Research Article Received: 16 January 2015,
Revised: 11 February 2015,
Accepted: 15 February 2015
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/jat.3150
Titanium dioxide nanoparticles increase plasma glucose via reactive oxygen species-induced insulin resistance in mice Hailong Hua, Qian Guoa, Changlin Wanga, Xiao Mab, Hongjuan Hea, Yuri Ohc, Yujie Fengd, Qiong Wua and Ning Gua* ABSTRACT: There have been few reports about the possible toxic effects of titanium dioxide (TiO2) nanoparticles on the endocrine system. We explored the endocrine effects of oral administration to mice of anatase TiO2 nanoparticles (0, 64 and 320 mg kg–1 body weight per day to control, low-dose and high-dose groups, respectively, 7 days per week for 14 weeks). TiO2 nanoparticles were characterized by scanning and transmission electron microscopy (TEM) and dynamic light scattering (DLS), and their physiological distribution was investigated by inductively coupled plasma. Biochemical analyzes included plasma glucose, insulin, heart blood triglycerides (TG), free fatty acid (FFA), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), total cholesterol (TC), tumor necrosis factor-alpha (TNF-α), interleukin (IL)-6 and reactive oxygen species (ROS)-related markers (total SOD, GSH and MDA). Phosphorylation of IRS1, Akt, JNK1, and p38 MAPK were analyzed by western blotting. Increased titanium levels were found in the liver, spleen, small intestine, kidney and pancreas. Biochemical analyzes showed that plasma glucose significantly increased whereas there was no difference in plasma insulin secretion. Increased ROS levels were found in serum and the liver, as evidenced by reduced total SOD activity and GSH level and increased MDA content. Western blotting showed that oral administration of TiO2 nanoparticles induced insulin resistance (IR) in mouse liver, shown by increased phosphorylation of IRS1 (Ser307) and reduced phosphorylation of Akt (Ser473). The pathway by which TiO2 nanoparticles increase ROS-induced IR were included in the inflammatory response and phosphokinase, as shown by increased serum levels of TNF-α and IL-6 and increased phosphorylation of JNK1 and p38 MAPK in liver. These results show that oral administration of TiO2 nanoparticles increases ROS, resulting in IR and increasing plasma glucose in mice. Copyright © 2015 John Wiley & Sons, Ltd. Keywords: TiO2 nanoparticles; plasma glucose; insulin; ROS; IR
Introduction During the past few decades, nanotechnology has become widely utilized in many fields, such as drugs, medicine and engineering technology (Hood, 2004; Nowwack and Bucheli, 2007). Titanium dioxide (TiO2) nanoparticles are a highly stable, anticorrosive and photoactive nanoparticles and are frequently used as an important industrial material for products such as pharmaceuticals, antibacterial, cosmetics and food additives (Helinor et al., 2009; Shi et al., 2013). Human exposures to TiO2 nanoparticles may occur during both manufacturing and use. For workers, gravimetric concentrations of TiO2 nanoparticles at workplaces ranged from 0.1 to 4.99 mg m–3, and for ordinary people, a typical diet may be the major exposure route that contributes 300–400 μg per day (Shi et al., 2013). TiO2 nanoparticles were previously considered to be a biologically inert material that is very safe for both humans and animals (Ophus et al., 1979; Lindenschmidt et al., 1990). However, some recent studies have suggested that the small size of TiO2 nanoparticles may have different chemical, optical, magnetic and structural properties of normal-sized TiO2 particles, and consequently, their toxicity profile may also be different (Maynard et al., 2006; Wu et al., 2009). For instance, cell studies revealed that most cells treated with TiO2 nanoparticles showed a series of morphological changes, including decreased cell size, membrane blebbing, peripheral chromatin condensation and apoptotic body formation (Gurevitch et al., 2012; Hussain et al., 2010). In vivo
J. Appl. Toxicol. 2015
studies showed that cutaneous exposure to TiO2 nanoparticles induced injury to both skin and organs (Wu et al., 2009), whereas inhalation of TiO2 nanoparticles induced alveolar epithelial lesions (Bermudez et al., 2004). Intravenous injection of TiO2 nanoparticles caused cytotoxicity to red and white cells in the bone marrow (Dobrzyńska et al., 2014), and oral administration of TiO2 nanoparticles harmed the digestive system (Wang et al., 2007; Shi et al., 2013). Although there is still no unanimous consensus on the mechanism of possible biological toxicity of TiO2 nanoparticles, studies
*Correspondence to: Ning Gu, School of Life Science and Technology, Harbin Institute of Technology, Harbin, China, Ning Gu, No. 92 West Da-zhi Street, Harbin, 150001, Heilongjiang, China. E-mail:
[email protected] a School of Life Science and Technology, State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin, China b Key Laboratory of Pu-erh Tea Science of Ministry of Education, Yunnan Research Center for Advanced Tea Processing, Yunnan Agricultural University, Kunming, China c
Faculty of Education, Wakayama University, Wakayama, Japan
d State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin, China
Copyright © 2015 John Wiley & Sons, Ltd.
