life also carry a potential risk of harm to human health and the environment; therefore, in order to avoid health crises and maintain public safety, it is necessary to accurately estimate the risk carried by exposures to chemical substances and drugs.
Through the accumulation of toxicogenomics data collected during a single exposure to a chemical substance or drug, the highly accurate impact assessment of chemical substances and drugs on the basis of their molecular mechanisms is reaching the stage of practical use. However, these data are derived from specific organs (mainly the liver), and comprehensive toxicity evaluation at the individual level is both costly and labor-intensive.
Recent reports demonstrate that EVs circulating in various fluids could be used as diagnostic biomarkers for various cancers (Logozzi et al. 2009; Lu et al. 2009; Rabinowits et al. 2009; Choi et al. 2011). These EV-associated biomarkers are more sensitive and accurate than biomarkers that are currently widely used, such as CEA for adenocarcinoma and PSA for prostate cancer. Furthermore, microRNAs contained in EVs secreted from various cell types and human samples are being identified as specific biomarkers for chemically induced inflammation (Mobarrez et al. 2014; Li et al. 2010; Baek et al. 2016; Bala et al. 2012; Cho et al. 2017). In addition, EV-associated miRNAs are well protected owing to the lipid bilayer membrane of EVs, even in EVs that have been purified from the circulating bloodstream (Yanez-Mo et al. 2015).
Therefore “next-generation type” toxicity tests for chemical substances and drugs were developed by using EV-associated miRNAs in blood as biomarkers (Figure 3.7).
Figure 3.7 Schematic representation of toxicity testing using EVs as biomarkers.
Isolation and Characterization of EVs from Mouse Blood
Whole blood was collected from C57BL/6J male mice (twelve weeks), and the serum was carefully separated, as previously described. The serum was centrifuged at 10,000 × g for ten minutes to remove cellular debris and subsequently ultracentrifuged at 210,000 × g for thirty minutes. The pellet was washed and resuspended in PBS solution. To characterize EVs collected from whole blood, nanoparticle-tracking analysis (NTA) was performed by using NanoSight (Figure 3.8). The peak size of the EVs was 67 nm, which was comparable with that described in previous reports (Lobb et al. 2015; Urabe et al. 2017).
Figure 3.8 Evaluation of EVs using NanoSight. Serum was separated after blood collection, and microparticle analysis by NanoSight was performed.
Carbon Tetrachloride (CCl4) Administration and Histology
C57BL/6J male mice were orally dosed with carbon tetrachloride (CCl4), because there are numerous reports that CCl4 induces hepatotoxicity in many experimental animals (Chopra et al. 1972). Whole blood and liver samples were collected twenty-four hours after the administration of CCl4 (0 mg/kg (vehicle control: corn oil), 7 mg/kg and 70 mg/kg) for the following experiments.
Hematoxylin and eosin (H&E) staining showed CCl4 (70 mg/kg)-induced histopathological changes in the liver, with significant degeneration and necrosis of hepatocytes in the centrilobular region and perivenular inflammatory infiltrates twenty-four hours after administration, while histological tissue sections of mice in the vehicle control group and CCl4 (7 mg/kg) showed a normal histological morphology of liver tissue samples (Figure 3.9).
Figure 3.9 Representative H&E micrographs of liver tissues collected from mice treated with oral administration of corn oil (control) (a), 7 mg/kg CCl4 (b), and 70 mg/kg CCl4 (c). The control section shows the normal histological structure of the central vein (cv) and surrounding hepatocytes (a). Twenty-four hours after 7 mg/kg CCl4 treatment, there were no histopathological changes by comparison with the control section (b). Twenty-four hours after 70 mg/kg CCl4 treatment, the hepatocytes around the central veins (cv) were vacuolized and necrotic (c).
Identification of Differentially Expressed miRNAs by RNA-Seq
To identify differentially expressed EV-associated miRNAs, we performed RNA-Seq on a size-selected EV-associated small RNA library for three doses of CCl4, namely 0, 7, and 70 mg/kg. We obtained forty-five differentially expressed EV-associated miRNAs, including forty-five upregulated genes and no downregulated genes, between the corn oil and 70 mg/kg CCl4 samples, while only one differentially expressed (upregulated) EV-associated was identified between the corn oil and 7 mg/kg CCl4 samples (see Figure 3.10; also Ono et al. 2020).
Figure 3.10 Differentially expressed EV-associated miRNAs. Individual normalized counts for four differentially expressed EV-associated miRNAs are shown. **P < 0.001, *P < 0.01 vs. control.
Conclusion
Liquid biopsy is a very powerful tool because it is rapid and non-invasive. In fact, AST and ALT are known to be very good biomarkers of liver damage; however, it is difficult to distinguish the cause of liver damage or the status of the liver only on the basis of elevated levels of AST and ALT. Thus, in some cases further examination is needed for detailed lesion assessment. Recently, using EV-associated miRNAs or circulating free miRNAs in serum as biomarkers has made it possible to diagnose thirteen types of cancer with an accuracy of 90% or more (Ogata-Kawata et al. 2014; Shimomura et al. 2016; Yokoi et al. 2018; Yao et al. 2019; Asano et al. 2019; Shiino et al. 2019; Usuba et al. 2019; Asakura et al. 2020). The development of next-generation toxicity tests using miRNA as a biomarker is therefore expected.
Our studies identified forty-two novel miRNAs—such as miR-122 and miR-192—as candidate liver damage biomarkers. Using these novel biomarkers, it may be possible to elucidate the mechanism of hepatotoxicity caused by the administration of drugs or chemical substances other than carbon tetrachloride. Since EVs in blood are secreted by a wide variety of cells, it is necessary to clarify the origin of the EV-associated miRNAs. In addition to their utility as markers of hepatotoxicity, EV-associated miRNAs are expected to be valuable as biomarkers of toxicity that targets other organs, such as the kidney, the lung, and the heart.
Although this study analyzed the effects of carbon tetrachloride administration at a single time point, twenty-four hours, evaluating the time courses of biomarkers in response to the repeated administration of carbon tetrachloride may be applicable if we want to shorten chronic toxicity tests and long-term carcinogenicity tests.
It may also be important to elucidate the function of the identified EV-associated miRNAs induced by hepatotoxicity in vivo.
Acknowledgments
The authors thank N. Moriyama, T. Momiyama, E. Tachihara, M. Uchiyama, and H. Aihara for excellent technical assistance. The images of mice, cows and cultured cells are from
