Carbon black

Specific investigations: other studies

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

mechanistic studies

Type of information:

experimental study

Adequacy of study:

key study

Study period:

16th of September 2020

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

test procedure in accordance with generally accepted scientific standards and described in sufficient detail

Qualifier:

no guideline required

Principles of method if other than guideline:

Inhalation exposure to particles may cause adverse lung reaction and the activation and/or damage of alveolar macrophage is assumed to be an early key event. The NR8383 rat alveolar macrophage cell line is a well-established in vitro model which has been successfully used to examine the biological activity of respirable particles (Wiemann et al. J. Nanobiotechnology 12:14). Four parameters are tested in the cell culture supernatant under serum-free conditions, namely lactate dehydrogenase (LDH), the lytic enzyme glucuronidase (GLU), H2O2, and tumour necrosis factor α (TNF). These effects are relevant for the biological response of alveolar macrophages (AM) and, if occurring also in vivo, for their clearance process of particles in the lung: Release of LDH reflects membrane damage and necrosis of AM. Glucuronidase is a representative of lytic enzymes released from (phago)lysosomes. These enzymes may damage surrounding tissue. H2O2 from AM may lead to oxidative damage of lipids, proteins and/or DNA putting lung cells in the neighbourhood of AM at risk. Finally, TNF is an important pro-inflammatory cytokine with a multitude of effects on various cell types. These include proliferation, apoptosis, and/or triggering further immune responses.
Quartz DQ12 is a well-accepted positive control as it acts progressively inflammatory and fibrotic within the lung. In contrast, corundum particles elicit no such effects in the rat lung even if a lung burden of 5 mg is applied, e.g. by intratracheal instillation. This remarkable difference between both types of mineral particles is reflected by in vitro testing with primary as well as cultured alveolar macrophages (NR8383) under serum-fee conditions. These particles were, therefore, used for positive and negative control experiments, respectively. Most often the response to unknown particle is intermediate to these controls. To test for the capacity of NR8383 cell to produce H2O2 and TNF, zymosan and lipopolysaccharide (LPS) were used as additional positive controls, respectively.
In case that agglomeration of particles leads to a complete sedimentation followed by an uptake into a defined number of cells, the experiments provide a first information which mean particle load may be tolerated by AM in vivo. Therefore, sedimentation and uptake of particles are routinely controlled and documented on micrographs. For nano-sized materials, which often need a well-adapted preparation of particle suspension according to established protocols, particle size distribution is monitored via particle tracking analysis in different relevant media.

GLP compliance:

no

Type of method:

in vitro

Endpoint addressed:

not applicable

Route of administration:

other: in vitro test

Positive control:

Corundum and quartz DQ12 particles in most tests served as negative and positive control, respectively

Details on results:

Sterility Testing: The suspension of N 990 did not give any positive results, neither on casein-peptone nor on malt agar during the 72h incubation period at 37°C. Light microscopic inspection of the diluted suspension at the end of the incubation period also gave no indication for a contamination of test materials with live germs.
Particle Settling and Uptake: The material was dispersed in distilled H2O (dH2O) following the NanoGenotox protocol. The hydrodynamic diameters (HD) was measurable by PTA in H2O only. Under cell culture conditions, i.e. in F-12K medium, there was a fine fraction of particles with a hydrodynamic diameter (mode value) of 315 nm. There was a homogeneous layer of micron-sized agglomerates/aggregates at the bottom of the cell culture vials. The size of these aggregates/agglomerates was < 5 μm. The density of these settled agglomerates, whose size is not included in the PTA measurements, correlated with the administered concentration. Importantly, the settled aggregates/agglomerates were completely engulfed by NR8383 cells most of which exhibited dark inclusions.
In vitro toxicity data: Control cells reacted as expected: non-particle treated, or LPS-treated cells (a control for TNF induction) were undamaged. Corundum treated cells were particle-laden but undamaged. Quartz DQ12 treated cells were particle laden and appeared granular and partly deteriorated.
NR8383 cells exposed to N990 completely cleared the settled fraction of particles from the bottom of the culture wells up to a concentration of 180 μg/mL. Dark inclusions appeared within the cells indicating successful particle uptake. N990 elicited some release of LDH upon the highest concentration of 180 μg/mL. Values for GLU, H2O2, and TNFα were not different from control.