H. Hu et al. indicate that production and accumulation of reactive oxygen species (ROS) is likely to be the most important mechanism (Colvin, 2003; Long et al., 2006). TiO2 nanoparticles have been shown to increase ROS levels when illuminated by ultraviolet (UV) irradiation (Burello and Worth, 2011). It has been confirmed that TiO2 nanoparticles can also increase ROS levels in bacteria and animals without UV irradiation, as evidenced by depletion of cellular antioxidants and an increase in oxidative products (Wang et al., 2007; Sohm et al., 2015). For instance, ROS produced by chondriosomes could be cleared by glutathione (GSH) and superoxide dismutase (SOD), but when tissues were exposed to TiO2 nanoparticles, SOD activity, and the GSH level were found to be significantly reduced in relevant organs (Sha et al., 2013), and at the same time, ROS productions such as methane dicarboxylic aldehyde (MDA) were significantly increased (Wu et al., 2009). Studies have focused on the morphological injuries to organs caused by a ROS increase induced by TiO2 nanoparticles, but few studies exist about the types of disease that can be caused by these ROS. ROS-induced changes in organ morphology and function are correlated with the incidence of certain diseases; for example, as reported in the literature, ROS play an important role in the pathogenesis of diabetes (Fridlyand and Philipson, 2006; Houstis et al., 2006). Diabetes is an endocrine system disease and is usually the product of two distinct abnormalities: abnormal β-cell function and insulin resistance (IR) (Barzilay et al., 2001). β-Cells are the insulin-producing cells in pancreatic islets (Hou et al., 2013), and increases in ROS induce ever-worsening abnormalities in the pancreatic β-cell, and therefore, in insulin synthesis and secretion (Robertson, 2006; Yeo et al., 2013). IR refers to the decrease in insulin sensitivity seen in patients with diabetes and is the cardinal feature of type 2 diabetes (Gurevitch et al., 2012; Lukic et al., 2014). According to the literature (Anabela and Carlos, 2006; Philippe et al., 2005), ROS can increase the phosphorylation of serine/ threonine residues and reduce the phosphorylation of tyrosine residues in insulin receptor substrate 1 (IRS1), inhibiting the ability of the insulin substrate to combine with the molecules downstream of IRS1. As a result, the insulin signal is interrupted, and IR is induced. In contrast, inhibition of ROS and remission of the antioxidant level can ameliorate IR (Fridlyand and Philipson, 2006; Houstis et al., 2006). All this evidence strongly suggests that ROS are a factor in disrupting plasma glucose homeostasis. Therefore, humans intake TiO2 nanoparticles via manufacturing and diet. For ordinary people, the diet is the major intake route. In this study, to explore whether the increase in ROS caused by TiO2 nanoparticles in the diet affects plasma glucose homeostasis, we administered TiO2 nanoparticles orally to mice and measured their plasma glucose and ROS levels. We also investigated the mechanism by which TiO2 nanoparticle-induced ROS might affect plasma glucose homeostasis.
Materials and Methods
dynamic light scattering (DLS; Brookhaven Instruments Corporation, Brookhaven, MS, USA). All of the above were tested in the dose of 1 mg ml–1 TiO2 nanoparticles.
Animals and treatments All animal experiments were reviewed and approved by Harbin Institute of Technology, and carried out according to the guidelines for the care and use of experimental animals approved by the Heilongjiang Province People’s Congress (http://www.nicpbp.org. cn/sydw/CL0249/2730.html). Six-week-old CD-1 (ICR) mice of males (25.37 ± 0.71 g) were obtained from Harbin Veterinary Research Institute (Harbin, China) and acclimated for 7 days after arrival at the study facility. Mice were housed in an animal room at a controlled temperature (21–24 °C) and light cycle (12 h light/dark). Autoclaved water and rodent diets (Keao Co., Ltd., Beijing, China) were provided ad libitum. To explore the endocrine effects of oral administration of anatase TiO2 nanoparticles, TiO2 nanoparticles should be absorbed and accumulated in mice tissues. TiO2 nanoparticle is a type of nanoparticle with a low absorption rate. Thus, we orally administered 0.52, 2.6, 13, 64, and 320 mg kg–1 body weight dose TiO2 nanoparticles suspension to mice to determine the absorption dose. Then, we selected two absorption doses from 0.52, 2.6, 13, 64 and 320 mg kg–1 body weight for low-dose (LD) and high-dose (HD) groups of the experiment that explores endocrine effects of TiO2 nanoparticles. The control group mice were given an equal volume of PBS. The body weights of mice were weighted every week, and food intakes of mice were weighed every 2weeks (7 days per week).
Blood collection and analysis Every 2 weeks, after mice were fasted for 16 h, and before the daily administration of TiO2 nanoparticles, blood was collected from the tail vein to measure plasma glucose. After being determined that plasma glucose increased in tail vein blood, mice plasma glucose levels were followed up for 4 weeks to test whether the increase in plasma glucose was not an isolated phenomenon. At the end of week 14, mice were fasted for 16 h, and then blood was collected from mouse hearts. Tail vein blood plasma glucose and heart blood plasma glucose were measured using a glucose assay kit (Wako Pure Chemical Industries, Ltd., Osaka, Japan). Heart blood plasma insulin was measured using a mouse insulin ELISA kit (Shibayagi Co., Ltd., Gunma, Japan). Heart blood triglycerides (TG), free fatty acid (FFA), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and total cholesterol (TC) were measured using each kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). Tumor necrosis factor (TNF)-α and interleukin (IL)-6 were measured using TNF-α kit and IL-6 kit (R&D Systems, Minneapolis, MN, USA).
Oral glucose tolerance test (OGTT) Nanoparticles and physicochemical characterization Food additives powder-form TiO2 nanoparticles were obtained from Veking Co., Ltd. (Hangzhou, China). Primary particle sizes and morphology were measured using transmission electron microscope (TEM, FEI200KV; FEI Co., Ltd., Hillsboro, OR, USA) and scanning electron microscopy (SEM, Quanta; FEI Co., Ltd., Hillsboro, OR, USA). The hydrodynamic size and zeta potential of nanoparticles in phosphate-buffered saline (PBS) were measured using
wileyonlinelibrary.com/journal/jat
At week 14, mice were fasted for 16 h and then orally administered glucose (1.5 g kg–1 body weight). Blood was collected for plasma glucose and insulin level measurement from the tail vein into capillary tubes each 10 μl at 0, 30, 60 and 120 min after administration of glucose. Plasma glucose was measured using a glucose assay kit (Wako Pure Chemical Industries, Ltd., Osaka, Japan). Plasma insulin was measured using a mouse insulin ELISA kit (Shibayagi Co., Ltd., Gunma, Japan).
Copyright © 2015 John Wiley & Sons, Ltd.
