Cerium dioxide

Acute Toxicity: other routes

Administrative data

Endpoint:

acute toxicity: other routes

Remarks:

Intravenous injection

Type of information:

experimental study

Adequacy of study:

other information

Study period:

Not applicable

Reliability:

2 (reliable with restrictions)

Cross-reference

Cross-reference 1

Reason / purpose for cross-reference:

reference to other study

Reference

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

Not applicable

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

study well documented, meets generally accepted scientific principles, acceptable for assessment

Remarks:

; nevertheless the test was not conducted under GLP conditions. Moreover, there were some discrepancies regarding the number of test animals described in the "Materials & Methods" section and the figure legends.

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Objective of study:

other: blood, clot and serum distribution; rate of clearance from blood; comparison between sizes and shapes of nano-CeO2, and to metal form

Qualifier:

no guideline followed

Principles of method if other than guideline:

- Unique intravenous (i.v.) injection of either a suspension of nanometric cerium dioxide (nano-CeO2) with various physico-chemical properties, or Ce ion (as chloride); sampling at 0.167, 0.5, 0.75, 1, 2, 4, 20, 168 (1 week) or 720 hours (1 month) post-infusion and sacrifice of animals 168 or 720 hours after exposure
- Blood distribution/clearance: inductively coupled plasma mass spectrometry (ICP-MS) and pharmacokinetic modelling

GLP compliance:

not specified

Remarks:

The GLP status was not specified in the article.

Radiolabelling:

no

Species:

rat

Strain:

Sprague-Dawley

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: University of Kentucky (USA)
- Age at study initiation: No data available
- Weight at study initiation: 325 ± 30 g
- Fasting period before study: No data available
- Housing: Individually
- Individual metabolism cages, diet, water, acclimation period: No data available

ENVIRONMENTAL CONDITIONS
No data available

IN-LIFE DATES: No data available

Route of administration:

intravenous

Vehicle:

water

Details on exposure:

Rats were prepared with two cannulas, surgically inserted into femoral veins under ketamine/xylazine anaesthesia, which terminated in the vena cava, and were connected to infusion pumps via a flow-through swivel. This enabled conduct of the study in the awake, mobile rat. All samples were sonicated to ensure dispersion prior to administration. The day after cannula implantation the non-anesthetized rat was infused via the longer cannula with cerium ion (Ce, as chloride), nano-CeO2 as a ~5 % nano-CeO2 suspension in water, or water.
Each of the materials was infused over 1 h, except for Ce ion that was infused over 2 h.
Control rats received water adjusted to the pH of the paired nano-CeO2 or Ce ion.
To compensate for the i.v. administration of a considerable volume of grossly hypotonic infusion, concurrent i.v. infusion of an equal volume and rate of 1.8% sodium chloride in water was delivered into the second, shorter, cannula. Each fluid was delivered at approximately 0.6 mL/h.

Duration and frequency of treatment / exposure:

A single exposure (sampling performed 0.167, 0.5, 0.75, 1, 2, 4, 20, 168 (1 week) or 720 hours (1 month) after i.v. infusion; sacrifice 168 or 720 hours after i.v. infusion)

Dose / conc.:

78 mg/kg bw (total dose)

Remarks:

for Ce55 in the preliminary study
basis: nominal nano-CeO2 dose

Dose / conc.:

100 mg/kg bw (total dose)

Remarks:

for Ce30 and Ce55 in the preliminary study
basis: nominal nano-CeO2 dose

Dose / conc.:

175 mg/kg bw (total dose)

Remarks:

for Ce5 in the preliminary study
basis: nominal nano-CeO2 dose

Dose / conc.:

250 mg/kg bw (total dose)

Remarks:

for Ce5, Ce15, and Ce55 in the preliminary study
basis: nominal nano-CeO2 dose

Remarks:

CeM was not included in the preliminary study.

Dose / conc.:

0 mg/kg bw (total dose)

Remarks:

(control)
in the main study

Dose / conc.:

100 mg/kg bw (total dose)

Remarks:

for Ce5, Ce15 and Ce30 in the main study
basis: nominal nano-CeO2 dose
This corresponded to an analytical dose of 85 mg/kg for Ce5 and Ce30, and 70 mg/kg for Ce15.

Dose / conc.:

50 mg/kg bw (total dose)

Remarks:

for Ce55 in the main study
basis: nominal nano-CeO2 dose
This corresponded to an analytical dose of 50 mg/kg.

No. of animals per sex per dose / concentration:

- Ce ion (for measurements of cerium concentration in whole blood):
* n = 3 for all time points

- Ce5 (for measurements of cerium concentration in whole blood):
* n = 9 at 0.167 h post-infusion
* n = 10 at 0.5 h post-infusion
* n = 5 at 0.75 h post-infusion
* n = 21 at 1 h post-infusion
* n = 10 at 2 h post-infusion
* n = 5 at 4 h post-infusion
* n = 12 at 20 h post-infusion
* n = 5 at 168 h post-infusion
* n = 7 at 720 h post-infusion

- Ce15 (for measurements of cerium concentration in whole blood):
* n = 10 at 0.167 and 0.5 h post-infusion
* n = 5 at 0.75 h post-infusion
* n = 10 at 1, 2, and 4 h post-infusion
* n = 5 at 20, 168, and 720 h post-infusion

- Ce30 (for measurements of cerium concentration in whole blood):
* n = 6 at 0.167, 0.5, and 0.75 h post-infusion
* n = 10 at 1 h post-infusion
* n = 6 at 2 and 4 h post-infusion
* n = 8 at 20 h post-infusion
* n = 3 at 168 h post-infusion
* n = 11 at 720 h post-infusion

- Ce55 (for measurements of cerium concentration in whole blood):
* n = 5 at 0.167, 0.5, and 0.75 h post-infusion
* n = 10 at 1 h post-infusion
* n = 5 at 2 and 4 h post-infusion
* n = 7 at 20 h post-infusion
* n = 5 at 168 and 720 h post-infusion

- CeM (for measurements of cerium concentration in whole blood):
* n = 5 for all time points

Control animals:

no

Details on study design:

A pilot study was conducted with the cerium ion and each of the nano-CeO2 to determine the tolerability following i.v. infusion. A 100 mg cerium ion/kg infusion was lethal. Rats were infused approximately 250 or 175 mg/kg of Ce5; three of eight died. A dose of approximately 250 mg of Ce15/kg was tolerated. Infusion of 100 mg/kg of the Ce30 resulted in some evidence of mild distress (tachypnea, skittish behaviour and not resting well). Dyspnoea and lethargy were seen in rats given 78, 100, or 250 mg of Ce55/kg. Therefore, the target doses were 100 mg/kg for the Ce5, Ce15, Ce30 and 50 mg/kg for the Ce55 and the cerium ion.

Details on dosing and sampling:

PHARMACOKINETIC STUDY (Absorption, distribution, excretion)
Blood was sampled at 0.167, 0.5, 0.75, 1, 2, 4, 20, 168, or 720 h post-injection. After infusion, blood (0.6 mL) was withdrawn at the different time points from the cannula that had not delivered Ce. The blood was immediately separated volumetrically into two aliquots to enable determination of Ce in whole blood in one and in the serum and clot fractions of the other.
To determine the distribution of Ce in the serum and clot fractions samples (0.3 mL) of whole blood was allowed to clot at room temperature, the clot given time to contract, and serum withdrawn into a digestion vessel. The clot was withdrawn and placed in another digestion vessel and the remaining serum added to the serum sample. This enabled determination of the mass amount of Ce in the two major compartments of blood; serum and, from the clot fraction, association with the formed elements (red cells, white cells and platelets).
To determine Ce in whole blood the sample was transferred into a polytetrafluoroethylene digestion vessel (CEM) to which trace-metal grade nitric acid and 30% H2O2 were added, digested in a microwave at 180°C for 10 min to convert all Ce to dissolved Ce4+, diluted in water, and subsequently further diluted dependent on the Ce concentration. Ce concentration was determined in whole blood, serum and clot by ICP-MS. A Method Detection Limit (MDL) of Ce for blood and serum samples of 0.018 mg/L was determined (see Yokel RA at al., 2009). No samples in this study had Ce concentrations below this MDL. Seven samples, containing whole blood, serum or clot, were analysed in duplicate and were spiked with 8.3 ng Ce/mL. Results of duplicate analysis showed a range of 0 to 4% between the two determinations. Recovery of the Ce spike ranged from 93 to 114%, averaging 105%.