Hydrodynamic diameters [nm] of test material in H2O.

particle name

Messkonz. [µg/mL]

diluent

camera level

concentration [particle/mL]

size (mean)

size (mode)

size (D10)

size (D50)

size (D90)

average

±

SEM

average

±

SEM

average

±

SEM

average

±

SEM

average

±

SEM

N990

294_H2O

3.6

H2O

8

5.73E+07

326.6

15.9

315.1

10.4

191.4

21.8

319.2

10.2

440.3

13.4

SEM: standard error of the mean; Values for D10, D50, D90 describe the cumulative particle size distribution at 10%, 50% and 90% of the maximum value.

Table showing in vitro effects of CORAX N990 on NR8383 macrophages in comparison to corundum and quartz DQ12 (n=3).

	internal sample number	LDH [% of pos. CTR]			2-way ANOVA	GLU [% of pos. CTR]			2-way ANOVA	ROS H2O2[µmol/L]			2-way ANOVA	TNF TNFa [pg/mL]			2-way ANOVA
[µg/mL]
N990	294
0		14.07	±	0.33		0.76	±	0.35		0.83	±	0.21		6.23	±	0.24
22.5		12.80	±	1.17		0.36	±	0.46		0.72	±	0.23		8.27	±	1.26
45		12.83	±	1.30		0.71	±	0.65		0.77	±	0.19		10.20	±	2.24
90		15.76	±	1.21		0.63	±	0.42		0.82	±	0.15		11.73	±	1.46
180		21.29	±	0.34	*	1.37	±	0.66		0.83	±	0.32		13.14	±	2.66
corundum	n.d
0		11.77	±	0.46		0.85	±	0.29		0.92	±	0.54		9.37	±	1.91
22.5		11.08	±	2.47		1.46	±	0.41		1.10	±	0.18		5.02	±	2.09
45		11.86	±	1.58		2.17	±	0.59		1.24	±	0.19		4.65	±	1.74
90		12.52	±	2.65		1.78	±	0.34		1.36	±	0.28		5.34	±	2.68
180		14.72	±	1.71		1.19	±	0.62		1.82	±	0.11		5.94	±	3.55
quartz DQ12	n.d
0		11.77	±	0.46		0.85	±	0.29		0.92	±	0.54		9.37	±	1.91
22.5		11.48	±	3.97		0.18	±	0.40		1.07	±	0.44		5.90	±	3.32
45		12.55	±	1.06		0.64	±	0.41		1.26	±	0.40		8.40	±	4.51
90		26.13	±	4.91	***	2.97	±	1.00	*	1.45	±	0.41		30.44	±	18.47
180		80.05	±	9.75	***	14.06	±	4.16	***	2.18	±	0.38	**	113.29	±	47.32	***

Mean values and standard deviations from three independent experiments. Values significantly different from cell control are marked by asterisks (*: P ≤ 0.05, , **: P ≤ 0.01, ***: P ≤ 0.001). One way analyses of variance (ANOVA) and Dunnett's test were used to compare means from the control and treated groups.

A particle is classified as to be active if the Low Observed Adverse Effect Concentration (LOAEC) multiplied by the specific BET value drops below the threshold value of 6000 mm2/mL for at least 2 out of the 4 tests.

Table of low observed adverse effect concentrations and active/passive calculation.

BET (m2/g)	LOAEC (µg/mL)				LOAEC (µg/mL) x BET (mm2/mL)				Assay numbers underscoring threshold	Classification
	LDH	GLU	TNF	H2O2	LDH	GLU	TNF	H2O2
9	180	n.s	n.s	n.s	1620	-	-	-	1	Passive

According to the active/passive classification criteria of Wiemann et al. 2016, the substance may be classified as passive.