J. Appl. Toxicol. 2015
TiO2 nanoparticle increases plasma glucose Titanium content analysis Mice were euthanized at the end of week 14. The liver (lobes), spleen (whole), small intestine (middle part), gastrocnemius muscle (whole), kidney (middle part) and pancreas (whole) were taken out. Next, 0.2 g of each tissue was weighed, digested and analyzed for titanium content. Briefly, the tissues were digested in nitric acid (ultrapure grade) overnight. After adding 0.5 ml of H2O2, the mixed solutions were heated at about 160 °C using a high-pressure reaction container in an oven chamber until the samples were completely digested. Then, the solutions were heated at 120 °C to remove the remaining nitric acid until the solutions were colorless and clear. At last, the remaining solutions were diluted to 3 ml with 2% nitric acid. Inductively coupled plasma (ICP, Optima 5300 DV, Perkin Elmer Inc., Richmond, CA, USA) was used to analyze the titanium concentration in the samples. Data are expressed as nanograms per gram fresh tissue. To assess whether TiO2 particles exist in each tissue after mice were orally administered with TiO2 nanoparticles, 0.1 g of each tissue was also weighed for SEM and energy dispersive X-ray analysis (EDXA). The tissues were homogenized in RIPA lysis buffer, filtered to discard cell debris using a 0.45-μm filter membrane and centrifuged for 30 min at 18 000 g to precipitate TiO2 nanoparticles. The precipitation was resuspended in alcohol and spread out on an aluminum sheet, sputter-coated with platinum and observed by SEM. The surface element was analyzed by EDXA. Reactive oxygen species assessment ROS levels were assessed using the levels of total SOD (T-SOD), GSH and MDA. Heart blood was centrifuged at 750 g for 10 min at 4 °C to separate serum and cell debris. The liver was homogenized in nine volumes (1:10 w/v) of PBS. Homogenates were centrifuged at 750 g for 10 min at 4 °C to discard cell debris. The T-SOD, GSH and MDA of serum and liver supernatant were measured using each kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). Immunofluorescence For pathological studies, the pancreas was fixed in 4% paraformaldehyde solution for 24 h, embedded in paraffin, cut into 5 μm sections, and placed onto glass slides. The β-cell mass was identified by performing immunohistochemical localization of insulin with a polyclonal anti-insulin antibody (Santa Cruz biotechnology, Inc., Santa Cruz, CA, USA). For the analysis of pancreatic β-cell apotosis, paraffin-embedded pancreatic sections were stained with the terminal deoxynucleotidyl transferase mediated dUDP nick endlabeling (TUNEL, KeyGen Biotech. CO., LTD., Nanjing, China). The slides were observed and the photos were taken using an optical microscope (OLYMPUS BX51; OLYMPUS Co., Tokyo, Japan). β-Cells pretreated with DNase I were used as a positive control, and these were colored green (positive) as expected. Western blot Next, 0.3 g of liver tissues was resuspended in RIPA lysis buffer. Lysates were centrifuged for 15 min at 14 000 g and 4 °C, and the protein contents of the supernatant were determined using a DC protein kit (Bio-Rad Laboratories, Hercules, CA, USA). Aliquots of the proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and transferred to PVDF membranes purchased from
J. Appl. Toxicol. 2015
Bio-Rad Laboratories (Hercules, CA, USA). PVDF membranes were incubated with antibodies against phospho-c-Jun aminoterminal kinase 1 ( JNK1), phospho-p38 mitogen-activated protein kinase (MAPK), phospho-IRS1 (Ser307), phospho-Akt (Ser473), JNK1, p38 MAPK, IRS1 and Akt. All antibodies were purchased from Cell Signaling Technology (Danvers, MA, USA). Bands were visualized using an ECL plus western blotting detection system (GE Healthcare, Milwaukee, WI, USA). Statistical analysis The results are expressed as the mean ± SEM. Multigroup comparisons of the means were carried out by the one-way analysis of variance (ANOVA) test. The statistical significance for all tests was set at P < 0.05.
Results Physicochemical characteristics of TiO2 nanoparticles The anatase phase TiO2 nanoparticles were characterized by DLS, TEM and SEM. The physicochemical properties of the TiO2 nanoparticles are summarized in Table 1. The primary sizes of TiO2 nanoparticles provided by the manufacturer were 15 nm, but the average size measured by TEM and SEM was 25.64 ± 6.63 nm. The average hydrodynamic size of TiO2 nanoparticles in PBS measured by DLS was 49.59 ± 0.41 nm. The zeta potential of the TiO2 nanoparticles in PBS was negative (Table 1). The soluble ionic form of Ti was less than the minimum detection limit of ICP (data not shown). Absorption and physiological distribution of TiO2 nanoparticles in mice To analyze the absorption of different concentrations of TiO2 nanoparticles, we treated mice with TiO2 nanoparticles (0, 0.52, 2.6, 13, 64 and 320 mg kg–1 body weight) by oral administration and collected their blood from hearts, respectively, at 0, 0.5, 1, 2, 4, 6, 9, 12 and 24 h to measure the absorption of TiO2 nanoparticles. The results showed that the change of titanium levels cannot be detected after mice were orally administered 0.52 and 2.6 mg kg–1 body weight TiO2 nanoparticles (Fig. 1A). Titanium levels were increased significantly after mice were orally administered 13, 64 and 320 mg kg–1 body weight TiO2 nanoparticles, and the absorption peak was at 1 h after oral administration (Fig. 1A). There were no significant differences in titanium levels in the blood among mice that were orally administered 13, 64, and 320 mg kg–1 body weight TiO2 nanoparticles. These results suggested that, among 0.52, 2.6, 13, 64 and 320 mg kg–1 body weight, the dose which was higher than or equal to the 13 mg kg–1 body weight can be absorbed via oral administration and enter into mice blood. In the experiment that explores endocrine effects of oral administration to mice of anatase TiO2 nanoparticles, mice were randomly divided into three groups: control, low dose (LD) and high dose (HD). In order to more clearly detect endocrine effects of TiO2 nanoparticles, we selected 64 and 320 mg kg–1 body weight dose TiO2 nanoparticles for mice of the LD group and HD group, and an equal volume of PBS for mice of the control group. At the end of week 14, mice were euthanized, and the liver, spleen, small intestine, gastrocnemius muscle, kidney, and pancreas were collected to analyze the physiological distribution of titanium. The
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jat
H. Hu et al. Table 1. Physico-chemical characteristics of titanium dioxide (TiO2) nanoparticles. Property
TiO2 nanoparticles
Shape: TEM & SEM
Average TEM&SEM Size (nm±sd) Hydrodynamic Size distribution
25.64±6.63
Average Hydrodynamic Size (nm ± SD) Zeta Potential (mV ± SD) Crystalline Structure Size (nm)
49.59±0.41 -24.51±3.92 Anatasea 15 nma
a
Manufacturer’s data. The particle hydrodynamic size and zeta potential was tested in the dose of 1 mg ml–1 TiO2 nanoparticles
results showed that titanium concentrations were considerably increased in the liver, spleen, small intestine, kidney and pancreas in both the LD and HD groups, and there was no significant difference between these two groups (Fig. 1B). SEM and EDXA results revealed that TiO2 nanoparticles (nanosized spherical white objects) were found in the above tissues homogenates (Fig. 1C).
