, liver, small and large intestine, and mesenteric lymph nodes. In brief, there was no statistically significant increase in titanium levels in any of the analysed tissues 24h after administration. No TiO2 NPs were found in the smooth section of the small intestine; merely some isolated cells in Peyer’s patches were found to contain freely distributed NPs in cytoplasm, not surrounded by vesicular membranes.

 

 

5.3 Tissue and blood levels of titanium

 

Blood titanium levels in “non-exposed” (thereby reflecting background dietary intake) individuals appear to vary considerably:

 

(1) In a study with human volunteers, normal Ti blood levels ranged from 5.9 – 18.1 ug/L (mean: 11.2 , SD 4,09). 5 of these were later sampled again before (background mean: 11.5 ug/L, SD 4.14) and after having been exposed orally to 22.9 and 45,8 mg of TiO2 in gelatine capsules, yielding peak blood values in the range 37.4-45.0 ug/L and 61.8-109.9 ug/L, respectively. The authors themselves however note that these levels were below other published data (Böckmann et al., 2000), see below.

 

(2) Based on a very extensive study of titanium levels in a wide range of foods as well as dietary intake data, Schroeder et al. (1963) derived a tentative estimate of daily dietary intake of man at 300 ug/day.

The same author reports on mice (CD-1), when exposed chronically (208-268 days) via drinking water (5 mg/L, as Ti oxalate), experienced elevated titanium tissue levels in several organs (Schroeder et al., 1963). The mice received titanium-free water and a low (< 0.06 ug/g) titanium diet:

 

 

Titanium tissue levels in laboratory mice exposed via drinking water (5 mg/L) in ug/g wet wt

Tissue	controls		exposed
	F	M	F	M
Heart	-	-	32.49	10.74
Lungs	4.00	-	18.07	7.68
Spleen	3.84	-	18.03	7.68
Kidneys	0.80	0.45	9.95	10.45
Liver	1.37	0.45	9.95	10.45
No. of animals	14	24	3	4

 

(3) Historical human biomonitoring data (summarised in an EU Commission document by Valentin & Schaller, 1980) document generally low levels of Ti in blood: Hamilton et al. (1972) report a mean blood level in test material from the UK of 70 ug/kg; Timakin et al. (1967) analysed blood of a total of 100 healthy men and women from the former USSR, yielding an average of 54 ug/kg blood; Maillard & Ettori (1936) state a value of 30 ug/kg. Because of lack of detail in reporting and limitations of the analytical methods employed at the time, these values should be regarded merely as indicative. Valentin & Schaller do not report any Ti levels in organs except for lungs.

 

 

5.4 Toxicokinetics – in vivo data on oral absorption

 

ICRP (1981) have tentatively assigned an oral absorption factor of 0.01%, based on a total body store of 10 mg of Ti (resulting from an estimated daily intake via food and fluids of 300 ug/d, as estimated by Schroder et al, 1963) and a urinary excretion of 10 ug Ti/d. This must however be taken with caution because of an unknown extent of potential biliary excretion. ICRP also assumes a fairly uniform distribution of Ti throughout all organs and tissues except for the lung (with higher levels due to ambient atmospheric exposures).

 

Jani et al. (1994) applied TiO2 particles (500nm) at a dose of 12.5 mg/kg to female Sprague Dawley rats via gavage for 10 consecutive days. Animals were kept in metabolic cages for 24h urine and faeces sampling directly after the last dose. After another 15h period (to clear the gut from any food particles), the following tissues were taken for Ti analysis (by ICP-AES): stomach, intestine, colon, peritoneal tissue, liver, spleen, kidney, heart and lungs. Interestingly, they claim that “background” Ti in tissues was found to be less than 0.25 ug/g in all samples, which is in clear contradiction to other studies (see below). They also report what they designate as “comparative observations” of uptake and presence on TiO2 particles in tissues based on histopathological inspection, thereby subjectively attributing high presence of such particles for example in (among others) spleen, peritoneal tissue and liver, with tabulated Ti concentrations in these organs (spleen: 8.1 ug/g; peritoneal tissue: 42.6 ug/g; liver: 87.3 ug/g). The major drawback of this publication is the lack of a vehicle control. The authors attempt to calculate a % uptake of dose from their data, and report a value of 11.9% for this. Given that most studies yield a percentage of absorption that is <<0.1%, these findings are highly contradictory.

