Calcium-deficient diet attenuates carbon tetrachloride-induced hepatotoxicity in mice through suppression of lipid peroxidation and inflammatory response

The aim of this study is to investigate whether a Ca-deficient diet has an attenuating effect on carbon tetrachloride (CCl4)-induced hepatotoxicity. Four-week-old male ddY mice were fed a Ca-deficient diet for 4 weeks as a part of the experimental protocol. While hypocalcemia was observed, there was no significant change in body weight. The CCl4-exposed hypocalcemic mice exhibited a significant decrease in alanine aminotransferase and aspartate aminotransferase activities at both 6 h and 24 h even though markers of renal function remained unchanged. Moreover, lipid peroxidation was impaired and total antioxidant power was partially recovered in the liver. Studies conducted in parallel with the biochemical analysis revealed that hepatic histopathological damage was attenuated 24 h post CCl4 injection in hypocalcemic mice fed the Ca-deficient diet. Finally, this diet impaired CCl4-induced inflammatory responses. Although upregulation of Ca concentration is a known indicator of terminal progression to cell death in the liver, these results suggest that Ca is also involved in other phases of CCl4-induced hepatotoxicity, via regulation of oxidative stress and inflammatory responses.


Introduction
Carbon tetrachloride (CCl 4 ) is widely used in experimental animal models that are meant to mimic human hepatotoxicity. The mechanism of CCl 4 hepatotoxicity has been thoroughly studied since the 1970s, by using in vivo models of acute and chronic CCl 4 poisoning, perfused livers, and isolated or cultured hepatocytes [1,2]. CCl 4 -induced toxicity is a multifactorial process involving the generation of free radicals [2,3,4,5]. The first step is the metabolic activation of CCl 4 by CYP2E1.
Subsequently, CCl 4 is converted to free radicals (trichloromethyl and trichloromethyl peroxy radicals). The second step is radical binding; the free radicals react with antioxidant enzymes and sulfhydryl groups such as those in glutathione (GSH) and the protein thiol. The third step involves the overexpression of these free radicals leading to several deleterious effects such as enhanced membrane lipid peroxidation, covalent binding to macromolecules, ATP depletion, generation of inflammatory cytokines, and loss of Ca homeostasis [6,7,8]. Since sulfhydryl groups are essential elements of the molecular arrangement responsible for the Ca transport access cellular membranes, loss of these proteins inhibits microsomal and mitochondrial regulation of cellular Ca levels.
Cadmium (Cd) has been classified by the International Agency for Research on Cancer as a group I carcinogen and is a ubiquitous contaminant of the environment and dietary product. Exposure to Cd is known to cause hepatic injury in acute toxicity and renal injury in chronic toxicity [9]. Cd-related toxicity is also a multifactorial process [10,11,12,13]. The first step of which is GSH depletion.
GSH depletion raises the level of lipid peroxidation in the cell membrane and mitochondrial dysfunction occurs as the next step. In addition, after these actions, disruption of calcium homeostasis and calcium uptake is observed. These mechanisms indicated that calcium uptake is the terminal phase of cell death. However, Acosta and Sorensen reported that Cd-induced cytotoxicity is impaired in a Ca-free medium in vitro [14]. This phenomenon was also observed by our investigations (data not shown, manuscript in progress). These data suggests that Ca is directly involved in Cd-induced toxicity, not only in the terminal phase, but also in other phases. These can be considered closely similar to the phases of C Cl 4 -induced toxicity. Therefore, we hypothesized that Ca could exacerbate CCl 4 -induced toxicity as well.
To address this, the current study was carried out to investigate whether hypocalcemia in mice decreases CCl 4 -induced toxicity or not. To examine this, we Article No~e00126 fed mice a Ca-deficient diet and determined plasma biochemical markers, hepatic lipid peroxidation, and the hepatic inflammatory response.