noticeably increased in both the LD and HD groups, but there was no change in values of insulin, TG, FFA, HDL-C, LDL-C, or TC (Table 2).
Oral administration of TiO2 nanoparticles dose did not affect body weights or food intakes, but increases plasma glucose in mice
To analyze whether oral administration of TiO2 nanoparticles increases ROS levels in mice, the activity of T-SOD, the levels of GSH and MDA in mouse serum and livers were measured after mice were euthanized. The results showed that in both the serum and liver of LD and HD groups, the activity of T-SOD and the level of GSH were significantly reduced (Fig. 3A and B), and the level of MDA was significantly increased (Fig. 3C). These results showed that oral administration of TiO2 nanoparticles increases ROS levels in mice.
Before mice were orally administrated with TiO2 nanoparticles, the mean body weights of the LD group (24.99 ± 4.92 g) and the HD group (25.88 ± 3.92 g) were not significantly different from the control group (25.25 ± 2.87 g). The mean fasting plasma glucose levels of the LD group (5.93 ± 1.43 mmol l–1) and the HD group (5.76 ± 1.50 mmol l–1) were also similar to the control group (6.06 ± 1.19 mmol l–1). Blood was taken from the tail vein of mice during the oral administration phase. Plasma glucose levels were measured, and found to be increased from week 10 in both the LD and HD groups (Fig. 2A). Body weights and food intakes were measured at the same period, and the results showed that there were no significant differences in body weight and food intake between the control, LD and HD groups (Fig. 2B and C). The measurement of mice plasma glucose levels of tail vein blood was kept for 4 weeks. After being determined that the increase in plasma glucose was not an isolated phenomenon, blood was collected from mouse hearts in the euthanasia procedure to measure plasma glucose and insulin levels, as well as a number of other markers. The plasma glucose of heart blood was also
wileyonlinelibrary.com/journal/jat
Oral administration of TiO2 nanoparticles increases ROS levels in mice
Oral administration of TiO2 nanoparticles does not induce β-cell apoptosis in mice Pancreatic β-cells were identified by insulin antibody staining, and we attempted to detect apoptotic β-cells using the TUNEL method. Apoptotic cells showed up as green cells. β-Cells pretreated with DNase I were used as a positive control, and these were colored green (positive) as expected. However, when the TUNEL assay was performed on mouse β-cells without DNase I pretreatment, no apoptotic β-cells were detected in any of the three groups (Fig. 4), indicating that oral administration of TiO2 nanoparticles does not induce β-cell apoptosis in mice.
Copyright © 2015 John Wiley & Sons, Ltd.
J. Appl. Toxicol. 2015
TiO2 nanoparticle increases plasma glucose
Figure 1. Titanium dioxide (TiO2) nanoparticles were absorbed via oral administration and accumulated in mice tissues. (A) Titanium levels of mice blood –1 –1 after mice were orally administered different concentrations of TiO2 nanoparticles. * P < 0.05 for 13 mg kg group vs. 0 h, # P < 0.05 for 64 mg kg group vs. –1 0 h, $ P < 0.05 for 320 mg kg group vs. 0 h. Results are the mean ± SE (n = 5). (B) Titanium concentrations of mice tissues after mice were orally administered –1 TiO2 nanoparticles 14 weeks (0, 64 and 320 mg kg body weight per day to the control, low-dose (LD) and high-dose (HD) groups, respectively, 7 days per week). * P < 0.05 vs. the control group. Results are the mean ± SE (n = 10). (C) SEM results (nanosized spherical white objects are TiO2 nanoparticles) and EDXA results (arrows indicate titanium) of tissues homogenates.
Oral administration of TiO2 nanoparticles induces IR in mice We conducted OGTT to examine glucose-dependent insulin secretion in mice at week 14 after oral administration of TiO2 nanoparticles. There was a significant increase in plasma glucose levels, but no difference in plasma insulin levels at 30, 60 and 120 min after oral administration of glucose in the LD and HD groups compared with the control group (Fig. 5A and B).
J. Appl. Toxicol. 2015
The area under the curve (AUC) for plasma glucose was higher in the LD and HD groups than in the control group, but there was no significant difference between the LD and HD groups (Fig. 5A). There was no significant difference in the AUC of plasma insulin between the three groups (Fig. 5B). These results indicated that oral administration of TiO2 nanoparticles does not change insulin secretion but does induce IR, and this in turn causes a significant increase in plasma glucose in mice.
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jat
A
Glucose (mmol/L)
H. Hu et al. 10 9 8 7 6 5 4 3 2 1 0 0
2
4
6
8
**
**
10
12
*
*
Cont LD HD
14
B
Body Weight (g)
Time (week) 40 38 36 34 32 30 28 26 24 22 20
Cont LD HD
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Time (week) 6
Food Intake (g)
C
Cont
5
LD
4
HD
3 2 1 0 0
2
4
6
8
10
12
14
Time (week) Figure 2. Oral administration of titanium dioxide (TiO2) nanoparticles increases mice plasma glucose, but does not affect mice body weights or food intakes. (A) The plasma levels of mice tail veins. (B) Body weights. (C) Food intakes. Mice were orally administered with TiO2 nanoparticles everyday –1 (0, 64 and 320 mg kg body weight per day to control, low-dose (LD) and high-dose (HD) groups, respectively, 7 days per week for 14 weeks). Plasma glucose levels and food intakes were measured every 2 weeks. Body weights were measured every week. * P < 0.05 vs. the control group. Results are the mean ± SE (n = 10).
