Effects of caffeine on brain antioxidant status and mitochondrial respiration in acetaminophen-intoxicated mice

Abstract

Hepatic encephalopathy is a pathophysiological complication of acute liver failure, which may be triggered by hepatotoxic drugs such as acetaminophen (APAP). Although APAP is safe in therapeutic concentration, APAP overdose may induce neurotoxicity, which is mainly associated with oxidative stress. Caffeine is a compound widely found in numerous natural beverages. However, the neuroprotective effect of caffeine remains unclear during APAP intoxication. The present study aimed to investigate the possible modulatory effects of caffeine on brain after APAP intoxication. Mice received intraperitoneal injections of APAP (250 mg/kg) and/or caffeine (20 mg/kg) and, 4 h after APAP administration, samples of brain and blood were collected for the biochemical analysis. APAP enhanced the transaminase activity levels in plasma, increased oxidative stress biomarkers (lipid peroxidation and reactive oxygen species), promoted an imbalance in endogenous antioxidant system in brain homogenate and increased the mortality. In contrast, APAP did not induce dysfunction of the mitochondrial bioenergetics. Co-treatment with caffeine modulated the biomarkers of oxidative stress as well as antioxidant system in brain. Besides, survival assays demonstrated that caffeine protective effects could be dose- and time-dependent. In addition, caffeine promoted an increase of mitochondrial bioenergetics response in brain by the enhancement of the oxidative phosphorylation, which could promote a better energy supply necessary for brain recovery. In conclusion, caffeine prevented APAP-induced biochemical alterations in brain and reduced lethality in APAP-intoxicated mice, these effects may relate to the preservation of the cellular antioxidant status, and these therapeutic properties could be useful in the treatment of hepatic encephalopathy induced by APAP intoxication.

Introduction

The brain is one of the tissues most susceptible to the oxidative damage and redox imbalance. Due to this susceptibility, brain damage may occur when other organs, such as liver, are affected [1, 2]. For instance, encephalopathy in acute liver failure has been demonstrated to be induced by high doses of acetaminophen (APAP) [2]. APAP is widely consumed because of its analgesic and antipyretic properties [3] and at high doses can induce a critical damage to the structure and cellular components of liver [4]. The neurotoxicity associated with APAP-induced acute liver failure may be related to the formation of the toxic metabolite, N-acetyl-p-benzoquinone imine (NAPQI), reactive species generation and accumulation of ammonia [2, 5, 6]. A common factor associated with this disorder is the redox imbalance, which could lead, at least in part, to the disruption of brain homeostasis, triggering a high mortality rate due to the deleterious effects to the central nervous system. Among these effects are development of brain edema and increased intracranial pressure, and often glutamate-mediated excitotoxicity, abrupt onset of delirium, seizures and coma [7, 8]. In addition, APAP is able to induce mitochondrial dysfunction, triggering reactive oxygen species (ROS) generation and mitochondrial transition permeability in brain [2].

As APAP hepatotoxicity is followed by significant enhance of oxidative stress, the consequent redox imbalance could be efficiently prevented by treatment with antioxidant [9, 10]. Caffeine (1,3,7-trimethylxanthine) is commonly found in beverages, food and in combination with APAP in popular analgesic products and cold medicine formulations. Caffeine has been studied due to anti-inflammatory, biochemical and physiological effects, which can protect the tissue during the liver injury [11, 12]. Moreover, it is able to prevent the oxidative damage induced by radiation, alcoholic and non-alcoholic hepatotoxicity [11, 13, 14]. In this context, caffeine have demonstrated modulatory effects on mitochondrial physiology, such as adaptation of mitochondrial glutathione (GSH) redox system, calcium homeostasis and bioenergetics parameters [15–17]. Many works have focused on the effects of caffeine on hepatic dysfunction associated to APAP intoxication [18, 19], whereas the caffeine effects on neurotoxicity associated to APAP overdose remains poorly understood.