Animal treatment
Male ddY mice were purchased from Japan SLC (Shizuoka, Japan) at 3 weeks of age. The mice were maintained under standard conditions of controlled temperature (24 ± 1°C), humidity (55 ± 5 %), and light (12:12 h light/dark cycles) with free access to water and food. After acclimatization to a normal diet (CE-2; Clea Japan, Inc., Tokyo, Japan; [protein (soybean waste, whitefish meal, yeast): 24.9 %; carbohydrate (wheat flour, corn, Milo): 51.0 %; fat (cereal germ, soybean oil): 4.6 %; Ca: 1.06 g; and the other: 3.59 kcal/g]) for 1 week, the mice were divided into 2 groups of 8 or 9 each. One group was fed the CE-2 diet, and the other group was fed a Ca-deficient diet based on AIN-93 G (Oriental Yeast Co., Tokyo, Japan) [protein (casein, L-cysteine): 20.0 %; carbohydrate (corn, maltodextrin, sucrose): 64.0 %; fat (soybean oil, t-butylhydroquinone): 7.0 %; Ca: 0 g; and the other: 3.90 kcal/g] for 4 weeks. Food intake was monitored and body weight was measured once per week throughout the study. We collected blood samples from each mouse every 2 weeks to confirm the effects of the Cadeficient diet on plasma Ca concentrations. After a final plasma Ca determination at 8 weeks of age, each mouse was injected intraperitoneally (i.p.) with 2 g/kg (at 5 mL/kg) CCl 4 (Wako Chemical, Osaka, Japan). After 6 h and 24 h, blood samples were collected from each group of mice. Whole blood was centrifuged (3000 g, 10 min) and the supernatant was tested for hepatic and renal injury markers. The liver from each animal was harvested 24 h after CCl 4 injection, and separate samples were stored at −80°C or fixed in 15 % neutral buffered formalin (pH 7.2). All experiments were approved by the Institutional Animal Care and Experiment Committee of Kinjo Gakuin University (NO.110).

Plasma biochemical analysis
Plasma Ca levels were measured using the calcium-E test (Wako Chemical) according to the manufacturer's instructions. Plasma samples (2.5 μL) were mixed with the substrate buffer (100 μL) and a coloring reagent (50 μL). The absorbance of the reaction mixture was measured at 610 nm.
Plasma alanine aminotransferase (ALT) and aspartate aminotransferase (AST) activities were measured using the Transaminase CII Test Wako (Wako Chemical) according to the manufacturer's instructions and as previously described [15,16]. Concentrations of plasma creatinine and blood urea nitrogen (BUN) were measured using the Creatinine Liquid Reagents Assay (DIAZYME, Poway, CA, USA) and the BUN Wako Test (Wako Chemical), according to the manufacturer's Article No~e00126 instructions and as previously described [17]. For relative quantification, calibration curves were prepared using a standard solution.

Measurement of thiobarbituric acid levels in the liver
The total liver thiobarbituric acid (TBA) levels and antioxidant power were examined by a colorimetric microplate assay (Oxford biochemical research, Oxford, MI, USA) according to the manufacturer's protocol and as previously described [16,17].

Histopathological findings
For histological analysis, a portion of the left lobe of the liver from each animal was perfused with 15 % phosphate-buffered neutral formalin (pH 7.2: Wako Chemical), dehydrated, and embedded in paraffin. Embedded tissues were sectioned at 4 μm thickness and stained with hematoxylin and eosin (H&E) or periodic acid Schiff (PAS) using standard methodologies. Histopathological features of the slices were observed using a light microscope.  Table 1. Relative expression of each mRNA was determined using the standard curve method. The amount of each target mRNA quantified was normalized against that of GAPDH-encoding mRNA.

Measurement of hepatic tumor necrosis factor (TNF)-a level by enzyme-linked immunosorbent assay (ELISA)
Aliquots (0.1 g each, including mixed cell types) of hepatic tissue were homogenized in 900 μL ice-cold phosphate-buffered saline (PBS) containing a protease inhibitor (Nacalai Tesque, Kyoto, Japan) and centrifuged at 18000 g for 20 min at 4°C. The resulting supernatant (diluted to yield consistent total protein concentrations) for each sample was used for further steps. Hepatic levels of TNF-α were determined using a commercially available ELISA kit (eBioscience, San Diego, CA, USA), according to the manufacturer's instructions. TNF-α concentrations were determined from a standard curve, and were expressed as pg/mL.