Table 2. Measurement of metabolic values of mice after oral administration with TiO2 nanoparticles (values are the mean ± SE; n = 10) Parameters Fasting plasma glucose (mmol l–1) Fasting plasma insulin (mIU l–1) TG (mmol l–1) FFA (mmol l–1) LDL-C (mmol l–1) HDL-C (mmol l–1) TC (mmol l–1)
Cont
LD
HD
6.99 ± 0.52 1.67 ± 0.11 0.33 ± 0.05 0.28 ± 0.07 0.87 ± 0.13 1.29 ± 0.21 2.12 ± 0.24
9.60 ± 0.58a 1.79 ± 0.07 0.43 ± 0.07 0.28 ± 0.07 0.72 ± 0.06 1.28 ± 0.18 1.90 ± 0.16
10.59 ± 0.52a 1.60 ± 0.06 0.38 ± 0.06 0.31 ± 0.08 1.06 ± 0.24 1.22 ± 0.14 2.36 ± 0.28
Significantly different from the control group (P < 0.05).
a
Oral administration of TiO2 nanoparticle-increased ROS induces IR via the inflammatory response pathway and MAPK pathway in mice The insulin-signaling pathway was examined to clarify whether TiO2 nanoparticle-increased ROS can induce IR in mice. Western blotting results showed that phosphorylation of IRS1 (Ser307) was increased and phosphorylation of Akt (Ser473) was reduced significantly in mouse liver in both the LD and HD groups in week 14 after oral administration of TiO2 nanoparticles (Fig. 6A and B).
wileyonlinelibrary.com/journal/jat
In order to clarify the mechanism by which TiO2 nanoparticleincreased ROS induces IR, the pathways of the inflammatory response and MAPK were examined. The levels of TNF-α increased significantly in the serum of mice in the LD and HD groups, and the levels of IL-6 increased significantly in the serum of mice in the HD group in week 14 after oral administration of TiO2 nanoparticles (Fig. 6C). Western blotting results showed the phosphorylation of JNK1 and p38 MAPK were increased significantly in the liver of mice in both the LD and HD groups in week 14 after oral administration of TiO2 nanoparticles (Fig. 6A and B).
Copyright © 2015 John Wiley & Sons, Ltd.
J. Appl. Toxicol. 2015
TiO2 nanoparticle increases plasma glucose
A 8
*
250
*
200 150 100
T-SOD of Liver (U/mg protein)
T-SOD of Serum (U/mL)
300
50
7
*
6
*
5 4 3 2 1
0
0 Cont
LD
Cont
HD
LD
HD
B 1.4
4
*
3
*
2 1
GSH of Liver (mg/mg protein)
GSH of Serum (mg/mL)
5
1.2
*
1
*
0.8 0.6 0.4 0.2
0
0 Cont
LD
HD
*
*
Cont
LD
HD
*
*
LD
HD
C 20 15 10 5
3
MDA of Liver (nmol/mg protein)
MDA of Serum (nmol/mL)
25
2.5 2 1.5 1 0.5
0
0 Cont
LD
HD
Cont
Figure 3. Oral administration of titanium dioxide (TiO2) nanoparticles increases mice reactive oxygen species (ROS) levels. (A) The activities of total SOD (T-SOD) in mice serum and the liver. (B) The levels of glutathione (GSH) in mice serum and the liver. (C) The levels of methane dicarboxylic aldehyde (MDA) in mice serum and the liver. At week 14, mice were euthanized and heart blood serum and livers were collected and the activities of T-SOD and contents of GSH and MDA were measured. * P < 0.05 vs. the control group. The results are the mean ± SE (n = 10).
These results indicated that TiO2 nanoparticles increase ROS and inhibit insulin signaling via two pathways that involve the inflammatory cytokines TNF-α and IL-6, and the phosphokinases JNK1 and p38 MAPK.
Discussion We administered different concentration of TiO2 nanoparticles (0, 0.52, 2.6, 13, 64 and 320 mg kg–1 body weight) orally to healthy male CD-1 (ICR) mice. The results showed that the absorption of TiO2 nanoparticles can not be detected after mice were orally administered 0.52 and 2.6 mg kg–1 body weight TiO2 nanoparticles, and there was a trend that the amount of accumulation of TiO2 nanoparticles was gradually increased as the oral administration dose increased after mice were orally administered 13, 64 and 320 mg kg–1 body weight TiO2 nanoparticles, but there was no significant difference among the 13, 64 and 320 mg kg–1 body weight TiO2 nanoparticles groups (Fig. 1A). It has been reported previously that the TiO2 nanoparticle is a type of nanoparticle with a low
J. Appl. Toxicol. 2015
absorption rate, and that few were detected in tissues but most were detected in feces after oral administration (Cho et al., 2013). There is a threshold of the absorption rate beyond which extra doses have no effect in this study, and this is the possible reason there was no significant difference between titanium levels in blood among the 13, 64 and 320 mg kg–1 body weight groups. In order to more clearly detect the endocrine effects of TiO2 nanoparticles, in the experiment that explore endocrine effects of oral administration to mice of anatase TiO2 nanoparticles, we selected 64 and 320 mg kg–1 body weight dose TiO2 nanoparticles (LD group, 64 mg kg–1 body weight per day, and HD group, 320 mg kg–1 body weight per day). TiO2 nanoparticles have been reported to be absorbed via oral administration and to accumulate in the liver, spleen and kidneys (Wang et al., 2007; Shi et al., 2013). In the present study, we found that after oral administration of TiO2 nanoparticles, titanium accumulated not only in the liver, spleen and kidney, but also in the pancreas and small intestine, and also found that there was no significant difference in titanium levels between the LD and HD group (Fig. 1B). SEM images also
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jat
H. Hu et al.
16 14 12 10 8 6 4 2 0
Cont
*# *
HD
*#
0
LD
#
*#
30
60
120
% AUC of Plamsa Glucose
A
Plamsa Glucose (mmol/L)
Figure 4. Oral administration of titanium dioxide (TiO2) nanoparticles does not induce mice β-cells apoptosis. Mice were euthanized at the end of week 14 and pancreas islets were collected and measured apoptosis. Insulin (red), TUNEL ( green) and Nucleus (blue) costaining. TUNEL and Nucleus are merged in Merge. DNase I treatment as a TUNEL-positive control.