OTHER (see below in “any information on materials and methods incl. tables)
- PHARMACOKINETIC MODELLING
- RATIO OF Ce CONCENTRATION IN WHOLE BLOOD FOLLOWING INFUSION OF NANO-CeO2 VERSUS Ce ION
- Ce PARTITIONING BETWEEN SERUM AND CLOT

Statistics:

Blood, serum and clot Ce concentrations were normalized to an infusion dose of 100 mg CeO2/kg for all rats, based on determination of the actual Ce content of the infusions by ICP-MS analysis, and because a few materials were given at doses other than 100 mg/kg. This enabled direct comparison of results from the Ce ion, nano-CeO2 sizes and shapes.
Outliers identified by the Grubb's test were not used in the data analysis. Four percentages of the 600 whole blood, serum and clot samples were outliers by this test.

Preliminary studies:

As detailed by Yokel RA et al. (2013), 3 of 8 rats that received Ce5 died. There was evidence of mild distress in rats that received Ce30 (tachypnea, skittish, not resting well) and significant toxicity in rats that received Ce55 (dyspnoea, lethargic).
Therefore, the target doses were 100 mg/kg (ca. 0.58 mM) for Ce5, Ce15, and Ce30 and 50 mg/kg for Ce55. This large CeO2 exposure was used to enhance the ability to detect Ce in the blood and organs up to the 30 days after the single nano-CeO2 infusion. The authors found that the % of Ce5 dose in the liver was comparable after a single i.v. administration of 11 mg CeO2/kg, 5 administrations of this dose given over consecutive days, and a single administration of 56 mg CeO2/kg (unpublished results).
Given the persistence of nano-CeO2 in the mammal model, it was anticipated that small frequent doses will accumulate so that a single large dose, as used in this study, models the result of this accumulation (e.g., multiple smaller doses).

Details on absorption:

Following a single i.v. administration, no animal died prior to their planned termination. Up to the times studied in this study, there was generally a decrease of blood cerium (Ce). Ce15, Ce30, Ce55, and CeM were rapidly cleared from blood since ≤ 2% of nano-CeO2 was found in blood 0.167 h after infusion. However, the clearance of Ce ion and Ce5 was less effective. Indeed, 14% of Ce ion and 32% of Ce5 remained in whole blood 0.167 h after exposure. This was calculated from the measured blood Ce concentration compared to the Ce dose, based on the i.v. infusion of 100 mg CeO2/kg into the rat’s vascular system (ca. 7% of its body volume), which would introduce approximately 1163 mg Ce/L blood. The authors suggested that the greater extent of citrate coating on 5-nm CeO2 (and therefore its greater hydrophilicity) may have contributed to its slower clearance. It has to be noted that the blood levels of Ce15 and Ce30 increased over the first 4 h after their infusion; according to Dan M et al., this suggested a redistribution of nano-CeO2 over time.

Metabolites identified:

not measured

Bioaccessibility (or Bioavailability) testing results:

No data available

Conclusions:

In conclusion, the kinetics and distribution of cerium ion did not predict those of the tested nano-CeO2. Clearance of nano-CeO2 from blood was size-dependent since 5-nm nano-CeO2, being resistant to agglomeration, was cleared much more slowly from blood than larger nano-CeO2 which were very rapidly cleared from circulating blood. The distribution of nano-CeO2 between serum and the blood clot was also size-dependent.
=> Interpretation of results: bioaccumulation potential cannot be judged based on study results

Executive summary:

Dan M et al. (2012) conducted an in vivo study to characterise distribution in, and clearance from blood of nanometric cerium dioxide (nano-CeO2), with various physico-chemical properties following a single intravenous (i.v.) injection to rats.

Five nano-CeO2 were used in this study. The nanomaterials were physico-chemically characterised:

Parameters	Results					Methods
	Ce5	Ce15	Ce30	Ce55	CeM
Primary particle size Expected value Measured average particle size	5 nm 4.6 ± 0.1 nm {4.8 nm}	15 nm 12.0 ± 0.2 nm {13.5 nm}	30 nm 31.2 ± 0.5 nm {51 nm}	55 nm 55.0 ± 0.2 nm {59 nm}	- Cubic = 30 nm and rod = 10 nm diameter and 2 µm length	- TEM
Particle size distribution in water - Number basis [Volume basis]	7 nm [98%, 7 nm; 2%, 18 nm]	25 nm [57%, 25 nm; 43%, 145 nm]	41 nm [36%, 41 nm; 64%, 273 nm]	No data	No data	TEM, dynamic light scattering (DLS)
Stability	Slight agglomeration/aggregation tendency	Slight agglomeration/aggregation tendency	Slight agglomeration/aggregation tendency	No data	No data	-
Specific surface area [mean diameter estimated from surface area]	121 m²/g [6.5 nm]	71 m²/g [11 nm]	15 m²/g [52 nm]	No data	Not determined	BET method
Surface charge in water at pH ~7.3 (zeta potential)	-53 ± 7 mV	-57 ± 5 mV	-56 ± 8 mV	-32 ± 2 mV	No data	DLS
Shape	Polyhedral	Polyhedral	Cubic	Polyhedral	Cubic and rod	TEM
Crystallinity	Crystalline (face-centred cubic)	Crystalline (face-centred cubic)	Crystalline (face-centred cubic)	Crystalline (face-centred cubic)	Crystalline (face-centred cubic)	X-ray diffraction (XRD), TEM
Cerium content in nano-CeO2 suspension	Analytical dose was 85% of the intended 100 mg/kg nominal dose.	70% of the intended 100 mg/kg nominal dose	85% of the intended 100, 100 mg/kg nominal	103% of the intended 50 mg/kg nominal dose	No data	ICP-MS
Surface properties – Extent of surface citrate coating	ca. 40% surface coated with citrate	ca. 27% surface coated with citrate	ca. 18% surface coated with citrate	ca. 15% surface coated with citrate	No data	Thermogravimetric analysis (TGA)
Oxidation degree	Enrichment of Ce3+ at the particle rim (surface layer)	Enrichment of Ce3+ at the particle rim (surface layer)	Enrichment of Ce3+ at the particle rim (surface layer)	Enrichment of Ce3+ at the particle rim (surface layer)	Enrichment of Ce3+ at the particle rim (surface layer)	Electron energy loss spectrometry (EELS)
Free Ce content in suspensions	11.6 ± 0.3%	<< 1%	<< 1%	<< 1%	No data	ICP-MS
pH of suspensions	pH 8-9	pH 3.5	pH 3.9	pH 7	No data	pH metry

{d} mean diameter estimated from TEM measurements

Male Sprague-Dawley rats (3 to 21/group) were i.v. infused with the tested nano-CeO2 (as a 5% suspension in water) at the nominal doses of 0 (control), 50 (Ce55) or 100 mg/kg (Ce5, Ce15, Ce30).Control rats received water, which was nano-CeO2 vehicle. In addition, Ce ion was administered to rats at a dose level of 50 mg/kg through i.v. infusion. Blood samples were collected at 0.167, 0.5, 0.75, 1, 2, 4, 20, 168 (1 week), and 720 h (1 month) after exposure. Animals were terminated 168 or 720 h later. Blood, serum and clot distribution of cerium (Ce) was assessed by inductively coupled plasma-mass spectrometry (ICP-MS). Moreover, a pharmacokinetic analysis of Ce concentrations in blood was performed, notably to compare the data for nano-CeO2 to those of Ce ion.