 

Onischenko et al. (2012) investigated the absorption and distribution of TiO2 E171 food pigment grade in comparison to two different TiO2 nanomaterials (one Anatase and one Rutile form, particle characterisation described elsewhere^[1]) in male Wistar rats (n=4/group) by gavage dosing at 100 mg/kg bw for 28 consecutive days. Liver titanium levels measured by ICP-MS in the E171 dosed group were not statistically different from non-dosed controls. Liver titanium levels in NP-dosed animals were contradictory: whereas Anatase-treated animals had lower Ti levels in liver than the controls, the Rutile-treated animals were reported to have a significant increase in liver titanium. Before the reliability of this source can be assessed, the two publications given in the footnote which apparently describe the particle characterisation need to be further evaluated.

 

Himmelstein (2014a,b; 2015) compared the absorption, excretion and distribution in m/f Crl:CD(SD) rats (n=3/group) after single dosing at 500 mg/kg bw per gavage of (i) ultrafine anatase/rutile TiO2, and (ii) pigment grade Anatase TiO2, plus a vehicle (water) control. In a second, experiment, groups of three animals per sex per time point received either a control diet or diets containing one of the two forms of titanium dioxide, at a nominal concentration of 200 mg/kg (equivalent to approximately 30 mg/kg bw). These diets were fed to rats for seven consecutive days and after the treated diets were replaced by control diet for three days. Groups of animals were sacrificed at 1, 24 and 72 hours after withdrawal from the treated diet for determination of the titanium content in tissues (liver, kidneys, muscle and whole-blood) and urine and faeces collected up to 72 hours by Inductively Coupled Plasma-Atomic Emission Spectrometry (ICP-AES). The main route of titanium excretion was via the faeces. The faecal excretion rate at each collection interval (0-24, 24-48, 48-72 hours) was similar in all treated groups. The mean total amounts of titanium excreted in the faeces in the 72 hours after withdrawal of titanium dioxide containing diet were in the range of 1.1-2.2 mg for male rats and 1.1-1.3 mg for female rats. Urinary excretion of titanium was generally below the limit of quantification (<0.04 mg/L). Whole blood concentrations of titanium in all groups were below the limit of quantification (<0.04 mg/L) and concentrations of titanium in liver, kidney and muscle were mainly below the limit of detection (<0.1-<0.2 mg/kg wet weight) or in the range of 0.1-0.3 mg/kg wet weight for most of the animals treated with either control or titanium dioxide containing diet.

 

Langford-Pollard (2003) studied the absorption, rates and routes of excretion and distribution of titanium in rats after dietary exposure to four different forms of pigment grade TiO2 (rutile (thin platelet), rutile (thick platelet), rutile (amorphous) and anatase, including a non-dosed diet control. The dietary concentration of TiO2 was approximately 200 ppm. The rats were fed for seven consecutive days after which the diets were replaced by control diet; sacrifice was at 1, 24 and 72h after withdrawal of diet. Titanium levels were analysed in liver, kidneys, muscle and whole-blood by ICP-AES. Ti was also measured in urine and faeces 0-72h after withdrawal of diets. Titanium was excreted via faeces during 0-72h corresponding to 39 - 63% of the estimated daily dose. Urinary excretion of titanium was equivalent to <2% daily dose/l for all groups. Whole-blood concentrations of titanium were <0.04 mg/l for all groups. Concentrations of titanium in liver, kidney and muscle were mainly below the limit of detection (0.1-0.2 mg/kg wet weight).

 

Jo et al. (2016) investigated the oral absorption of two pigment grade TiO2 anatase materials (designated by the authors as nanomaterials, but with their specifications suggesting that they were rather pigment grade) in male Sprague-Dawley rats. The oral absorption of food pigment grade TiO2 varied somewhat with the composition of the vehicle, i.e. 0.01% from distilled water, 0.3% from 1% glucose and 0.8% form 1%albumin. Food pigment grade f-TiO2 elevated tissue levels (controls 2-7 ug/g in kidney, liver, lung and spleen) only minimally (approx. <10%) and only at 6h p.a., with urinary excretion (0-72h p.a.) given with 0.3% of dose.

