The Effects of Ionotropic GABA Receptor Blockage on Brain in Sepsis-induced Rats

Encephalopathy develops following sepsis, which is dened as life-threatening organ failure due to the irregular response of a host to infection. It has high mortality and morbidity rates. In this study, we aimed to investigate the effects of inammation on brain tissue, and the effects of the GABA A receptor antagonist bicuculline in rats with sepsis. Sepsis was experimentally generated in rats using LPS. The rats were divided into four groups as control, LPS (10 mg/kg i.p.), bicuculline (1.5 mg/kg bicuculline methiodide s.c.), and LPS + bicuculline. Electrophysiologic recordings and body temperature measurements were completed at the 24th hour after injection, and blood samples were taken from the heart for measurements of biochemical parameters. TNF-α, IL-10, and GABA levels were measured using ELISA, and MDA levels were measured using the Bouge method from brain tissue. Tissue imaging was performed with S100-ß, NEUN, and synaptophysin antibody using immunouorescence staining. One-way ANOVA and the Tukey test were used in statistical analysis. Inammatory parameters increased in brain tissue in the LPS group compared with the control group. The immunouorescence staining results in brain tissues were as follows: S-100ß involvement increased, and NeuN and synaptophysin involvement decreased in the LPS group. In electrophysiologic recordings, activity consistent with acute non-focal seizures was observed in the LPS group; however, it was consistent with the resting status in other groups. We suggest that the GABA A antagonist bicuculline methiodide may be a prophylactic agent in sepsis, which caused the impaired neurotransmitter balance, increased pro-inammatory cytokine and lipid peroxidation, and decreased anti-inammatory cytokine levels.


Introduction
Sepsis and its complications are among the main causes of mortality in approximately 50% of deaths in intensive care units (ICUs) (Brun-Buisson 2000;Tran et al. 1990). In recent years, in ammatory cascades that cause death with multiple organ failure have been extensively investigated. A large number of studies focused on peripheral organs such as the lung, liver, intestine, and kidney; however, few studies investigated brain during sepsis. Septic encephalopathy represents central nervous system (CNS) dysfunction due to sepsis and systemic in ammatory response syndrome (SIRS) (Sprung et al. 1990;Young et al.1990). In addition, encephalopathy, which is the most common nding in ICUs, has a high mortality rate in patients with sepsis (Bleck et al. 1993). Studies on pathophysiologic mechanisms that lead to septic encephalopathy are ongoing.
Microorganisms or various toxin-induced macrophage release pro-in ammatory cytokines such as tumor necrosis factor-α (TNF-α), and interleukins (ILs), which induce an endotoxic response. The environmentally produced in ammatory signal stimulates endogenous expression of pro-in ammatory cytokines of brain cells by crossing the blood-brain barrier (Well et al. 2010). During septic encephalopathy, pro-in ammatory cytokines have been shown to increase the activity of gamma (γ)amino butyric acid (GABA) in the CNS (Rossi et al. 2011). The ionotropic and metabotropic receptors of GABA are commonly found in the CNS, including those receiving cardiovascular and autonomic control.
In hepatic encephalopathy and endotoxic shock, in ammatory molecules such as nuclear factor-κß and mitogen-activated protein kinase ERK (MAPKerk), showed an increase in GABA receptor activity and in the upregulation of central GABAergic transmission in cardiovascular and in ammatory reactions (Hellstrom et al. 2005;Serantes et al. 2006). In studies, the stimulation of macrophage GABA receptors in experimental animals with endotoxemia was reported to activate nuclear factor-κß and MAPKerk pathways. Bicuculline is a well-known blocker of the GABA A receptor, and studies showed that lipopolysaccharide (LPS) was reduced with both hypotensive and in ammatory effects in studies where the receptor was blocked by bicuculline (Sallam et al. 2006). In addition, bicuculline was reported to inhibit acetylcholinesterase in the literature (Olsen et al. 1976). GABA A receptor antagonists were reported to be effective on septic brain tissue. However, few studies have tried to describe the molecular mechanism.
In the present study, we aimed to investigate the effects of ionotropic GABA A receptor blockage in brain tissue on molecular, morphologic, and electrophysiologic levels in rats with LPS-induced sepsis.