Regarding blood distribution and clearance of nano-CeO2, Ce5 was cleared much more slowly than larger nano-CeO2. Nano-CeO2 larger than 5 nm were rapidly cleared from circulating blood. All nano-CeO2 seemed to display a prolonged retention throughout the rat. Traditional pharmacokinetic modelling showed best fit for 5-nm nano-CeO2 and Ce ion. After initially declining, whole blood Ce15 and Ce30 increased over 4 h after infusion; these results were not well-described by traditional pharmacokinetic modelling). The distribution of nano-CeO2 between serum and the blood clot was size-dependent. Ce ion, and Ce5 and Ce55 did not preferentially distribute into serum or clot, CeM was predominantly in the clot, and Ce15 and Ce30 migrated into the clot over 4 h, suggesting a change in surface properties of Ce15 and Ce30. The latter nanomaterial showed the greatest distribution in the clot fraction. The authors suggested that reticulo-endothelial organs may not readily recognize Ce5. Increased cerium distribution into the clot over time may be due to opsonisation.

In conclusion, traditional pharmacokinetic analysis was not very informative and nano-CeO2 pharmacokinetics were quite different from that of Ce ion.

Cross-reference 2

Reason / purpose for cross-reference:

reference to other study

Reference

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Study period:

Not applicable

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

study well documented, meets generally accepted scientific principles, acceptable for assessment

Remarks:

; but test not conducted under GLP conditions.

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Objective of study:

other: hepatic biotransformation

Qualifier:

no guideline followed

Principles of method if other than guideline:

- Unique intravenous (i.v.) injection of a suspension of nanometric cerium dioxide (nano-CeO2) in rats; sacrifice of animals 1 and 20 hours, and 30 or 90 days after injection
- Biotransformation in liver: Light microscopy (LM), transmission electron microscopy (TEM), scanning transmission electron microscopy (STEM), electron energy loss spectroscopy (EELS) , high-resolution transmission electron microscopy (HRTEM)

GLP compliance:

not specified

Remarks:

The GLP status was not specified in this article.

Radiolabelling:

no

Species:

rat

Strain:

Sprague-Dawley

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: No data available
- Age at study initiation: Described as adult
- Weight at study initiation: 316 ± 27g
- Fasting period before study, housing, individual metabolism cages: No data available
- Diet: Ad libitum (2018 Harlan diet)
- Water: Ad libitum (reverse osmosis water)
- Acclimation period: No data available

ENVIRONMENTAL CONDITIONS
- Temperature: 21°C
- Humidity: 30-70%
- Air changes: No data available
- Photoperiod: 12 hrs dark / 12 hrs light

IN-LIFE DATES: No data available

Route of administration:

intravenous

Vehicle:

water

Remarks:

adjusted to pH 3.9

Details on exposure:

A cannula was surgically inserted, under sterile conditions, into each of the femoral veins and extended into the vena cava. The following day, non-anaesthetised rats were intravenously (i.v.) infused via the shorter cannula with a nano-CeO2 suspension or water (controls) adjusted to pH 3.9 with citric acid, concurrently with an equal volume and rate of 1.8% saline infused into the second cannula to achieve net infusion of iso-isomotic solution. Each fluid was i.v.-infused at rates of 2 mL/kg/h over a duration of 1 h. The cannulas were surgically removed from 30- and 90-day rats.

Duration and frequency of treatment / exposure:

A single exposure (sacrifice 1 h, 20 h, 30 days or 90 days after i.v. infusion)

Dose / conc.:

0 mg/kg bw (total dose)

Remarks:

(control)

Dose / conc.:

85 mg/kg bw (total dose)

Remarks:

basis: analytical nano-CeO2 dose

No. of animals per sex per dose / concentration:

Control rats terminated at:
* 1 h: n = 5
* 20 h: n = 8
* 30 days: n = 10
* 90 days: n = 6

Exposed rats terminated at:
* 1 h: n = 5
* 20 h: n = 5
* 30 days: n = 14
* 90 days: n = 7

Control animals:

yes, concurrent vehicle

Details on study design:

- Dose selection rationale: The high dose was mainly used because of a previous study that was centred on characterising the safety and toxicity of nano-CeO2 (Yokel RA et al., 2012 and 2013).
- Rationale for animal assignment: No data available

Details on dosing and sampling:

CHARACTERISATION OF HEPATIC NANO-CeO2 CONTENT
Characterisation of nano-CeO2 in liver tissue was done using light microscopy and also using transmission electron microscopy (TEM).
Light microscopy was used to tabulate size and density of lesions inside liver where nano-CeO2 was located. Thin tissue sections (ca. 3 mm²) were immersion fixed, dehydrated and then embedded. Sections were stained with hematoxylin & eosin (H&E) and screened for nano-CeO2-containing Kupffer cells. Nano-CeO2-induced cellular degeneration and proliferation in liver tissue were determined by TUNEL and proliferating cell nuclear antigen (PCNA) immunostaining.
TEM imaging was performed using thin sections (800 Å) without staining. Scanning transmission electron microscopy (STEM) was used to locate nano-CeO2 in liver tissue. Electron Energy Loss Spectroscopy (EELS) analyses were performed at the nanoscale to study the oxidation degree of Ce as a function of particle size and were done in combination with high resolution TEM (HRTEM).

Statistics:

No data available

Preliminary studies:

No data available

Details on absorption:

No data available

Details on distribution in tissues:

According to Graham UM et al., spleen and liver had sequestered approximately the same amount of nano-CeO2. The authors mentioned that details on nano-CeO2 distribution, clearance and biopersistence were described in the articles of Yokel RA et al. (2012 and 2013).

Details on excretion:

No data available

Details on metabolites:

- CHARACTERISATION OF HEPATIC NANO-CeO2:
Nano-CeO2 formed agglomerates in the liver up to day 30, but otherwise showed no indication of breakdown or chemical transformation. However, after 90 days inside the liver, i.v.-infused cube-shaped nano-CeO2 had become highly fragmented and rounded along their edges, which indicated that in vivo processing of the particles had occurred. Moreover, accumulations of copious ultrafine crystallites (1 to 3 nm) that formed in the liver within close proximity to the rounded nano-CeO2 were observed and labelled as CeO2 clouds. These second-generation CeO2 particles (crystallites in CeO2 clouds) seemed to be trapped in "groups or swarms" owing to fast precipitation.

Very large polycrystalline CeO2 (up to hundreds of nm) with rounded, partially missing corners was observed in the liver (not seen in the spleen) after 90 days. Since the as-synthesised nano-CeO2 was dense crystals, the authors suggested that the large polycrystals formed in situ of the liver. Many of the large polycrystalline CeO2 grains had rounded corners suggesting there was post-formation dissolution occurring (high energy edge sites) and in some cases the corners had formed pits or cavities. According to Graham UM et al., this further evidenced that nano-CeO2 in the liver underwent a dynamic transformation and bioprocessing.