Caffeine presents ergogenic properties, which is observed by the increase of mitochondrial metabolic rate [20]. As the brain have a high bioenergetics demand due to energetic consumption by the ATP-dependent Na+, K+-pump [2], the impairment of mitochondrial functioning could be fatal for the maintenance of normal brain functions. Mitochondrial bioenergetics maintenance is associated to the oxidative phosphorylation (OXPHOS) and the electron transport system [21, 22]. The combination of caffeine and APAP was demonstrated to enhance the APAP efficacy as well as the liver microsome metabolism, although this was accompanied by enhance of APAP hepatotoxicity [23]. On the other hand, some studies demonstrated that caffeine consumption could reduce the hepatotoxicity biomarkers in serum [24]. Thus, in order to understand the mechanism of caffeine during the neurotoxicity, this work aimed to evaluate the changes in the oxidative stress biomarkers and redox system, as well as the mitochondrial bioenergetics functioning in brain through the high-resolution respirometry (HRR) during the neurotoxicity associated with APAP intoxication.

Materials and Methods

Materials

Caffeine, thiobarbituric acid (TBA), 2′-7′-dichlorofluorescein diacetate (DCFH-DA), trichloroacetic acid (TCA) and nucleotides were purchased from Sigma Chemical Co. (St. Louis, MO, USA). All other chemicals were of analytical grade and obtained from standard commercial suppliers.

Animals

Seven-week-old male adult Swiss albino mice (30–40 g) from our own breeding colony were used. The animals were kept on a separate animal room, at 12-h light/dark cycle, and at temperature of 22 ± 2°C. Mice were acclimated for 7 days before initiation of any procedures. This study was approved by the Ethics and Animal Welfare Committee of Federal University of Santa Maria, Brazil (Permit Number: 3208150915). Animals had free access to water and rodent chow before initiation of any treatment. However, mice were fasted before treatment with APAP as indicated below.

Experimental protocol

Briefly, the mice were randomly divided into following four groups: control (vehicle), caffeine (Caf), APAP treated (APAP) and Caf + APAP. All the solutions were administered by the intraperitoneal (i.p.) route and prepared in saline 0.9%. Injections were administered at 9:00 a.m. in order to remove any confounding factors of circadian rhythm. Mice were fasted for 16 h before injection of APAP beginning with the removal of food at 5:00 p.m. on the previous day. First, mice in the control and caffeine groups received saline 0.9% (20 ml/kg) and caffeine 20 mg/kg, respectively, and the APAP and Caf + APAP groups received APAP 250 mg/kg (15 ml/kg). APAP and caffeine doses have been used in previous studies [25, 26] and animals received first caffeine and 30 min after APAP. At 4 h following APAP administration, animals were killed by cervical dislocation, brain was collected and blood also was collected by cardiac puncture using heparin-rinsed 1-ml syringes (20 gauge needles) and centrifuged. The plasma was used for determination of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) activities using a commercial kit (Labtest^®, Diagnostica S.A., Minas Gerais, Brazil).

Brain homogenate preparation

At the end of the treatment period the animals were euthanized by cervical dislocation, perfused in petri dish with 5 ml of cold Tris buffer (10 mM Tris-HCl, pH 7.4) to remove the contaminating blood, in according to previously described protocol with some modifications [27]. So, a sagittal section was made in brain, cerebellum and brain stem were removed and one brain hemisphere was collected and homogenized in Tris buffer with a pre-cooled glass potter. Brain homogenates were centrifuged at 2000 × g for 10 min to separate large tissue fragments and yield the supernatant fractions with cellular components, which were used for the biochemical assays. The protein content of homogenate brain preparation was quantified by Bradford’s assay [28].

APAP survival challenge

For survival studies, mice were injected with caffeine and APAP in according to protocol described earlier, then 30 min later the mice were returned to their cages and fed with food and water ad libitum. To determine the effect of caffeine on mortality of APAP-administered mice, the survival rate after APAP administration was evaluated for 72 h.

Measurement of lipid peroxidation

Brain homogenate lipid peroxidation (LPO) was quantified measuring the malondiadehyde (MDA) levels. In summary, brain homogenate were incubated in 300 μl of a medium consisting of 175 mM KCl and 10 mM Tris-HCl, pH 7.4, and then were added to color reaction. TBA reactive substances levels were measured at 532 nm using a standard curve of MDA [29].

Measurement of ROS production

ROS generation was determined spectrofluorimetrically in vitro using H₂DCF-DA levels as an index of the peroxide production by cellular components (1 μM) [30]. Briefly, the brain homogenate was added to standard medium and the fluorescence was determined at 488 nm for excitation and 525 nm for emission, with slit widths of 3 nm. Results are expressed as increase of fluorescence regarding fluorescence measured in control group, which was used as standard.

