Novel miRNA biomarkers for genotoxicity screening in mouse

Article abstract of DOI:10.1016/j.tox.2018.05.009

The genotoxic potential of drugs is a serious problem, and its evaluation is one of the most critical processes of drug development. Although the comet assay of compound-exposed tissue is a frequently used genotoxicity test, its high false-positive rate is a major complication, and we consistently obtained false-positive results using the comet assay of mouse liver for nine hepatotoxic non-genotoxins (NGTXs). To identify novel genotoxin (GTX)-specific biomarkers, we screened the expression of 750 microRNAs (miRNAs) in the livers of mice treated with GTXs or NGTXs. Three miRNAs, miR-22-3p, miR-409-3p, and miR-543-3p, were significantly down-regulated in GTX-treated mouse liver. In contrast, these three miRNAs were significantly up-regulated in plasma. A discrimination model based on the expression levels of these biomarkers successfully identified GTXs and NGTXs. This novel biomarker expression-based discrimination model analysis using both liver and plasma is effective for detecting genotoxicity with high sensitivity and reliability to support drug development.

Full text of DOI:10.1016/j.tox.2018.05.009

Accepted Manuscript
Title: Novel miRNA biomarkers for genotoxicity screening in
mouse
Authors: Hiroyuki Oka, Koichi Masuno, Takeki Uehara, Toru
Okamoto, Yoshiharu Matsuura, Toru Nakano, Shinpei
Yamaguchi
PII:
DOI:
Reference:
S0300-483X(18)30085-4
https://doi.org/10.1016/j.tox.2018.05.009
TOX 52035
To appear in:
Toxicology
Received date:
Revised date:
Accepted date:
7-2-2018
8-5-2018
13-5-2018
Please cite this article as: Oka, Hiroyuki, Masuno, Koichi, Uehara, Takeki,
Okamoto, Toru, Matsuura, Yoshiharu, Nakano, Toru, Yamaguchi, Shinpei,
Novel miRNA biomarkers for genotoxicity screening in mouse.Toxicology
https://doi.org/10.1016/j.tox.2018.05.009
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Research Article
Novel miRNA biomarkers for genotoxicity screening in mouse
Hiroyuki Okaa,c, #, Koichi Masunoa, Takeki Ueharab, Toru Okamotod, Yoshiharu
Matsuurad, Toru Nakanoc, e, Shinpei Yamaguchic,f,#
aDrug Developmental Research Laboratories, Shionogi & Co., Ltd., 3-1-1 Futaba-cho,
Toyonaka, Osaka 561-0825, Japan, bGlobal Project Management, Shionogi & Co., Ltd.,
c
Kita-ku, Osaka, Japan. Department of Pathology, Medical School, Osaka University,
d
2-2 Yamada-oka Suita, Osaka 565-0871, Japan. Department of Molecular Virology,
Research Institute for Microbial Diseases, Osaka University, 3-1 Yamada-oka, Suita,
e
f
Osaka 565-0871, Japan. CREST and PRESTO, Japan Agency for Medical Research
and Development (AMED).
#To whom correspondence should be addressed
e-mail: hiroyuki.oka@shionogi.co.jp
e-mail: yamaguchi@patho.med.osaka-u.ac.jp
Abstract
The genotoxic potential of drugs is a serious problem, and its evaluation is one of the
most critical processes of drug development. Although the comet assay of

compound-exposed tissue is a frequently used genotoxicity test, its high false-positive
rate is a major complication, and we consistently obtained false-positive results using
the comet assay of mouse liver for nine hepatotoxic non-genotoxins (NGTXs). To
identify novel genotoxin (GTX)-specific biomarkers, we screened the expression of 750
microRNAs (miRNAs) in the livers of mice treated with GTXs or NGTXs. Three
miRNAs, miR-22-3p, miR-409-3p, and miR-543-3p, were significantly down-regulated
in GTX-treated mouse liver. In contrast, these three miRNAs were significantly
up-regulated in plasma. A discrimination model based on the expression levels of these
biomarkers successfully identified GTXs and NGTXs. This novel biomarker
expression-based discrimination model analysis using both liver and plasma is effective
for detecting genotoxicity with high sensitivity and reliability to support drug
development.
Keywords Comet assay Genotoxicity MicroRNA Mouse Toxicogenomics
Introduction
Rigorous toxicological evaluation is critical for the development of drugs and cosmetics.
Genotoxicity screening is particularly important, because genotoxicity affects genetic

fidelity and can cause various diseases such as cancer, immune deficiency, and
neurodegeneration (Hoeijmakers, 2009; Jackson and Bartek, 2009). The safety guideline
of the International Council for Harmonisation of Technical Requirements for
Pharmaceuticals for Human Use (ICH, 2011) recommend to perform multiple tests,
such as bacterial reverse mutation test and in vivo genotoxicity test(s), and judge the
genotoxic potential by taking all results into consideration. The comet assay (also
referred to as single-cell gel electrophoresis) is one of the genotoxicity test to detect
DNA damage, including single-strand breaks and alkali-labile sites, in individual cells
(Ostling and Johanson 1984; Singh et al. 1988). In this assay, cells are embedded in an
agarose gel, lysed, and subjected to electrophoresis. The tail intensity represents the
magnitude of DNA damage of each cell, where fragmented DNA migrates farther,
forming a tail. The alkaline comet assay of rodent tissues is widely used because it is
sensitive, relatively inexpensive and easy to perform, and is one of the recommended
methods in the guideline to detect DNA damage.
However, DNA migration in the comet assay is sometimes increased in tissues exposed
to compounds with no clear genotoxic potential, yielding false-positive results. This can
be caused by DNA fragmentation induced by apoptosis, necrosis, or inflammation of
tissue exposed to cytotoxic compounds (Choucroun et al., 2001; Hartmann et al., 2003;