The redox potential of nano-CeO2 both after hydrothermal synthesis and after in vivo processing in the liver was assessed with EELS. The core region of both freshly synthesised nano-CeO2 cubes and in vivo processed grains always showed characteristic satellite peaks, which suggested a highly equilibrated content of Ce4+ while surfaces of the same particles were high in Ce3+. Thus, a significant enrichment in Ce3+ was systematically observed in surface layers whereas the core regions were primarily Ce4+; reduction potential of as-synthesised CeO2 cubes and that of processed rounded grains in the liver were essentially the same. However, EELS analyses of ultrafines (1 to 3 nm) from CeO2 clouds indicated a reduced valence state (Ce3+) for the entire grains, with a range from 38 to 70% Ce3+. According to the authors, this indicated that CeO2 cloud formation shifted the overall redox activity of CeO2 to a more reduced state and, thus, higher free-radical scavenging activity.

Much lesser amounts of nano-CeO2 seemed to have been bioprocessed inside the spleen which was evidenced by minor occurrence of ultrafines or CeO2-clouds in the spleen. Also, there was less rounding of the sequestered nano-CeO2 in the spleen suggesting that dissolution of nano-CeO2 was less favoured in the spleen environment compared with the liver. Graham UM et al. suggested that dissolution of nano-CeO2 could be linked to the occurrence of the CeO2-clouds.

According to the authors, rod-shaped crystallites were also formed in the liver near in vivo processed nano-CeO2; they were composed of Ce, O, and P and were of comparable size and morphology as sol-gel-derived Ce phosphates. Since significantly more CeO2 clouds formed in the liver than in the spleen (even though the spleen had sequestered approximately the same amount of nano-CeO2), the liver environment could promote the processing of nano-CeO2 through its localized liver acid-base physiology (cellular pH is tightly regulated and nano-CeO2 would primarily encounter a low pH in lysosomes). The in situ formation of CeO2 clouds therefore could be governed by the release of Ce ions derived from dissolving i.v.-infused CeO2 in the liver since the CeO2 dissolution rate is a function of local pH and electrochemical potential.

Thus, in vivo processing led to a much greater reactive nano-CeO2 surface area in the second-generation CeO2 (CeO2 clouds). EELS of the cloud particles demonstrated that the oxidation degree, when compared with the original nano-CeO2, was more reduced and Graham UM et al. suggested that in vivo processing was a function of 1) the acidic environment, 2) the oxidizing potential of the environment, and 3) the exposure time (it took 90 days before the effects of in vivo processing were recognized in tissue samples).

Graham UM et al. then set out to determine how fast the second-generation nano-CeO2, after exposure to an oxidant would be able to return back to the reduced state (Ce4+ => Ce3+), since EELS demonstrated that all particles in the second-generation CeO2 clouds were in a reduced state. As XPS could not be performed directly on CeO2 clouds because nanoparticles could not be isolated because of strong liver tissue interactions, the authors synthesised nano-CeO2 of 2 to 4 nm to resemble the crystallites in CeO2 clouds. To mimic the redox activity of CeO2 clouds, Graham UM et al. first performed oxidation experiments (using 0.5m H2O2 that quickly evaporated) and oxidized the particles. The 2-4 nm nano-CeO2 (as-synthesised) had approximately 40% Ce3+ before H2O2 exposure and approximately 0% Ce3+ afterwards, and furthermore in less than 1 h transitioned back to the pre-H2O2 baseline composition (i.e., around 40% Ce3+). Thus, oxidized ultrafine CeO2 returned quickly and completely to the reduced state without the presence of reducing agents.

The authors proposed the following predictive model for the in vivo transformation and reducing potential of ultrafine CeO2 in the: 1) nanoparticle sequestration in a complex acid–base-controlled environment (liver) with some dissolution of original particles; 2) in situ formation of second-generation CeO2 (CeO2 clouds); 3) greater reactive surface area for the second-generation CeO2 inside CeO2 clouds; and 4) higher Ce3+ associated with second-generation CeO2. This in vivo transformation of CeO2 could be a dynamic system.
According to the authors, the model established a temporal link between cellular processes in the brain (Hardas SS et al., 2014) and the in vivo transformation of CeO2 in the liver. Prior to in vivo processing of nano-CeO2, oxidative stress gradually increased in CeO2-treated rats and brain cellular stress (stage S1; 1 h-1 d; antioxidant defence) was then followed by brain inflammation (stage S2; 1-7 d; pro-inflammatory response) and then brain-cell death (stage S3; 7-30 days; mitochondrial-mediated cytotoxic effects). However, after 90 days an unexpected brain cellular clearance response was reported (Hardas SS et al., 2014), which corresponded in the model to stage S4 (30-90 days) and indicated a reversal or decrease in oxidative stress. Brain oxidative stress was shown to have increased as early as 1 h after nano-CeO2 infusion (signalling pathway: Nrf-2), and 1 day later triggered brain inflammation (signalling pathway: TNF-α (NF-κB IL-1β)). The model also showed that between 7 and 30 days, inflammation was significantly increased in the hippocampus, cortex, and cerebellum, and brain-cell death occurred at 30 days (signalling pathway: pro-apoptotic; Pro-Caspase-3), and then followed by autophagy (LC-3AB), and finally returned to normal levels at 90 days. Importantly, the late cellular and protein-clearance response previously observed in the rat brain (Hardas SS et al., 2014) coincided with the inception of in vivo processing of nano-CeO2. Since the processed nanoparticles in the liver (second-generation CeO2) were shown to have high Ce3+, they may also have a high affinity for reactive oxygen species and free-radical scavenging activity. Although the mechanisms were not established by the authors, they anticipated that the in vivo processing of nano-CeO2 had far-reaching protective effects, as was observed with the reversal in pro-oxidant brain effects.

Bioaccessibility (or Bioavailability) testing results:

No data available

Conclusions:

Intravenously-infused nano-CeO2, after prolonged residence time in the liver (90 days), underwent in vivo processing. Nano-CeO2 undergoes partial dissolution inside the liver that causes a shift towards smaller particle size and an increased reactive surface area with enhanced Ce3+ activity that leads to greater free-radical scavenging potential of the ultrafines. The authors concluded that the breakdown and redistribution could be a step towards improving ceria biocompatibility in vivo.
=> Interpretation of results: bioaccumulation potential in reticulo-endothelial organs and biotransformation in liver

Executive summary:

Graham UM et al. (2014) investigated the in vivo transformation of nanometric cerium dioxide (nano-CeO2) in the rat liver by evaluating nano-CeO2 chemical and structural stability and solubility once sequestered inside the liver.

A water suspension of 5% crystalline citrate-coated nano-CeO2 was used in this study. The in-house synthesised nano-CeO2 had a primary particle size of ca. 30 nm, a cubic shape, and a crystalline structure. The particles showed no or few agglomeration/aggregation tendency in aqueous suspension due to citrate coating which stabilised the suspended particles at a mean size of 31 ± 4 nm. The nano-CeO2 suspension had a pH of 3.9.

The nano-CeO2 suspension was intravenously (i.v.) infused into Sprague-Dawley male rats at 0 (control), or 85 mg/kg nano-CeO2. Animals were terminated 1 h, 20 h, 30 days and 90 days after infusion. CeO2 distribution in liver, lung, kidney, spleen, heart, thymus, and brain was by high resolution transmission electron microscopy (HRTEM) and inductively coupled plasma mass spectrometry (ICP-MS) (for details, see the publications of Yokel RA et al., 2012 and 2013). In addition, the authors characterised the nano-CeO2 content in liver using transmission electron microscopy (TEM), scanning transmission electron microscopy (STEM), electron energy loss spectroscopy (EELS) and HRTEM.