 

 

(II) Oral absorption studies on TiO2 nanomaterials

 

Janer et al. (2014) investigated the in vivo absorption after oral gavage of a dose of 100 mg/kg of uncoated TiO2 NPs (average size range 18+8 nm, synthesized by flame spray pyrolysis) in male Sprague-Dawley rats. TiO2 NPs were dispersed at a concentration of 10mg/mL in 2mM sodium citrate solution. Ti analysis in tissues was via ICP-MS, and particles in tissues were visualised with TEM. 24 hours p.a., there was no statistically significant increase in Ti levels in any of the tissues sampled (liver, spleen, mesenteric lymph nodes, Peyer’s patches, small intestine and caecum) compared to vehicle controls. With TEM, no TiO2 NPs could be detected in smooth sections of the small intestine, but there were some TiO2 particles in at least one cell in a Peyer’s patch.

 

Jones et al. (2015) dosed human volunteers (n=9) orally at 5 mg/kg bw (single dose) with all three different TiO2 particle sizes (10nm, 70 nm, 1.8µm), and analysed Ti by ICP-MS in (i) urine 24h-pre and 72h-post dose and in (ii) blood 2, 4, 24 and 48h post-dose. None of the volunteers showed typical absorption curves following dosing, and the variation of Ti levels in urine and erythrocytes appeared to reflect normal background fluctuation:

 

 

 

Jones et al. conclude that even with the smallest dose (93 mg of 15 nm particles), 0.1% absorption would have been expected to give rise to a demonstrable increase in Ti blood levels. Unfortunately, they do not state a LoD, so that a more precise estimate of max. uptake rate could be calculated.

McNicoll et al (2015) investigated the uptake of four different nanoforms (15, 25, 40, 40-50 and 120nm) and one micron-sized (< 5 µm) TiO2 form in-vivo after a single oral (gavage) dose of 5 mg/kg bw to groups of (n=6) Sprague Dawley (m) rats, which were housed in metabolic cages for collection of urine and faeces, and blood for up to 96 hours. After sacrifice (96h post treatment), tissue samples (liver, brain, heart, kidney, spleen and GI tract) were collected for Ti tissue levels via ICP-MS. The following experimental result can be summarised here briefly:

- there was no detectable Ti in any of the urine samples. Interestingly, they observed a high dietary background in normal laboratory diet, which is why the main study was run with a low (0.44 ng/g) Ti diet. It was however later found that the source of the high background were cardboard “environment enricher” toys place in the rat cages (the abolishment of which led to low Ti background levels).

- the statistical analysis showed (with the exception of faecal excretion) that there were no significant differences between any of the TiO2 treatments.

- in view of the dose given (5 mg/kg bw), the tissue levels were generally very low (in ng/g): liver (1.43-3.74), kidney (<LoD-2.42), spleen (10.2-19.5), heart (1.49-20.8) and brain (<LoD-4.43); it is worthy of note that the standard variation (cited as being driven by analytical uncertainty) was actually quite high, for example 40-160% in liver across all six different treatment regimes.

- the tissue titanium levels (in ng/g) of the control animals (n=6) were in liver (2.46), in kidneys (<LoD), in spleen (10.2) in heart (13.7) and in brain (<LoD).

 

Conclusion and discussion (oral absorption):

It is interesting to compare the above results to those of Tassinari et al. (2014), who treated male and female rats orally with either 1 or 2 mg/kg bw of TiO2 NPs (<25nm) for 5 consecutive days. They analysed only spleen tissue for Ti levels 24h after the last treatment, yielding 36+9 ng/g for controls, 40+9 ng/g for the 1 mg/kg dose level and 46+8 ng/g for the 1 mg/kg dose level, seemingly indication a dose-related increase.