Determination of reduced (GSH) and oxidized glutathione (GSSG)

For the measurement of GSH and GSSG levels, the brain homogenate was treated with 0.5 ml of 13% TCA and centrifuged at 13 000 rpm for 10 min at 4°C. Aliquots (100 μl) of the supernatant were mixed with 2 ml of 100 mM NaH₂PO₄ buffer, pH 8.0, containing 5 mM EDTA. One hundred microliters of O-phthalaldehyde (1 mg/ml) was added and fluorescence was measured 15 min later using the 350/420 nm excitation/emission wavelength pair, with slit widths of 3 nm [31]. For measurement of GSSG levels, a 250 μl of the brain supernatant was incubated at room temperature with 100 μl of N-ethylmaleimide (0.04 M) for 30 min at room temperature, and after that 140 μl of the mixture were added to 1760 μl of NaOH (0.1 N) buffer, following of added 100 μl OPT and incubated for 15 min, using the procedure outlined above for GSH assay. Results are expressed in nmol of GSH and corrected by protein content of each sample.

Total superoxide dismutase activity

Brain homogenate total superoxide dismutase (SOD) activity was measured by the capacity of inhibiting auto-oxidation of adrenaline to adrenochrome at 480 nm [32]. The brain supernatant (5 mg protein) was added to a medium containing 2 mM EDTA, 50 mM NaHCO₃/Na₂CO₃ buffer (pH 10.3) and adrenaline (4 mM). The SOD activity was analysed in spectrophotometer at 480 nm. The results were expressed in U SOD/mg protein.

Catalase activity

The catalase (CAT) enzyme activity was determined in brain homogenate in according to the method previously proposed [33]. Brain homogenate (5 mg protein) was added to a medium containing potassium phosphate buffer (50 mM KH₂PO₄, 50 mM K₂HPO₄; pH 7.4) and H₂O₂ (1 mM). The kinetic analysis of CAT was started after H₂O₂ addition. The CAT activity was determined using the molar extinction coefficient 36 M⁻¹ cm⁻¹ and the reaction was measured at 240 nm.

Glutathione-S-transferase activity

The glutathione-S-transferase (GST) activity was determined spectrophotometrically [34]. GST activity was quantified in brain homogenates (5 mg protein) in a reaction mixture containing 1 mM 1-chloro-2,4-dinitrobenzene (CDNB) and 1 mM glutathione as substrates in 0.1 M sodium phosphate buffer, pH 6.5, at 37°C. Enzyme activity was calculated by the change in the absorbance value from the slope of the initial linear portion of the absorbance time curve at 340 nm for 5 min. Enzyme activity was determined using the molar extinction coefficient 9.6 mM⁻¹ cm⁻¹ and expressed as nmol CDNB/min/mg protein.

Glutathione peroxidase activity

Glutathione peroxidase (GPx) activity was determined spectrophotometrically at 340 nm by NADPH consumption for 2 min at 30°C [35]. The liver homogenate supernatant (5 mg protein) was added to medium containing 0.1 M phosphate buffer (0.1 M KH₂PO₄, 0.1 M K₂HPO₄ and 5 mM EDTA, pH 7.0), 1 mM GSH, 0.15 mM NADPH, 0.1 U/ml glutathione reductase (GR) and 1 mM sodium azide. So, the reaction was initiated by adding the H₂O₂ to a final concentration of 0.4 mM. The GPx activity was determined using the molar extinction coefficient 6220 M⁻¹ cm⁻¹ and expressed as nmol/min/mg protein.

GR activity

For the measurement of GR activity, the liver homogenate supernatant (5 mg protein) was added to medium containing 0.15 M phosphate buffer (0.15 M K₂HPO₄ and 1.5 mM EDTA, pH 7.0) and 0.15 mM NADPH. The measurements were made at 340 nm and initiated with addition of 20 mM GSSG at 30°C for 2 min [36]. GR activity was determined using the molar extinction coefficient 6220 M⁻¹ cm⁻¹ and expressed as nmol/min/mg protein.