According to Graham UM et al., spleen and liver had sequestered approximately the same amount of nano-CeO2. The general cytoarchitecture of liver tissue was not altered by the sequestered nano-CeO2, but Kupffer cells contained many nano-CeO2 which formed small agglomerates and underwent fusion with lysosomes to form phagolysosomes. Moreover, Kupffer cells with mononucleated lymphocytes had formed large granulomata in the tissue parenchyma, while surrounding hepatocytes rarely showed cytoplasmic inclusion of CeO2 agglomerates.

Nano-CeO2 formed agglomerates in the liver up to day 30, with no indication of breakdown or chemical transformation. However, after 90 days inside the liver, the i.v.-infused cube-shaped nano-CeO2 had become highly fragmented and rounded along their edges, which indicated that in vivo processing of the particles had occurred. Moreover, accumulations of copious ultrafine crystallites (1 to 3 nm) that formed in the liver within close proximity to the rounded nano-CeO2 were observed and labelled as CeO2 clouds. This represented a second generation of CeO2. EELS of the cloud particles demonstrated that the oxidation degree, when compared with the original nano-CeO2, was more reduced (38 to 70% Ce3+). This measurable change in the valence reduction of the second-generation CeO2 could be linked to an increased free-radical scavenging potential.

Much lesser amounts of nano-CeO2 seemed to have been bioprocessed inside the spleen.

In conclusion, i.v.-infused nano-CeO2, after prolonged residence time in the liver (90 days), underwent in vivo processing. Nano-CeO2 undergoes partial dissolution inside the liver that causes a shift towards smaller particle size and an increased reactive surface area with enhanced Ce3+ activity that leads to greater free-radical scavenging potential of the ultrafines. The authors concluded that the breakdown and redistribution could be a step towards improving CeO2 biocompatibility in vivo.

Data source

Reference

Reference 1

Reference Type:

publication

Title:

Unnamed

Year:

2012

Reference 2

Reference Type:

publication

Title:

Unnamed

Year:

2014

Materials and methods

Test guideline

Qualifier:

no guideline followed

Principles of method if other than guideline:

- Single intravenous (i.v.) injection of a suspension of nanometric cerium dioxide (nano-CeO2) in rats; sacrifice at 1 and 20 h, 30 and 90 days post-exposure
- Hepatic cerium (Ce) Ce content: inductively coupled plasma-mass spectrometry (ICP-MS)
- Nano-CeO2 localization and size distribution in liver: light (LM) and electron microscopy (EM)
- General observations: mortality, body weight, excretion
- Hepatotoxicity:
* Cell proliferation and degeneration by proliferating cell nuclear antigen (PCNA) immunostaining
* Apoptosis by terminal transferase-mediated dUTP nick end-labelling (TUNEL) method
* Liver function by measuring serum levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT)
* Oxidative stress by measuring hepatic levels of protein carbonyls (PC), 3-nitrotyrosine (3-NT), 4-hydroxy-2-trans-nonenal (4-HNE), catalase (CAT), glutathione reductase (GR), glutathione peroxidase (GPx), manganese superoxide dismutase (MnSOD), heme oxygenase-1 (HO-1)
- Histopathology: liver (tip of the right middle lobe), lung, kidney, spleen, heart, thymus, and brain
- Immunohistochemistry analysis: presence of T cells in liver parenchyma, cellular degeneration/apoptosis (TUNEL) and proliferation (PCNA) in liver tissue

GLP compliance:

not specified

Remarks:

The GLP status was not specified in the article.

Limit test:

no

Test material

Test material information

Constituent 1

Reference substance name:

Cerium dioxide

EC Number:

215-150-4

EC Name:

Cerium dioxide

Cas Number:

1306-38-3

Molecular formula:

CeO2

IUPAC Name:

cerium dioxide

Test material form:

other: crystalline nanomaterial in suspension

Details on test material:

- Name of test material: Nanometric cerium dioxide (nano-CeO2)
- Supplier: None (in-house synthesis by hydrothermal approach)
- Substance type: Monoconstituent substance
- Substance form: Nanoparticulate substance / crystalline nanomaterial in suspension
- Primary particle size (TEM, DLS, HRTEM): ca. 5 nm based on TEM and ca. 7.6 nm based on DLS (2012); 5 to 100 nm with the majority between 20 and 50 nm and average particle size of 30 nm (2014)
- Particle size distribution (DLS): Bimodal distribution with one peak centred at 7.3 nm (98%) and another at 17.2 nm (2%) (2012)
- Stability: No agglomeration/aggregation of nano-CeO2 in water based on particle size distribution (probably due to citrate coating). DLS analysis showed no change in average particle hydrodynamic diameter in 10% sucrose or in saline after 20 h at room temperature, or after 1 h at physiological temperature (37°C) (2012).
- Specific surface area (BET): 121 m²/g (2012)
- Surface charge (electrophoretic mobility method): Zeta potential of -53 ± 7 mV at pH 7.35 (2012)
- Isoelectric point: No data available
- Shape (HRTEM, HRSTEM): Polyhedral shape (2012); cubic shape (2014)
- Crystallinity (HRTEM, HRSTEM, XRD): Face-centred cubic (2012)
- Analytical purity / impurities: No data available
- Number density of nano-CeO2 in the suspension: No data available
- Cerium content in nano-CeO2 suspension (ICP-MS): 4.35 ± 0.20% (2012); 5% (2014)
- Solubility: No data available
- Oxidation degree (EELS): Ce(III) and Ce(IV) oxidation degrees for both the as-synthesized 5-nm CeO2 (collected powder) and CeO2 agglomerates in liver 30 days after infusion into rats (2012); enrichment of Ce(III) at the crystal's outer surface, or face, of the internalised nanoparticles (2014)
- Surface properties: No data available (2012); citrate coating to minimise agglomeration otherwise seen with uncoated nano-CeO2 introduced into high ionic strength solutions, such as blood (2014).
- Lot/batch No.: No data available
- Expiration date of the lot/batch: No data available
- pH: 3.9 (2014)

Further explanations on the physico-chemical characterisation of CeO2 nanoparticles are presented below in "any other information on materials and methods incl. tables".

Test animals

Species:

rat

Strain:

Sprague-Dawley

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Harlan Laboratories Inc., USA (2012)
- Age at study initiation: No data available
- Weight at study initiation: 324 ± 29 g (2012); 316 ± 27 g (2014)
- Fasting period before study: No data available
- Housing: Individually (2012)
- Diet: Ad libitum (2018 Harlan diet)
- Water: Ad libitum (reverse osmosis water)
- Acclimation period: No data available

ENVIRONMENTAL CONDITIONS
- Temperature: ca. 21.1°C
- Humidity: 30-70 %
- Air changes: No data available
- Photoperiod: 12 hrs dark / 12 hrs light

IN-LIFE DATES: No data available

Administration / exposure

Route of administration:

intravenous

Vehicle:

water

Details on exposure:

PREPARATION OF DOSING SOLUTIONS:
In both studies, an aqueous nano-CeO2 suspension (4-5%) was synthesised by hydrothermal approach, centrifuged and then washed with deionised ultra-filtered water. In the study from 2014, the suspension was citrate-stabilised.

INTRAVENOUS EXPOSURE:
A cannula was surgically inserted into the femoral veins of rats intraperitoneally anesthetised with ketamine and xylazine. Both femoral cannulae terminated in the vena cava, and one cannula was deliberately left 2 cm longer than the other. They were connected to an infusion pump via a flow-through swivel connector.
The day after cannulation, non-anaesthetised rats were infused intravenously (i.v.) via the shorter cannula with the citrate-stabilised nano-CeO2 suspension, delivered in a volume of 2 mL/kg over 1 h.
To compensate for i.v. administration of a considerable volume of water, rats were given a concurrent i.v. infusion of an equal volume and rate of 1.8% sodium chloride solution in water through the longer cannula.

