Carbon nanomaterial technology, including the development of sp²-based carbon nanotubes, has advanced greatly during the last 25 years. The nanoscale size and extraordinary physicochemical properties of multiwalled carbon nanotubes (MWCNTs) have proven very useful in a wide variety of applications, including nanocomposites, sensors, energy storage, energy-conversion systems, field-emission displays, radiation sources, medical applications, and other nanodevices.¹,² MWCNTs can be mass-produced by chemical vapor deposition methods, especially the floating reactant method.³ Simultaneously, occupational and public exposure to manufactured nanomaterials has increased significantly. Carbon materials boast high biocompatibility. However, because of their unique nanoscale size and high aspect ratio (>100), MWCNTs, like asbestos, may present important human health hazards.⁴–⁷

The human body can be exposed to MWCNTs through five possible routes, ie, inhalation of airborne carbon nanotubes, dermal penetration by skin contact, injection of engineered MWCNTs, implantation of composite MWCNTs, and ingestion with drinking water or food additives. Once in the body, MWCNTs are thought to be biopersistent. ⁸ Safety evaluations of MWCNTs have been performed using in vitro and in vivo models.⁹–¹² However, there is no consensus as to whether or not MWCNTs affect the human body. Thus, it is important to identify biomarkers that clarify whether MWCNTs are taken into the system, regardless of their ultimate effect. Such biomarkers will also prove useful for the development of biocompatible MWCNTs.

DJ-1 protein, encoded by the PARK7 gene, belongs to the peptidase C56 family. It functions as a redox-sensitive chaperone (a sensor for oxidative stress), ultimately protecting neurons against oxidative stress and cell death.¹³–¹⁵ Moreover, it has proven to be a biomarker for Parkinson’s disease and some types of cancer.¹⁶–¹⁸ Using a proteomic approach, we recently revealed that expression of this protein decreased when U937 cells were exposed to MWCNTs for 4 days. However, we did not detect biological responses, such as pathological or histological changes, by conventional methods.¹⁹,²⁰

In the present study, we investigated whether DJ-1 could serve as a biomarker for biocompatible MWCNTs. We used the highly purified carbon nanotube, HTT2800, and two different types of cells (characterized by their contrasting biological responses to HTT2800).

To identify potential biomarkers for MWCNTs in vitro, we used two different cell lines, ie, THP-1 and MESO-1. We recently reported that DJ-1 protein expression decreased when U937 cells were exposed to MWCNTs for 4 days; however, carbon black levels were not altered in response to DJ-1.²⁰ Similar to U937 cells, THP-1 human acute monocytic leukemia cells have the ability to differentiate into macrophages. ²² The objective of the present study was to confirm whether the expression of DJ-1 in U937 cells is universal. We recently also reported that exposure of BEAS-2B cells (derived from human bronchial epithelium) to HTT2800 proved cytotoxic,²³ causing arrest of proliferation. One of the most significant differences between U937 and BEAS-2B cells is their surface adhesiveness. Therefore, in the present study, we used adherent MESO-1 cells (derived from human malignant pleural mesothelioma) to evaluate DJ-1 protein expression upon exposure to MWCNTs. Pulmonary mesothelial cells are thought to constitute the crucial cell type for asbestos toxicity.

The proliferation of THP-1 and MESO-1 cells exposed to HTT2800 for 4 days was clearly different (Figure 1A). While the propagation of MESO-1 cells was suppressed by HTT2800 in a concentration-dependent manner (with a few dead cells observed in the culture dish), THP-1 cells were not affected by the nanomaterial. Figure 1B depicts the cell images after 4 days of culture with or without HTT2800. In the case of MESO-1 cells, endocytosis of MWCNTs was clearly observed. These results were in accordance with a previous report, in which a similar effect was observed in the BEAS-2 cell line.²³ By contrast, in the case of THP-1 cells, MWCNTs adhered around the cells, but were not ingested. This response was similar to that previously observed in U937 cells.¹⁹

Figure 2A shows the expression levels of DJ-1 protein, determined by means of Western blot analysis. In comparison with the vehicle control, the quantity of DJ-1 detected in THP-1 cells exposed to 100 μg/mL of HTT2800 was reduced to 67%. This finding is in accordance with the results obtained with U937 cells exposed to MWCNTs for 4 days, analyzed using the proteomic approach.¹⁹ The decrease rate observed in the present study was lower than that previously documented,²⁴ because our Western blot quantitative analysis was performed by summing oxidized and reduced DJ-1, whereas Duan et al separated the protein using two-dimensional electrophoresis. Correspondingly, in MESO-1 cells exposed to 10 μg/mL of HTT2800, DJ-1 expression was reduced to 68% compared with the vehicle control.