Jaiswal et al., 2000; Sasaki et al., 1998; Shimoda et al., 1994). As such, additional
cytotoxicity tests of comet assay-positive compounds are usually required. However,
false-positive results cannot be eliminated in cases where compounds retain both
genotoxic and cytotoxic potentials (OECD, 2014). Therefore, a novel deterministic
indicator of genotoxicity would be useful for improving the efficiency of drug
development.
microRNAs (miRNAs) are small non-coding RNA molecules about 22 nucleotides in
length. The major role of miRNA is post-transcriptional silencing of multiple targets via
either degradation or translational inhibition of mRNA. Since miRNAs are highly
sensitive to cellular and physiological perturbation, they are used widely as a diagnostic
biomarker for cancer, cardiovascular diseases, and psychiatric disorders (Denk et al.,
2015; Lan et al., 2015; Mihelich et al., 2015; Min and Chan, 2015; Mitchell et al., 2008;
Wang et al., 2016). One notable advantage of miRNA as a biomarker is its stability,
enabling its detection in body fluids such as peripheral blood, which is the best material
for diagnosis and animal models (Farid et al., 2012; Mitchell et al., 2008; Nishimura et
al., 2015; Wang et al., 2009).
miRNAs have been reported to have an important role in the repair process after DNA
damage (Chowdhury et al., 2013; Francia et al., 2012; Luan et al., 2014). For example,

the miR-34 family, a downstream factor of p53, is activated after DNA damage and
regulates the cell cycle and apoptosis (He et al., 2007; Kato et al., 2009; Zhou et al.,
2014). It has been reported that miR-21, miR-24, and miR-182 regulate the DNA repair
factors MSH2, H2AX, and BRCA1, respectively (He et al., 2007; Moskwa et al., 2011;
Yu et al., 2010). These reports suggest that miRNA expression analysis may be a
promising method to detect genotoxicity. In this study, we searched for miRNA
biomarkers that could detect the genotoxicity of compounds in vivo. Through miRNA
expression profiling of mice treated with genotoxins (GTXs) or non-genotoxins
(NGTXs), we identified miR-22-3p, miR-409-3p, and miR-543-3p, which were
significantly down-regulated in GTX-treated liver and up-regulated in plasma.
Materials and methods
2.1.
Chemicals
Table 1 summarizes the information of the chemicals used in this study, including their
names, classes, and dosing methods. Representative GTXs were chosen from each
genotoxicity category, and typical hepatotoxic compounds and carcinogens were
selected as NGTXs. The dose was set at either a sufficient dose to induce
genotoxicity/hepatotoxicity in mouse, or one half the median lethal dose (LD50), with an

upper limit of 2,000 mg/kg/day.
2.2. Animal treatment
Seven-week-old male Crl:CD1 (ICR) mice were purchased from Charles River
Laboratories Japan Inc. (Kanagawa, Japan). The animals were acclimated for 1 week
before use under the following conditions. Mice were kept in plastic cages and provided
pelleted food and water ad libitum in an animal room maintained under controlled
conditions (12 h light, 12 h dark; 23 ± 3°C; 50 ± 20% humidity). All compounds and
their vehicles (shown in Table 1) were administered to three to six mice, three times (at
0, 24, and 45 h). Three hours after the last administration, mice were euthanized by
cervical dislocation and the left lobe of the liver was harvested for analysis. For plasma
collection, mice were anesthetized with isoflurane and blood was collected in an
EDTA-containing Venoject tube (Terumo, Tokyo, Japan). All animals were treated
according to the ethical guidelines for the use and care of laboratory animals established
by Shionogi & Co., Ltd.
2.3.
Pathological examination
Liver was collected and fixed in 10% buffered formalin and embedded in paraffin. Each
section was cut into 3-μm-thick segments, stained with hematoxylin and eosin,
examined under a light microscope, and graded for toxicologically significant findings

(necrosis, inflammation, and mitosis). Lesions were classified into the following five
categories according to morphological severity: 0, normal; 1, minimal; 2, mild; 3,
moderate; and 4, marked.
2.4.
TUNEL assay
Sections were deparaffinized and apoptotic nuclei were stained using the In Situ Cell
Death Detection Kit, Fluorescein (Roche Diagnostics, Basel, Switzerland) according to
the manufacturer’s protocol. Nuclei were visualized under a fluorescence microscope
and analyzed with ImageJ (National Institutes of Health, Bethesda, MD, USA). More
than 1,000 nuclei were analyzed per animal.
2.5.
Comet assay
The alkaline comet assay was conducted as reported previously (JaCVAM, 2009; Recio
et al., 2010; Singh et al., 1988). Briefly, a portion of liver in cold mincing buffer (Mg2+,
Ca2+, and phenol-red-free Hank’s balanced salt solution containing 20 mM Na2EDTA
and 10% (v/v) dimethyl sulfoxide, pH 7.5) was cut into small pieces with a pair of
scissors. The cell suspension was filtered through a nylon mesh (pore size: 40 μm). The
filtered cell suspension was mixed with low-melting-point agarose (final concentration:
0.45% (w/v)) and spread on a glass slide printed with highly water-repellent mark
(Matsunami Glass Ind., Ltd., Osaka, Japan). All slides were immersed in cold lysis