Doses:

0 (control) or 85 mg/kg (analytical nano-CeO2 conc.)

No. of animals per sex per dose:

- TSENG ET AL. (2012):
* Rats terminated at 1 h
i) Nano-CeO2 treated: n = 11
ii) Water-treated (control): n = 7

* Rats terminated at 20 h
i) Nano-CeO2 treated: n = 12
ii) Water-treated (control): n = 8

* Rats terminated at 30 days
i) Nano-CeO2 treated: n = 10
ii) Water-treated (control): n = 10

- TSENG ET AL. (2014):
* Rats terminated at 1 h
i) Nano-CeO2 treated: n = 5
ii) Water-treated (control): n = 5

* Rats terminated at 20 h
i) Nano-CeO2 treated: n = 5
ii) Water-treated (control): n = 8

* Rats terminated at 30 days
i) Nano-CeO2 treated: n = 14
ii) Water-treated (control): n = 10

* Rats terminated at 90 days
i) Nano-CeO2 treated: n = 7
ii) Water-treated (control): n = 6

Control animals:

yes

Remarks:

(concurrent vehicle)

Details on study design:

- Duration of observation period following administration: 1 h, 20 h, 30 days or 90 days
- Frequency of observations and weighing: The cannulas were surgically removed from the 30- and 90-day rats several days after nano-CeO2 or vehicle infusion. Respiratory, neurological, ophthalmic, gastrointestinal, and other physiological statuses of the control and treated rats held for 30 and 90 days were assessed by daily cage-side observations. Rats assigned to day-30 and -90 groups were housed in metabolic cages from day 1 to day 14 after infusion to allow collection of urine and faeces (2014).
- Necropsy of survivors performed: yes
- Other examinations performed:


TISSUE SAMPLING AND HISTOPATHOLOGICAL ANALYSIS:
In the study from 2012, after termination of ketamine-anaesthetized rats liver samples harvested from the tip of the median lobe were fixed in 10% neutral buffered formalin and processed for histopathology analysis. Sections (5 µm) were stained with hematoxylin and eosin, and were examined for qualitative and quantitative changes. The size of hepatic granuloma was digitally recorded. The lesions were assessed on coded slides.
In the second study (2014), blood was retrieved and nine organs were excised and weighed. Tissues were harvested from the liver (tip of the right middle lobe), lung, kidney, spleen, heart, thymus, and brain and were fixed in formalin and processed for histopathology analysis. Sections (5 µm) stained with hematoxylin and eosin were examined for qualitative and quantitative changes.
Nano-CeO2-containing Kupffer cells were detected by their enlarged size and sinusoidal location; their density determined and expressed as number per mm².
Hepatic granulomata were digitally recorded and their size measured with a tracing software in an optical microscope.
The lesions were assessed on coded slides by an observer unaware of the experimental setting.

IMMUNOHISTOCHEMISTRY (IHC) (2012, 2014):
The presence of T cells in liver parenchyma was determined by IHC using anti-cluster of differentiation 3 (CD3) antibody. Liver sections (5 µm) were deparaffinised, boiled in hot citrate buffer, and incubated with anti-CD3 antibody for 60 min at room temperature (RT). Immunoreactive complexes were detected using a avidin-biotin affinity system and visualised with 3,3'-diaminobenzidine-tetrahydrochloride (DAB) as a substrate. The sections were counterstained with Mayer's hematoxylin and mounted. Controls included substitution of primary antibody with phosphate-buffered saline (PBS) and human thymus tissue as a positive tissue control.

Nano-CeO2-induced cellular degeneration and proliferation in liver tissue were determined by the terminal transferase-mediated dUTP nick end-labelling (TUNEL) method and proliferating cell nuclear antigen (PCNA) immunostaining, respectively.
For TUNEL assay, positive reaction control was obtained by using DNase treatment, negative reaction control by deleting the TdT enzyme, and positive tissue control with active apoptotic components.
Additionally, apoptosis was evaluated by immunofluorescence labelling of caspase 3. Deparaffinised liver sections were hydrated and blocked, and then incubated overnight at 4°C with anti-caspase 3 antibody. Primary antibody was detected with Alexa Fluor reagent. Tissues were counterstained with diamidino-2-phenylindole (DAPI) and images were captured on a confocal microscope.
PCNA IHC, which detects nuclear antigen synthesised in early G1 and S phases of the cell cycle, was used to assay for possible nano-CeO2-induced cellular proliferation. Liver sections were incubated overnight at 4°C with an anti-PCNA antibody. The signal was visualised using a secondary antibody and a DAB kit. Controls included substitution of primary antibody with PBS and human melanoma tissue as a positive tissue control.

SERUM BIOCHEMISTRY AND LIVER FUNCTION:
After termination of anesthetising rats with ketamine/xylazine, blood was obtained from which serum was collected by centrifugation within 1 h after blood sampling, aliquoted, and stored frozen (at -80°C).
To assess possible hepatotoxicity serum aspartate aminotransferase (AST) and serum alanine aminotransferase (ALT) were measured in blood samples from terminated animals by spectroscopy. In the study from 2014, only ALT was measured.

OXIDATIVE STRESS:
Liver samples obtained from rats at 30 days after nano-CeO2 or saline treatment were evaluated for oxidative stress by evaluating levels of proteins (western blot analyses): protein carbonyls (PC), 3-nitrotyrosine (3-NT), and 4 -hydroxy-2-trans-nonenal (4 -HNE), catalase (CAT), glutathione reductase (GR), glutathione peroxidase (GPx), manganese superoxide dismutase (MnSOD), heme oxygenase 1 (HO-1), and Hsp70.

TISSUE Ce CONTENT:
CeO2 concentrations were determined by measuring total Ce ion analysis ICP-MS after microwave-assisted nitric acid/hydrogen peroxide digestion. The mean method limit (MDL) of Ce in tissue was reported in that study as 0.089 mg/kg. Cerium concentrations that did not reach the MDL were reported here as 50% of the MDL for the purpose of data analysis.

ELECTRON MICROSCOPY (2012, 2014):
Immediately after excision, thin strips of liver samples were cut and immersion fixed in formalin for 24 h before dissection into 3 mm3 pieces, post-fixed in osmium tetraoxide, dehydrated, and embedded in epoxy resin. Thick sections (1 µm) were stained with toluidine blue for light microscopy observation and screening. Selected blocks were sectioned, mounted on grids, and examined in an electron microscope (EM) operated at 80 kV. For high-resolution microscopy, sections were collected on Formvar/carbon-coated copper grids, without staining, for transmission electron microscopy (TEM)/scanning TEM (STEM)/electron energy loss spectroscopy (EELS) analysis (at 200 kV).

IN SITU LOCALIZATION AND SIZE DISTRIBUTION OF NANO-CeO2:
The in situ bioaccumulated nano-CeO2 were characterised in Kupffer cells and hepatocytes by high-resolution transmission electron microscopy (HRTEM) as described in EM in liver section. For the in situ nano-CeO2 size distribution, particles were measured from randomly chosen, digitised phagolysosome images taken from all study groups using a microscope/imaging software. A total of 19,229 nano-CeO2 from 55 phagolysosomes were measured and used to calculate a monomodal continuous distribution at each time point studied.

Statistics:

In the study from 2012, for cell proliferation and degeneration analysis, statistical differences between two groups of parametric data were determined by one-way ANOVA and Tukey's multiple comparison post hoc tests. Comparison of oxidative stress results between groups was analysed by one-tailed Student's test (2012).
In the second study from 2014, for cell proliferation, degeneration analysis, CD3+ IHC, and ALT studies, results were statistically analysed using Student's t test. Comparison between groups was analysed by the one-tailed Student's t test.
Differences were considered to be statistically significant when the p values were less than 0.05.

