Lung inflammation and lack of genotoxicity in the comet and micronucleus assays of industrial multiwalled carbon nanotubes Graphistrength © C100 after a 90-day nose-only inhalation exposure of rats

General

The present studies were conducted according to the OECD Principles of Good Laboratory
Practice 68] and the OECD test guidelines no. 413 29], 474 69] and 489 70]. The design of the 90-day inhalation toxicity study was developed taken into account
the OECD recommendations 30] for the revision of the tests guidelines applicable to the inhalation toxicity testing
of nanomaterials. A graphical representation of the design of the 5- and 90-day studies
is presented in Fig. 4. The studies were performed in an AAALAC-accredited laboratory in accordance with
the Swiss Animal Protection Law.

Fig. 4. Graphical representation of the design of the 5-day exposure/28-day recovery and 90-day
exposure/90-day recovery studies. Abbreviations: balf: bronchoalveolar lavage fluid;
bp: blood pressure; cbf: clinical signs, body weight and food consumption; gt: genotoxicity
tests (comet and micronucleus assays); hcu: hematology, blood chemistry and urinalysis;
fob: functional observation battery; pa: pathology (organ weight, macroscopic and
microscopic observations); op: ophthalmology; sp: sperm analysis; vs: vaginal smears

Test materials

Graphistrength
^©
C100, are exclusively produced by Arkema France, Colombes, France. SEM and TEM images
(Fig. 5) show that Graphistrength
^©
C100 is made of tightly bound agglomerates constituted with entangled MWCNT. These
agglomerates can be spherical, ovoid or irregular shaped and have a granulometry centred
on 400 ?m, with fragments of pellets of less than 15 ?m representing a volume of under
0.23 % 27]. The MWCNT of Graphistrength
^©
C100 are synthesized at high temperature using a fluidized bed with ethylene as a
carbon feedstock and iron oxide (Fe
₂
O
₃
, ?5 %, [1309-37-1]) on alumina (Al
₂
O
₃
, ?7 %, [1344-28-1]) as catalytic source. Batches nos. 8287 and 110329-018 of Graphistrength
^©
C100 were used for the 5-day and the 90-day exposure studies, respectively.

Fig. 5. Electron microscopic images of Graphistrength
^©
C100. SEM of the commercial Graphistrength
^©
C100. (a) Magnification: 22 fold. (b) Magnification: 120 fold. (c) Magnification: 10’000 fold

Positive control substances for the genotoxicity assays were selected as recommended
by the OECD test guidelines 69], 70]. Cyclophosphamide monohydrate (CPA, batch A0302605, purity 97 %) from Fisher Scientific
GmbH for the micronucleus assay and methyl methanesulfonate (MMS, batch MKBL6789V,
purity 99.9 %) from Sigma-Aldrich for the comet assay.

Physico-chemical characterisations

The original Graphistrength
^©
C100 batches and samples taken at different steps of our aerosol generation process
during the method development (see the additional file 1) were analyzed by SEM (ZEISS LEO 1530 VP Scanning Electron Microscope, LEO Elektronenmikroskopie
GmbH, Oberkochen, Germany equipped with an X-rays analyzer Oxford type “Energy 200”)
for the morphology of the particles, by TEM (Philips-FEI CM 200 transmission electron
microscope, FEI, Hillsboro, Oregon, USA working under electron beam accelerating voltages
from 20 kV to 200 kV and with a point-to-point resolution of 0.27 nm) for the walls
number, diameters, length size and ends of the nanotubes, by laser method for the
particle size, by porosimetry with mercury intrusion for the apparent density, by
BET method for specific area (surface to volume ratio was calculated), by calcination
for ash content and the elementary organic analysis, by XPS for the chemical surface
analysis and by lithium tetraborate fusion method for metal content. Further descriptions
of the methods are included in Additional file 1: Table S3.

As reported in the additional file, the effect of the micronisation and aerosolisation
methods used for these 5-day and 90-day exposure studies on the integrity and the
physico-chemical properties of Graphistrength
^©
C100 was carefully examined during the method development. It was also compared to
the effect of a brush dust generator as used by Ma-Hock et al.16] during another 5-day inhalation toxicity study with Graphistrength
^©
C100 (see the additional file 1).