Experimental Groups:
Ethics committee approval for the study was obtained from the Local Ethics Committee of Animal Experiments of Istanbul Bagcilar Training and Research Hospital (2017/63). Male Sprague Dawley rats weighing 180-230 g were divided into four groups as controls, LPS, bicuculline, and LPS + bicuculline. The animals were fed with a commercial diet and tap water ad libitum, housed in cages kept at a controlled temperature (22 ± 2°C) and humidity (55-60%) with a 12-hour light/dark cycle.

Experimental Procedures:
LPS from Escherichia coli O55: B5(Sigma Aldrich, Product No: L2880) was dissolved in 1 mL of sterile saline solution, and a single intraperitoneal dose of 10 mg/kg was injected.
Bicuculline methiodide was dissolved in 1 mL of sterile saline solution, and a single subcutaneous dose of 1.5 mg/kg was administered.
Animals were anesthetized using ketamine (90 mg/kg) (i.p.) + xylazine (10 mg/kg) (i.p.) 24 hours after the LPS injection. Under anesthesia, non-invasive electrophysiologic recordings were performed, and rectal body temperatures were measured. Blood samples were taken from the heart, and the animals were sacri ced.
To investigate the levels of TNF-α, IL-10, and GABA using enzyme-linked immunosorbent assay (ELISA), and also to determine malondialdehyde (MDA) levels using the Bouge method, dry total brain tissue was taken and placed in 10% formaldehyde for histopathologic evaluations.

Lipid peroxidation procedure:
Lipid peroxide levels in brain tissue were determined by measuring the MDA levels using the Bouge method. In this method, the end products of lipid peroxidation were spectrophotometrically measured using thiobarbituric acid (TBA, Sigma, Aldrich) in warm, acidic conditions, read at 540 nm.
Immuno uorescence protocol: Brain tissue was xed in 10% formaldehyde overnight. In follow-up procedures, tissue was processed to remove any remaining xative in a series of increased alcohol concentration solutions for 24 hours. The alcohol was removed from the tissue samples, which were then treated with xylene for 1 hour to ensure good penetration of para n wax. Tissues were kept in 58°C pure para n for 2 hours. When the tissues were ready for blocking, they were embedded in para n.
For staining, 10-µm-thick sections were taken from the blocks using a microtome and transferred to polyl-lysine-coated slides.
Multiple uorescence staining was performed on the sections. After the sections were puri ed with para n, xylene was evaporated until the sections became white. Subsequently, the sections were allowed to stand for 10 minutes in cold methanol at -20°C. Finally, for rehydration, the sections were immersed in distilled water for 2 minutes. The epitopes were then incubated with 0.2% Tween 20 for 5 minutes and washed with distilled water, and then boiled in microwave oven with 1 x citrate buffer (citrate buffer, pH 6.0) for 20 minutes. Sections were washed twice at room temperature for 10 min using phosphatebuffered saline (PBS) (pH 7.6). Prior to the primary antibodies being optimized for dilution, NEUN (diluted with PBS containing 2% serum) was dropped onto the sections, which were then covered with a coverslip and kept overnight at + 4°C. The next day, the NEUN secondary antibody was incubated at room temperature for 2 hours, washed with PBS, and then dropped onto sections with SYN (diluted with PBS containing 10% goat serum) or S100 (diluted with 0.3% TritonX-100, PBS containing 0.3% serum), covered, and kept at + 4°C for one night. The following day, the secondary antibodies of the primary antibodies were incubated at room temperature for 2 hours and washed with PBS. The core dye was incubated with Dapi and the preparations were sealed using anti-fading and mounting medium. Histologic changes were evaluated and photographed using uorescence microscopy with FITC, H&L, and DAPI-compatible lters.

Statistics:
Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) [one-way analysis of variance (ANOVA)] software program. The post-hoc Tukey test was used to determine the group that caused the difference. The level of signi cance was accepted as p < 0.05.