solution
(100
mM
Na2EDTA,
2.5
M
NaCl,
and
10
mM
tris(hydroxymethyl)aminomethane containing 1% (v/v) Triton X-100 and 10% (v/v)
DMSO, pH 10.0) overnight at 4°C. Two slides per sample were incubated in alkaline
solution (300 mM sodium hydroxide and 1 mM Na2EDTA, pH > 13) for 20 min at 4°C,
followed by electrophoresis in the same buffer for 20 min at 0.7 V/cm with a constant
voltage of approximately 300 mA. After electrophoresis, the slides were immersed in
neutralization buffer (0.4 M tris(hydroxymethyl)aminomethane, pH 7.5) for 5 min. The
slides were dehydrated by immersion into absolute ethanol for 5 min and air-dried at
room temperature. DNA was stained with SYBR Gold, observed under a fluorescence
microscope, and analyzed with the Comet Assay IV (Perceptive Instruments Ltd.,
Suffolk, UK). Two replicate slides were prepared for each animal, and 50 randomly
chosen cells on each slide were analyzed. The tail intensity of each sample was
calculated as the mean of the median values of the slides.
2.6.
miRNA extraction
miRNA was isolated using the miRNeasy Kit (Qiagen, Hilden, Germany) according to
the manufacturer’s protocol. For plasma, the miRNeasy Serum/Plasma Kit was used and
samples were spiked with synthetic Caenorhabditis elegans miRNA (cel-miR-39).

2.7.
miRNA analysis by PCR array
In total, 150 ng of pooled miRNA of liver samples from each administration group was
converted into cDNA using Megaplex RT Primers, Rodent Pool Set v3.0, and the
TaqMan MicroRNA Reverse Transcription Kit (Thermo Fisher Scientific Inc., Waltham,
MA, USA). The cDNA was mixed with TaqMan Universal PCR Master Mix, No
AmpErase UNG (Thermo Fisher Scientific), and PCR arrays were performed using
TaqMan Array Rodent MicroRNA A+B Cards Set v3.0 and a 7900HT Fast Real-Time
PCR System (Thermo Fisher Scientific). Expression levels were normalized to
snoRNA202.
2.8.
Reverse transcription quantitative real-time PCR analysis of individual miRNA
miRNA expression was analyzed by reverse transcription quantitative real-time PCR
(RT-qPCR) using the TaqMan MicroRNA Reverse Transcription Kit and TaqMan
MicroRNA Assay. Expression levels were normalized to snoRNA202 (liver) or
cel-miR-39 (plasma).
2.9.
Statistical analysis
The comet assay data were statically analyzed by Bonferroni-adjusted Student’s t-test
(for the comparison of paired data) or Tukey’s method (for the comparison of multiple
data). The RT-qPCR data were statically analyzed by Bonferroni-adjusted Welch’s t-test.

GraphPad Prism (ver. 6.0; GraphPad Software, Inc., San Diego, CA, USA) and R
software (ver. 3.3.2; R Foundation for Statistical Computing, Vienna, Austria) were used
for the general statistical analysis. Logistic regression model was constructed by R
software using glm function with the training set (Table S1). Parameters were set as
defaults except for the family parameter (set as binomial). Validation of the model was
performed using the test set shown in Table S2.
Results
3.1.
Comet assay of mouse liver
First, we evaluated the sensitivity and specificity of the comet assay to detect the
genotoxic potential of compounds. Mouse liver samples were harvested after
administration of nine representative GTXs (2-acetamidofluorene (2AAF),
N-nitrosodiethylamine (DEN), N-nitrosodimethylamine (DMN), ethyl methanesulfonate
(EMS), N-ethyl-N-nitrosourea (ENU), cisplatin (CDDP), mitomycin C (MMC),
methotrexate (MTX) and teniposide (TNP)) and nine representative NGTX
hepatotoxicants (1-naphthyl isothiocyanate (ANIT), acetaminophen (APAP), carbon
tetrachloride (CCl4), bis(2-ethylhexyl) phthalate (DEHP), diclofenac (DFNa),
D(+)-galactosamine hydrochloride (GalN), phenobarbital (PB), phenytoin (PHT), and

thioacetamide (TAA)) (Table 1). The tail intensity significantly increased for five of the
GTXs (2AAF, DEN, DMN, EMS, and ENU) and all nine NGTXs (Fig. 1A). The
increases in tail intensity for the NGTXs were assumed to be false positives, because no
detectable genotoxicity has been reported for any of these compounds. Several of the
false-positive results were consistent with previous reports (Erenberk et al., 2014;
Oshida et al., 2008; Parolini et al., 2009; Sasaki et al., 1998). Histopathological tests are
recommended during the drug development process when an increase in DNA migration
is detected by the comet assay (OECD, 2014). However, it is uncertain whether
histopathological findings correlate with tail intensity and are useful for identifying
false-positive results. To test this, livers showing positive results in the comet assay
were subjected to histopathological analysis (Fig. 1B). Although grades 1 to 4 necrosis
were observed in the mouse livers treated with some of the GTXs (DEN and DMN) and
NGTXs (ANIT, CCl4, GalN, and TAA), no clear correlation between the grade and tail
intensity was detected. In addition, the livers treated with some NGTXs (e.g., APAP,
DFNa, and PHT) showed a high tail intensity but no detectable necrosis. Similarly, no
clear correlation between tail intensity and inflammation grade, mitosis grade, or
apoptosis rate was observed. These results indicate that histopathological tests are
inappropriate for judging false-positive results of the comet assay. Therefore, alternative