HRR

For the respirometry determination, the mouse brain homogenate (400 mg) was weighed and transferred to 2 ml of cold homogenization buffer containing 5 mM Tris-HCl, 250 mM sucrose and 2 mM EGTA (pH = 7.4), and brain homogenate (0.1 mg/ml) was used to the HRR. Oxygraph-2 k (O2k, Oroboros Instruments, Innsbruck, Austria) was employed for all used for measurements of respiration. The experiments were performed in MiR05 [110 mM sucrose, 60 mM K-lactobionate, 0.5 mM EGTA, 3 mM MgCl₂, 20 mM taurine, 10 mM KH₂PO₄, 20 mM HEPES and 0.1% bovine serum albumin (BSA) essentially fatty acid free, pH 7.1] [22, 37, 38]. All experiments were performed at 37°C using DatLab 4.0 software (Oroboros Inc., Austria), with continuous stirring at 750 rpm, and all experiments started by registering the endogenous substrate supported respiration, following protocols established in the literature [39]. Titration protocols of multiple substrates and inhibitors were used to assess mitochondrial function in terms of different respiration states. Complex I (CI)-mediated “leak” respiration was determined using 2 mM malate and 10 mM glutamate. CI-mediated “OXPHOS” was determined using saturating ADP (2.5 mM). The convergence electron flow, CI and complex II (CII) OXPHOS, was then determined by the addition of 10 mM succinate. CII-mediated “OXPHOS” was determined using 0.5 μM rotenone. Addition of 2.5 μM antimycin A (Ama) inhibited complex III, resulting in non-mitochondrial respiration with small contributions from electron leak in the uncoupled state. Cytochrome c oxidase activity was evaluated using 0.25 mM TMPD and 2 mM ascorbate.

Protein determination

Protein content was determined using BSA as standard [28].

Statistical analysis

Statistical analysis was performed using GraphPad (version 5.0 for Macintosh OSX, GraphPad Software, San Diego, CA, USA). Significance was assessed by two-way analysis of variance, followed by Newman–Keuls Test for post hoc comparison. Values of P ≤ 0.05 were considered statistically significant.

Results

Liver injury biomarkers

In order to analyse the liver damage, the transaminase activity in plasma was determined. APAP treatment induced a significant increase of ALT and AST activities compared with the control, and caffeine did not reduce the transaminase activities levels when compared with the APAP groups (Table 1).

Table 1.

Analyses of transaminase levels in plasma after ALF induced by APAP

	AST	ALT
Control	66.2 ± 1.9 n.s.	45.6 ± 1.0 n.s.
Caffeine	67.2 ± 1.9 n.s.	48.5 ± 0.5 n.s.
APAP	116.3 ± 11.9*	98.0 ± 17.0*
Caf ± APAP	115.9 ± 16.6*	114.4 ± 21.1*

Values are presented as means ± SEM. AST, aspartate aminotransferase (IU/l); ALT, alanine aminotransferase (IU/l); ALF, acute liver failure; n.s., not significant.

^* P ≤ 0.05 from control group by Newman–Keuls Multiple Comparison Test.

^# P ≤ 0.05 from APAP group by Newman–Keuls Multiple Comparison Test.

Markers of oxidative damage in brain

As the oxidative stress plays an important role in the development of several diseases, ROS generation and LPO was investigated in vitro. LPO levels were increased in APAP group as well as the Caf + APAP group when compared with control group (Fig. 1A). APAP group presented significant increase of ROS in relation to the control; on the other hand, the ROS generation levels remained unchanged in the Caf + APAP group (Fig. 1B).

Figure 1.

Effects of treatment with caffeine and APAP on oxidative damage markers in brain homogenate: (A) lipid peroxidation and (B) ROS generation. Mice were given caffeine (20 mg/kg; i.p.) and APAP (250 mg/kg, i.p.) and were killed at 4 h after the APAP treatment. Data are expressed as means ± SEM (n = 5). Significance was assessed by one-way analysis of variance, followed by Newman–Keuls Test for post hoc comparison. Significant differences are indicated by ^*P ≤ 0.05 when compared with control group. Significant difference is indicated by ^#P ≤ 0.05 when compared with APAP group.

CAT and SOD system in brain

Both CAT and SOD activities were significantly reduced during the APAP treatment when compared with control group, whereas the Caf + APAP group presented similar levels to the control group (Fig. 2A and B).

Figure 2.

Effects of treatment with caffeine and APAP on antioxidant enzyme system in brain homogenate: (A) catalase (CAT) activity and (B) total superoxide dismutase (SOD) activity. Mice were given caffeine (20 mg/kg; i.p.) and APAP (250 mg/kg, i.p.) and were killed at 4 h after the APAP treatment. Data are expressed as means + SEM (n = 5). Significance was assessed by one-way analysis of variance, followed by Newman–Keuls Test for post hoc comparison. Significant differences are indicated by ^*P < 0.05 when compared with control group. Significant difference is indicated by ^#P < 0.05 when compared with APAP group.