Findings of septic conditions:
In the LPS group, the animals were detected to be either hypothermic or hyperthermic in accordance with their metabolic tendencies. The body temperatures of animals in the bicuculline, control, and LPS + bicuculline groups were determined in standard values.
In the LPS group, the leucocyte count was found signi cantly higher compared with the bicuculline (P < 0.01), control, and LPS + bicuculline (P < 0.05) groups. There were no signi cant differences between the other experimental groups (P > 0.05) (Fig. 1).

Cytokine ndings:
In the LPS group, the TNF-α level was higher than in the other groups; however, the result was not signi cant (P > 0.05).
IL-10 levels were signi cantly higher in the LPS group compared with the control (P < 0.05). There were no signi cant differences between the other experimental groups (P > 0.05) (Fig. 2).

GABA ELISA ndings:
In the LPS group, TNF-α levels were found higher compared to other experimental groups. However, the result was not signi cant (P > 0.05) Lipid peroxidation ndings: In the LPS group, MDA levels were higher than in the control group, but the result was not signi cant (P > 0.05) (Fig. 3).

Immuno uorescence ndings:
In the LPS group, S100-ß immunoreactivity was found higher; however, NeuN and synaptophysin immunoreactivity were found lower than in the other experimental groups (P < 0.01). No signi cant changes were detected between the other groups (P > 0.05) (Fig. 4-9).
Neuron-speci c enolase (NSE) ndings: In the LPS group, NSE levels were signi cantly higher than the levels in other groups (P < 0.01). No signi cant change was detected between the other groups (P > 0.05) (

Electrophysiologic ndings:
In the LPS group, signi cant increases were detected in the amplitudes of the electroencephalography waves at the min-max value (P < 0.001). No signi cant changes were detected between the other groups (P > 0.05) ( Table.2).

