Targeting glycolysis for neuroprotection in early LPS-induced neuroinflammation Authors: Adriana Fernanda K Vizuete, Fernanda Fróes, Marina Seady, Aline Castro Caurio, Osmar Vieira Ramires Junior, Ana Karla Oliveira Leite, Clarissa Penha Farias, Angela TS Wyse, Carlos-Alberto Gonçalves Published: 2024 ABSTRACT AbstractNeuroinflammation is a pathophysiological feature of numerous neurological and psychiatric disorders. The immune response in the central nervous system, driven by microglia and astrocytes, leads to metabolic reprogramming towards aerobic glycolysis, a phenomenon known as the Warburg effect. The control of metabolic reprogramming via immunomodulation may represent a potential therapeutic target for providing protection against neuroinflammation, which contributes to neuronal dysfunction and death in several neurological disorders. For this purpose, we investigated putative neuroprotective effects of the downregulation of aerobic glycolysis using the 3PO inhibitor, and the downregulation of neuroinflammation using MCC950, in the early LPS-induced neuroinflammation model. The LPS-induced shift towards glycolysis, inflammatory and glial changes (IL-1β, NF-κB, COX2, Iba1, GFAP) were reversed by 3PO, which improved animal behavior. Additionally, MCC950 (an NLRP3 inhibitor) downregulated TLR4/Akt/p38 MAPK/NF-κB/STAT3 signaling, expressions of COX2 and IL-1β, and the astrocyte reactivity (decreasing GFAP) induced by early neuroinflammation, resulting in low glucose uptake. Our data support the occurrence of the Warburg effect during early neuroinflammation and suggest potential new approaches for the treatment of brain injury, given the role of neuroinflammation in such events.Keywords: Neuroinflammation, LPS, Aerobic glycolysis, 3PO, MCC950 Introduction Neuroinflammation is a key communication process between the immune system and the central nervous system (CNS) that maintains brain homeostasis and nervous function in the face of immune challenges, such as harmful stimuli and/or damaged cell signaling (Glass et al., 2010). However, acute and chronic exacerbated immune response is detrimental to nervous tissue and is related to dysfunction and neuronal death (Glass et al., 2010; Ellwardt and Zipp, 2014). Indeed, neuroinflammation is a common neurochemical hallmark in several neuronal disorders and neurodegenerative diseases, and it is unclear whether it is a cause or a consequence of the development of these pathologies (Ellwardt and Zipp, 2014; Kwon and Koh, 2020). Immunostimulation by pathogen-associated molecular patterns (PAMPs) and damage associated molecular patterns (DAMPs) leads to reactive gliosis and phenotypic polarization of glial cells, such as microglia and astrocytes, which are considered innate immune cells in the CNS (Colonna and Butovsky, 2017; Colombo and Farina, 2016). Lipopolysaccharide (LPS) is an immunostimulatory endotoxin component of gram negative bacteria walls (Qin et al., 2007) that triggers the immune signaling cascade in cells expressing toll-like receptor 4 (TLR4) (Cunningham et al., 2009). Distinct LPS models induce morphological, metabolic and molecular changes in glial cells (Guerra et al., 2011; Liddelow et al., 2017), and systemic LPS-induced inflammation leads to microglia and astrocyte activation in the CNS (Moreira et al., 2022). The immune response of glial cells is followed by a neurometabolic shift to enhance the glycolytic pathway, constituting changes in energy metabolism that are similar to those of the Warburg phenomenon observed in tumor cells and the peripheral immune system (O'Neill and Pearce, 2016; Vizuete et al., 2022). However, metabolic changes and the role of neuroimmune metabolism in brain injury and neuronal dysfunction requires clarification. Recently, this neuroimmune metabolism shift was associated with neuronal dysfunction during the development of Alzheimer's disease (Traxler et al., 2022). The regulation of the glycolytic pathway in the immune response has been suggested as a neuroprotective strategy in order to reduce activation of neuroinflammation and protect nervous tissue (Vizuete et al., 2022). One potential inhibitor of glycolysis is the small molecule inhibitor 3-(3-pyridinyl)-1-(4-pyridinyl)-2-propen-1-one (3PO), which acts on the substrate-binding domain of the bifunctional enzyme 6-phosphofructose-2-kinase/fructose-2,6-bisphosphatase isoform 3 isozyme (PFKFB3) (Clem et al., 2008). PFKFB3 is an inducible enzyme of the family of bifunctional 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatases (PFK-2/FBPases), which controls the concentration of fructose-2,6-bisphosphate (Fru-2,6-BP), a positive allosteric effector of PFK1, regulating a major point in cellular metabolism and a commitment step in the glycolytic pathway (Yalcin et al., 2009). Therefore, the 3PO inhibitor decreases levels of the allosteric molecule, Fru 2,6-BP, and consequently downmodulates PFK1 activity. Additionally, the downregulation of neuroinflammation, such as via the inhibition of nucleotide-binding domain and leucine-rich repeat protein 3 (NLRP3) inflammasome assembly, represents a potential strategy for preventing the neuroimmunometabolic changes induced by LPS-immune challenge (Vizuete et al., 2022). The NLRP3 is a cytosol multiprotein complex that controls the initiation and progression of the innate immune response in the presence of PAMPs and DAMPs (Abderrazak et al., 2015). MCC950, a diarylsulfonylurea compound, is a selective inhibitor of the NLRP3 inflammasome and is recognized as an important potential anti-inflammatory strategy for mitigating the effects of neurodegenerative diseases by blocking ATP hydrolysis and NLRP3 activation (Zahid et al., 2019). Our group previously reported the effect of several inhibitors in acute hippocampal slice LPS model (Vizuete et al., 2022). However, their potential neuroprotective roles after in vivo neuroinflammation model, especially on neuroimmunometabolic response, needs clarification. Thus, the purpose of this study was to evaluate the downregulation of glycolysis aerobic, using the 3PO inhibitor, and the downregulation of neuroinflammation, using MCC950 during the early LPS-induced neuroinflammation model. Methods Animals and in vivo neuroinflammation model Forty male Wistar rats were obtained from our breeding colony (Department of Biochemistry, UFRGS), and maintained under controlled light and environmental conditions (12 h light/12 h dark cycle at a constant temperature of 22 ± 1 °C) (project number 38546). At postnatal day 25, the animals were anesthetized with ketamine (75 mg/kg) and xylazine (10 mg/kg) via intraperitoneal injection (i.p.) and placed in a stereotactic apparatus in order to insert guide cannulae bilaterally (22 gauge; Plastics One, Roanoke, VA) into the ventricles. A sagittal incision was made in the midline of the scalp. The lateral ventricles were accessed by the following coordinates from the bregma: anteriorposterior [AP] - 0.9 mm; mediolateral [ML] ± 1.5 mm; dorsoventral [DV] - 3.6 mm beneath the surface of the brain (Paxinos and Watson, 2007). After recovery from surgery, the animals were subjected to the in vivo LPS neuroinflammation model (Vizuete et al., 2022), whereby rats were treated with LPS (1 mg/kg, i.p.) or saline (0.9% NaCl, i.p.). After 3 h, bilateral infusions (2 μl, icv) of vehicle (DMSO 0.01%/Kg), the 3PO inhibitor (20 μM/kg) or the MCC950 inhibitor (10 μM/kg) were administered. The doses of 3PO and MCC950 compounds were used in accordance with an ex vivo model of neuroinflammation in acute hippocampal slices (Vizuete et al., 2022). Animals were divided into the following groups: 1) Sham + vehicle (saline i.p., vehicle icv); 2) sham + inhibitor (saline i.p., 3PO/MCC950 icv); 3) LPS + vehicle (LPS i.p., vehicle icv); and 4) LPS + inhibitor (LPS i.p., 3PO/MCC950 icv) (Supplementary Fig. 1). Sickness behavior Sickness behavior was evaluated by measuring rectal body temperature and spontaneous locomotor activity was assessed by the open field task (Clark et al., 2015). Rectal temperature was measured using a digital rectal probe at time 0 h (basal) and at 3 h after saline/LPS ip and/or saline/3PO/MCC950 icv injection. The total distance covered in the open field by animals and their immobility latency were evaluated by the same person, who was blind to group treatment. Brain tissue, serum and CSF samples Rats were anesthetized by i.p. injection of ketamine (75 mg/kg) and xylazine (10 mg/kg). For ventricular access, the anesthetized rats were placed in a stereotaxic apparatus and CSF was obtained by carefully puncturing the cisterna magna. Blood was collected by cardiac puncture and serum was obtained by centrifuging at 1000×g for 10 min (Eppendorf 5402, Hamburg, Germany), before storing at −80 °C. Hippocampi were dissected and transverse slices of 0.3 mm were obtained using a McIlwain Tissue Chopper. Samples were stored at −80 °C until biochemical and immunological assays. Glucose uptake The glucose uptake assay was performed with hippocampal slices, as previously described (Pellerin