160 140 120 100 80 60 40 20 0
*
#
Cont
LD
HD
Cont
LD
HD
Time after treatment with glucose (min)
Insulin (mIU/L)
25
Cont LD
20
HD
15 10 5 0 0
30
60
% AUC of Plasma Insulin
B
160 140 120 100 80 60 40 20 0
120
Time after treatment with glucose (min) Figure 5. Oral administration of titanium dioxide (TiO2) nanoparticles increases mice plasma glucose but does not affect mice insulin secretion in the oral glucose tolerance test (OGTT). (A) Time course of changes and area under the curve (AUC) in plasma glucose levels during the OGTT. (B) Time course of –1 changes and AUC in plasma insulin levels during the OGTT. In week 14, mice were fasted for 16 h and then orally administered glucose (1.5 g kg body weight). Plasma insulin and glucose levels were measured at 0, 30, 60 and 120 min. * P < 0.05 for the low-dose (LD) group vs. the control group; # P < 0.05 for the high-dose (HD) group vs. the control group. Results are the mean ± SE (n = 10).
obviously showed white nanoparticles in these tissues of the LD and HD groups, and the EDXA proved these nanoparticles are TiO2 nanoparticles (Fig. 1C).
wileyonlinelibrary.com/journal/jat
To analyze the effect of TiO2 nanoparticles on plasma glucose, we collected blood from the tail vein every 2 weeks after we began treating the mice with TiO2 nanoparticles. Plasma glucose was
Copyright © 2015 John Wiley & Sons, Ltd.
J. Appl. Toxicol. 2015
TiO2 nanoparticle increases plasma glucose
Cont
LD
B
HD
p-JNK1 Total JNK1 p-P38 Total P38 p-IRS1(Ser307) Total IRS1 p-Akt(Ser473)
Cont
Phosphorylated protein/total protein
A
5
LD
*
4 3
HD
**
* **
2 1
** 0
Total Akt
C
60
*
Content (pg/mL)
* 40
Cont LD
50
HD
*
30 20 10 0 TNF-α
IL-6
Figure 6. Oral administration of titanium dioxide (TiO2) nanoparticle-increased reactive oxygen species (ROS) induces insulin resistance (IR) via the inflammatory response pathway and the MAPK pathway in mice. (A) The phosphorylation of JNK1, p38 MAPK, IRS1 (Ser307) and Akt (Ser473) in liver tissues. (B) The p-JNK1/total JNK1, p-P38/total P38, p-IRS1/total IRS1, p-Akt/total Akt ratios. (C) The tumor necrosis factor-alpha (TNF-α) and interleukin (IL)-6 levels in serum. Mice heart blood serum and liver were collected at the end of week 14 after mice were orally administered with TiO2 nanoparticles. The phosphorylation of JNK1, p38 MAPK, IRS1(Ser307) and Akt (Ser473) were measured by Western blot. The bars represent the p-JNK1/total JNK1, p-P38/total P38, p-IRS1/total IRS1, p-Akt/total Akt ratios were normalized to control which were defined as one unit. * P < 0.05 vs. the control group. Results are the mean ± SE (n = 10).
significantly increased from week 10 in both the LD and HD groups, and there was no significant difference between LD and HD (Fig. 2A). The measurement of mice plasma glucose levels of tail vein blood was kept for 4 weeks. After being determined that the increase in plasma glucose was not an isolated phenomenon, we also analyzed plasma glucose levels in heart blood, and found that the plasma glucose of heart blood was also noticeably increased (Table 2). An increase in food intake will induce obesity, and thus increase plasma glucose (Anabela and Carlos, 2006; Robertson, 2006); however, the body weight and food intake results, which were measured when mice were treated with TiO2 nanoparticles by oral administration, showed that there was no change in food intake or body weight in the LD and HD groups compared with the control group (Fig. 2B and C). Therefore, the increase in plasma glucose was caused by long-term TiO2 nanoparticles treatment, and not food intake or body weight. TiO2 nanoparticles have been shown to have biological toxicity with the ability to increase ROS, which are factors inducing tissue damage (Colvin, 2003; Long et al., 2006). Normally, ROS can be cleared by some antioxidants, such as SOD and GSH. However, as the report showed (Wu et al., 2009), after the skin of hairless mice was exposed to TiO2 nanoparticles, the activity of SOD and the levels of GSH in the skin and liver were greatly reduced, and the levels of MDA, which is a ROS production, was significantly increased. In line with this, we found in the present study that the activity of T-SOD and the level of GSH were significantly reduced (Fig. 2A and B), the level of MDA was significantly increased
J. Appl. Toxicol. 2015
(Fig. 2C). These results suggest that TiO2 nanoparticles can be absorbed via oral administration and can then increase ROS. ROS increases plasma glucose via two mechanisms. The first is ROS-induced apoptosis of pancreatic islet β-cells and a reduction in insulin secretion (Robertson, 2006; Hou et al., 2013). The TUNEL is a commonly used method for detecting apoptosis (Yeo et al., 2013). We, therefore, evaluated β-cell apoptosis by TUNEL, but did not find any TUNEL-positive β-cells in either the LD or HD groups, indicating that TiO2 nanoparticle-increased ROS did not induce β-cell apoptosis in this study (Fig. 4). In addition, compared with the control group, there was no significant difference in insulin secretion by mice in the fasting or OGTT states in either the LD or HD group (Table 2, Fig. 5B). These results indicate that the method by which TiO2 nanoparticles increase plasma glucose does not involve insulin secretion. The second mechanism is ROS impairment of insulin action in peripheral target tissues, such as the liver (Gurevitch et al., 2012; Lukic et al., 2014). In IR models (both human and animal), insulin secretion is at normal levels, but it cannot upregulate plasma glucose intake effectively (Fridlyand and Philipson, 2006; Gurevitch et al., 2012). The OGTT experiment results showed that there was no significant difference in the insulin secretion levels of mice in the control, LD, or HD groups (Fig. 5B), but plasma glucose increased significantly and reduced more slowly in the LD and HD groups (Fig. 5A), indicating that oral administration of TiO2 nanoparticle increases ROS-induced IR in mice. Normally, when the insulin receptor receives an insulin signal, IRS1 is activated by