Discussion
In our study, we aimed to investigate the effects of in ammation on the brain tissue of rats with LPSinduced sepsis, and how these effects could be changed using an ionotropic GABA A receptor antagonist, bicuculline.
The mechanism of neurologic complications in sepsis-induced encephalopathy has not been fully elucidated. Researchers in many studies reported that the brain GABA receptor density and serum GABA levels were increased in experimental septic encephalopathy models, bacteria were reported to possibly produce large amounts of GABA, and bacterial sepsis was a cause of hepatic encephalopathy. Furthermore, recent studies reported that a GABA agonist, benzodiazepine, exacerbated septic encephalopathy (Baraldi et al. 1982;Ferenci et al. 1983;Jones et al. 1984;Minuk et al. 1985;Sallam et al. 2006;Ramirez-Jarquin et al. 2017). Acute brain dysfunction was reported in LPS-induced sepsis, especially with the increase in GABA levels in the hippocampus, and with a decrease in the level of consciousness and neuronal activity (Hellstrom et al. 2005;Serantes et al. 2006;Jacob et al. 2011;Kadoi et al. 1996). Rossi et al. reported that the released proin ammatory molecules increased the central GABA activity during autoimmune encephalomyelitis (Rossi et al. 2011). Sugiura et al. showed that macrophage GABA receptors increased the susceptibility of mice to endotoxemia through regulation by increasing the NF κß, and MAPKERK (Sugiura et al. 2011).
For blocking GABAergic activity using the GABA A receptor blocker bicuculline, and GABA B receptor blocker baclofen, Salam et al. showed that administration of bicuculline improved cardiac, hemodynamic, and in ammatory ndings compared with a baclofen-treated group, and ionotropic GABA A receptors were reported to be more effective in the in ammatory process (Sallam et al. 2006).
Variable levels of neurotransmitter in sepsis led to alterations of cognitive processes, in addition to the damage of neuronal mechanisms and brain tissue (Ramirez-Jarquin et al. 2017). In ammatory mediators were reported to cause increased receptor activity by binding to GABA A receptors by modulating the expression of GABA A receptor subunits with infectious factor, hypoxia or stress (Wang et al. 2012). In our study, we observed that GABA levels increased in total brain tissue; however, the increase was not statistically signi cant (p > 0.05). In the bicuculline and LPS + bicuculline group, we found similar GABA levels to those in the control group.
In their experimental study, Winder et al. observed that GABA levels were elevated in the brain tissue of rats with sepsis. They reported that GABA levels increased, especially in the sub-brain region; however, in a total brain investigation, this increase was not as signi cant as in our study (Winder et al. 1988). Our ndings of GABA levels in total brain tissue were similar with the levels in other studies in the literature.
Recent studies showed that acetylcholine and bicuculline methiodide could activate nicotinic acetylcholine receptors (Hsu et al. 2004). Wang et al. demonstrated that activation of nicotinic acetylcholine receptors inhibited TNF-α synthesis and decreased in ammatory response by increasing IL-10 release (Wang et al. 2003). In addition, activation of nicotinic acetylcholine receptors was reported to be associated with a decrease in the production of NO and reactive oxygen species (Shimohama et al. 1996;Li et al. 2000). Bicuculline methiodide was suggested to have a protective effect after cecal ligation and perforation in a study examining the change of cytokines with bicuculline methiodide. TNF-α is known to trigger the production of free radicals in neonatal rat hepatocytes and reduce oxidative metabolism (New et al. 2001). This activity is positively controlled by various up-regulatory stimuli, including TNF-α and IL-1ß, and is negatively controlled by pro-in ammatory cytokines such as IL-10 Thiemermann et al. 1993).
In our study, we investigated the effects of bicuculline methiodide on IL-10 levels from TNF-α and antiin ammatory cytokines from proin ammatory cytokines in brain tissue in the sepsis group. The TNF-α levels of the bicuculline and LPS + bicuculline groups were in close proximity to the control group. We observed that TNF-α levels increased in total brain tissue in the sepsis group; however, it was not statistically signi cant in the sepsis group (P > 0.05). TNF-α is known to reach peak values at minute 90 and its half-life is shorter. In addition, studies showed that TNF-α concentrations were decreased by stimulating a number of pathways such as anti-in ammatory pathways in long-term sepsis models . This explains why the increase in our ndings was not signi cant. We found that IL-10 levels signi cantly decreased in the sepsis group compared with the other experimental groups (P < 0.05). Our other experimental groups were similar to the control group. The decrease in IL-10 levels was reported as an indicator of the in ammatory pathway induced by TNF-α in the literature (Munford et al. 2005).
Although the NF-κß pathway plays an important role in the regulation of cytokine production, it has been supported in studies that bicuculline methiodide might decrease proin ammatory cytokine production by inducing activation of nicotinic acetylcholine receptors instead of using the NF-κß pathway (Loop et al. 2003). However, researchers reported that nicotinic acetylcholine receptors might be associated with the anti-in ammatory effect of bicuculline methiodide in sepsis (Wang et al. 2003). In another study, bicuculline methiodide was suggested to inhibit the pro-in ammatory cytokine release through nicotinic α-7 receptors (Sallam et al. 2006;Wang et al. 2003).