Aerosol generation

Graphistrength
^©
C100 was ground in a ceramic ball mill for 20 h under air (5-day study) or 12 h under
an argon atmosphere to reduce oxidation (90-day study) and the ground test material
was sieved before the aerosol generation (see the additional file 1 for details).

The highest aerosol concentrations used for the 5-day and 90-day exposure studies
(1.25 and 5.0 mg/m
³
, respectively) were generated from the milled and sieved Graphistrength
^©
C100 using a SAG 410 Solid Aerosol Generator (Topas GmbH, Dresden, Germany) connected
to a micronizing jet mill (CR Equipement SA, Coppet, Switzerland) and a cyclone and
two elutriators (GlasKeller, Basel, Switzerland) thereafter. The effect of the milling
duration on the structure (by TEM) and oxidation (by XPS) of the MWCNT was evaluated
in a pilot study as reported in the additional file 1. The aerosols generated were then discharged into the exposure chamber through a

⁶³
Ni charge neutralizer. The generated test aerosols were diluted as necessary with
compressed air to achieve the concentrations required for the 5-day and 90-day inhalation
exposures. The aerosol concentrations for the low and intermediate groups were achieved
by serial dilution with compressed, filtered, dry air of the higher aerosol concentrations
(mid and high concentrations, respectively) using a TD190 H and TD190 M air vacuum
device (Air-Vac Engineering, Seymour, USA). The aerosol was discharged constantly
through the exposure system and exhausted using a tubing/filter system. The exposure
system ensured a uniform distribution and provided a constant flow of test material
to each exposure tube. The flow of air at each tube was between 0.73 to 1.0 L/min,
which was sufficient to minimize re-breathing of the test aerosol as it was more than
twice the respiratory minute volume of a rat.

Animals and husbandry

Healthy male and female Rats, RccHan
^©
: WIST(SPF) were supplied by Harlan Laboratories, B.V. (5961 NM Horst, The Netherlands).
After an acclimatization period of at least 7 days, the animals were 11 and 8 weeks
old at the start of 5-day and 90-day exposures, respectively. On the first day of
the 5-day and 90-day exposures, the body weights ranged from 291 to 347 g and 243
to 296 g for males and from 176 to 214 g and 135 to 228 g for females, respectively.
The rats were randomly allocated by sex to the control and the test groups in groups
of maximally five in Makrolon type-4 cages with wire mesh tops and sterilized standard
softwood bedding (“Lignocel” J. Rettenmaier Söhne GmbH Co. KG, 73494 Rosenberg,
Germany) including paper enrichment (Enviro-dri from Lillico, Biotechnology, Surrey,
UK). The animal room was air-conditioned with 10-15 air changes/h, a 12 h light-12 h
dark cycle, the temperature ranging from 20 to 24 °C with relative humidity ranging
from 30 to 70 %. Except during exposure, pelleted standard Harlan Teklad 2914C rodent
maintenance diet (Provimi Kliba AG, 4303 Kaiseraugst, Switzerland) and water were
provided ad libitum.

Animal exposure

Groups of 20 male or 10 female rats or 35 male and 35 female rats were exposed nose-only
6 h per day for 5 days or 5 days per week for 90 days, respectively. Target concentrations
were 0.05, 0.25 and 1.25 mg/m
³
air and 0.05, 0.25 and 5.0 mg/m
³
air of Graphistrength
^©
C100 as low, mid and high concentrations for the 5-day and 90-day inhalation exposures,
respectively. The choice of the concentrations for the 5-day exposure study was based
on published evidence on other MWCNT 40], 44]. Considering the limited effects observed in the 5-day range-finding study, the same
low and mid concentrations were used for the 90-day exposure study and the high concentration
was increased up to the top concentrations tested in previous 90-day exposure studies
40], 44].