Glutathione redox system in brain

Homeostasis of glutathione redox system was evaluated. APAP treatment did not change the GSH levels, but it was able to promote the increase of the GSSG levels in vitro, resulting in the reduction of brain GSH/GSSG ratio. GSH levels were reduced in the Caf + APAP group; in contrast, the GSSG levels were similar to the control as well as the GSH/GSSG ratio (Table 2). In addition, levels of GSSG (% of total GSH) were significantly higher than control in APAP group, whereas the Caf + APAP remain unchanged (Table 2).

Table 2.

Analyses of glutathione redox system in brain after ALF induced by APAP

	GSH	GSSG	GSH/GSSG ratio	GSSG (% of total GSH)
Control	36.8 ± 2.9 n.s.	6.1 ± 0.8 n.s.	5.0 ± 0.1 n.s.	13.9 ± 0.4 n.s.
Caffeine	29.7 ± 1.8 n.s.	6.1 ± 0.5 n.s.	5.0 ± 0.5 n.s.	14.6 ± 1.0 n.s.
APAP	34.5 ± 1.9 n.s.	8.8 ± 0.6*	3.5 ± 0.1*	17.8 ± 0.5*
Caf ± APAP	32.2 ± 0.6 n.s.	6.4 ± 0.3#	5.4 ± 0.3#	14.2 ± 0.5#

Values are presented as means ± SEM. GSH levels (nmol GSH/mg protein); GSSG levels (nmol GSSG/mg protein); ALF, acute liver failure; n.s., not significant.

^* P ≤ 0.05 from control group by Newman–Keuls Multiple Comparison Test.

^# P ≤ 0.05 from APAP group by Newman–Keuls Multiple Comparison Test.

Enzymes glutathione redox system in brain

In order to evaluate the enzymatic antioxidant system associated with GSH, GPx, GR and GST activities were determined in vitro. GPx and GST activities were higher than control in APAP group (Fig. 3A and C), whereas the GR activity presented a significant reduction (Fig. 3B). On the other hand, GPx, GR as well as GST remain similar to control levels in Caf + APAP group (Fig. 3A–C).

Figure 3.

Effects of treatment with caffeine and APAP on enzymes of glutathione redox system in brain homogenate: (A) glutathione peroxidase (GPx) activity, (B) glutathione reductase (GR) activity and (C) glutathione S-transferase (GST) activity. Mice were given caffeine (20 mg/kg; i.p.) and APAP (250 mg/kg, i.p.) and were killed at 4 h after the APAP treatment. Data are expressed as means ± SEM (n = 5). Significance was assessed by one-way analysis of variance, followed by Newman–Keuls Test for post hoc comparison. Significant differences are indicated by ^*P ≤ 0.05 when compared with control group. Significant difference is indicated by ^#P ≤ 0.05 when compared with APAP group.

APAP survival challenge

Mice were monitored for 72 h to determinate the effects of treatment. In this way, mice that received APAP presented a significant mortality when compared with control, which was 100% in ~48 h, and the treatment with caffeine (20 mg/kg) demonstrated a significant extension of survival, which was ~33% in 72 h (Fig. 4).

Figure 4.

Mortality of the treatment with caffeine on APAP intoxication (250 mg/kg; i.p.). Survival curve in mice treated with low caffeine dose (20 mg/kg). Survival was followed for 72 h, n = 10 per group.

Mitochondrial bioenergetics parameters in brain

To evaluate the mitochondrial function, O₂ flux consumption was determined by respiratory control ratio (RCR). In this way, APAP intoxication did not presented any significant effect on RCR levels (Table 3).

Table 3.

Analyses of bioenergetics parameters in brain after ALF induced by APAP

	Respiratory control ratio (RCR) for complex I
Control	3.7 ± 0.02 n.s.
Caffeine	4.2 ± 0.1 n.s.
APAP	4.0 ± 0.2 n.s.
Caf ± APAP	3.6 ± 0.3 n.s.

Values are presented as means ± SEM. ALF, acute liver failure; n.s., not significant.