tools such as GTX-specific biomarkers with higher sensitivity and accuracy are required
to detect drug genotoxicity.
3.2.
Screening for miRNAs with altered expression in GTX-treated mouse liver
miRNA expression changes were analyzed by PCR array using mouse liver after
administration of four GTXs (CDDP, EMS, ENU, and MMC) or two NGTXs (APAP
and DFNa). miRNA expression levels decreased globally in all GTX-treated livers, but
no such tendency was observed in NGTX-treated livers among 403 miRNAs that were
highly expressed in control liver (control liver Ct value: < 35) (Fig. 2A and 2B).
In total, 84 miRNAs were categorized as GTX-specific down-regulated genes based on
the following evidence. First, three or four GTX-treated livers showed down-regulation
compared to the control (fold change < -1.5). Second, such down-regulation was not
observed in NGTX-treated livers. Moreover, no miRNAs were up-regulated based on a
similar criterion (fold change > 1.5) (Fig. 2C and 2D, Table S3). Cancer-related genes
were enriched in 84 candidates through gene ontology analysis using ingenuity pathway
analysis. Among them, we selected 13 miRNAs reportedly associated with DNA
damage repair pathways or differentially expressed in cancer as GTX-specific
biomarker candidates.

The GTX-specific expression change of biomarker candidates was validated using
RT-qPCR analysis of mouse liver after the administration of four GTXs (CDDP, EMS,
ENU, and MMC) or four NGTXs (APAP, DFNa, PHT, and TAA), and ΔΔCt values
were analyzed by principal component analysis (PCA) (Fig. S1, Fig. 2E). GTXs and
NGTXs were not well-separated in PC 1 and 3, indicating that some unknown factors
other than genotoxicity would be presumably responsible for these PCs. In contrast, the
effects of GTX and NGTX showed the greatest separation based on PC2. Three
miRNAs, miR-22-3p, miR-409-3p, and miR-543-3p, showed particularly high absolute
loading values for PC2, as well as significant down-regulation in GTX-administrated
mouse liver (Fig. 2F, Fig. S1). In contrast, these three miRNAs were up-regulated in
NGTX-treated mouse liver (Fig. S1). Based on these results, we concluded that
miR-22-3p, miR-409-3p, and miR-543-3p showed potential as genotoxicity biomarkers
and constructed a discriminant model based on logistic regression analysis using the
ΔΔCt values of these miRNAs (Table S1). The developed logistic regression equation
was: Logit (p) = 4.4634 * (miR-22-3p) + 2.4673 * (miR-409-3p) + 0.6958 *
(miR-543-3p) - 1.722.
Next, in order to validate the discrimination model, we administrated another five GTXs
(2AAF, DEN, DMN, MTX, and TNP) or five NGTXs (ANIT, CCl4, DEHP, GalN, and

PB) and performed RT-qPCR analysis. Down-regulation of biomarker miRNAs, except
miR-22-3p for TNP and miR-543-3p for 2AAF and DMN, was observed in all
GTX-treated mouse livers (Fig. 3A, Table S2). The combined ΔΔCt values of the three
biomarkers showed drastic differences between the GTX- and NGTX-treated livers (Fig.
3B, Table S2). The constructed logistic model discriminated all GTXs and NGTXs
consistently with their previous classifications, except for one NGTX, DEHP (Fig. 3B).
These results were further confirmed by receiver operating characteristic (ROC)
analysis using training and test sets. The highest area under the ROC curve (AUC) was
obtained when using the three biomarkers in combination (Fig. 3C).
3.3.
Genotoxicity of DEHP in mouse liver
Previous studies have classified DEHP as NGTX based on reverse mutation test in
bacteria and chromosome aberration test in mammalian cells (Douglas et al., 1986;
Melis et al., 2014; Phillips et al., 1982; Tomita et al., 1982; Watanabe et al., 2012;
Yoshikawa et al., 1983). However, the discriminant model based on miRNA biomarker
expression change in liver raised the possibility that DEHP could be genotoxic (Fig. 3B).
DEHP is hydrolyzed in vivo into mono(2-ethylhexyl) phthalate (MEHP) by lipases, and
MEHP is genotoxic (Albro and Thomas, 1973; Lhuguenot and Cornu, 1993; Phillips

et al., 1982; Pollack et al., 1985; Sjoberg et al., 1985; Tomita et al., 1982) (Fig 4A).
Lipase activity has been reported to be particularly high in mouse small intestine, kidney,
and liver (Ito et al., 2005). Therefore, it is conceivable that the increased tail intensity in
the comet assay and biomarker-based genotoxic prediction in DEHP-administered
mouse liver was due to the genotoxic metabolite of DEHP, MEHP. To verify this, DEHP
was administered simultaneously with a lipase inhibitor, cetilistat. When cetilistat was
concomitantly administrated with DEHP, no increase in tail intensity was observed (Fig.
4B). Similarly, the expression of the three miRNAs in liver treated with DEHP and
cetilistat exhibited an NGTX pattern in the discriminant model (Fig. 4C and 4D). These
data strongly suggest that the genotoxicity of DEHP is due to its metabolite, MEHP. It is
notable that the proposed biomarker-based discrimination model was useful for
identifying this.
3.4.
Changes in biomarker miRNA expression in plasma
Next, we examined the expression change of the three miRNA biomarkers for GTX
detection in mouse plasma after administering GTXs or NGTXs. In contrast to
down-regulation in the liver, expression was significantly increased in plasma after
GTX administration (Fig. 5A and 5B, Table S4). We performed PCA to evaluate the

