Depressive-Like Behavior Induced by Long-Term Ouabain Administration Accompanied Alteration in Neuroin ammation Parameters in Rats

Samira S. Valvassori (  samiravalvassori@unesc.net ) Unesc: Universidade do Extremo Sul Catarinense https://orcid.org/0000-0003-4824-7742 Jorge M. Aguiar-Geraldo Unesc: Universidade do Extremo Sul Catarinense Taise Possamai-Della Unesc: Universidade do Extremo Sul Catarinense Dayane D. da-Rosa Unesc: Universidade do Extremo Sul Catarinense Samira Menegas Unesc: Universidade do Extremo Sul Catarinense Gustavo C. Dal-Pont Unesc: Universidade do Extremo Sul Catarinense José H. Cararo Unesc: Universidade do Extremo Sul Catarinense João Quevedo The University of Texas Health Science Center at Houston


Introduction
Bipolar disorder (BD) is one of the most prevalent forms of mental illness comprising different mood shifts, ranging from episodes of mania or hypomania, depression, euthymia, and mixed states. 1 Euphoria, expansive mood, high energy, psychomotor agitation or hyperactivity, and increased risk-taking behavior characterize the manic episodes, while deep sadness and anhedonia usually represent the depressive phase. 2,3 BD is a common and disabling condition associated with increased mortality in comparison to the general population, whose world prevalence rate stands at 1-2.5%. 4,5 Precise BD pathological mechanisms are not fully understood, but several biological systems were supposed to act perpetuating the condition. 6 For example, oxidative stress, mitochondrial dysfunction, the release of in ammatory cytokines, and, consequently, the changes in ionic regulation channels can play a pivotal role in the pathophysiology. [7][8][9][10] The advance in our understanding of BD pathophysiology can ultimately improve the therapeutic arsenal used for this condition. 11 In this regard, lithium (Li) is a mood stabilizer drug used for more than 60 years, able to reduce acute symptoms of mania and also used in bipolar depression and BD maintenance therapy, avoiding new mood episodes. 8, 12 Studies have shown that Li induces widespread effects on different cellular pathways, with antioxidant, neuroprotective, and neurotrophic properties. 13 Besides, the drug collaborates with the in ammatory system modulation in bipolar patients and animal models of mania. 8,12 Research using animal models of psychiatric disorders has been relevant to unravel the pathophysiology of these conditions. 14 To illustrate, research conducted with animal models of BD has helped to discover the pathways involved in its development. 9 In this context, the present study uses as a platform the animal model of BD induced by the intracerebroventricular (ICV) administration of ouabain (OUA), a wellknown glycoside that inhibits the sodium/potassium adenosinetriphosphatase enzyme (Na + /K + -ATPase; EC 3.6.3.9). 15 Recently, the Valvassori and coworkers provided all validities required for a proper animal model of BD using this same platform since it mimics manic and depression behaviors in the same animal, as well as the alterations concerning pathophysiology and remission of the symptoms through traditional pharmacotherapy. 16 Indeed, evidence has shown a signi cantly impaired Na + /K + -ATPase activity in bipolar patients in manic and depressive episodes. 17 Additionally, recent research shows that this enzyme inhibition can lead to increased cytokine release, oxidative stress, immune response, and in ammation. [17][18][19][20] Cytokines are small peptides and proteins responsible for cell interactions and are included in this class of molecules the cytokine-induced neutrophil chemoattractant 1 (CINC-1), interleukin(IL)-1β, IL-6, IL-10, and tumor necrosis factor α (TNF-α). 21 IL-6 is a multifunctional cytokine and a trophic factor for hybridoma and myeloma cells that mediates the transformation of activated B cells into antibody-synthesizing ones. 22 IL-10 limits neuroin ammation by altering the resident glia and in ltrating leukocytes' response and by decreasing the synthesis of mediators by such cells. 23 IL-1β is a potent pro-in ammatory cytokine mostly synthesized macrophages and other innate immune system cells, usually in response to the recognition of pathogen-associated molecular patterns by these cells. 24 CINC-1 is a pro-in ammatory cytokine and a relevant chemoattractant to neutrophils, collaborating to in ltration of these cells in tissues. 25 TNF-α is a pro-in ammatory mediator that drives immune pathways to prevent in ltration or damage to tissues and rules the survival or apoptosis of its target cells. 26 Therefore, as an additional step towards a better insight on the proin ammatory imbalance of the disorder and validation of the BD model induced by OUA, the present study evaluates the behavior of rats and investigates the levels of important in ammatory mediators, i.e., IL-1β, IL-6, IL-10, CINC-1, and TNF-α in the frontal cortex and hippocampus, seven and 14 days after a single ICV administration of the Na + /K + -ATPase inhibitor. Besides, it is also presented the Li effect in this experimental scenario.