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jat
H. Hu et al. phosphorylation of its tyrosine residues and dephosphorylation of its serine/threonine residues, thus allowing the insulin signal to pass downstream, and to upregulate plasma glucose intake and glycogen synthesis (Lukic et al., 2014; Philippe et al., 2005). ROS lead to the activation of multiple serine/threonine kinase signaling cascades, and to phosphorylation of the serine/threonine residues of IRS1, thus inducing IR in mice (Anabela and Carlos, 2006; Robertson, 2006). In this study, in order to investigate whether the increase in ROS owing to TiO2 nanoparticles induced IR in mice, we analyzed the insulin signaling pathway, and found that phosphorylation of IRS1 (Ser307) was significantly increased in the livers of mice in the LD and HD groups, indicating that the increase in ROS due to TiO2 nanoparticles did indeed induce IR in mice (Fig. 6A and B). In addition, to test whether ROS can inhibit the passage of insulin signals downstream, we evaluated the phosphorylation of Akt, a serine/threonine kinase located downstream of IRS1 in the insulin signaling cascade. Dephosphorylation of Akt (Ser473) results in suppression of the insulin signal (Hou et al., 2013; Schubert et al., 2000). We found that phosphorylation of Akt (Ser473) was significantly decreased in the livers of mice in the LD and HD groups, indicating that the TiO2 nanoparticle-increased ROS was able to induce IR and inhibit the passage of the insulin signal downstream in mice (Fig. 6A and B). Therefore, the mechanism by which TiO2 nanoparticles increase plasma glucose in mice is by increasing ROS-induced IR, thus reducing plasma glucose uptake and resulting in increased plasma glucose. Finally, we investigated the mechanism by which TiO2 nanoparticles can increase ROS-induced IR in mice. ROS has been reported to activate inflammatory cytokines such as TNF-α and IL-6 (Evans et al., 2003; Houstis et al., 2006), both of which promote serine phosphorylation of IRS1, thereby impairing insulin signaling and resulting in IR (Gurevitch et al., 2012; Lukic et al., 2014). The level of TNF-α was significantly increased in the serum of mice in the LD and HD groups after oral administration of TiO2 nanoparticles, and the level of IL-6 was higher in the HD group (Fig. 6C). In addition, ROS induces IR by phosphorylating IRS1 (Ser307) via activation of MAPK pathways, including JNK/stress-activated protein kinase (SAPK) and p38 MAPK (Anabela and Carlos, 2006; Gurevitch et al., 2012). The western blotting data showed that JNK1 and p38 MAPK were highly phosphorylated in the liver of mice in the LD and HD groups (Fig. 6A and B). These results imply that TiO2 nanoparticle-increased ROS can activate the inflammatory response and MAPK pathways, and thus phosphorylate the serine of IRS1, inhibiting the insulin signal from passing downstream, and resulting in IR in mice.
Conclusions TiO2 nanoparticles, which are widely utilized in many fields such as pharmaceuticals, cosmetics, paints and paper, have been reported to cause a toxicology effect on human and animals. In our daily lives, the primary exposure route is diet. Thus, in this study, we administered TiO2 nanoparticles orally to mice and detected the endocrine effects. The results show that, when exceed a certain concentration, TiO2 nanoparticles can be absorbed via oral administration and accumulated in the liver, spleen, kidney, pancreas and small intestine of mice. The accumulation of TiO2 nanoparticles increases ROS levels. Most importantly, this study finds that oral administration of TiO2 nanoparticles increases plasma glucose in mice. The mechanism that the increase in ROS caused by TiO2 nanoparticles affects plasma glucose homeostasis is not the
wileyonlinelibrary.com/journal/jat
pancreatic β-cell apoptosis or insulin secretion, but it is that TiO2 nanoparticle-increased ROS activate the inflammatory response and MAPK pathways, thus inducing IR and resulting in increased plasma glucose in mice. Acknowledgments This work was supported by funds of the National Natural Science Foundation of China (Grant No. 31271593, 31040054), the Natural Science Foundation of Heilongjiang Province (Grant No. LC2012C07), the Open Project of State Key Laboratory of Urban Water Resource and Environment of Harbin Institute of Technology (Grant No. ES201115, ES201512), the National Funds for Creative Research Group of China (Grant No. 51121062).
