Neurochemical Research

, Volume 31, Issue 9, pp 1141–1151

Chronic Lithium Treatment has Antioxidant Properties but does not Prevent Oxidative Damage Induced by Chronic Variate Stress

Authors

    • Programa de Pós-Graduação em NeurociênciasInstituto de Ciências Básicas da Saúde, UFRGS
    • Departamento de BioquímicaInstituto de Ciências Básicas da Saúde, UFRGS
  • Fabiane Battistela Nieto
    • Departamento de BioquímicaInstituto de Ciências Básicas da Saúde, UFRGS
  • Leonardo Machado Crema
    • Programa de Pós-Graduação em NeurociênciasInstituto de Ciências Básicas da Saúde, UFRGS
  • Luisa Amália Diehl
    • Departamento de BioquímicaInstituto de Ciências Básicas da Saúde, UFRGS
  • Lúcia Maria de Almeida
    • Departamento de BioquímicaInstituto de Ciências Básicas da Saúde, UFRGS
  • Martha Elisa Prediger
    • Departamento de BioquímicaInstituto de Ciências Básicas da Saúde, UFRGS
  • Elizabete Rocha da Rocha
    • Programa de Pós-Graduação em NeurociênciasInstituto de Ciências Básicas da Saúde, UFRGS
    • Departamento de BioquímicaInstituto de Ciências Básicas da Saúde, UFRGS
  • Carla Dalmaz
    • Programa de Pós-Graduação em NeurociênciasInstituto de Ciências Básicas da Saúde, UFRGS
    • Departamento de BioquímicaInstituto de Ciências Básicas da Saúde, UFRGS
Original Paper

DOI: 10.1007/s11064-006-9139-2

Cite this article as:
de Vasconcellos, A.P.S., Nieto, F.B., Crema, L.M. et al. Neurochem Res (2006) 31: 1141. doi:10.1007/s11064-006-9139-2

Abstract

This study evaluated the effects of chronic stress and lithium treatments on oxidative stress parameters in hippocampus, hypothalamus, and frontal cortex. Adult male Wistar rats were divided into two groups: control and submitted to chronic variate stress, and subdivided into treated or not with LiCl. After 40 days, rats were killed, and lipoperoxidation, production free radicals, total antioxidant reactivity (TAR) levels, and superoxide dismutase (SOD) and glutathione peroxidase (GPx) activities were evaluated. The results showed that stress increased lipoperoxidation and that lithium decreased free radicals production in hippocampus; both treatments increased TAR. In hypothalamus, lithium increased TAR and no effect was observed in the frontal cortex. Stress increased SOD activity in hippocampus; while lithium increased GPx in hippocampus and SOD in hypothalamus. We concluded that lithium presented antioxidant properties, but is not able to prevent oxidative damage induced by chronic variate stress.

Keywords

LithiumChronic variate stressOxidative stressSODGPxAntioxidant enzymesLipoperoxidationFree radicals

Introduction

Stress is known to influence a wide range of neuronal systems and, in its acute phase, may result in beneficial endocrine and behavioral responses; however, repeated or severe stress can lead to adverse effects on neuronal function [1]. Several brain structures are involved in the stress response, e.g., hypothalamus, frontal cortex, and hippocampus, and may be affected by chronic exposure to stress [2, 3]. Among these structures, the hippocampus has been the most extensively studied with regard to the actions of stress and depression, and hippocampal neurons are reported to be damaged by chronic exposure to stress or activation of the hypothalamic-pituitary-adrenal (HPA) axis and increased levels of glucocorticoids (GCs). Dysfunction of the hippocampus could result in some of the vegetative and endocrine abnormalities, as well as cognitive and memory deficits, observed after prolonged stress exposure and in depressed patients [4].

The neuroendangerment observed after stress exposure or elevated GC levels have been linked to an increased generation of reactive oxygen species (ROS) [5]. ROS are believed to be involved in tissue damage resulting from a wide variety of insults. These substances can directly damage cellular proteins, DNA, and lipids, and thereby affect all cellular functions [6]. The nervous system is extremely sensitive to peroxidative damage, since it is rich in oxidizable substrates, has a high oxygen tension and low antioxidant capacity [7, 8]. Membrane lipids are highly susceptible to this kind of injury, and this event may damage cell membranes and interfere with the activity of membrane-associated enzymes. Furthermore, these alterations induce changes in membrane fluidity and potential and in its permeability to ions [9]. Moreover, the localization of major antioxidant defense systems in glial cells, rather than in neurons, may cause the nerve cells to be more susceptible to oxidants present in the brain [10].