Results and discussion

Mortality:

No mortality was observed after nano-CeO2 infusion over the duration of the experiment (2014).

Clinical signs:

No data available

Body weight:

The treated rats showed a transient reduction in body weight gain compared with controls during the first week after nano-CeO2 infusion (2014).

Gross pathology:

Splenomegaly (i.e., spleen enlargement) was consistently observed in nano-CeO2-infused rats at necropsy, with no gross abnormality in major organs including the liver and lung (2014).

Other findings:

- Organ weights: No data available
- Histopathology, potential target organs: See below
- Other observations: The authors mentioned a small decrease of cerium in the sampled sites to 90 days which was consistent with the less than 1% nano-CeO2 excretion in urine and faeces previously described (Yokel et al., 2012) in the first 2 weeks after nano-CeO2 infusion.

HISTOPATHOLOGICAL ANALYSIS:
According to both studies, the general histological features of the hepatocellular cords in the liver were not altered by nano-CeO2 administration at all time points measured. In the study from 2012, no nano-CeO2 was observed in liver by light microscopy either in Kupffer cells or hepatocytes cytoplasm. However, in rats terminated on day 30 post-infusion, activated Kupffer cells had clearly visible accumulations of cytoplasmic particles and were intermingled with surrounding mononucleated cells to form granulomatous lesions. The size of the granulomata ranged from 132 to 4780 µm², with an average area of 1201 ± 104 µm². The average frequency of the granuloma was 0.89 ± 0.59 granuloma/mm². Some solitary Kupffer cells filled with nano-CeO2 were identified without the encircling mono-nucleated cells and were present in the hepatic sinusoid of 30-day rats. The nature of the encircling mono-nucleated cells were partially revealed by IHC with a pan-T-cell antibody, anti-CD3.

In contrast, in the study from 2014, as early as 1-h post-infusion, nano-CeO2 internalisation in Kupffer cells was observed. The captured nano-CeO2 caused cellular enlargement and bulging of the Kupffer cell surface. The adjacent hepatocytes appeared normal and contained a full complement of cytoplasmic organelles. The number of nano-CeO2-containing Kupffer cells dispersed throughout the liver parenchyma rose in the first hour after nano-CeO2 infusion from 0 in control rats to 30.8 ± 2.5 cells/mm² but declined by 20 h (11.8 ± 1.2 cells/mm²) and remained steady at 30 days (14.0 ± 3.0 cells/mm²), before increasing again at 90 days (36.5 ± 5.7 cells/mm²) (data not shown). Statistical difference was found between all paired groups, except for 20-h and 30-day rats.
Adjacent to the sinusoidal space, occasional hepatic stellate (or perisinusoidal) cells exhibiting hyperchromatic nuclei with occasional lipid inclusions were observed. The hepatocellular cords surrounding the sinusoidal space showed only occasional and low-level nano-CeO2 storage 1- or 20-h post-infusion.
Over time, the nano-CeO2 laden Kupffer cells intermingled with mononucleated cells in the sinusoids to form granulomata, which were observed in the 30- and 90-day nano-CeO2-treated rats and appeared to coexist with the hepatocytes without imparting hepatocyte injury. The observed granulomata were of non-necrotising type with nano-CeO2 accumulation in the phagocytes. The average size of the granulomata measured 1206 ± 374 and 1654 ± 835 µm² for days 30 and 90, respectively (data not shown). In spite of the slight increase in granuloma size, their number expressed as per 100 mm² remained stable (1.42 ± 1.16 and 1.42 ± 0.75/mm²) (data not shown).
IHC revealed the presence of CD3+ T cells in the granulomata. The average number of CD3+ labelled cells per granuloma for 30- and 90-day rats was 2.4 ± 1.6 and 3.9 ± 1.7, respectively. The numbers of CD3+ cells in liver tissue parenchyma of nano-CeO2-infused rats increased from 4.6 ± 1.8 cells/mm² at 1 h, to 32.8 ± 18.9 cells/mm² by day 90.
Collagen fibre bundles stained by Masson Trichrome were observed in the periphery of granulomata and perivascular spaces without signs of generalised, diffuse fibrosis.

ELECTRON MICROSCOPY - IN SITU LOCALIZATION AND SIZE DISTRIBUTION OF NANO-CeO2
In the study from 2012, at the ultrastructural level (electron microscopy), uptake of nano-CeO2 by mainly Kupffer cells, hepatocytes and some stellate cells was detected at 1 h, 20 h and 30 days post-exposure. The circulating nano-CeO2 particulates appeared not to be retained by circulating erythrocytes. One hour after infusion, the uptake of nano-CeO2 in Kupffer cells ranged from small nanometre-sized clusters that included only a few nano-CeO2 to agglomerates larger than 1 µm. Careful examination of the surface of hepatocytes revealed endocytosis of nano-CeO2 through flask-shaped caveolar pits. The cytoplasmic distribution of the accumulated CeO2 agglomerates appeared random and CeO2 particles were not often membrane enclosed and very few were found in lysosomes. Moreover, necrotic foci within the liver were not observed. Furthermore, particles were not retained in mitochondria or within the cell nucleus. Although many agglomerates appeared near the bile canaliculi, active translocation of CeO2 particles into biliary system was not observed.
According to the study from 2014, nano-CeO2 was only occasionally observed in the sinusoidal lumen at the time points measured. The majority of the nano-CeO2 continued to appear in Kupffer cell cytoplasm throughout the time points studied. The internalised nano-CeO2 appeared as aggregates of varying dimensions and was mostly enclosed by the cell membrane. The size of the nanoparticles ranged from a few nanometres to cubes exceeding 200 nm. Once internalised, nano-CeO2 aggregates underwent fusion with lysosomes to form phagolysosomes. The in situ nano-CeO2 cubes showed a shift toward smaller-sized particles over the 90-day period. While nearly 50% of all the particles were less than 30 nm size in the as-synthesised CeO2, the size distribution shifted toward a decreased particle size with less than ~10 nm particles accounting for nearly 50% of the particles counted.
Nano-CeO2 penetration into organelles including the nucleus or mitochondria was not observed.
Within the hepatocellular cords, the lining endothelial cells and perivascular stellate cells showed occasional nano-CeO2 uptake. Little nano-CeO2 uptake and retention was observed in hepatocytes. However, occasional solitary CeO2 inclusions, containing highly compacted crystalline particles, reaching 5 µm in diameter, were observed in hepatocytes. The surrounding endoplasmic reticulum arrays, mitochondria, and the cell nucleus of these hepatocytes appeared normal and unaffected.
Active discharge of electron-dense material into bile canaliculi was not observed, and the bile duct lining cells in the periportal region were devoid of CeO2-like inclusion.
Granulomata found at days 30 and 90 were formed by a mixture of CeO2-laden Kupffer and a few non-CeO2-accumulating mononuclear cells. Phagocytic cells in the granulomata enclosed an array of different nano-CeO2 crystal sizes.


OXIDATION DEGREE OF CeO2 (EELS):
In the study from 2012, both the as-synthesized 5-nm CeO2 (collected powder) and CeO2 agglomerates in liver 30 days after infusion into rats showed a similar ratio of Ce(III) and Ce(IV) oxidation degrees. In contrast, in the study from 2014, an enrichment of Ce(III) was observed at outermost surface of the internalised nanoparticles.