Animals
Thirty-two adult male Wistar rats (Rattus norvegicus, heterogenic strain), with 250-350 g body weight, were obtained from the breeding colony at the University of Southern Santa Catarina (UNESC). Animals were housed as ve per cage, with ad libitum food and water and controlled conditions (12 h-light/dark cycle; lights on at 6:00 a.m.; 22 ± 1°C). Experimental procedures started following approval by the UNESC's Ethical Committee on Animal Use for Research (record 66/2010), under the guidelines from the National Institutes of Health (US) "Guide for the Care and Use of Laboratory Animals" 27 and the Brazilian Society for Neuroscience and Behavior.

Surgical procedure
Animals were submitted to anesthesia by intramuscular injection of ketamine (80 mg/kg body weight) and xylazine (10 mg/kg) and accommodated in a stereotaxic apparatus, followed by removing the skin and hair covering the head. After this, it was placed a 27-gauge guide cannula (9 mm) on the surface of the cranial bone, according to the following coordinates: 0.9 mm posterior to bregma; 1.5 mm right from the midline; and 1 mm above the lateral brain ventricle. 28 After marking, it was carrying out a 2 mm ori ce on the animal skull, and a cannula was ventrally implanted 2.6 mm to the superior surface of the bone and xed with dental acrylic cement.
Animals recovered from the procedure within three days, time in which they subcutaneously received tramadol hydrochloride (10 mg/kg body weight) each 12 h to mitigate the postoperative pain, by following the manufacturer's instructions (União Química Farmacêutica Nacional S. A., São Paulo, Brazil).

Experimental design
On the fourth day after the surgical procedure, a 30-gauge cannula was placed inside the guide cannula and linked to a microsyringe by using a polyethylene tube. The tip of the infusion cannula extended 1.0 mm beyond the guide cannula to reach the right lateral brain ventricle. Each animal received a single ICV injection of arti cial cerebrospinal uid (aCSF, 5 µL) or OUA (10 −3 M) dissolved in aCSF (5 µL). Each infusion lasted 30 seconds to prevent the liquid e ux. 18,29 From the day following the aCSF or OUA injection, animals received intraperitoneal injections of saline (Sal, 1 mL/kg) or Li (47.5 mg/kg) twice a day, for 14 days. 30 Therefore, the rats were randomly assigned to four groups (n = 8 per group): 1) aCSF + Sal, 2) aCSF + Li, 3) OUA + Sal, and 4) OUA + Li ( Figure 1).

Behavioral tests Open eld
The open eld test occurred in two periods: seven and 14 days after ICV infusions, to provide insightful information on the locomotion, exploratory, risk-taking, and stereotypical behaviors of the animals. Brie y, the test ran in an apparatus with a 40 cm × 60 cm oor surrounded by 50 cm-high walls made of brown plywood with a frontal wall made of glass. The oor was divided into nine equal rectangles by black lines and covered with a glass base. Each animal was gently placed in the left rear rectangle to explore the arena for 5 minutes (min). The hyperlocomotion indicates a manic-like behavior through the total number of crossings and rearings during the entire test period. 31 Forced swimming The procedure was carried out 13 days following ICV infusions to investigate animal immobility (total immobility or movements to keep the heads out of the water), which is a depressive-like behavior. The test occurred in two days of procedures, where each rat is placed in a cylinder with water (23 ºC) in su cient quantity so that it cannot support its paws on the bottom. On the rst day, animals swam inside the apparatus for 15 min (training session). Twenty-four hours following training, the test session took place in which was assessed the times of swimming, climbing, and immobility for 5 min. 32 Biochemical analysis

Brain samples
On the 14th day following OUA or aCSF infusion, after the behavioral tests, the rats were killed with a guillotine and the brain dissected to obtain the frontal cortex and hippocampus. Procedure occurred on a Petri dish placed on ice, followed by immersion of the samples in liquid nitrogen and subsequent storing at −80°C for subsequent biochemical analysis.

Neuroin ammation parameters
Levels of TNF-α, IL-1β, IL-6, IL-10, and CINC-1 The hippocampus and frontal cortex were homogenized in an extraction solution containing aprotinin (100 mg tissue per 1 mL). The concentration of cytokines/chemokine was determined in the brain structures using commercially available ELISA assays, following instructions supplied by the manufacturer (DuoSet kits, R&D Systems, MN, United States). Data are expressed as pg/100 mg tissue.