The possible damage induced by ROS in cells is normally held in check by natural enzymatic and non-enzymatic antioxidant systems [11]. These cellular defenses reduce the steady-state concentrations of free radical species and repair oxidative cellular damage. The antioxidant defense system includes enzymes, such as superoxide dismutase (SOD), which converts superoxide radicals into H2O2, and glutathione peroxidase (GPx), which breaks down peroxides, notably those derived from the oxidation of membrane phospholipids. The removal of superoxide and H2O2 reduces the generation of hydroxyl radicals, which are formed by the iron-catalyzed Fenton reaction or by the Haber–Weiss reaction [12]. Non-enzymatic antioxidants (carotenoids, vitamin E, glutathione) also play important roles in defense mechanisms; exposure to stress situations has been proposed to impair antioxidant defenses, leading to oxidative damage by changing the balance between oxidant and antioxidant factors [1315].

Lithium salts constitute the first line of therapeutic drugs used to treat affective disorders, mainly bipolar disorder [16, 17], and increasing evidence supports the notion that lithium has neuroprotective effects in a variety of insults, such as glutamate-induced excitotoxicity, in cultured cells, and animal models of diseases [16, 1820]. Chronic lithium treatment has been demonstrated to markedly increase the levels of the neuroprotective and anti-apoptotic protein bcl-2 in the rat frontal cortex, hippocampus, and striatum [18, 20]. Lithium also inhibits glycogen synthase kinase 3beta (GSK-3beta) activity, which is an apoptotic promoter and is involved in neurodegenerative diseases [21], and both bcl-2 and GSK-3beta seem to be involved in the prevention and/or induction of the deleterious effects induced by oxidative stress processes [2224]. Therefore, it is conceivable that lithium treatment may exert some of its neuroprotective effects by attenuating oxidative stress. The aim of this work was to evaluate the effects of chronic variate stress and lithium treatments on different parameters of oxidative stress in brain structures such as hippocampus, hypothalamus, and cerebral frontal cortex.

Experimental procedures

Chemicals

Thiobarbituric acid (TBA) and Trolox were obtained from Merck, 2,2′-azobis (2-amidinopropane) dihydrochloride (ABAP) was obtained from Wako Chemicals USA, Inc., 2′-7′-dichlorofluorescein diacetate (DCFH-DA), 2′-7′-dichlorofluorescein (DCF), trichloroacetic acid (TCA), 2,4-dinitrophenylhydrazine, guanidine hydrochloride, 5-amino-2,3-dihydro-1,4-phtalazinedione (luminol), and H2O2 stock solution were purchased from Sigma Chemical Co.

Animals

Sixty adult male Wistar rats (60 days at the beginning of the treatment) weighing 160–230 g were used. Experimentally naive animals were housed in groups of four or five rats in home cages made of Plexiglas (65 cm × 25 cm × 15 cm) with the floor covered with sawdust. They were maintained under a standard dark–light cycle (lights on between 7:00 a.m. and 7:00 p.m.), with a room temperature of 22 ± 2°C. Rats had free access to food and water, except during the period when restraint stress or forced swimming were applied. All animal treatments were in accordance with the institutional guidelines and followed the recommendations of the International Council for Laboratory Animal Science.

Experimental groups

The animals were divided in two groups. One group received standard rat chow and the other group had lithium chloride (2.5 mg LiCl/g of chow) and NaCl (17 mg/g) added to the food, as previously described [25]. This treatment has been previously used, and at the end of a period of 4 weeks or more, animals remain healthy and present lithium levels in the range of 0.6–1.2 mM [2527], similar to the levels observed in treated patients. These groups were subdivided into two other groups: control and submitted to a chronic variate stress paradigm.