utility of the combined analysis of miRNA expression in liver and blood (Fig. 5C).
Although the GTXs and NGTXs were separated in PC1 based on the individual
expression analyses of liver and plasma, the distances of some compounds (e.g., DMN
and GalN in liver and EMS and TAA in plasma) were small. In contrast, the distances
between GTXs and NGTXs were much greater after applying the combined biomarker
expressions from blood and liver. Therefore, we conclude that the combined analysis of
these biomarkers in plasma and liver would be useful for the identification of
genotoxicity in vivo.
Discussion
Both GTXs and NGTXs (e.g., cytotoxic chemicals) can cause increased DNA migration
in the comet assay (Choucroun et al., 2001; Erenberk et al., 2014; Hartmann et al.,
2003). The main cause of such false-positive results of NGTXs is DNA fragmentation
by generated reactive oxidative species, necrosis, or apoptosis due to cytotoxicity (Fang
et al., 2015; Ghaleb A. Husseini, Kim L. O’Neill, 2006). In this study, we confirmed
that hepatotoxic chemicals caused false-positive results in liver. Although
histopathological examination is often used to check whether the positive results of the
comet assay are caused by secondary effects of cytotoxicity, it did not provide a

definitive index to detect false-positive results in this study. Next, we examined whether
the expression changes of miRNAs could detect the genotoxic potential of compounds,
and identified miR-22-3p, miR-409-3p, and miR-543-3p as effective biomarkers.
Some miRNAs have important roles in cellular responses and DNA repair after damage
(Chowdhury et al., 2013; Francia et al., 2012; Luan et al., 2014; Wan et al., 2011; Zhang
et al., 2011). The three genotoxic biomarker miRNAs identified in this study have been
reported to have roles in the cellular response after DNA damage. For example, miR-22
suppresses the expression of the mediator of DNA damage checkpoint protein 1
(MDC1) gene, and knockdown of miR-22 promoted DNA double-stranded break repair
(Lee et al., 2017, 2015). Meanwhile, miR-409 is predicted to target mutS homolog 6
gene (MSH6), a critical factor for DNA mismatch repair (Alani, 1996; Fishel and
Wilson, 1997; Tulay and Sengupta, 2016). In addition, miR-543 directly binds and
suppresses the translation of aminoacyl-tRNA synthetase-interacting multifunctional
protein 3 (AIMP3/p18), the tumor suppressor gene controlling p53 (Lee et al., 2014;
Park et al., 2006). Decreases in miR-543 and increases in AIMP3/p18 have been
observed in MMC-treated human mesenchymal stem cells, suggesting the potential role
of miR-543 in the DNA damage response. Since down-regulation of miR-22, miR-409,
and miR-543 is likely to promote the DNA repair pathway via the up-regulation of

target genes, decreases in these miRNAs by GTXs might represent a DNA repair
process in liver. The physiological significance and underlying mechanism of the
down-regulation of these miRNAs should be clarified in future studies.
We found that several known GTXs (CDDP, MMC, MTX, and TNP) did not show a
significant increase in tail intensity in the comet assay using mouse liver (Fig. 1A).
These compounds have been shown to possess both mutagenicity and DNA damage
potential in in vitro studies (CDDP (Benedicl et al., 1977; Hashem and Sinden, 2002;
Sakai et al., 2014), MMC (Hashem and Sinden, 2002; Sakai et al., 2014; Sofuni et al.,
1996), MTX (Benedict et al., 1977; Sofuni et al., 1996) , TNP (“Teniposide Injection
Package Insert,” 2015)). Since DNA–DNA or DNA–protein cross-link formation
reduces the migration of fragmented DNA in the comet assay (Merk and Speit, 1999),
the negative results of CDDP, MMC, and TNP are likely caused by such cross-linking
effects. Therefore the comet assay is inappropriate for the detection of cross-linker.
Meanwhile, MTX is a nucleotide synthesis inhibitor. While DNA fragmentation was
increased in MTX-treated mice, statistical significance was not obtained due to the large
variation among samples. Although the comet assay failed to detect the genotoxicity of
these compounds, the novel miRNA biomarkers identified in this study were sensitive
enough to detect their genotoxic potential (Fig. 3A).

The expression levels of the three biomarker miRNAs were significantly increased in
plasma after GTX administration, in contrast to the decreased expression in liver (Fig. 5,
Fig. 6). Some toxic compounds and diseases have been reported to induce opposing
effects on miRNA expression between tissue and blood (Matamala et al., 2015; Wang et
al., 2009; Zhu et al., 2014). For example, decreased and increased miR-122 expression
has been reported in liver and plasma, respectively, after hepatic injury (Wang et al.,
2009). miR-122 is likely packed in exosomes and released from liver cells into the
blood in response to liver injury (Bala et al., 2012; Farid et al., 2012). Since miR-122
has a role in cell cycle regulation, active excretion of miR-122 out of hepatocytes might
be important for controlling the cell cycle for liver regeneration (Farid et al., 2012). The
three biomarker miRNAs identified in this report may also be secreted from liver cells
into the blood, regulating the DNA damage response of liver cells. However, further
studies are required to clarify the physiological significance of this hypothesis.
In conclusion, we found that the genotoxicity of several compounds could be identified
through the expression analysis of three miRNAs, miR-22-3p, miR-409-3p, and
miR-543-3p, in both liver and plasma, and combined analysis of their changes further
strengthened the reliability of this method. On the other hand, we must admit that this
miRNA method has a few downside compared with the comet assay. One is that this