In order to understand the possible changes in the mitochondrial bioenergetics parameters, the OXPHOS capacity was evaluated. APAP and Caf + APAP groups presented a significant increase of CI + CII_Oxphos when compared with the control group (Fig. 5). Caffeine group promoted the enhance of OXPHOS capacity associated to the convergent electron flow simultaneously through complexes I and II (CI + CII_Oxphos) when compared with the control, whereas the APAP and Caf + APAP groups remained unchanged (Fig. 5). On the other hand, both caffeine and APAP enhanced the CII_Oxphos when compared with the control group, and the Caf + APAP group was significantly different from control (Fig. 5). The residual oxygen consumption (ROX) was determined with addition of Ama, the levels of ROX were increased in APAP and Caf + APAP groups (Fig. 5). The APAP intoxication did not present significant effects on COx activity; in contrast, Caf + APAP group had significant increase of COx activity when compared with the control group (Fig. 5).

Figure 5.

HRR representative in brain homogenate of the treatment with caffeine on APAP intoxication (250 mg/kg; i.p.). Mitochondrial function are presented with the abbreviation(s) of the complex(es) involved followed by the state of respiration measured in presence of glutamate + malate (CI_Leak), + ADP (CI_Oxphos), + succinate (CI + CII_Oxphos), + rotenone (CII_Oxphos), + antimycin A (Ama). Ama was used for correction of residual O₂ consumption. Significance was assessed by one-way analysis of variance, followed by Newman–Keuls Test for post hoc comparison. Results are expressed as means ± SEM for four different preparations. ^*Significant difference is indicated by ^#P ≤ 0.05 when compared with APAP group.

Discussion

This work was performed with the purpose to understand the mechanism of caffeine during the neurotoxicity induced by APAP. To our knowledge, this is the first study demonstrating that caffeine is able to reduce the neurotoxicity induced by APAP, being the beneficial effects associated, at least in part, with the modulation of the redox homeostasis. In addition, the mitochondrial function remained unchanged during the neurotoxicity induced by APAP, but the caffeine treatment promoted a significant increase of CI + CII_Oxphos and CII_Oxphos, suggesting a bioenergetics adaptation in brain, which could be associated with the ergogenic properties of caffeine. It is interesting that according to our data mitochondrial respiration is not affected in brain homogenates from APAP-treated mice in 4 h, which would exclude the early mitochondrial dysfunction associated with APAP toxicity. However, as recognized by other authors, in vitro and in vivo experiments, the brain mitochondrial dysfunction had been noticed to be closely related with APAP doses and exposure time dependent, with a progressive caspase 3 activation and cytochrome c releases 6 h after APAP intoxication [2, 4, 40, 41].

Despite the importance of APAP hepatic metabolism for the hepatotoxicity [2], previous studies demonstrated that brain cytochrome P450 plays a significant role during neurotoxicity induced by APAP [42], resulting in oxidative damage. The increase of ROS and LPO is associated with the inhibition of CAT and SOD activities, which could enhance the oxidative damage in brain during the neurotoxicity induced by APAP. In contrast, the caffeine treatment was able to reduce the ROS generation with maintenance of the CAT and SOD activities, essentials for the ROS detoxification. In addition, the antioxidant properties of caffeine have been highlighted and the efficacy to scavenger ROS was demonstrated to be similar or higher than classical antioxidants [43, 44]. Pharmacokinetic properties of caffeine, such as fast absorption and widely distribution among the organism, could be an advantage for the treatment against the neurotoxicity induced by APAP [45]. The metabolism of caffeine promote the generation of paraxanthine and theophylline, which could contribute to the benefit action [46]. In addition, previous work demonstrated that chronic treatment with caffeine (20 mg/kg/day) reduces the LPO and promotes the upregulation of antioxidant enzymes [47].

Another important point is the impairment of GSH redox homeostasis due to the direct and/or indirect effects of reactive metabolite generation, which induce the oxidative damage in cellular components including protein and lipids in cellular membrane [48]. Here, treatment with APAP increased GST and GPx activities. The inhibition of GR activity in group treated with APAP could be associated with the higher levels of GSSG, which indicates the inability of the brain to restore the GSH pool. Caffeine consumption has demonstrated significant biochemical adaptive response in brain, such as increase of antioxidant enzyme activity (SOD and GR) and GSH levels [49]. Undoubtedly, these beneficial effects increase the cytoplasmic antioxidant capacity, reducing the impact of ROS generation and redox imbalance associated with the neurotoxicity induced by APAP.