By comparison, McNicoll actually measured faecal excretion (µg/g) in their control rats in 24h intervals post treatment, yielding 0-24h: 8.96; 24-48h: 9.48; 48-72h: 10.9; 72-96h: 8.17; this corresponds to an average of 9.4 ug/g (+16%). This can be considered reflective of the day-to-day variation in dietary intake of Ti of non-exposed rats, which should also be reflected in tissue concentrations. Tissue levels in spleen 96h after treatment (note: after 72h, there were still appreciable amounts of TiO2 in the GI tract of treated animals) averaged at 13.9 ng/g+14% (range 10.2-15.8)

The Tassinari et al (2014) spleen tissue levels appear to be approx. 4-fold higher than in the McNicoll study even in untreated animals, with a standard deviation of 17 -25% of the measured averages. Given the similarity in particle sizes used for the treatment by Tassinari and McNicoll, and assuming little or no difference in absorption rate between the six different treatments of McNicoll, then one may conclude that dietary background variation in untreated controls is of the same order of magnitude as the Ti levels in spleen vary in both investigations. Notwithstanding that McNicoll administered only a single dose (5 mg/kg) vs. 5 and 10 mg/kg bw (cumulated over 5 days), then the findings by Tassinari et al. do not qualify as a proof of tissue accumulation of Ti in spleen tissue.

Kim & Park (2014) administered high doses (10 and 100 mg/kg bw/d) of TiO2 NPs (“P-25 Aeroxide”, av. particle diameter 21 nm, source: Evonik, Ham, France) orally for five consecutive days to Sprague-Dawley rats (n=4; male only). Rats were placed in metabolic cages for the entire duration of the study for 24h collection of urine and faeces - in addition, blood and tissues (liver, kidney, lungs) were sampled for Ti analysis upon sacrifice, the analytical procedure involved an oxidative dissolution in nitric acid and Ti quantification by ICP-MS. There was no statistically significant increase of Ti levels in any of the organs sampled. Excretion was almost exclusively via faeces, and negligible via urine (results presented only graphically, no tabulated data).

Geraets et al. (2014) administered four different TiO2 nanomaterials to male (n=3/group, all four test materials) and female (n=3/group, only one test material) Wistar rats by gavage on five consecutive days at a dose of 2.3 mg/animal (corresponding to 6.8-8.6 mg/kg bw). Upon sacrifice, samples from liver, spleen and mesenteric lymph nodes were taken for titanium analysis by ICP-MS. Liver and spleen tissue samples contained very low titanium levels, mostly below the limit of detection of 0.03 μg/g. Twenty-eight measurements were below the LOD, one was at the LOD and only one was above the LOD; it is worthy of note that even in control animals two measurements at the LOD were reported. Whereas all samples of MLN tissue contained amounts above the LOD, the authors themselves note that levels in MLN of TiO2 exposed animals were except for one material similar to control animals, and overall conclude that “the results indicate that after oral administration, absorption of TiO2 is very low.”

Cho et al. (2013) administered TiO2 nanomaterial (source: ABC Nanotech, Korea; 80:20 anatase:rutile, 21nm acc. to manufacturer, 26nm by SEM) particles orally by gavage (vehicle: drinking water) for 13 weeks (7d/week) consecutively to male and female Sprague-Dawley rats (n=11/group) at doses of 0 (drinking water), 260, 521 and 1042 mg/kg bw/d. There were no body weight changes at any of the dose level employed. Upon sacrifice, blood was sampled from the abdominal vein, and tissue samples were taken from liver, spleen, kidney and brain for Ti analysis via ICP-MS (LoD: 0.1-1 ng Ti/L). Five animals from each group were placed into metabolic cages for 24h urine and faees collection immediately upon gavaging. Ti blood levels were approx. 0.5 ug/g at all dose levels including vehicle control, indicating low to negligible systemic uptake. Ti levels in all tissue ssamples showed no significant increase at any dose level, and the urinary excretion likewise was not significantly different from controls in any of the dose groups, with correspondingly high concentration in faeces.

Wang et al. dosed young as well as adult male Sprague-Dawley rats (n=7/group) with TiO2 NM (source: Shanghai Reagant Co.; 75nm) at doses 0, 10, 50 and 200 mg/kg bw/d for a period of 30 consecutive days via gavage. There were no statistically siginifant differences in levels of titanium either in blood or in livers, kidneys or in spleen.