method is more expensive than comet assay and the other is that underlying mechanism
of the down regulation of these miRNAs is unclear. Although further validation would
be necessary, our findings will contribute to improve the genotoxicity tests to reduce the
number of experimental animals required for drug development.
Conflict of interest
The authors declare no conflict of interest.
Acknowledgments
We thank Dr. Tamio Fukushima and Hirofumi Miyajima for their support in this
research.

References
Alani, E., 1996. The Saccharomyces cerevisiae Msh2 and Msh6 proteins form a
complex that specifically binds to duplex oligonucleotides containing mismatched DNA
base pairs. Mol. Cell. Biol. 16, 5604–15.
Albro, P.W., Thomas, R.O., 1973. Enzymatic hydrolysis of di-(2-ethylhexyl) phthalate
by
lipases.
Biochim.
Biophys.
Acta
306,
380–90.
doi:https://doi.org/10.1016/0005-2760(73)90176-8
Bala, S., Petrasek, J., Mundkur, S., Catalano, D., Levin, I., Ward, J., Alao, H., Kodys,
K., Szabo, G., 2012. Circulating microRNAs in exosomes indicate hepatocyte injury
and inflammation in alcoholic, drug-induced, and inflammatory liver diseases.
Hepatology 56, 1946–1957. doi:10.1002/hep.25873
Benedicl, W.F., Baker, M.S., Ames, B.N., 1977. Mutagenicity of Cancer
Chemotherapeutic Agents in the Salmonella/Microsome Test. Cancer Res. 37, 2209–
2213.
Benedict, W.F., Banerjee, A., Gardner, A., Jones, P.A., 1977. Induction of
Morphological Transformation in Mouse C3H/10T1/2 Clone 8 Cells and Chromosomal
Damage in Hamster A(T1)C1-3 Cells by Cancer Chemotherapeutic Agents. Cancer Res.
37, 2202–2208.

Choucroun, P., Gillet, D., Dorange, G., Sawicki, B., Dewitte, J.., 2001. Comet assay and
early apoptosis. Mutat. Res. Mol. Mech. Mutagen. 478, 89–96.
doi:10.1016/S0027-5107(01)00123-3
Chowdhury, D., Choi, Y.E., Brault, M.E., 2013. Charity begins at home: non-coding
RNA functions in DNA repair. Nat. Rev. Mol. Cell Biol. 14, 181–9.
doi:10.1038/nrm3523
Denk, J., Boelmans, K., Siegismund, C., Lassner, D., Arlt, S., Jahn, H., 2015.
MicroRNA Profiling of CSF Reveals Potential Biomarkers to Detect Alzheimer`s
Disease. PLoS One 10, e0126423. doi:10.1371/journal.pone.0126423
Douglas, G.R., Hugenholtz, A.P., Blakey, D.H., 1986. Genetic toxicology of phthalate
esters: mutagenic and other genotoxic effects. Environ. Health Perspect. 65, 255–62.
Erenberk, U., Dundaroz, R., Gok, O., Uysal, O., Agus, S., Yuksel, A., Yilmaz, B., Kilic,
U., 2014. Melatonin attenuates phenytoin sodium-induced DNA damage. Drug Chem.
Toxicol. 37, 233–239. doi:10.3109/01480545.2013.838777
Fang, L., Neutzner, A., Turtschi, S., Flammer, J., Mozaffarieh, M., 2015. Comet Assay
as an Indirect Measure of Systemic Oxidative Stress. J. Vis. Exp. e52763.
doi:10.3791/52763
Farid, W.R.R., Pan, Q., Van Der Meer, A.J.P., De Ruiter, P.E., Ramakrishnaiah, V., De

Jonge, J., Kwekkeboom, J., Janssen, H.L.A., Metselaar, H.J., Tilanus, H.W., Kazemier,
G., Van Der Laan, L.J.W., 2012. Hepatocyte-derived microRNAs as serum biomarkers
of hepatic injury and rejection after liver transplantation. Liver Transplant. 18, 290–297.
doi:10.1002/lt.22438
Fishel, R., Wilson, T., 1997. MutS homologs in mammalian cells. Curr. Opin. Genet.
Dev. doi:10.1016/S0959-437X(97)80117-7
Francia, S., Michelini, F., Saxena, A., Tang, D., de Hoon, M., Anelli, V., Mione, M.,
Carninci, P., d’Adda di Fagagna, F., 2012. Site-specific DICER and DROSHA RNA
products
control
the
DNA-damage
response.
Nature
488,
231–5.
doi:10.1038/nature11179
Ghaleb A. Husseini, Kim L. O’Neill, and W.G.P., 2006. The Comet Assay to
Determine the Mode of Cell Death for the Ultrasonic Delivery of Doxorubicin to
Human Leukemia (HL-60 cells) from Pluronic P105 Micelles. Technol. Cancer Res.
Treat. 4, 707–711. doi:10.1177/153303460500400616
Hartmann, A., Agurell, E., Beevers, C., Brendler-Schwaab, S., Burlinson, B., Clay, P.,
Collins, A., Smith, A., Speit, G., Thybaud, V., Tice, R.R., 2003. Recommendations for
conducting the in vivo alkaline Comet assay. Mutagenesis 18, 45–51.
doi:10.1093/mutage/18.1.45