Chronic variate stress model

Chronic variate stress was modified from other models of variate stress [2831]. The following stressors were used: (a) inclination of the home cages at a 45° angle for 4–6 h, (b) 10–15 min of noise, (c) 1–3 h of restraint, as described below, (d) 1.5–2 h of restraint at 4°C, (e) forced swimming for 10 or 15 min, as described below, (f) flashing light during 2–4 h, (g) isolation (2–3 days). Animals were exposed to only one stressor every day, with stress starting at a different time each day, in order to minimize its predictability. The exposure to stress situations continued during 40 days. Please see Table 1 for the sequence of stressors applied during all the experimental procedures.
Table 1

Stressors applied during chronic treatment

Day of treatment

Stressor applied

1

Restraint (2 h)

2

Flashing light (4 h)

3

Forced swimming (15 min)

4

Cold restraint (1 h 30 min)

5

Isolation

6

Isolation

7

Isolation

8

Inclination of home cages (6 h)

9

Noise (10 min)

10

Forced swimming (3 min, cold water)

11

Cold restraint (1 h)

12

Restraint (2 h)

13

Flashing light (6 h)

14

No stressor applied

15

Forced swimming (15 min)

16

Inclination of home cages (5 h)

17

Cold restraint (1 h 30 min)

18

Noise (10 min)

19

Isolation

20

Isolation

21

Isolation

22

Restraint (2 h)

23

Forced swimming (3 min, cold water)

24

Flashing light (6 h)

25

Noise (10 min)

26

Cold restraint (1 h)

27

Inclination of home cages (5 h)

28

No stressor applied

29

Forced swimming (15 min)

30

Restraint (1 h)

31

Flashing light (6 h)

32

Cold restraint (1 h 30 min)

33

Isolation

34

Isolation

35

Isolation

36

Inclination of home cages (6 h)

37

Noise (10 min)

38

Forced swimming (3 min, cold water)

39

Flashing light (5 h)

40

Restraint (2 h)

Restraint was carried out by placing the animal in a 25 cm × 7 cm plastic tube and adjusting it with plaster tape on the outside, so that the animal was unable to move. There was a 1-cm hole at the far end for breathing. Forced swimming was carried out by placing the animal in a glass tank measuring 50 cm × 47 cm × 40 cm with 30 cm of water at 23 ± 2°C. Exposure to flashing light was achieved by placing the animal in a 50 cm high, 40 cm × 60 cm open field made of brown plywood with a frontal glass wall. A 100-W lamp, flashing at a frequency of 60 flashes per minute, was used.

Rats were submitted to chronic variate stress and/or treated with lithium chloride during 40 days.

Tissue preparation

Rats were killed by decapitation, on the day following the last day of stress exposure. Hippocampus, hypothalamus, and cerebral cortex were dissected out and instantaneously placed in liquid nitrogen and stored at −70°C until biochemical measurements, when the tissues were homogenized in 10 volumes of ice-cold phosphate buffer (0.1 M, pH 7.4) containing 140 mM KCl and 1 mM EDTA. The homogenate was centrifuged at 960g for 10 min and the supernatant was used.

ROS formation

To assess the free radical levels, DCFH-DA was used as a probe. This method does not determine the presence of specific free radicals, since DCFH may be oxidized by several reactive intermediates [32]. An aliquot of the sample was incubated with DCFH-DA (100 μM) at 37°C for 30 min; chilling the reaction mixture in ice terminated the reaction. The formation of the oxidized fluorescent derivative (DCF) was monitored at excitation and emission wavelengths of 488 and 525 nm, respectively, using a fluorescence spectrophotometer (Hitachi F-2000). The free radicals content was quantified using a DCF standard curve and results were expressed as nmol of DCF formed/mg protein. All procedures were performed in the dark, and blanks containing DCFH-DA (no homogenate) and homogenate (no DCFH-DA) were processed for measurement of autofluorescence. Data were expressed as percentages of the control group.

Assay of lipid peroxidation

The formation of thiobarbituric acid reactive substances (TBARS) was used as an indicator of lipoperoxidation. Malondialdehyde (MDA), a product of lipoperoxidation, reacts with two molecules of TBA at low pH and high temperature to form a pink-colored complex. Therefore, the formation of TBARS was expressed as MDA equivalents/mg of protein. This test was based on the method described by Buege and Aust [33]. Aliquots of samples were incubated with 10% TCA and 0.67% TBA. The mixture was heated (30 min) in a boiling water bath. Afterwards, n-butanol was added and the mixture was centrifuged. The organic phase was collected to measure fluorescence at excitation and emission wavelengths of 515 and 553 nm, respectively. 1,1,3,3-Tetramethoxypropane, which is converted to MDA, was used as standard. Data were expressed as percentages of the control group.