Animals of control groups (0 mg/m
³
) were exposed to compressed air under the same conditions as animals exposed to Graphistrength
^©
C100. During nose-only inhalation exposure, rats have a breathing pattern that results
in a more realistic internal exposure than a single high dose in i.t. instillation
and oropharyngeal aspiration or possible additional oral exposure in whole-body exposure
chambers. Inhalation exposure was performed using a flow-past system. The animals
were confined separately in restraint tubes which were positioned radially around
the flow-past, nose-only exposure chamber as described by Cannon et al.71].

As positive control groups for the genotoxicity assays, 5 males and 5 females were
treated once by oral gavage with 20 mg/kg bw cyclophosphamide (CPA) approximately
24 h before tissue sampling for the micronucleus assay, and 5 males were treated by
oral gavage with methyl methane sulfonate (MMS) two times with 100 mg/kg bw and then
once with 70 mg/kg at approximately 24 h intervals for the comet assay.

Monitoring and characterization of the test aerosol

The oxygen concentration, relative humidity and temperature in the exposure device
were measured continuously during each exposure using a calibrated device. The nominal
concentration was determined for the high concentrations by weighing the generator
reservoir containing the test material before and after exposure and dividing the
used amount by the total air-flow volume. These data were used to monitor the performance
of the generation system. During the 5-day and 90-day exposures, gravimetric determinations
of the aerosol concentrations were respectively performed once and 4 to 7 times daily
for high concentrations and twice and 1 to 3 times daily for the low and mid concentrations
using Millipore® durapore filters, type HVLP, loaded in a 47 mm in-line stainless
steel filter sampling device. The sampling flow rates were respectively 3 to 4 and
1 L/min per exposure port. For the low concentration of the 90-day exposure study,
filters were taken at the inlet pipe with airflow of 3 to 4 L/min. The same device
was used to collect aerosol samples for TEM analysis.

For both exposure periods, the cumulative particle size distribution of the test aerosol
was determined using Mercer cascade impactors (In-Tox Products, Moriarty, USA). The
test aerosol was impacted at each stage (covered with grease) and the particle size
distribution of the test material in the generated aerosol was measured by gravimetric
analysis of the test material deposited on each stage of the cascade impactor.

During the 5-day exposure, measurements were performed twice for the low and high
concentrations and once for the mid concentration using an impactor Model 200 at a
sampling rate of 9 L/min in order to collect sufficient material for this short study
period.

During the 90-day exposure, measurements were performed at least once per week for
the high concentration and five times for the mid concentrations using a standard
impactor Model 02-005 at a sampling rate of 1 L/min, matching the air flow rate at
the animal port to avoid a possible imbalance of the aerosol flow within the exposure
chamber. Accordingly, impactor samples for the mid concentration of the 90-day exposure
were collected over several days. No samples were collected for the low concentration
of the 90-day exposure as the aerosol concentration was too low to obtain reliable
results with an air flow rate of 1 L/min.

The mass median aerodynamic diameter (MMAD) and the geometric standard deviation (GSD)
were calculated on the basis of the results from the impactors, using Microsoft Excel®
software (Microsoft Corporation, USA). In addition, during the 90-day inhalation exposure,
the aerosol was analyzed once per week with a Wide Range Particle Spectrometer
^©
(WRPS, Model 1000XP, MSP Corporation, Shoreview, USA) in the size range of 5 nm to
10 ?m to determine the count median aerodynamic diameter (CMAD). The sampling airflow
rate was approximately 1 L/min.

Ante mortem observations

Cage-side clinical observations were recorded twice daily before and after exposures.
During the 90-day study, a careful clinical examination of each animal was performed
once weekly in a standard arena. Food consumptions and body weights were recorded
twice during the 5-day exposure and weekly during the 28-day recovery and the 90-day
exposure and 90-day recovery periods.

Functional observation battery, locomotor activity, grip strength, landing foot splay,
body temperature and ophthalmoscopic investigations were recorded during the last
week of the 90-day exposure period. In animals fasted in metabolism cages for approximately
18 h, hematology, blood chemistry and urinalysis (parameters as per OECD test guideline
no. 413 29] and listed in Tables 2 and 3) were performed during the last week of the 90-day exposure. Hematology and blood
chemistry were repeated during the last week of the 90-day recovery period. Tail cuff
blood pressure was measured during acclimatization and during the last week of the
90-day exposure period. Vaginal smears for estrous cycle evaluation were taken for
14 days from all females during treatment weeks 11 and 12.