It has been reported that oxidative stress induces intracellular redistribution of DJ-1.²⁵ Therefore, we performed immunofluorescence staining of DJ-1. However, we observed no changes in intracellular localization of the protein (Figure 2B).

We further investigated the expression of DJ-1 mRNA in MESO-1 cells by quantitative real-time polymerase chain reaction. MESO-1 cells are adherent and separation of the MWCNTs is easy. Thus, in comparison with cell lines characterized by nonadherent cells, MESO-1 cells are more advantageous when developing an evaluation method for highly biocompatible MWCNTs in vitro. The decrease in DJ-1 mRNA expression was time-dependent and statistically significant after day 2 (Figure 3). Our quantitative real-time polymerase chain reaction results indicate approximately 50% reduction in expression, which is larger than that observed in the case of Western blotting. However, this finding is in accordance with the decrease in DJ-1 expression previously demonstrated using the proteomic approach.¹⁹

Duan et al reported that total and reduced DJ-1 expression in human pneumocytes was decreased by hydrogen peroxide. ²⁴ Moreover, we recently demonstrated that purified MWCNTs tended to increase the production of intracellular hydrogen peroxide in human bronchial epithelial cells, one hour and 24 hours after exposure. This was accompanied by the accumulation of other intracellular reactive oxygen species after 24 hours (submitted for publication). MWCNT endocytosis does not occur as early as one hour after exposure. Nevertheless, augmentation of the intracellular hydrogen peroxide level at this time point may indicate the requirement for cells to come into contact with MWCNTs. In the present study, THP-1 cells only contacted MWCNTs, and did not endocytose them. Thus, the decrease in DJ-1 expression may have been caused by an intracellular hydrogen peroxide boost, generated by the contact between THP-1 cells and MWCNTs.

We evaluated the influence of other types of carbon nanomaterials on DJ-1 mRNA levels in MESO-1 cells (Figure 4). Cup-stacked carbon nanotubes did not change the expression level. By contrast, Sumi black increased it by 1.8-fold. At a concentration of 10 μg/mL, neither of the carbon nanomaterials showed cytotoxicity, but both were endocytosed (data not shown). Moreover, the pattern of cytokine secretion in THP-1 cells exposed to Sumi black differed from that observed in THP-1 cells exposed to MWCNTs (submitted for publication). Our results indicate that different physical properties of the carbon nanomaterials used were recognized by the cell membrane, resulting in distinctive responses. Our previous finding that HTT2800 converted the differentiation pathway of C2C12 myoblasts into that of adipoblast-like cells²⁶ further supports this hypothesis.

DJ-1 exists abundantly in the human body, with extremely diverse functions. However, it remains to be determined whether it has a tissue-specific or organ-specific role. We propose that, by measuring the expression of DJ-1 in blood cells, it may be possible to determine the extent of MWCNT exposure. In the present study, we observed a reduction in DJ-1 protein expression, regardless of the cytotoxic activity of intracellular HTT2800. Moreover, exposure to HTT2800, but not to Sumi black or cup-stacked carbon nanotubes, caused a reduction in DJ-1 mRNA levels. Thus, we conclude that DJ-1 is a novel biomarker for the development of biocompatible MWCNTs.

Disclosure

The authors report no conflicts of interest in this work.

We thank the staff of the Division of Instrumental Analysis at the Research Center for Human and Environmental Sciences of Shinshu University for their help. This research was supported by the Program for Fostering Regional Innovation in Nagano, and a grant-in-aid (No. 19002007) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.