Hashem, V.I., Sinden, R.R., 2002. Chemotherapeutically induced deletion of expanded
triplet repeats. Mutat. Res. - Fundam. Mol. Mech. Mutagen. 508, 107–119.
doi:10.1016/S0027-5107(02)00190-2
He, L., He, X., Lim, L.P., De Stanchina, E., Xuan, Z., Liang, Y., Xue, W., Zender, L.,
Magnus, J., Ridzon, D., Jackson, A.L., Linsley, P.S., Chen, C., Lowe, S.W., Cleary,
M.A., Hannon, G.J., 2007. A microRNA component of the p53 tumour suppressor
network. Nature 447, 1130–1134. doi:10.1038/nature05939
Hoeijmakers, J.H.J., 2009. DNA Damage, Aging, and Cancer. N. Engl. J. Med. 361,
1475–1485. doi:10.1056/NEJMra0804615
ICH, 2011. ICH harmonized tripartite guideline S2(R1): Guidance on Genotoxicity
Testing and Data Interpretation for Pharmaceuticals Intended for Human Use.
Ito, Y., Yokota, H., Wang, R., Yamanoshita, O., Ichihara, G., Wang, H., Kurata, Y.,
Takagi, K., Nakajima, T., 2005. Species differences in the metabolism of
di(2-ethylhexyl) phthalate (DEHP) in several organs of mice, rats, and marmosets. Arch.
Toxicol. 79, 147–54. doi:10.1007/s00204-004-0615-7
Jackson, S.P., Bartek, J., 2009. The DNA-damage response in human biology and
disease. Nature 461, 1071–1078. doi:10.1038/nature08467
JaCVAM (Japanese Center for the Validation of Alternative Methods), 2009.

International validation of the in vivo alkaline comet assay for the detection of
genotoxic carcinogens version 14.2.
http://www.mms-j.org/wp_files/wp-content/upLoads//Protocol_V14.2_final1.pdf.
Jaiswal, M., Larusso, N.F., Burgart, L.J., 2000. Inflammatory Cytokines Induce DNA
damage and Inhibit DNA repair in Cholangiocarcinoma Cells by a Nitric
Oxide-dependent Mechanism Inflammatory Cytokines Induce DNA damage and Inhibit
DNA repair in Cholangiocarcinoma Cells by a Nitric Oxide-dependent Mecha. Cancer
Res 60, 184–190.
Kato, M., Paranjape, T., Ullrich, R., Nallur, S., Gillespie, E., Keane, K.,
Esquela-Kerscher, A., Weidhaas, J.B., Slack, F.J., 2009. The mir-34 microRNA is
required for the DNA damage response in vivo in C. elegans and in vitro in human
breast cancer cells. Oncogene 28, 2419–2424. doi:10.1038/onc.2009.106
Lan, H., Lu, H., Wang, X., Jin, H., 2015. MicroRNAs as potential biomarkers in cancer:
Opportunities and challenges. Biomed Res. Int. doi:10.1155/2015/125094
Lee, J.-H., Park, S.-J., Jeong, S.-Y., Kim, M.-J., Jun, S., Lee, H.-S., Chang, I.-Y., Lim,
S.-C., Yoon, S.P., Yong, J., You, H.J., 2015. MicroRNA-22 Suppresses DNA Repair
and Promotes Genomic Instability through Targeting of MDC1. Cancer Res. 75, 1298–
310. doi:10.1158/0008-5472.CAN-14-2783

Lee, J.-H., Park, S.-J., Won Kim, S., Hariharasudhan, G., Jung, S.-M., Jun, S., Yong, J.,
Jin You, H., 2017. c-Fos-dependent miR-22 targets MDC1 and regulates DNA repair in
terminally differentiated cells. Oncotarget. doi:10.18632/oncotarget.18389
Lee, S., Yu, K.R., Ryu, Y.S., Oh, Y.S. un, Hong, I.S., Kim, H.S., Lee, J.Y. oung, Kim,
S., Seo, K.W., Kang, K.S., 2014. miR-543 and miR-590-3p regulate human
mesenchymal stem cell aging via direct targeting of AIMP3/p18. Age (Dordr). 36, 9724.
doi:10.1007/s11357-014-9724-2
Lhuguenot, J.C., Cornu, M.C., 1993. Metabolism of di (2-ethylhexyl)-phthalate (DEHP)
and di (2-ethylhexyl) adipate (DEHA) and their relationship to peroxisome proliferation
in different species, in: Peroxisomes: Biology and Importance in Toxicology and
Medicine. pp. 465–485.
Luan, Y., Qi, X., Xu, L., Ren, J., Chen, T., 2014. Absence of mature microRNAs
inactivates the response of gene expression to carcinogenesis induced by
N-ethyl-N-nitrosourea in mouse liver. J. Appl. Toxicol. doi:10.1002/jat.2973
Matamala, N., Vargas, M.T., González-Cámpora, R., Miñambres, R., Arias, J.,
Menéndez, P., Andrés-León, E., Mez-López, G.G., Yanowsky, K., Calvete-Candenas, J.,
Inglada-Pérez, L., Martínez-Delgado, B., Benítez, J., 2015. Tumor MicroRNA
expression profiling identifies circulating MicroRNAs for early breast cancer detection.