Conclusions:

Passive in accordance to the classification criteria of Wiemann et al. 2016.

Executive summary:

The gravitationally settled fraction of N990 up to a concentration of 180 μg/mL was completely ingested by alveolar macrophages (NR8383 cells). Minor cytotoxic were observed as indicated by a small release of LDH upon 180 μg/mL; there were no significant activating (GLU), H2O2, or pro-inflammatory effects (TNFα). Given the BET value of N 990 and according to the active/passive classification criteria of Wiemann et al. 2016, N990 is classified as passive.

Table of low observed adverse effect concentrations and active/passive calculation.

BET (m2/g)	LOAEC (µg/mL)				LOAEC (µg/mL) x BET (mm2/mL)				Assay numbers underscoring threshold	Classification
	LDH	GLU	TNF	H2O2	LDH	GLU	TNF	H2O2
9	180	n.s	n.s	n.s	1620	-	-	-	1	Passive

Reference 2

Endpoint:

mechanistic studies

Type of information:

experimental study

Adequacy of study:

key study

Study period:

August - Spetember 2020

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

test procedure in accordance with generally accepted scientific standards and described in sufficient detail

Qualifier:

no guideline required

Principles of method if other than guideline:

The SOP, which described multi-dose protocol of the FRAS assay (Ferric Reduction Ability of Serum) and was published in 2017 by BASF (Gandon et al. 2017), was used for reactivity testing of carbon black samples under physiological conditions.
Briefly, carbon black samples were incubated with HBS for 3 h at 37 °C. Before incubation, bath sonication for 1 min was applied to prevent forming large agglomerates and access hole surface area. CBs separated from HBS via ultracentrifugation (AUC-Beckman XL centrifuge (Brea, CA, USA) at 14,000 G for 150 min). Subsequently, a 100 μL of CB-free HBS supernatant was incubated in the FRAS reagent that contains the Fe3+ complex. The total antioxidant depletion as a measure of the oxidative potential of CBs was determined by using UV-vis spectrum of the iron complex solution. Trolox, a water-soluble analogue of vitamin E, was used as an antioxidant to calibrate the FRAS results. Different Trolox concentration (from 0.001 to 0.1 g/L) was tested by FRAS assay to obtain FRAS absorption signals that were linearly fitted. Finally, the oxidative damage induced by carbon blacks was calculated in Trolox equivalent units (TEUs).

Reference:
Gandon, Arnaud, Kai Werle, Nicole Neubauer, and Wendel Wohlleben. 2017. "Surface reactivity measurements as required for grouping and read-across: An advanced FRAS protocol." In Journal of Physics: Conference Series, 012033. IOP Publishing.

GLP compliance:

no

Endpoint addressed:

not applicable

Examinations:

The concentration range was arranged for each sample regarding its reactivity. For low reactive samples, the dose range of 0,1-100 g/L was applied. On the other hand, the dose range of 0,05-10 g/L was utilized for high reactive samples. Furthermore, surface-based concentrations of samples were calculated using BET values have been already provided by Orion. FRAS mass-metric Biological Oxidative Damage (mBOD) and surface-metric Biological Oxidative Damage (sBOD) values for all CBs were calculated. Mass metric FRAS results on all CB samples were plotted graphically. Mass-based and surface-based dose response curves of all samples were compared. The error bars of triplicate are indicated in graphs and some of them are smaller than the size of the symbol and thus often not visible in the plots. Background FRAS signal level is up to 5000 nmol TEU/L and saturation of FRAS signals occurred at the level of about 250,000 nmol TEU/ L, indicating that all antioxidants contained in the human serum are consumed during the incubation.

Details on results:

N990 was not potent in generating reactive species in the FRAS assay with FRAS signals for concentration between 0.22 - 74.9 g/L still at background levels (up to 5000 nmol TEUL/mL). Only from 100 g/L was there an appreciable increase over background signals.