Statistical analysis
All data from the control and treatment groups were obtained from the same numbers of replicated experiments. All experiments were performed independently at least twice. Comparisons between the two groups were made using Student's t test or Welch's t test and multiple comparisons were analyzed using One Way ANOVA with post-hoc Tukey-Kramer's test. All statistical analyses were performed using SPSS 19.0J software (Chicago, IL, USA). Values of P < 0.05 were considered statistically significant.

Effect of the Ca-deficient diet on biochemical markers and body weight in CCl 4 -induced toxicity
Plasma Ca concentrations decreased after 2 weeks of feeding and 28 % suppression was observed after 4 weeks (Fig. 1A). In addition to plasma Ca concentrations, the levels of ALT, AST (markers of hepatic injury), creatinine, and BUN (markers of renal injury) were compared between the normal diet and Ca-deficient diet groups (data not shown). In this study, no significant changes in body weight gain, or food intake were observed (Fig. 1B, Table 2).

Effect of the Ca-deficient diet on hepatic and renal injury markers in acute CCl 4 toxicity
To determine how CCl 4 -induced toxicity is impaired under hypocalcemic conditions, we examined hepatic injury markers ALT and AST, whose activities increase in CCl 4 -induced toxicity. As shown in Fig. 2, mice feeding on a

Article No~e00126
Ca-deficient diet had significantly reduced ALT and AST activities at both 6 h and 24 h post injection.
In addition to ALT and AST, creatinine and BUN were also evaluated as markers of renal injury. As shown in Table 3, although CCl 4 increases both creatinine and BUN, the levels of these parameters in our hypocalcemic mice were comparable.

Evaluation of Ca-deficient diet against CCl 4 acute toxicity on TBA and total antioxidant levels in the liver
To further investigate hypocalcemia-induced attenuation of CCl 4 liver toxicity, we measured TBA levels as a marker of lipid peroxidation. CCl 4 treatment significantly increased hepatic TBA levels in mice on the normal diet while a partial reduction in the upregulated TBA levels was observed in mice on the Cadeficient diet (Fig. 3A).
Studies suggest that total antioxidant power may be used as an indicator of oxidative stress levels. As shown in Fig. 3B, CCl 4 treatment markedly decreased the total antioxidant power in the mouse liver. However, mice on a Ca-deficient diet recovered the lost antioxidant power.
[ ( F i g . _ 1 ) T D $ F I G ]  [ ( F i g . _ 2 ) T D $ F I G ]  In parallel with the measurement of functional markers ( Fig. 2 and Fig. 3, Table 3), we conducted histopathological studies. Liver sections obtained from the control group and stained with H&E showed normal cell morphology, well-preserved cytoplasm, and a clear, plump nucleus (Fig. 4A). The CCl 4 -injected mice on a normal diet showed signs of extensive necrosis (especially in the acinus, zone 3) (Fig. 4B), while those on a Ca-deficient diet counteracted some, but not all, of this liver necrosis (Fig. 4C). In the normal livers, glycogen granules accumulated diffusely in the hepatocytes, as shown by PAS staining (Fig. 4D). However, intrahepatic glycogen was almost completely depleted in the liver sections of CCl 4exposed mice in the normal diet group (Fig. 4E), while the livers of the CCl 4exposed animals in the Ca-deficient diet group recovered some hepatic glycogen content (Fig. 4F).

Estimation of hypocalcemic effect on CCl 4 -induced inflammatory response and CYP induction
It has been reported that inflammation plays an important role in CCl 4 -induced liver injury. In order to confirm involvement of Ca in this response, we determined the mRNA levels of inflammatory cytokines. As shown in Fig. 5, CCl 4 injection increased both tumor necrosis factor-α (TNF-α) (A) and interleukin-6 (IL-6) (C) mRNA levels. Mice on a Ca-deficient diet showed a decrease in some, but not all, of these parameters. In addition, the protein levels of TNF-α (B) showed a similar trend. In parallel with the measurement of inflammatory cytokines, we evaluated CYP2E1 mRNA expression (D). CCl 4 injection significantly decreased the levels of CYP2E1 mRNA in both feeding groups, but no marked differences between them were observed.