Total antioxidant reactivity assay

This assay is based on luminol-enhanced chemiluminescence (CL) measurement, induced by an azo initiator [3436]. The reaction mixture contained 2 mM ABAP, a source of peroxyl radicals, and 6 mM luminol in glycine buffer (0.1 M, pH 8.6). The CL generated was measured in a scintillation counter (Beckman) working out of coincidence mode. The addition of Trolox (antioxidant standard, 200 nM) or samples decreases CL levels, and total antioxidant reactivity (TAR) values were determined by assessing the initial decrease of luminescence calculated as the ratio “Io/I,” where “Io” is the CL in the absence of additives, and “I” is the CL after addition of the 200 nM Trolox, or the TAR values of the samples were expressed as equivalents of Trolox concentration per mg of protein. Data were expressed as percentages of the control group.

Enzymatic activities

Superoxide dismutase activity

Superoxide dismutase activity was determined using a RANSOD kit (Randox Laboratories Ltd., UK) based on a procedure previously described by Delmas-Beauvieux et al. [37]. This method employs xanthine and xanthine oxidase to generate superoxide radicals that react with 2-(4-iodophenyl)-3-(4-nitrophenol)-5-phenyltetrazolium chloride to form a red formazan dye that is assayed spectrophotometrically at 505 nm at 37°C. One unit of SOD activity is defined as the amount of enzyme that inhibits the rate of the formazan dye formation by 50% and the results were expressed as units/μg protein.

GPx activity

Glutathione peroxidase activity was determined according to Wendel [38]. The reaction was carried out at 25°C in 600 μl of solution containing 100 mM pH 7.7 potassium phosphate buffer, 1 mM EDTA, 0.4 mM sodium azide, 2 mM GSH, 0.1 mM NADPH, 0.62 U of GSH reductase. The activity of selenium-dependent GPx was measured taking tert-butylhydroperoxide (0.8 μM, final concentration) as the substrate at 340 nm. The contribution of spontaneous NADPH oxidation was always subtracted from the overall reaction rate. GPx activity was expressed as units (nmol NADPH oxidized/min)/μg protein.

Protein assay

Total protein concentrations were determined using the method described by Lowry et al. [39] with bovine serum albumin as the standard.

Statistical analysis

Data were analyzed using two-way analysis of variance (ANOVA) when evaluating possible interactions between chronic variate stress and lithium, with post hoc analysis performed by the Duncan’s multiple range test. A difference was considered significant when P < 0.05. Results are expressed as mean ( standard error of the mean (SEM).

Results

There were no significant effects of chronic variate stress and lithium treatments on oxidative stress in the frontal cortex. Table 2 shows the effects of stress and lithium on the ROS formation, as evaluated by the DCF test [two-way ANOVA, F(1, 20) = 2.75, P > 0.1 for stress effects, F(1, 20) = 0.745, P > 0.1 for lithium effects, and F(1, 20) = 2.927, P > 0.1 for the interaction], on lipid peroxidation, evaluated by TBARS test [two-way ANOVA, F(1, 20) = 1.747, P > 0.1 for stress effects, F(1, 20) = 4.213, P > 0.05 for lithium effects, and F(1, 20) = 2.464, P > 0.1 for interaction] and on the TAR [two-way ANOVA, F(1, 20) = 1.260, P > 0.1 for stress effects, F(1, 20) = 1.222, P > 0.1 for lithium effects, and F(1, 20) = 3.083, P > 0.1 for interaction], indicating that neither of the treatments alter the oxidative status of this structure.
Table 2

Effects of chronic variate stress and chronic lithium treatments on oxidative stress parameters in frontal cortex of rats

 

Control

Lithium

Stressed

Stressed + lithium

TBARS

100.13 ± 3.5

103.07 ± 6.4

82.55 ± 6.22

104.58 ± 7.48

DCF

100.12 ± 2.7

121.36 ± 11.2

104.83 ± 5.2

99.99 ± 7.68

TAR

100.01 ± 8.5

111.61 ± 7.6

146.13 ± 20.7

110.90 ± 13.7

Data are expressed as mean ± SEM, N of six animals/group. There was no effect of chronic variate stress and lithium treatments on lipid peroxidation (TBARS; mean ± SEM for 100% values correspond to 0.33 ± 0.01 MDA equivalents nmol/mg protein), on free radicals formation (DCF; mean ± SEM for 100% values correspond to 0.50 ± 0.03 nmol of DCF formed/mg protein) nor on antioxidant reactivity (TAR; mean ± SEM for 100% values correspond to 63.0 ± 10.9 equivalents of nmol Trolox/mg protein) in this structure (two-way ANOVA, P > 0.05 for all parameters)