Post mortem observations

Twenty-four hours and 28 days after the 5-day exposure, 5 rats/sex/group were anaesthetized
by i.p. injection of pentobarbitone and killed by exsanguination for histological
examinations and another 5 males/group served for BALF investigations on both occasions.
Twenty-four hours and 90 days after the 90-day exposure, 10 rats/sex/groups were sacrificed
for organ weights, and macroscopic, histological and BALF examinations.

Bronchoalveolar lavage fluid analysis

For broncho-alveolar lavage sampling, the complete lungs (5-day study) or the left
lobes of the lungs (90-day study) were washed six times with physiological saline
(4.0 ml for the complete lungs or 1 x 2.5 mL?+?5 x 1.5 mL for the left lobes) at room
temperature by slow instillation and withdrawal of fluid. The lavage fluid of animals
from the positive control group for the comet assay was discarded. After washing,
the left lung lobes were preserved in mincing solution at 5?±?3 °C for the comet assay.
The lavage fluids from the first (5-day study) or the first two lavages (90-day study)
were collected in a centrifugation tube on ice. The recovered volume from this wash
was recorded. Approximately 0.75 or 1 mL of the fluid was placed into a second tube
containing 37.5 or 50 ?L of 20 % BSA in PBS, respectively. Both tubes were centrifuged
at approximately 300 g for 10 min between 2 and 8 ° C. Thereafter, the supernatant
(without BSA) was taken in one tube and immediately set on ice and stored at 2-8 °
C until analysis. The first aliquot (without BSA) was used for the determination of
the enzymatic activity of lactate dehydrogenase (LDH), alkaline phosphatase (ALP),
and ?-glutamyltransferase (GGT) as well as for the determination of total protein
and/or phospholipids (storage at 2-8 ° C until analysis). The second aliquot from
the 90-day exposed and 90-day recovery animals was split into three tubes, two of
them containing approximately 0.1 mL each, the third one containing 0.3 mL and were
analyzed for contents of natural immunity mediator cytokines (TNF-?, IL-1? and IL-1?)
and adaptive immunity mediator cytokines (IL-5). IL-5, IL-1-? and TNF-? were analyzed
using the Proinflammatory Panel 1 (rat) Kit V-Plex
^TM
from MSD. IL-? was analyzed using the MSD® Multi-Spot Assay System, Rat Demonstration
7-Plex Ultra-sensitive Kit.

The lavage fluid from the four further lavages was pooled separately for each animal
in a centrifugation tube, and centrifuged at approximately 300 g for 10 min between
2 and 8 °C. The supernatant was discarded. The cell pellets were each suspended with
physiological saline and pooled with the suspended cell pellet from the first two
lavages. The total volume was gently shaken for mixing. Aliquots of the cell suspension
were taken for total cell count, cell viability and differential cell counting. A
total cell count was performed by diluting an aliquot of the cell suspension in Türk’s
stain (Diagnostik Merck No. 9277, Merck AG, Darmstadt, Germany). An aliquot of this
mixture was introduced into a hemocytometer chamber, and the cells were counted (at
least 1 mm
³
). A differential cell count was performed. According to the results of the total
cell count, an aliquot of cell suspension was diluted with physiological saline solution
to give an end concentration of at least 10
⁶
cells/ml. From this suspension a smear was prepared using a Cytocentrifuge (Shandon,
Instrument Gesellschaft AG, Switzerland) and stained with Diff-Quick (Baxter Dade,
Switzerland). At least 500 cells per smear were counted by light microscopy. The number
of each cell type (macrophages, neutrophils, lymphocytes, eosinophils, epithelial
cells, and other cells) was counted.

Macroscopic and organ weight examinations at necropsy

All animals were weighed at necropsy and descriptions of all macroscopic abnormalities
were recorded. The weights of adrenal glands, heart including auricles, kidneys, liver,
lungs, and spleen were determined in all animals exposed for 5 days and 90 days (including
recovery animals). In addition, weights of brain, epididymis, ovaries, testes, thymus
and uterus were recorded in all animals exposed for 90 days (including recovery animals).
The organ to terminal body weight ratios was calculated.