Discussion
Our current study demonstrates that a Ca-deficient diet attenuates CCl 4 -induced hepatotoxicity, but does not decrease renal toxicity. Findings from other researchers and from our previous investigation demonstrated that CCl 4 induced severe hepatotoxicity and renal toxicity [15,16,17]. In the current study, hypocalcemia-induced attenuation of CCl 4 toxicity is only observed in hepatic biochemical analysis (ALT and AST), while the levels of plasma markers of renal injury (creatinine and BUN) remained unchanged. As a Ca-deficient diet significantly decreased plasma Ca levels, this suggests that plasma Ca might preferentially effect change in the liver.
CCl 4 is metabolized to its active form by CYPs including CYP2E1 and CYP2B family [18,19,20]. In fact, pretreatment with a CYP2E1 inhibitor, like an antibody or a natural product (Antrodia camphorata) attenuates CCl 4 -induced hepatotoxicity [21,22,23,24]. In addition, Wong et al. reported that the CYP2E1 KO mouse is resistant to CCl 4 toxicity [4]. Although CYP2E1 is also expressed in the kidney, the level of expression is much lower than that in the liver [25,26]. We hypothesized that different CYP2E1 levels may be one cause for the relatively higher hepatic sensitivity to CCl 4 toxicity. However, despite the markedly decreased levels of CYP2E1 mRNA upon CCl 4 injection, no significant difference [ ( F i g . _ 4 ) T D $ F I G ] in these levels was observed between the normal diet and the Ca-deficient diet groups. Thus, Ca-deficient diet does not seem to be involved in altering CYP induction.
CCl 4 is widely used to investigate hepatic injury associated with oxidative stress and free radicals. The reactive oxygen species induced by CCl 4 not only cause direct tissue damage, but also initiate inflammation through the activation of various cytokines [24,27,28,29]. Oxidative stress has been postulated to be a major molecular mechanism in acute liver injury induced by CCl 4 [2,30,31].
Increased TBA, a lipid peroxidative product of cell membranes, was partially attenuated by feeding the mice a Ca-deficient diet. This suggests that the protective effects of a Ca-deficient diet may be partly due to counteraction of oxidative stress in acute liver injury. In addition to oxidative stress, an inflammatory response was shown to be involved in the process of CCl 4 -induced acute chemical liver injury [28,29,32]. In the present study, CCl 4 -intoxicated mice on a Ca-deficient diet exhibited significant reduction in the inflammatory response compared to the mice [ ( F i g . _ 5 ) T D $ F I G ] in the normal diet group, suggesting that the beneficial effect of a Ca-deficient diet may be partly due to impairment of inflammatory response caused by CCl 4 .
The extracellular plasma Ca concentration is tightly controlled by hormones and by a complex homeostatic mechanism involving fluxes of Ca between the extracellular fluid, kidney, and bones. It has been reported that CCl 4 disrupts hepatic Ca homeostasis [33,34]. In our study, hepatic Ca concentration is not significantly altered by a Ca-deficient diet although there is a 28 % reduction in plasma Ca levels. There has been evidence that the upregulation of cytosolic Ca concentration is a terminal event in the progression to cell death in toxic liver injury [35]. However, the present study suggests that plasma Ca might also be a candidate trigger for mediating CCl 4 -induced toxicity. Further investigations will be needed to clarify how Ca is involved in CCl 4 -induced hepatotoxicity in other phases before the terminal phase. To investigate this, we are currently working on a vitamin D3-induced hypercalcemia model.
In conclusion, we demonstrated that a Ca-deficient diet attenuates CCl 4 -induced hepatotoxicity via suppression of lipid peroxidation and inflammatory response. To our knowledge, this is the first study to provide evidence that Ca is involved in CCl 4 -induced hepatotoxicity in a mouse model, not only in the terminal phase but also in other phases. These findings may have potential application in studies of other hepatotoxic compounds.

Declarations
Author contribution statement

Funding statement
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Additional information
No additional information is available for this paper.