Hypothalamic oxidative stress was also assessed by the measurements of lipid peroxidation and TAR. In this case, we observed an absence of effect of both treatments on lipid peroxidation values [two-way ANOVA, F(1, 18) = 2.44, P > 0.1 for stress effects, F(1, 18) = 0.008, P > 0.1 for lithium effects, and F(1, 18) = 0.055, P > 0.1 for the interaction; see Fig. 1A], however, lithium treatment significantly increased TAR in this structure [two-way ANOVA, F(1, 18) = 4.253, P < 0.05; Fig. 1B], with a mean increase of 40% over groups not receiving lithium. There was no effect of stress [two-way ANOVA, F(1, 18) = 0.001, P > 0.1] nor interaction between treatments [two-way ANOVA, F(1, 18) = 0.848, P > 0.1] in this parameter. There were no effects of lithium or stress treatments on the DCF test in this structure [two-way ANOVA, F(1, 16) = 0.199, P > 0.1 for stress effects, F(1, 16) = 0.368, P > 0.1 for lithium effects, and F(1, 16) = 0.103, P > 0.1 for the interaction, Fig. 1C].
https://static-content.springer.com/image/art%3A10.1007%2Fs11064-006-9139-2/MediaObjects/11064_2006_9139_Fig1_HTML.gif
Fig. 1

Effects of chronic variate stress and chronic lithium treatments on oxidative stress parameters in hypothalamus of rats. Data are expressed as mean ± SEM, N of five to six animals/group. (A) TBARS in hypothalamus homogenates of chronically stressed and lithium-treated rats. Mean ± SEM for 100% values correspond to 2.34 ± 0.24 MDA equivalents nmol/mg protein. There were no significant differences among groups. (B) TAR in hypothalamus homogenates of chronically stressed and lithium-treated rats. Mean ± SEM for 100% values correspond to 202.8 ± 32.3 equivalents of nmol Trolox/mg protein. *Significantly different from groups not receiving lithium (two-way ANOVA, P < 0.05). ( C) ROS formation, assessed by the DCF test, in hypothalamus homogenates of chronically stressed and lithium-treated rats. Mean ± SEM for 100% values correspond to 18.3 ± 3.3 nmol of DCF formed/mg protein. There were no significant differences among groups

Considering the altered TAR observed in lithium-treated animals, we decided to measure the activity of antioxidant enzymes, i.e., SOD and GPx in the hypothalamus of stressed and lithium-treated rats. We observed that SOD activity was significantly increased by lithium treatment [two-way ANOVA, F(1, 16) = 5.654, P < 0.05; Fig. 2A], and that there was a significant interaction between stress and lithium treatments, since the stressed + lithium group presented values higher than those of the other treatments alone [two-way ANOVA, F(1, 16) = 4.727, P < 0.05]: the percent increase in lithium-treated group was 21%, while in the stress + lithium group there was a fivefold increase. There was no effect of stress in this parameter [F(1, 16) = 3.462, P > 0.05]. GPx activity was not significantly altered by stress or lithium treatments [two-way ANOVA, F(1, 19) = 0.015, P > 0.1 for stress and F(1, 19) = 0.959, P > 0.1 for lithium treatment; Fig. 2B]; however there was a significant interaction between stress and lithium treatments [two-way ANOVA, F(1, 19) = 6.282, P < 0.05].
https://static-content.springer.com/image/art%3A10.1007%2Fs11064-006-9139-2/MediaObjects/11064_2006_9139_Fig2_HTML.gif
Fig. 2

Effects of chronic variate stress and chronic lithium treatments on the activity of antioxidant enzymes in hypothalamus of rats. Data are expressed as mean ± SEM, N of five to six animals/group. (A) Superoxide dismuthase activity in hypothalamus of chronically stressed and lithium-treated rats. One unit of SOD activity is defined as the amount of enzyme that inhibits the rate of the formazan dye formation by 50%. There was a significant effect of lithium treatment (two-way ANOVA, P=0.03) and a significant interaction between treatments (two-way ANOVA, P < 0.05). *Significantly different from other groups (Duncan multiple range test, P < 0.05). (B) Glutathione peroxidase activity in hypothalamus of chronically stressed and lithium-treated rats. Units correspond to nmol NADPH oxidized/min. There were no significant differences among groups, but a significant interaction was observed between stress and lithium treatments (two-way ANOVA, P < 0.05)