Histopathology

Almost all organs were collected from all rats exposed for 5-day (including recovery
animals). All gross lesions, all organs as per OECD test guideline no. 413 29], olfactory bulb and pleura were collected from all rats exposed for 90 days (including
recovery animals). Lung lobes were instilled via trachea with neutral phosphate buffered 4 % formaldehyde solution at approximately
30 cm H
₂
O pressure. The organs were fixed in formalin (excepted testes fixed in modified Davidson’s
solution), processed, embedded and cut at an approximate thickness of 2 – 4 ?m and
stained with hematoxylin and eosin.

For the 5-day exposure study, all gross lesions, heart including auricles, kidneys,
larynx, liver, spleen, mediastinal lymph nodes, nasal cavity, nasopharyngeal duct
and pharynx, spleen and trachea (adjacent to larynx and carina and bifurcation) of
the main and recovery rats from the control and 1.25 mg/m
³
groups were examined microscopically. Lungs from all main and recovery rats from all
groups were examined.

For the 90-day exposure study, the microscopic examination of the respiratory tract
and associated lymph nodes was performed on all control and exposed animals at both
sacrifice periods (except the nasal cavity and larynx of the recovery rats exposed
to 0.05 mg/m
³
). The other organs, as per OECD test guideline no. 413 29], were examined for the control and 5.0 mg/m
³
groups rats sacrificed 24 h post-exposure (this microscopic examination was not performed
on the 90-day recovery rats as no microscopic changes was observed in the 24-h sacrificed
rats).

A part of left lobe of lungs (also used for BALF), a caudal part of kidney and the
middle lobe of liver of male rats exposed for 90 days were preserved for the comet
assay.

Semiology and spermatid count

Sperm motility, spermatid count in testis and sperm count in cauda epididymis were
performed on all males exposed for 90 days and sacrificed 24 h and 90 days post exposure
(expect the sperm count in recovery males exposed to 0.05 and 0.25 mg/m
³
). Sperm morphology was analyzed on all males at the 24-h post exposure sacrifice
and on control and 5.0 mg/m
³
exposed males after the 90-day recovery period.

Micronucleus test

The 5 last male and female rats (in the numbering order) out of 10 of sacrificed 24 h
after the 90-day exposure were used for the micronucleus assay. The femora were removed,
the epiphyses cut off and the marrow was flushed out with fetal calf serum, using
a syringe (3 mL per femur). The nucleated cells were separated from the erythrocytes
using the method of Romagna 72]. Briefly, the cell suspensions were passed through a column consisting of ?-Cellulose
(Sigma) and Cellulose (Sigmacell type 50) (1:1 mixture). The columns were then washed
with Hank?s buffered saline. The cell suspensions were centrifuged at 1500 rpm (390
× g) for 10 min and the supernatant was discarded. The pellet was suspended in a small
drop of fetal calf serum and spread on slides. The smears were air-dried, fixed in
methanol. The slides were stained with May-Grünwald/Giemsa. For staining, the slides
were first incubated one minute in a 5 % May-Grünwald solution followed by one minute
in a 1:1 mixture of May-Grünwald/phosphate buffer (pH 7.4). After 20 min in 14 % Giemsa
solution the slides were washed twice in phosphate buffer and 10 s in deionized water.
Cover slips were mounted with EUKITT (Kindler, 79110 Freiburg, Germany). At least
one slide was made from each bone marrow sample. The slides were coded using a computer
generated coding list. Evaluation of the slides was performed using NIKON microscopes
with 100x oil immersion objectives. Six thousand polychromatic erythrocytes (PCE)
were analyzed per animal for micronuclei.

The test is considered to be positive if there is either a dose-related increase in
the number of micronucleated polychromatic erythrocytes or a statistically significant
positive response for at least one of the test points. In the absence of a dose-related
increase in the number of micronucleated polychromatic erythrocytes or a statistically
significant positive response at any of the test points the substance is considered
non-genotoxic in this system. To describe a cytotoxic effect the ratio between polychromatic
and normochromatic erythrocytes was determined in the same sample and expressed in
polychromatic erythrocytes per 2000 erythrocytes.