Protein quanti cation
Total protein was measured by the Lowry and coworkers' method, 33 with slight modi cations. 34 Bovine serum albumin was used as a standard.

Statistical analysis
All data are present as mean ± standard error of the mean. Variables were analyzed according to their distribution, with the Shapiro-Wilk's test for normality used for this purpose. Differences between groups determined by two-way analysis of variance (ANOVA) followed by Tukey's post hoc test. The software used in the analyzes was Statistica 7 (StatSoft, Inc., Tulsa, OK, United States). Differences were statistically signi cant as p ≤ 0.05.

Behavioral analyses
The open eld test was performed to evaluate the rat locomotion. Assessed parameters included the number of crossings and rearings. Seven days after the rats received a single ICV injection of OUA, they presented a marked pattern of locomotion in comparison to aCSF group animals ( Figure 2). Li administration reversed the increase in the crossings and rearings induced by OUA. Two-way ANOVA revealed signi cant OUA effects in the following seven days after ICV infusions

Discussion
In ammation has been increasingly reported in the research focusing on the BD pathophysiology and therapy, both in clinical trials 35,36 and animal studies. 37 In bipolar patients, in ammatory mediators have been implicated in the disorder, potentially in uencing the progression or severity. 38,39 For instance, the cytokines whose levels are prone to an increase in BD patients, in comparison to healthy controls, include IL-4, IL-6, IL-10, and TNF-α. 40 Additionally, the use of anti-in ammatory drugs as adjunctive therapy to mood stabilizers in the context of bipolar mania has demonstrated promising outcomes. 35,37,41 Accordingly, it will be of great interest that a BD animal model, besides mimic the behavioral alterations of the disorder, also recapitulates the in ammatory alterations that may occur in bipolar patients, which highlights the need for further research in this eld.
In this regard, the present study sought a better insight into the OUA model of BD, which was recently validated by the paper of Valvassori and coworkers. 16 By corroborating data from these authors, the present study demonstrated that OUA administration induces an increase in the number of crossings and rearings seven days after OUA administration in rats, and these alterations are preventable by Li. Additionally, such alterations do not perpetuate until 14 days following the ICV infusions, while the forced swimming test showed a depressive-like behavior ("hopelessness") in the same animals at this period (also preventable by Li). Therefore, our behavioral ndings reinforce the consistency of the OUA administration to model BD, which is a step forward to its validation. Indeed, in bipolar patients, manic episodes last at least seven days, while depressive symptoms are present within 14 days in the same subjects. 2 This pattern is accurately represented in the OUA model. Besides the study carried out by Valvassori et al.,16 our ndings are similar to data provided by other papers, especially in the context of manic-like behaviors following injection of the Na + /K + -ATPase inhibitor. 37,42,43 Moreover, the present study focused on the need to model the pro-in ammatory perspective of BD. 10 To this aim, we measured the levels of relevant in ammatory mediators (IL-1β, IL-6, IL-10, and TNFα) 14 days after OUA ICV infusions. In all parameters evaluated, OUA elicited a signi cant increase, while the Li administration reversed these alterations. Intriguingly, a similar study performed by our research group measuring the same cytokines did not nd any alterations in these parameters, except IL-6, whose levels were decreased in the striatum of rats receiving OUA. 21 Nonetheless, that paper investigated the OUA effect in an acute context, and here the ndings are presented in a subchronic/chronic perspective, with the inclusion of the Li treatment for validation purposes. Data provided by our study mimic the in ammatory pathophysiological alterations reported in patients, 44 which reinforces the validity of the model to simulate the pathophysiological aspects of the disorder. Further studies are required to unravel the precise mechanisms underlying these alterations in the in ammatory parameters. One possibility is that OUA can activate neuronal signaling pathways ruling the release of cytokines, such as the nuclear factor kappa B (NF-κB), which is activated by the glycoside according to prior studies. 45,46 Also, NF-κB upregulation was detected in bipolar patients, potentially contributing to the pro-in ammatory balance of the disorder. 