IMMUNOHISTOCHEMISTRY (IHC):
According to the study from 2012, nano-CeO2 affected hepatocellular programmed cell death in liver as revealed by TUNEL assay. The persistence of nano-CeO2 in the liver 30 days post-infusion was associated with significant elevation of apoptotic cell counts. While the majority of apoptotic cells was hepatocytes, degenerating mononucleated cells were also found in the granulomata.
In spite of the CeO2-induced apoptosis, PCNA cell proliferation analysis showed no clear trend and level comparable to control at day 30.

In the study from 2014, according to the TUNEL results, acute presence of nano-CeO2 caused a statistically significant increase in the number of apoptotic cells by 30 days that was sustained to 90 days after infusion, confirmed by caspase-3 IHC. Concurrently, cell proliferation analysis (PCNA staining) showed a similar trend, although the number of G1/S phase positively stained cells was not statistically significantly different from controls until 90 days when the results showed a statistically significant elevation in proliferating cells.

SERUM BIOCHEMISTRY AND LIVER FUNCTION:
According to the study from 2012, serum AST levels 1 and 20 h post-exposure were statistically significantly higher than controls, meaning that nano-CeO2 infusion induced an acute hepatotoxicity. However, the AST elevation subsided by day 30 in CeO2-infused rats, suggesting that the acute hepatotoxicity was transient.

According to the study from 2014, serum ALT levels were slightly modified, reaching a peak of 30.33 ± 16.08 U/L by 20 h (control level was 20.71 ± 15.39 U/L) and subsequently decreased over time reaching the lowest level on day 90 in nano-CeO2-infused rats. The differences observed were not statistically significant.

OXIDATIVE STRESS:
According to the study from 2012, 30 days post-infusion, a significant increase was observed in PC levels only, among the markers assessed. Moreover, there was a slight but significant decrease in liver CAT and GPx activities. Among the six antioxidant-related proteins analysed, only HO-1 (Hsp32) levels were significantly elevated in the liver following nano-CeO2 treatment compared to controls.

TISSUE Ce CONTENT:
See in Table 1 below in "any other information on results incl. tables".

Any other information on results incl. tables

Table 1: Summary of Ce contents in liver 1 h, 20 h and 30 days post-infusion (Tseng MT et al., 2012)

	Ce content
	1 h	20 h	30 d
In liver (mg/kg wet weight)	424 ± 297 (22%)	1007 ± 264 (51%)	578 ± 336 (44%)
In blood (mg/mL)	270 ± 79 (27%)	12 ± 9 (1.3%)	0.11 ± 0.16 (0.01%)

 

These data showed clearance of nano-CeO2 from the blood into liver, and the persistence of nano-CeO2 in the liver. The remaining nano-CeO2 was dispersed in other organs, including the spleen. Of the 25 control rats, 5 animals had liver cerium concentration that exceeded the MDL, with levels ranging from 0.24 to 1.5 mg/kg.

Applicant's summary and conclusion

Conclusions:

The large single intravenous administration of nano-CeO2 was well tolerated in rats since no mortality occurred. However, once nano-CeO2 gained systemic entrance to an organ rich in reticulo-endothelial cells like the liver, mainly in Kupffer cells and in hepatocytes, the particles accumulated and persisted for an extended period in liver. Except for a relatively mild but persistent hepatic inflammatory response, a relative lack of hepatic injury was found.

Executive summary:

Tseng MT et al. (2012, 2014) examined the systemic biodistribution of nanometric cerium dioxide (nano- CeO2) and the subsequent acute and subacute adverse hepatic responses in rats.

In the first study (Tseng MT et al., 2012), nano-CeO2 of 5-7 nm diameter was in-house synthesised. Once suspended in water, the nanoparticles showed a bimodal distribution with one peak at 7.3 nm (98%) and another at 17.2 nm (2%). Thus, the nano-CeO2 displayed no agglomeration/aggregation in water. Although it was not explained in the publication, this absence of agglomeration/aggregation could be due to a citrate coating of nano-CeO2 as already demonstrated in previous articles of the same research’s team (i.e., Yokel RA et al.). Further, this crystalline nano-CeO2 had a specific surface area of 121 m²/g, a zeta potential of -53 ± 7 mV at neutral pH 7.35, and a mixed oxidation degree (i.e., Ce(III) and Ce(IV)).

In the second publication (Tseng MT et al., 2014), the in-house synthesised nano-CeO2 had an average size of 30 nm. The aqueous suspension contained ~5% of citrate-coated nano-CeO2 and had a pH of 3.9. These nanoparticles displayed a cubic shape and an enrichment of Ce(III) at its surface.

Suspended in water, the tested nano-CeO2 was intravenously (i.v.) infused at 85 mg/kg to male Sprague- Dawley rats (5 to 14/group), which were terminated 1 h, 20 h, 30 days or 90 days later. Water-infused animals served as control (5 to 10/group). General observations were performed daily until 90 days and excretion was estimated from day 1 to day 14 after infusion. The hepatic content in cerium was determined using inductively coupled plasma-mass spectrometry (ICP-MS). Hepatotoxicity was assessed by performing histopathological examination of the liver, immunohistochemistry (T cell, cell apoptosis, cell proliferation, cell degeneration), measurement of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels in serum, and evaluation of several oxidative parameters at all different time points. Moreover, the localization and size distribution of nano-CeO2 in liver were evaluated at the different time points using light and electron microscopy.

In the first study (Tseng MT et al., 2012), nano-CeO2 was found mainly sequestrated by Kupffer cells. The internalised nano-CeO2 appeared as spherical agglomerates of varying dimension without specific organelle penetration. Indeed, nano-CeO2 was rarely found in lysosome and never in the nucleus and mitochondria. In hepatocytes, the agglomerated nano-CeO2 frequently localized to the plasma membrane facing bile canaculi. Sustained nano-CeO2 bioretention was associated with granuloma formations. A statistically significant elevation of serum AST level was seen at 1 and 20 h, but subsided by 30 days after nano-CeO2 administration. Furthermore, elevated apoptosis was observed on day 30. However, nano- CeO2 induced a relative small increase in oxidative stress markers. The lack of organelle penetration by nano-CeO2 may contribute to the relative lack of long-term cytotoxicity.

In the second study (Tseng MT et al., 2014), rats tolerated without mortality the single i.v. administration of a large nano-CeO2 dose. The particles seemed to be very slightly excreted through urine and faeces; nano- CeO2 was rather accumulated in liver. Ultrastructural analysis revealed again nano-CeO2 accumulations in Kupffer cells (as early as 1 h after infusion), and, in lesser extent, in stellate cells and hepatocytes. The internalised nano-CeO2 was mainly found in large phagolysosomes and in smaller lysosomes. However, the particles did not penetrate organelles such as mitochondria and the nucleus. As described above, there were observations of liver granulomata formation 30 days after nano-CeO2 infusion; these granulomata persisted to 90 days post-exposure. The nano-CeO2-induced granulomata were described as small, of non-necrotising type, and composed of Kupffer cell aggregates surrounded by T lymphocytes. Absence of necrotic areas and a lack of fibrosis in the liver suggested that the continuous presence of nano-CeO2 might be tolerated by rats. Furthermore, a non-significant variations in serum ALT levels were observed over time as a rapid elevation, consistent with an initial acute, although mild, liver injury followed by a slow return to the baseline by day 90 suggesting a functional stabilization of injury in the liver. However, increased cellular proliferation 90 days after infusion in parallel with increased apoptosis from 30 days post-infusion suggested that nano-CeO2 could alter the liver cytoarchitecture over longer survival time. An enrichment of Ce(III) at the surface was observed in this study in CeO2-retained cells over the 90 day observation period.

The authors concluded that a single large vascular nano-CeO2 infusion induced only a relatively mild but persistent inflammatory response in the liver where the particles accumulated and persisted for an extended period, mainly in Kupffer cells and hepatocytes.