Comet assay

As suggested by the OECD test guideline no. 489 70], when there was no difference in toxicity between males and females, the comet assay
could be performed in either sex. The 5 first male rats (in the numbering order) out
of 10 sacrificed 24 h after the 90-day exposure were assigned for the comet assay.
Three to 6 h after the last treatment for positive control group or 22 to 26 h after
the last exposure for air control and Graphistrength
^©
C100 exposed groups, the lung (left lobe used for BALF), the kidney (caudal part)
and the liver (middle lobe) were collected from each animal and maintained in cold
mincing buffer (Mg
²⁺
and Ca
²⁺
free Hanks’ Balanced Salt Solution (Gibco, CA, USA) with 20 mM Na EDTA (EDTA) (Sigma,
USA) and 10 % (v/v) dimethylsulfoxide (DMSO) (Sigma)). Immediately after dissection,
the organs were minced using fine scissors in cold mincing buffer. The cell suspension
was stored on ice for 15-30 s to allow large clumps to settle, and the supernatant
was used to prepare comet slides. An aliquot of single cell suspension (3 x 10
⁴
cells) was mixed per 75 ?L of 0.5 % low melting agarose (Invitrogen, USA) and spread
on comet assay slides (Travigen, USA). The slides were immersed in cold lysis solution
(2.5 M NaCl, 100 mM Na
₂
EDTA, 10 mM Tris-base, 10 % DMSO, 1 % Triton-X (pH 10)) overnight. For the hOGG1-modified
comet assay, following lysis as described, slides were washed two times for 5 min
with the hOGG1 incubation buffer (40 mM Hepes, 100 mM KCl, 0.5 mM EDTA and 0.2 mg/mL
BSA) at room temperature and then incubated for 10 min at 37 °C with 0.12 x 10
^-3
U/slide of hOGG1 (Biolabs) in the hOGG1 incubation buffer. After this incubation period,
slides were placed in electrophoresis solution (0.3 M NaOH, 1 mM EDTA (pH??13)) for
20 min to allow for DNA unwinding. Subsequently, electrophoresis was conducted at
25 V, 300 mA for 20 min. The slides were immersed in a neutralization solution (0.4 M
Tris-base, pH 7.5) for at least 5 min and dehydrated with absolute ethanol to fix.
The cells were stained with propidium iodide (20 ?g/mL) (Invitrogen). All slides were
independently coded before the microscopic analysis. The comet was observed via fluorescence microscope (Leica Microsystems SAS – DM 2000, Heerbrugg, Switzerland)
at magnification of x200 and analyzed by COMET ASSAY IV software (Perceptive Instruments,
UK). For each sample (animal/tissue), fifty comets per slide were analyzed, with 3
slides scored per sample.

A positive response is defined as a statistically significant change in the % tail
DNA in at least one dose group in comparison with the vehicle control value. The positive
control should produce a statistically significant increase. Cytotoxicity was evaluated
by the enumeration of hedgehogs. Indeed, important fragmentation of the DNA can be
induced not only by genotoxicity but also during the process of cell death, i.e. apoptosis and necrosis.

Statistical analysis

The food consumption, blood pressure, grip strength, landing foot play, body temperature,
body weight, food consumption, blood clinical laboratory and BALF parameters, organ
weights, and sperm parameters were analyzed for statistical significance by the Dunnett-test
(many to one t-test) based on a pooled variance estimate, if the variables can be assumed to follow
a normal distribution for the comparison of the treated groups and the control groups
for each sex. The Steel-test (many-one rank test) was applied for the locomotor activity,
urinalysis, and BALF biochemical parameters instead of the Dunnett-test when the data
cannot be assumed to follow a normal distribution. Fisher’s exact test was applied
to the ophthalmoscopy and macroscopic findings. Nonparametric Mann-Whitney test was
applied to the micronucleus, comet and to cytokines test results as data did not follow
a normal distribution.