Conflict of Interest The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References Anabela PR, Carlos MP. 2006. Diabetes and mitochondrial function: Role of hyperglycemia and oxidative stress. Toxicol. Appl. Pharm. 212: 167–178. Barzilay JI, Abrabam L, Heckbert SR, Cushman M, Kuller LH, Resnick HE, Tracy RP. 2001. The relation of markers of inflammation to the development of glucose disorders in the elderly: the cardiovascular health study. Diabetes 50: 2384–2389. Bermudez E, Mangum JB, Wong BA, Asgharian B, Hext PM, Warheit DB, Everitt JI. 2004. Pulmonary responses of mice, rats, and hamsters to subchronic inhalation of ultrafine titanium dioxide particles. Toxicol. Sci. 77: 347–357. Burello E, Worth AP. 2011. A theoretical framework for predicting the oxidative stress potential of oxide nanoparticles. Nanotoxicology 5: 228–235. Cho WS, Kang BC, Lee JK, Jeong J, Che JH, Seok SH. 2013. Comparative absorption, distribution, and excretion of titanium dioxide and zinc oxide nanoparticles after repeated oral administration. Part. Fibre Toxicol. 10: 9. Colvin VL. 2003. The potential environmental impact of engineered nanomaterials. Nat. Biotechnol. 21: 1166–1170. Dobrzyńska MM, Gajowik A, Radzikowska J, Lankoff A, Dušinská M, Kruszewski M. 2014. Genotoxicity of silver and titanium dioxide nanoparticles in bone marrow cells of rats in vivo. Toxicology 315: 86–91. Evans JL, Goldfine ID, Maddux BA, Grodsky GM. 2003. Are oxidative stressactivated signaling pathways mediators of insulin resistance and β-cell dysfunction? Diabetes 52: 1–8. Fridlyand LE, Philipson LH. 2006. Reactive species and early manifestation of insulin resistance in type 2 diabetes. Diab. Obes. Metab. 8: 1362–1451. Gurevitch D, Shuster-Meiseles T, Nov O, Zick Y, Rudich A, Rudich Y. 2012. TiO2 nanoparticles induce insulin resistance in liver-derived cells both directly and via macrophage activation. Nanotoxicology 6: 804–812. Helinor JJ, Gary RH, Frans MC, Sheona P, Steve H, Vicki S. 2009. Identification of the mechanisms that drive the toxicity of TiO2 particulates: the contribution of physicochemical characteristics. Part. Fibre Toxicol. 6: 33. Hood E. 2004. Nanotechnology: looking as we leap. Environ. Health Perspect. 112: A740–A749. Hou GJ, Li CN, Liu SN, Huan Y, Liu Q, Sun SJ, Li LY., Hou SC, Shen ZF. 2013. Long-term treatment with EXf, a peptide analog of Exendin-4, improves β-cell function and survival in diabetic KKAy mice. Peptides 40: 123–132. Houstis N, Rose ED, Lander ES. 2006. Reactive oxygen species have a causal role in multiple forms of insulin resistance. Nature 440: 9442–9481. Hussain S, Thomassen LC, Ferecatu I, Borot MC, Andreau K, Martens JA, Fleury J. 2010. Carbon black and titanium dioxide nanoparticles elicit distinct apoptotic pathways in bronchial epithelial cells. Part. Fibre Toxicol. 7: 10. Lindenschmidt RC, Driscoll KE, Perkins MA, Higgins JM, Mourer JK, Belfiore KA. 1990. The comparison of a fibrogenic and two non-fibrogenic dusts by bronchoalveolar lavage. Toxicol. Appl. Pharmacol. 102: 268–281. Long TC, Saleh N, Tilton RD, Lowry GV, Veronesi B. 2006. Titanium dioxide (P25) produces reactive oxygen species in immortalized brain microglia (BV2): implications for nanoparticle neurotoxicity. Environ. Sci. Technol. 40: 4346–4352.
Copyright © 2015 John Wiley & Sons, Ltd.
J. Appl. Toxicol. 2015
TiO2 nanoparticle increases plasma glucose Lukic L, Lalic NM, Rajkovic N, Jotic A, Lalic K, Milicic T, Seferovic JP, Macesic M, Gajovic JS. 2014. Hypertension in Obese Type 2 Diabetes Patients is Associated with Increases in Insulin Resistance and IL-6 Cytokine Levels: Potential Targets for an Efficient Preventive Intervention. Int. J. Environ. Res. Public Health 11: 3586–3598. Maynard AD, Aitken RJ, Butz T, Colvin V, Donaldson K, Oberdörster G, Philbert MA, Ryan J, Seaton A, Stone V, Tinkle SS, Tran L, Walker NJ, Warheit DB. 2006. Safe handling of nanotechnology. Nature 444: 267–269. Nowwack B, Bucheli TD. 2007. Occurrence, behavior and effects of nanoparticles in the environment. Environ. Pollut. 150: 5–22. Ophus EM, Rode LE, Gylseth B, Nicholson G, Saeed K. 1979. Analysis of titaniumpigments in human lung tissue. Scand. J. Work Environ. Health 5: 290–296. Philippe G, Yannick LM, Jean FT. 2005. Positive and negative regulation of insulin signaling through IRS1 phosphorylation. Biochimie 87: 99–109. Robertson RP. 2006. Oxidative stress and impaired insulin secretion in type 2 diabetes. Curr. Opin. Pharmacol. 6: 615–619. Schubert KM, Scheid MP, Duronio V. 2000. Ceramide inhibits protein kinase B/Akt by promoting dephosphorylation of serine 473. J. Biol. Chem. 275: 13330–13335.
J. Appl. Toxicol. 2015
Sha B, Gao W, Wang S, Li W, Liang X, Xu F, Lu TJ. 2013. Nano-titanium dioxide induced cardiac injury in rat under oxidative stress. Food Chem. Toxicol. 58: 280–288. Shi H, Magaye R, Castranova V, Zhao J. 2013. Titanium dioxide nanoparticles: a review of current toxicological data. Part. Fibre Toxicol. 10: 15. Sohm B, Immel F, Bauda P, Pagnout C. 2015. Insight into the primary mode of action of TiO2 nanoparticles on Escherichia coli in the dark. Proteomics 15: 98–113. Wang JX, Zhou GQ, Chen CY, Yu HW, Wang TC, Ma YM, Jia G, Gao YX, Li B, Sun J, Li YF, Jiao F, Zhao YL, Chai ZF. 2007. Acute toxicity and biodistribution of different sized titanium dioxide particles in mice after oral administration. Toxicol. Lett. 168: 176–185. Wu JH, Liu W, Xue CB, Zhou SC, Lan FL, Bi L, Xu H, Yang X, Zheng FD. 2009. Toxicity and penetration of TiO2 nanoparticles in hairless mice and porcine skin after subchronic dermal exposure. Toxicol. Lett. 191: 1–8. Yeo SH, Noh JR, Kim YH, Gang GT, Kim SW, Kim KS, Hwang JH, Shong M, Lee CH. 2013. Increased vulnerability to β-cell destruction and diabetes in mice lacking NAD(P)H: quinone oxidoreductase 1. Toxicol. Lett. 219: 35–41.
Copyright © 2015 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jat