In hippocampus, lithium treatment significantly decreased the ROS formation, with a mean decrease of 24%, compared to groups not receiving lithium [two-way ANOVA, F(1, 19) = 8.038, P < 0.01, Fig. 3A], and no effect of chronic stress nor interaction between treatments were observed [two-way ANOVA, F(1, 19) = 0.221, P > 0.1 for stress and F(1, 19) = 0.318, P > 0.1 for the interaction]. On the other hand, chronic stress induced an increased lipid peroxidation [two-way ANOVA, F(1, 17) = 6.524, P < 0.05, Fig. 3B], with a mean increase of 28% over groups not subjected to stress. No effect of lithium, nor interaction, were observed in this parameter [two-way ANOVA, F(1, 17) = 3.299, P > 0.05 for lithium and F(1, 17) = 0.321, P > 0.1 for interaction].
https://static-content.springer.com/image/art%3A10.1007%2Fs11064-006-9139-2/MediaObjects/11064_2006_9139_Fig3_HTML.gif
Fig. 3

Effects of chronic variate stress and chronic lithium treatments on oxidative stress parameters in hippocampus of rats. Data are expressed as mean ± SEM of percentage of the control group, N of six animals/group. (A) ROS formation, assessed by the DCF test, in hippocampus homogenates of chronically stressed and lithium-treated rats. Mean ± SEM for 100% values correspond to 0.89 ± 0.13 nmol of DCF formed/mg protein. There was a significant effect of lithium treatment in this parameter. *Significantly different from groups not receiving lithium (two-way ANOVA, P < 0.01). (B) TBARS in hippocampus homogenates of chronically stressed and lithium-treated rats. Mean ± SEM for 100% values correspond to 2.57 ± 1.10 MDA equivalents nmol/mg protein. There was a significant effect of stress on this parameter. *Significantly different from groups not subjected to stress (two-way ANOVA, P < 0.03). (C) TAR in hypothalamus homogenates of chronically stressed and lithium-treated rats. Mean ± SEM for 100% values correspond to 145.2 ± 23.5 equivalents of nmol Trolox/mg protein. There was a significant interaction between stress and lithium treatments (two-way ANOVA, P < 0.001). *Significant difference from control and lithium + stress groups (Duncan’s multiple range test, P < 0.05)

When evaluating the TAR in the hippocampus of chronically stressed and lithium-treated rats, we observed a strongly significant interaction, since both treatments separately increased the antioxidant reactivity (lithium-treated group presented an increase of 97%, while the stressed group showed an increase of 84%, compared to controls); however, when both treatments were applied together, the animals presented values similar to those of the control group [two-way ANOVA, F(1, 19) = 0.060, P > 0.1 for stress effects, F(1, 19) = 1.325, P > 0.1 for lithium effects, and F(1, 19) = 32.489, P < 0.001 for the interaction, Fig. 3C]. A post hoc analysis showed that both stress- and lithium-treated animals have increased TAR values when compared to the control and stress + lithium groups (Duncan’s multiple range test, P < 0.05).

To better characterize the effects of stress and lithium in the hippocampus, we also measured SOD and GPx activities. We observed that both stress and lithium treatments induced an altered activity of SOD, and that there was a significant interaction between treatments [two-way ANOVA, F(1, 17) = 22.076, P < 0.001 for stress effects, F(1, 17) = 6.549, P < 0.02 for lithium effects, and F(1, 17) = 8.413, P < 0.01 for the interaction, Fig. 4A], since the stressed group presented an increase of 120% over the control group, while the stress + lithium group presented an almost fivefold increase. GPx activity was also increased by lithium treatment [two-way ANOVA, F(1, 18) = 6.013, P < 0.03], with a mean increase of about 10%, but there was no effect of stress, nor interaction, on this enzyme activity [F(1, 18) = 0.842, P > 0.1 for stress effects, and F(1, 18) = 0.233, P > 0.1 for the interaction, Fig. 4B].
https://static-content.springer.com/image/art%3A10.1007%2Fs11064-006-9139-2/MediaObjects/11064_2006_9139_Fig4_HTML.gif
Fig. 4