We observed that TNF and IL-10 cytokine values decreased in the sepsis group in the LPS + bicuculline group. This may be explained by the fact that bicuculline blocks GABA A ionotropic receptors and inhibits acetylcholinesterase and blocks the in ammatory process. The alpha-7 nicotinic Ach receptor agonist GTS-21 was reported to decrease cytokine levels in studies conducted with volunteers who were given endotoxin treatment. Some studies reported that cholinergic neurons were particularly sensitive to systemic in ammation (Dal-Pizzol et al. 2014). These ndings are in line with our results. Bicuculline methiodide, which is known to have acetylcholinesterase inhibition, was suggested to possibly be a good anti-in ammatory agent as a result of stimulating cholinergic pathways, and was supported by other studies (Sallam et al. 2006;Hsu et al. 2004;Thomsen et al. 2012).
Increased ROS production in sepsis causes neuropathy and myopathy leading to destruction of brain and tissue. In studies with lipid peroxidation in sepsis, thiobarbituric acid reactive products were analyzed, and increased MDA levels were reported to have effects such as impaired blood-brain barrier (Martins et al. 2003). In our study, we observed that MDA levels in brain tissue increased in the sepsis group compared with the control group; however, the increase was not statistically signi cant (P > 0.05). In addition, the values in other experimental groups were similar to values in the control group. In experimental studies, MDA levels were reported to increase in the cortex, hippocampus, and cerebellum at hour 6 in sepsis-induced rats. However, this increase was not re ected in the total brain tissue (Takezawa et al. 1983). Our ndings are in parallel with data in the literature. A signi cant increase was shown in MDA levels in the sepsis group; however, no statistically signi cant increase was detected in brain tissues. We suggest that, with this difference, brain tolerates lipid peroxidation using its own immunity.
High serum levels were reported in cases of neuron-speci c enolase (NSE) levels in the brain quanti cation of cardiac injury, cardiac arrest in the clinic, septic shock, brainstem, neuroendocrine tumors, and malignancy (Isgro et al. 2015). Researchers reported that NSE levels in patients with sepsis in ICUs could be used to predict the clinical course and might show high sensitivity and speci city (Weigand et al. 2000). Yao et al. reported that NSE and S-100 β levels were signi cantly higher in patients with septic encephalopathy (Isgro et al. 2015;Yao et al. 2014). In our study, we found that NSE levels signi cantly increased in the sepsis group compared with the other experimental groups (P < 0.01). Our ndings support the data in the literature.
We obtained non-focal epileptic wave-like recordings from our rats with sepsis under mild anesthesia in an investigation of the non-invasive total brain electrical activity records. We found a higher degree of signi cance in the LPS group compared with the other groups in investigations of the minimummaximum ratios of wave amplitudes between the groups. The other experimental groups had similar wave characteristics and minimum-maximum ratios to the control group. In our group, where we used bicuculline methiodide, we detected no convulsion waves, supporting that it may be a suitable agent in GABA blockade according to other GABA antagonists and bicuculline derivatives, in accordance with the literature (Sallam et al.2006).
In the immuno uorescence staining of the neurotransmitter through the synaptophysin antibody, which is a speci c glycoprotein for the synaptic terminals, we observed a small involvement in the brain tissue in the sepsis group compared with the other groups. This shows that neurotransmitter regulation was impaired in the sepsis group. These results indicate that the rats are also affected by sepsis.
We observed that the results were similar in the bicuculline and LPS + bicuculline groups. This suggests that bicuculline had a positive effect on the neurotransmitter regulation of sepsis. S100-ß is one of the markers used to demonstrate neuronal damage in the brain. Studies have shown increased serum levels of S100-ß in sepsis, brain trauma, cerebral stroke, hypoxic ischemia, and encephalopathy. The exact mechanism for secretion of S100-ß has not been fully elucidated; however, it was reported to cause neural tissue damage and was associated with oxidative stress (Isgro et al. 2015;Yao et al. 2014).
In the comparison of the number of damaged cells stained with S100ß antibody in the sepsis group and the number of cells stained with NeuN antibody, which stains live neurons, we observed increased counts of live neurons in our study in the control, bicuculline, and LPS + bicuculline groups compared LPS group.
Our ndings suggest that LPS causes neuronal damage, and bicuculline reduces this damage. However, detailed studies are required to explain the mechanism (Krıijff et al. 2011;Hamed et al. 2009).

Conclusion
In conclusion, we suggest that the GABA A antagonist bicuculline methiodide, which is known as an acetylcholinesterase inhibitor, can be used a curative agent on in ammatory processes observed in lipopolysaccharide-induced sepsis such as imbalanced neurotransmitter levels, increased proin ammatory cytokine levels, decreased anti-in ammatory cytokine levels, and increased lipid peroxidation levels in brain tissue.

Declarations
Compliance with Ethical Standards

Ethic Approval
Ethics committee approval for the study was obtained from the Local Ethics Committee of Animal Experiments of Istanbul Bagcilar Training and Research Hospital (2017/63).

Acknowledgment
We would like to thank Vakur Olgac for histological preparation, and also David F. Chapman for English editing.

Declaration of Con icting Interests
The author(s) declared no potential con icts of interest with respect to the research, authorship, and/or publication of this article.

Funding
This research did not receive any speci c grant from funding agencies in the public, commercial, or notfor-pro t sectors.
Data Availability Statement; The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.