Summary of all FRAS data

Carbon Black	Surface treatment	BET m^2/g	Concen-trations (g/L)	Concen-trations (m²/L)	FRAS signal (nmol TEU /mL)	FRAS signal    Std dev	mBOD [nmol TEU / mg ENM]	sBOD [nmol TEU / m² ENM]
N 990 (bulk)	no	9.00	0.22	2.00	6142.00	293.00	27.92	3071.00
			1.04	9.40	7062.00	206.00	6.79	751.28
			4.90	44.10	5986.00	627.00	1.22	135.74
			9.69	87.20	4543.00	132.00	0.47	52.10
			24.48	220.30	4439.00	243.00	0.18	20.15
			74.90	674.10	5744.00	724.00	0.08	8.52
			100.30	902.70	13694.00	474.00	0.14	15.17
	Average						0.14	15.17
	STD						0.52	57.44

*numbers in red and in italics indicate that the oxidative damage is not significant above background (of approx. 5,000 nmol TEU/mL), or approaches saturation (approx. 250,000 nmol TEU/mL). These dose ranges should not be interpreted.

Executive summary:

To study the oxidative potential of the nanoforms of this set, four treated nanoforms (Special Black 100, Monarch 1000, Colour Black FW 200, Raven 5000 Ultra II), representative nanoforms of the lower, mid and upper end of “set_treated” were subjected to in vitro testing using the Ferric reduction ability of serum (FRAS) as described in (Gandon et al. 2017). The carbon black samples were tested at concentrations ranging from 0.05 – 100 g/L. The samples were sonicated (1 min) and then incubated with HBS for 3h at 37 °C. CB particles and HBS were separated in the next step via ultracentrifugation (14,000 G for 150 min). Subsequently, a 100 μL of CB-free HBS supernatant was incubated in the FRAS reagent that contains the Fe3+ complex. The total antioxidant depletion, as a measure of the oxidative potential, was determined by using UV-vis spectrum of the iron complex solution. Trolox, a water-soluble analogue of vitamin E, was used as an antioxidant to calibrate the FRAS results. The oxidative damage induced by the nanoforms was calculated in Trolox equivalent units (TEUs). 

N990 was not potent in generating reactive species in the FRAS assay with FRAS signals for concentration between 0.22 - 74.9 g/L still at background levels (up to 5000 nmol TEUL/mL). Only from 100 g/L was there an appreciable increase over background signals.

Description of key information

To study the oxidative potential of N990, a sample was subjected to in vitro testing using the Ferric reduction ability of serum (FRAS) as described in (Gandon et al. 2017). The carbon black samples were tested at concentrations ranging from 0.05 – 100 g/L. The samples were sonicated (1 min) and then incubated with HBS for 3h at 37 °C. CB particles and HBS were separated in the next step via ultracentrifugation (14,000 G for 150 min). Subsequently, a 100 μL of CB-free HBS supernatant was incubated in the FRAS reagent that contains the Fe3+ complex. The total antioxidant depletion, as a measure of the oxidative potential, was determined by using UV-vis spectrum of the iron complex solution. Trolox, a water-soluble analogue of vitamin E, was used as an antioxidant to calibrate the FRAS results. The oxidative damage induced by the nanoforms was calculated in Trolox equivalent units (TEUs). 

N990 was not potent in generating reactive species in the FRAS assay with FRAS signals for concentration between 0.22 - 74.9 g/L still at background levels (up to 5000 nmol TEUL/mL). Only from 100 g/L was there an appreciable increase over background signals.