47,48 One potential limitation of the present study is that Na + /K + -ATPase activity was not measured to con rm that OUA effects are related to an inhibition of this enzyme. However, this parameter was not altered by the glycoside in the Valvassori and coworkers' paper, 16 which provided the experimental platform for the present study, although OUA is a well-known inhibitor. That study showed no inhibition on the 14th day after ICV infusion of the drug, time in which were carried out the measurements here. By regarding that the enzyme activity is pivotal for many neuronal roles, a stable decrease in this parameter might generate neurological disturbances, 49,50 so that the protein can conform to the high OUA concentration to prevent disruption of its functions. Another possibility is the fact glycoside effect on the Na + /K + -ATPase may be independent on the ion transport activity of the protein; instead, the drug can activate a signaling complex comprising the enzyme, cluster determinant 36 (CD36), and Toll-like receptor 4, subsequently triggering in ammation driven by  It is worthy to note that experiments from the present study timely mimicked the behavioral alterations of the disorder (face validity), as well as its pathophysiological in ammatory alterations (construct validity). Additionally, Li, a standard drug in BD therapy, was able to mitigate the behavioral and pathophysiological alterations induced in the model (predictive validity), which is the nal criterion to validate OUA administration as a BD model.
In summary, OUA elicited a manic-like behavior seven days after its ICV administration, which was followed by depressive-like alterations several days later. Additionally, this drug induced an increase in all measured in ammatory mediators (IL-1β, IL-6, IL-10, and TNFα) 14 days after ICV infusions. Li administration alleviated or reversed all these disturbances in behavior and neurochemistry, indicating that OUA administration is a valid model to study neuroin ammation in the context of BD. Since some cytokines are also signi cantly altered in bipolar patients, this experimental model might be useful to the screening of drug candidates for the therapy of this condition.   Animals were subjected to the assessments seven and 14 days following OUA infusion. Data were analyzed by two-way analysis of variance followed by Tukey's test when p was signi cant. Values are expressed as mean ± standard error of the mean (arbitrary units or time, in seconds [s]). *p ≤ 0.05, as compared to aCSF + saline (Sal) group; #p ≤ 0.05, as compared to OUA + Sal group  Animals were subjected to the assessments 14 days following OUA infusion. Data were analyzed by twoway analysis of variance followed by Tukey's test when p was signi cant. Values are expressed as mean ± standard error of the mean (picograms per 100 milligrams [pg/100 mg] tissue). *p ≤ 0.05, as compared to aCSF + saline (Sal) group; #p ≤ 0.05, as compared to OUA + Sal group Effect of the intracerebroventricular (ICV) administration of ouabain (OUA) on the levels of interleukin 6 (IL-6) in the frontal cortex (4A) and hippocampus (4B) of adult male Wistar rats (n = 8 per group). Animals were subjected to the assessments 14 days following OUA infusion. Data were analyzed by two-way analysis of variance followed by Tukey's test when p was signi cant. Values are expressed as mean ± standard error of the mean (picograms per 100 milligrams [pg/100 mg] tissue). *p ≤ 0.05, as compared to aCSF + saline (Sal) group; #p ≤ 0.05, as compared to OUA + Sal group Effect of the intracerebroventricular (ICV) administration of ouabain (OUA) on the levels of interleukin 10 (IL-10) in the frontal cortex (5A) and hippocampus (5B) of adult male Wistar rats (n = 8 per group).
Animals were subjected to the assessments 14 days following OUA infusion. Data were analyzed by twoway analysis of variance followed by Tukey's test when p was signi cant. Values are expressed as mean ± standard error of the mean (picograms per 100 milligrams [pg/100 mg] tissue). *p ≤ 0.05, as compared to aCSF + saline (Sal) group; #p ≤ 0.05, as compared to OUA + Sal group Figure 6 Effect of the intracerebroventricular (ICV) administration of ouabain (OUA) on the levels of the tumor necrosis factor α (TNF-α) in the frontal cortex (6A) and hippocampus (6B) of adult male Wistar rats (n = 8 per group). Animals were subjected to the assessments 14 days following OUA infusion. Data were analyzed by two-way analysis of variance followed by Tukey's test when p was signi cant. Values are expressed as mean ± standard error of the mean (picograms per 100 milligrams [pg/100 mg] tissue). *p ≤ 0.05, as compared to aCSF + saline (Sal) group; #p ≤ 0.05, as compared to OUA + Sal group Effect of the intracerebroventricular (ICV) administration of ouabain (OUA) on the levels of the cytokineinduced neutrophil chemoattractant 1 (CINC-1) in the frontal cortex (7A) and hippocampus (7B) of adult male Wistar rats (n = 8 per group). Animals were subjected to the assessments 14 days following OUA infusion. Data were analyzed by two-way analysis of variance followed by Tukey's test when p was signi cant. Values are expressed as mean ± standard error of the mean (picograms per 100 milligrams [pg/100 mg] tissue). *p ≤ 0.05, as compared to aCSF + saline (Sal) group; #p ≤ 0.05, as compared to OUA + Sal group