In addition, our results demonstrated that the treatment with one dose of caffeine were able to extent the survival of mice treated with APAP. Other study demonstrates that caffeine modulates hyperammonemia-induced behavioral changes in patients with liver cirrhosis, decreasing subjective sleepiness and helping improve the vigilance state [50]. However, literature evidence demonstrated that ammonia levels in plasma were elevated only after 15 h to the acute liver failure induced by APAP (250 mg/kg) [51]. Beside that caffeine inhibits production of pro-inflammatory cytokines and promotes production of anti-inflammatory cytokines [52, 53], this could improve survival in APAP intoxication. Previous works have demonstrated the pro-oxidant activity of some caffeine concentrations, which can react with metals and, consequently, would contribute for oxidative damage, exacerbating the APAP toxicity. Caffeine is a non-selective adenosine receptors antagonist; recently, a protective role of moderate chronic caffeine consumption against neurodegenerative diseases has been discussed [54, 55]. The antagonization of caffeine on adenosine receptors modulates inflammation pathways in a concentration-dependent way [26]. Caffeine effects also can include reduction of oxygenated hemoglobin and increase of deoxygenated hemoglobin in cerebral blood flow [56].

Our results demonstrated that the caffeine concentration should be considered important during the neurotoxicity induced by APAP. In addition, higher concentration of caffeine could affect the CYP450 activity, accelerating the metabolism of APAP in liver and causing the enhance of NAPQI generation responsible for hepatoxicity and its consequences for organism [18].

Although APAP intoxication caused an increase in oxidative damage, our results did not demonstrate mitochondrial bioenergetics dysfunction, and besides cerebral oxidative stress, other factors may contribute to APAP-induced mortality. In this context, it has been demonstrated that the neurotoxicity induced by APAP is a process associated to the increase of toxic metabolite generation [6]. This factor could be important to cause the mitochondrial dysfunction, but the glutathione system helps to reduce the brain toxicity induced by APAP intoxication, increasing the transaminase levels and inflammation markers at liver [5]. While there is no direct evidence for normal ammonia levels in the brain under the experimental conditions of the present study, according to other references low doses of APAP, such as used in this study, are not able to increase the ammonia levels in blood serum [57], although there are evidences in literature demonstrating liver toxicity [58].

HRR analysis of brain mitochondrial function revealed that caffeine caused the stimulation of OXPHOS respiration when the convergent pathway (CI + CII_OXPHOS) and CII_OXPHOS were activated. In addition, the CIV activity was increased by caffeine treatment, CIV is a terminal oxidase of mitochondrial respiratory chain, which can regulate the aerobic ATP generation. Based upon these data, we postulate that concentration of caffeine (20 mg/kg) produces a bioenergetics adaptation and the respiration depends on the substrate applied, since the smallest stimulation was noticed when OXPHOS respiration was stimulated with only CI substrates (glutamate + malate), these changes may be associated with increased substrate oxidation in mitochondrial respiratory chain. Caffeine increases the oxidative metabolism without inducing a futile cycle associated with the mitochondrial uncoupling. It has been suggested that caffeine mobilizes the intracellular calcium concentration in brain [59, 60]. This intracellular signal modulates the regulatory enzymes of tricarboxylic acid cycle, increasing the NADH and FADH₂ production by dehydrogenases, and, consequently, leading to increase of the electron flow and H⁺-pumping in mitochondrial respiratory chain culminating with the mitochondrial membrane potential (∆Ψm). Given the lack of evidence demonstrating caffeine-induced increase of mitochondrial OXPHOS, this is the first work demonstrating that caffeine plays an important role for brain metabolism increasing the bioenergetics capacity. In this way, caffeine could be a useful tool for the re-establishment of cellular energy, since caffeine is able to increase the mitochondrial OXPHOS.

In conclusion, caffeine presents a modulatory effect on brain during the APAP intoxication. In addition, the APAP intoxication does not seems to induce bioenergetics impairment in brain; however, caffeine treatment promotes an increase in brain mitochondrial OXPHOS, which is an indicator of metabolic adaptation important for maintenance of vital functions in this tissue.

Declarations

Conflict of Interest

None declared.

Acknowledgments

This study was supported by the CNPq/FAPERGS/DECIT/SCTIE-MS/PRONEM #11/2029-1. The authors acknowledge CNPq, Brazil (404603/2015-7, 482313/2013-7), FAPERGS (2380-2551/14-8) and PROPESQ-Unipampa for the financial support. J.L.F. is recipient of a CNPq research fellowship (310861/2014-4). F.A.S. received a fellowship by CNPq.