Clin. Chem. 61, 1098–1106. doi:10.1373/clinchem.2015.238691
Melis, J.P.M., Derks, K.W.J., Pronk, T.E., Wackers, P., Schaap, M.M., Zwart, E., van
Ijcken, W.F.J., Jonker, M.J., Breit, T.M., Pothof, J., van Steeg, H., Luijten, M., 2014. In
vivo murine hepatic microRNA and mRNA expression signatures predicting the
(non-)genotoxic carcinogenic potential of chemicals. Arch. Toxicol. 1–12.
doi:10.1007/s00204-013-1189-z
Merk, O., Speit, G., 1999. Detection of crosslinks with the comet assay in relationship
to genotoxicity and cytotoxicity. Environ. Mol. Mutagen. 33, 167–172.
doi:10.1002/(SICI)1098-2280(1999)33:2<167::AID-EM9>3.0.CO;2-D
Mihelich, B.L., Maranville, J.C., Nolley, R., Peehl, D.M., Nonn, L., 2015. Elevated
serum microRNA levels associate with absence of high-grade prostate cancer in a
retrospective cohort. PLoS One 10, e0124245. doi:10.1371/journal.pone.0124245
Min, P.-K., Chan, S.Y., 2015. The biology of circulating microRNAs in cardiovascular
disease. Eur. J. Clin. Invest. 45, 860–874. doi:10.1111/eci.12475
Mitchell, P.S., Parkin, R.K., Kroh, E.M., Fritz, B.R., Wyman, S.K.,
Pogosova-Agadjanyan, E.L., Peterson, A., Noteboom, J., O’Briant, K.C., Allen, A., Lin,
D.W., Urban, N., Drescher, C.W., Knudsen, B.S., Stirewalt, D.L., Gentleman, R.,
Vessella, R.L., Nelson, P.S., Martin, D.B., Tewari, M., 2008. Circulating microRNAs as

stable blood-based markers for cancer detection. Proc. Natl. Acad. Sci. U. S. A. 105,
10513–8. doi:10.1073/pnas.0804549105
Moskwa, P., Buffa, F.M., Pan, Y., Panchakshari, R., Gottipati, P., Muschel, R.J., Beech,
J., Kulshrestha, R., Abdelmohsen, K., Weinstock, D.M., Gorospe, M., Harris, A.L.,
Helleday, T., Chowdhury, D., 2011. MiR-182-Mediated Downregulation of BRCA1
Impacts DNA Repair and Sensitivity to PARP Inhibitors. Mol. Cell 41, 210–220.
doi:10.1016/j.molcel.2010.12.005
Nishimura, Y., Kondo, C., Morikawa, Y., Tonomura, Y., Torii, M., Yamate, J., Uehara,
T., 2015. Plasma miR-208 as a useful biomarker for drug-induced cardiotoxicity in rats.
J. Appl. Toxicol. 35, 173–180. doi:10.1002/jat.3044
OECD, 2014. Report of the JaCVAM initiative international validation of the in vivo
rodent alkaline ocmet assay for the detection of genotoxic carcinogens. No. 196. OECD
Publ. Paris.
Oshida, K., Iwanaga, E., Miyamoto-Kuramitsu, K., Miyamoto, Y., 2008. An in vivo
comet assay of multiple organs (liver, kidney and bone marrow) in mice treated with
methyl methanesulfonate and acetaminophen accompanied by hematology and/or blood
chemistry. J. Toxicol. Sci. 33, 515–524. doi:10.2131/jts.33.515
Ostling, O., Johanson, K.J., 1984. Microelectrophoretic study of radiation-induced DNA

damages in individual mammalian cells. Biochem. Biophys. Res. Commun. 123, 291–8.
Park, B.-J., Oh, Y.S., Park, S.Y., Choi, S.J., Rudolph, C., Schlegelberger, B., Kim, S.,
2006. AIMP3 haploinsufficiency disrupts oncogene-induced p53 activation and
genomic stability. Cancer Res. 66, 6913–8. doi:10.1158/0008-5472.CAN-05-3740
Parolini, M., Binelli, A., Cogni, D., Riva, C., Provini, A., 2009. An in vitro biomarker
approach for the evaluation of the ecotoxicity of non-steroidal anti-inflammatory drugs
(NSAIDs). Toxicol. Vitr. 23, 935–942. doi:10.1016/j.tiv.2009.04.014
Phillips, B.J., James, T.E.B., Gangoli, S.D., 1982. Genotoxicity studies of
di(2-ethylhexyl)phthalate and its metabolites in CHO cells. Mutat. Res. Toxicol. 102,
297–304. doi:10.1016/0165-1218(82)90139-2
Pollack, G.M., Li, R.C.K., Ermer, J.C., Shen, D.D., 1985. Effects of route of
administration and repetitive dosing on the disposition kinetics of di(2-ethylhexyl)
phthalate and its mono-de-esterified metabolite in rats. Toxicol. Appl. Pharmacol. 79,
246–256. doi:10.1016/0041-008X(85)90346-1
Recio, L., Hobbs, C., Caspary, W., Witt, K.L., 2010. Dose-response assessment of four
genotoxic chemicals in a combined mouse and rat micronucleus (MN) and Comet assay
protocol. J. Toxicol. Sci. 35, 149–62.
Sakai, R., Kondo, C., Oka, H., Miyajima, H., Kubo, K., Uehara, T., 2014. Utilization of