Effects of chronic variate stress and chronic lithium treatments on the activity of antioxidant enzymes in hippocampus of rats. Data are expressed as mean ± SEM, N of five to six animals/group. (A) Superoxide dismuthase activity in hippocampus of chronically stressed and lithium-treated rats. One unit of SOD activity is defined as the amount of enzyme that inhibits the rate of the formazan dye formation by 50%. There was a significant effect of chronic stress (two-way ANOVA, P < 0.001) and a significant interaction between stress and lithium treatments (two-way ANOVA, P < 0.01). *Significantly different from other groups (Duncan multiple range test, P < 0.05). (B) Glutathione peroxidase activity in hippocampus of chronically stressed and lithium-treated rats. Units correspond to nmol NADPH oxidized/min. There was a significant effect of lithium treatment on this parameter (two-way ANOVA, P < 0.03). *Significantly different from control group (Duncan multiple range test, P < 0.05)

Discussion

This study was performed to evaluate the effects of chronic variate stress and lithium treatments on oxidative stress parameters in different brain structures. We observed that chronic exposure to stress induced an increase in the oxidative status in hippocampus, as shown by the increased lipid peroxidation, and that lithium treatment presented an antioxidant effect, demonstrated by decreased ROS generation in hippocampus and increased antioxidant reactivity. The effects of both treatments were region-specific and the hippocampus was the most affected structure; these results agree with several other studies showing that the hippocampus, as well as hippocampus-related behavior, is strongly affected by stress exposure and lithium treatment [25, 27].

2′–7′-Dichlorofluorescein has been used as a probe to study free radical production in a variety of biological systems. DCFH is oxidized to fluorescent dichlorofluorescein by hydrogen peroxide, as well as by other reactive intermediates [40]. It is usually hard to detect ROS directly, since they are extremely reactive and have very short lifetime. Radicals present in the living tissue are even more difficult to detect after freezing, storage, and homogenization of the tissue. Therefore, here we detect the production of oxidants by the system at the moment of the test, and differences are due to differences in the ability of the tissue to produce ROS. It is also important to consider that an increase in formation of ROS does not necessarily mean that it will cause oxidative damage to cellular macromolecules, since they can be detoxified by antioxidant defense systems [41].

The results of this study agree with other reports showing that exposure to GCs or to stress may lead to oxidative injury in various tissues, and an increased oxidative damage has been proposed as one of the mechanisms through which GCs increase the vulnerability of different brain regions, particularly the hippocampus, to metabolic insults [5, 13, 14]. Besides the reported effects of GCs, exposure to stress has also been shown to increase lipid peroxidation in the plasma of acutely stressed rats [42, 43], and in hippocampus, after repeated restraint stress [15]. Effects of a different model of chronic variate stress on TBARS in brain have also been observed, where distinct brain regions show different responses to stress [44]. Therefore, different models of chronic stress apparently lead to different results with regard to the induction of oxidative stress. The present findings support the idea that stress produces oxidants, and it is possible that the oxidative damage in stress could contribute, at least in part, to stress-related diseases, such as depression.

Exactly how increased lipoperoxidation occurs in the hippocampus of stressed animals, in the absence of any alteration in free radicals content (evaluated by the DCF test) is not clear. Measurements were made one day after the final stress procedures, in order to verify alterations induced by chronic stress, and not responses to any particular stress session. Therefore, it is possible that increased free radical production occurs during exposure to some determined stressor (s) (since different stressors were used in this model), and that these levels may return to normal at the time of the sacrifice. In addition, GCs (which have been reported to increase ROS production, as measured by the DCF test, see ref. [5]) had already returned to basal levels at the time of the sacrifice. Similar results, i.e., increased TBARS levels without increased ROS formation in the DCF test, were also observed in another model of chronic stress [15].

On the other hand, lithium treatment decreased free radical generation (as measured by DCF test) in hippocampus, with no effects on the other structures examined. Additionally, an increased TAR was observed in hippocampus and hypothalamus after lithium treatment; in hippocampus, however, a significant interaction with stress treatment occurred, and the group stress + lithium presented TAR levels similar to those of the control group. These results are in agreement with other studies, which suggest an antioxidant effect of lithium [45, 46]. Despite these effects, lithium treatment was not able to prevent a stress-induced increase in damage to macromolecules, such as lipoperoxidation, in hippocampus.

The analysis of antioxidant enzymes indicates that lithium treatment was able to increase GPx activity in hippocampus homogenates, as well as increase SOD activity in the hypothalamus, further suggesting an antioxidant effect of lithium chronic treatment. GPxs constitute a family of enzymes, capable of reducing a variety of organic and inorganic hydroperoxides to the corresponding hydroxy compounds. Several GPx subtypes have been identified; major types in brain are the cytosolic GPx (cGPx, GPx-1) and the phospholipid hydroperoxide GPx (phGPx, GPx-4) [47]; at the moment, we cannot indicate which of these isoforms is altered by lithium treatment. The altered activity of these enzymes in one structure, and the lack of alteration in another, after in vivo lithium administration, is difficult to explain; however other studies have also observed variable alterations in SOD or GPx activities after chronic lithium treatment in rats when different tissues were studied [48].