Pursuant to the recommendations of the DF4Nano- and nanoGRAVUR frameworks for grouping nanomaterials CORAX N990 (a non-nanoform/Bulk form of Carbon black) was tested in a cellular in vitro assay using rat NR8383 alveolar macrophages (Wiemann et al. 2016). Wiemann and co-authors proposed an approach in their 2016 paper, which has been endorsed by the DF4Nano and nanoGRAVUR frameworks for grouping that allows an assessor to predict effect outcome in the rat lung using data generated in vitro for four parameters indicative of adverse pulmonary effects (“what they do”). They validated their approach with several metal oxides and graphene showing with a 95 % accuracy that, they could predict whether a substance would have a NOAEC less (designated as active) or greater (designated as passive) than 10 mg/m³in thein vivoshort-term inhalation study (STIS) (Ma-Hock et al. 2009). In the macrophage assay, test materials are incubated with the cells in a protein-free culture medium (Ham's F-12K). Lactate dehydrogenase, glucuronidase, and tumour necrosis factor alpha are assessed after 16 h. In parallel, H₂O₂is assessed after 1.5 h. A particle surface area-based threshold of <6000 mm²/mL is used to determine the biological relevance of the lowest observed in vitro effect with statistical significance. Significant effects that are recorded above this threshold are assessed as resulting from test material-unspecific cellular 'overload'. A test material is classified as active in the in vitro assay if the Low Observed Adverse Effect Concentration (LOAEC) multiplied by the specific BET value drops below the threshold value of 6000 mm²/mL for at least 2 out of the 4 measured endpoints. They were assessed as passive if 0 or 1 parameter was altered.

 

Agglomeration and sedimentation to the bottom of the well was recorded. Applying the above criteria by Wiemann et al 2016 to results of the NR8383 alveolar macrophage assay, N990 is assessed as passive.

 

Nanoform

Suraface treatment?

BET (m2/g)

LOAEC (µg/mL)

LOAEC (µg/mL) x BET (mm2/mKL)

Number of assays underscoring threshold of 6000 mm2/ml

Classification

LDH

GLU

TNF

H2O2

LDH

GLU

TNF

H2O2

N990

no

9

180

n.s

n.s

n.s

1620

1

Passive

Additional information

As per DF4Nano Framework the group MG3 is designated for Passive nanomaterials (NMs): Bio-persistent, non-fibrous nanomaterials that (a) do not exhibit specific bio-interactions (low surface reactivity); (b) do not possess toxic potential (chemical composition devoid of active components; no specific cellular effects); and (c) are not mobile (agglomeration in biological fluids). In vivo, the 'passive state' of NMs is confirmed in that they do not elicit apical toxic effects and are not biodistributed from the site of contact or outside the mononuclear phagocyte system (MPS). At high concentrations, they may elicit effects on account of their particulate nature, especially by dust inhalation, just as non-nanosized particles may also do. Accordingly, NMs assigned to MG3 are expected to induce adverse effects in the lung only if aerosol concentrations, lung deposition and impaired clearance result in lung overload conditions (ECETOC, 2013).

References.

Arts, J. H., M. Hadi, M. A. Irfan, A. M. Keene, R. Kreiling, D. Lyon, M. Maier, K. Michel, T. Petry, U. G. Sauer, D. Warheit, K. Wiench, W. Wohlleben, and R. Landsiedel. 2015. "A decision-making framework for the grouping and testing of nanomaterials (DF4nanoGrouping)." Regul Toxicol Pharmacol71 (2 Suppl):S1-27. doi: 10.1016/j.yrtph.2015.03.007.

Wiemann, M., A. Vennemann, U. G. Sauer, K. Wiench, L. Ma-Hock, and R. Landsiedel.2016. "An in vitro alveolar macrophage assay for predicting the short-term inhalation toxicity of nanomaterials." J Nanobiotechnology14:16. doi: 10.1186/s12951-016-0164-2.Wohlleben, W., B. Hellack, C. Nickel, M. Herrchen, K. Hund-Rinke, K. Kettler, C. Riebeling, A. Haase, B. Funk, D. Kühnel, D. Göhler, M. Stintz, C. Schumacher, M. Wiemann, J. Keller, R. Landsiedel, D. Broßell, S. Pitzko, and T. A. J. Kuhlbusch.2019. "The nanoGRAVUR framework to group (nano)materials for their occupational, consumer, environmental risks based on a harmonized set of material properties, applied to 34 case studies." Nanoscale11 (38):17637-17654. doi: 10.1039/c9nr03306h.