Wang et al. (2007) administered TiO2 NPs (25 and 80nm) as well as a “fine grade (150nm) in a single dose of 5 mg/kg bw to male and female CD-1 mice via gavage in an acute toxicity study. Among other investigations, they analysed liver, spleen and lungs for Ti levels, but only upon sacrifice after the 14d observation period. Given this time lag, the lack of significant changes in Ti levels in spleen and kidneys do not have a lot of weight. Apart from this, the authors themselves acknowledge a high level of mortality in their animals associated with inadequate gavaging, which renders doubt on the reliability of some of the other reported findings.

Kreyling et al (2017a-c) compared the toxicokinetics of 48V-radiolabeled, pure TiO2 anatase nanoparticles ([48V]TiO2NP) with a median aggregate/agglomerate size of 70 nm in aqueous suspension after intravenous injection as well as intratracheal and oral administration in female Wistar rats:

- after a single i.v. administration, the highest [48V]TiO2NP accumulations were found in liver (95.5%ID after one day), followed by spleen (2.5%), carcass (1%), skeleton (0.7%) and blood (0.4%). Detectable nanoparticle levels were found in all other organs. The [48V]TiO2NP content in blood decreased rapidly after 24 h while the distribution in other organs and tissues remained rather constant until day-28.

- after oral application of a single dose of an aqueous [48V]TiO2-nanoparticle suspension by gavage, faecal excetion was the predominant route of elimination. About 0.6% of the administered dose passed the gastro-intestinal-barrier after one hour, and about 0.05% were still distributed in the body after 7 days, with quantifiable [48V]TiO2-nanoparticle organ concentrations present in liver (0.09 ng·g_1), lungs (0.10 ng·g_1), kidneys (0.29 ng·g_1), brain (0.36 ng·g_1), spleen (0.45 ng·g_1), uterus (0.55 ng·g_1) and skeleton (0.98 ng·g_1).

- after intratracheal instillation of a single dose of an aqueous [48V]TiO2-nanoparticle suspension, about 4% of the initial peripheral lung dose passed through the air-blood-barrier after 1 h and were retained mainly in the carcass (4%); 0.3% after 28 d. Highest organ fractions of [48V]TiO2-nanoparticles present in liver and kidneys remained constant (0.03%). [48V]TiO2-nanoparticles which entered across the gut epithelium following fast and long-term clearance from the lungs via larynx increased from 5 to 20% of all translocated/absorbed [48V]TiO2-nanoparticles. This contribution may account for 1/5 of the nanoparticle retention in some organs.

 

After normalizing the fractions of retained [48V]TiO2-nanoparticles to the fraction that reached systemic circulation, the biodistribution obtained after intratracheal, intravenous and oral administration were compared. The authors concluded that the toxicokinetic distribution and kinetics after intratracheal instillation and gavage administration were similar, but were distinctly different from the pattern after intravenous injection, disproving the latter to be a suitable surrogate of the former applications.

 

 

Toxicokinetics – data on dermal absorption

 

The absorption of titanium dioxide through porcine skin has been measured in an in vitro percutaneous study (Gamer, AO, et al. 2006). Under the present testing conditions the total dermal absorption is estimated to 0 %, for the two tested titanium dioxide samples, retention in the skin was estimated to be 0.1-0.5%. Therefore, based on this study a value of 0% for dermal absorption is suggested. This value is supported by various supportive data in which the retention of titanium dioxide from sunscreen formulations in human skin has been investigated. It has been shown that titanium dioxide is retained in the outmost layers of the stratum corneum.

 

Furthermore, titanium dioxide was tested in various percutaneous absorption tests which have been reviewed by the Scientific Committee on cosmetic products and non-food products (SCCNFP/0005/98, 2000) and which concluded: “extensive tests for percutaneous absorption, mostly in vitro, indicate that absorption does not occur, either with coated or uncoated material; one experiment found some evidence that a little of the material could be found in the openings of the follicles. [...] The toxicological profile of this material does not give rise to concern in human use, since the substance is not absorbed through the skin. In view, also, of the lack of percutaneous absorption, a calculation of the margin of safety has not been carried out.”

 

It is therefore concluded that dermal absorption of titanium dioxide is negligible.

[1]Raspopov, R.V. et al. (2010): Vopr. Pitaniya 79(4), 21-30; Sheveleva, S.A. et al. (2010): Vopr. Pitaniya 79(5), 29-3