In addition, chronic stress induced an increased SOD activity in hippocampus, maybe as an adaptation to this chronic situation. There are three isoforms of SOD: mitochondrial manganese SOD (MnSOD), cytosolic copper–zinc SOD (CuZnSOD), and extracellular SOD [49]. Both MnSOD and CuZnSOD have been shown to be induced by different pathological conditions, such as diabetes [50], cerebral ischemia [51], or quinolinic acid-induced lesion [52]. GC treatment of adrenalectomized rats during 3 days has previously been shown to cause reductions in hippocampal Cu/Zn-SOD activity [14]. These results differ from the present ones, using a model of variate stress, however, we must point to the fact that this treatment involved 40 days of different stressors and, in a stress model, different factors will be acting, besides GCs.

The most intriguing result, however, was the interaction observed between lithium and stress treatments on SOD activity, which was observed both in hippocampus and hypothalamus. In both structures, SOD activity levels were potentiated when both treatments were applied together. Interestingly, lithium added in vitro (in the concentration range of 0.1–10 mM) did not affect this enzyme activity (data not shown), suggesting that the in vivo action of lithium on SOD activity is probably not a simple enzyme activation induced by the presence of this salt in the medium, but possibly involves an indirect effect, which is induced in vivo.

Although the precise mechanisms involved in the unexpected potentiation of SOD activity in rats receiving both lithium and stress treatments, as observed in this study, are not known, it has been reported that lithium increases plasmatic levels of corticosterone [53, 54], suggesting that a further activation of the HPA axis in chronically stressed rats may be induced by lithium, possibly causing this interaction. However, this potentiation may also suggest that these two treatments act through different mechanisms to induce this increased SOD activity. These effects could be either on the number of molecules and/or on the degree of activation of this enzyme. We cannot exclude effects of stress exposure on the permeability of the blood–brain barrier [55], resulting in an improved entry of metals in the brain [56]. It is also of interest the relationship between SOD and neurotrophic factors. For example, brain-derived neurotrophic factor (BDNF) increases SOD activity and expression [57, 58], and lithium treatment is also known to induce increased BDNF [59].

The balance between SOD and GPX activities has been suggested to be more important than the absolute amount of any one of these activities for the protection against free radicals [60, 61]. In vitro, GPx was demonstrated to offer greater protection than SOD against oxidative stress [62, 63]. An important point, however, is that SOD transforms O2-, producing H2O2, which requires detoxification by another enzymes, such as GPx. An imbalance between these two activities, such as that observed in the stressed and stressed + lithium group, may result in the accumulation of H2O2, generating hydroxyl radicals that may lead to lipid peroxidation [64]. This increased production of free radicals, particularly in the stressed + lithium group (in which the interaction induces a greater increase in SOD activity) could be responsible for the consumption of antioxidants in the tissue, leading to a decreased TAR, as measured by the TAR test in the hippocampus.

The results obtained in the present study suggest regional differences in the vulnerability to stress effects, a fact also reported by other authors [44, 65]. This different vulnerability can be attributed to differences in the antioxidant capacity [65], cellular oxidative metabolism, production of compounds which can lead to enhanced formation of free radicals (e.g., dopamine; refs. [31, 66]), or to a heterogeneously distribution of iron in the rat brain [67]. Although iron may be related to the increased oxidative stress observed in the present study, it is not probable the main factor for differences between structures, since its content is not higher in hippocampus of adult rat brain when compared to cortex (where no increased oxidative stress was observed) [68].

In conclusion, the present findings suggest that lithium has antioxidant properties, while chronic stress presents pro-oxidant effects, mainly in the hippocampus. Nevertheless, chronic lithium was not able to prevent lipoperoxidation induced by stress exposure, and presented an interaction with chronic stress, leading to an imbalance in SOD and GPx enzymatic activities, which should be taken into account when considering lithium therapy. Further research is needed to determine the mechanisms underlying lithium and stress effects on oxidative status, as observed in the present study.

Acknowledgment

Supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico, CNPq.

Copyright information

© Springer Science+Business Media, Inc. 2006