and Magistretti, 1994). Briefly, slices were incubated at 35 °C in HBSS. The assay was initiated by the addition of 0.1 μCi/well D-[3-3H] glucose. The incubation was stopped after 30 min by removing the medium and rinsing the slices twice with ice-cold HBSS. Hippocampus tissue was then lysed in a 0.5 M NaOH solution. Glucose uptake was calculated by subtracting the non-specific uptake, obtained by the glucose transport inhibitor, cytochalasin B (25 μM), from the total uptake. Results are expressed as nmol/mg protein/min. Determination of mitochondrial parameters Activities of succinate:DCIP-oxidoreductase (complex II) and succinate dehydrogenase (SDH) Enzyme activities were determined by the decrease of absorbance of 2,6-dichloroindophenol (DCIP) at 600 nm, with 700 nm as the reference wavelength (complex II), and in the presence of phenazine methosulfate (PMS) (Fischer et al., 1985). To access the mitochondrial enzymes, the mitochondrial membranes were broken by thawing and refreezing of supernatants three times prior to the test. Briefly, the supernatant was preincubated in a reactional medium containing 16 mmol/L succinate, 40 mmol/L PBS (pH 7.4), and 8 μmol/L DCIP, at 30 °C for 20 min. Afterwards, 7 μmol/L rotenone, 4 mmol/L sodium azide, and 40 μmol/L DCIP were added to the medium and the activity of Complex II was measured for 5 min. Briefly, 1 mmol/L PMS was added to the reaction and the activity was measured for 5 min more for the SDH activity assay (dos Santos et al., 2019). Results are expressed as units/mg protein. Complex IV (cytochrome c oxidase) activity Complex IV activity was determined by the decrease in absorbance due to the oxidation of previously reduced cytochrome c at 550 nm, with 580 nm as a reference wavelength (Rustin et al., 1994). Four μg of protein homogenate was incubated with 10 mmol/L PBS (pH 7.4), 0.6 mmol/L n-dodecyl-β-D-maltoside and 7.0 μg of reduced cytochrome c for approximately 10 min, at 25 °C (dos Santos et al., 2019). Activity was measured for 5 min. Results are expressed as units/mg protein. Glycolytic enzymatic activity assays Hexokinase (HK) activity Hexokinase activity was evaluated by NADH formation at 340 nm. The assay was performed in medium containing 10 mM Tris-HCl, pH 7.4, and 0.1 mM Ap5A as an inhibitor of AK, 5 mM D-glucose, 10 mM MgCl2, 1 mM ATP, 1 mM NAD+, and 1 unit/mL Glc-6-P dehydrogenase (Leuconostoc mesenteroides) (Da-Silva et al., 2004). Results are expressed as μM/min/mg protein. Phosphofructokinase-1 (PFK1) activity PFK1 activity was measured using a kit and according to the manufacturer's instructions (Sigma, Saint Louis, MO, USA). Results are expressed as μmol of NADH/min/mg protein. Piruvate kinase (PK) activity PK activity was assayed by the incubation of the mitochondria-free supernatant with 0.1 M Tris/HCl buffer (pH 7.4), 10 mM MgCl2, 0.16 mM NADH, 75 mM KCl, 5.0 mM ADP, 7 units of L-lactate dehydrogenase (LDH), 0.1% (v/v). Triton X-100, and 10 μL. After a pre-incubation of 30 min at 37 °C, the reaction was started by the addition of 1.0 mM phosphoenolpyruvate. NADH oxidation was recorded for 2 min at 340 nm (Leong et al., 1981). Results are expressed as μM of pyruvate/min/mg protein. Oxidative stress parameters Reactive oxygen species (ROS) assay ROS generation was estimated using 2′7′- dichlorodihydrofluorescein diacetate (DCFH-DA), which is converted into highly fluorescent DCF by ROS (Lu et al., 2013). The resulting hippocampal homogenate was incubated with DCFH-DA (5 μM) for 30 min at 37 °C. Data are expressed as AU/mg protein. Thiobarbituric acid reactive substances measurement (TBA-RS) TBA-RS levels were measured in 200 μL of samples that were incubated for 1 h with 50 μL of 8.1% SDS; 375 μL of 20% acetic acid in aqueous solution (v/v), pH 3.5; and 375 μL of 0.8% thiobarbituric acid (Ohkawa et al., 1979). The mixture was then allowed to cool down for 5 min and was centrifuged at 750 ×g for 10 min. The reaction product was measured at 532 nm. Results are expressed as nmol of TBA-RS/mg protein. Total sulfhydryl content Sulfhydryl content was measured based on the reduction of 5,5′-dithio-bis (2-nitrobenzoic acid) (DTNB) by thiols, generating a yellow derivative (TNB) whose absorption is measured at 412 nm (Aksenov and Markesbery, 2001). Briefly, 0.1 mM DTNB was added to 120 μL of supernatants. This medium was incubated for 30 min at room temperature in a dark room. Total sulfhydryl content is inversely correlated to oxidative damage to proteins. Data are expressed as nmol of TNB/mg protein. Carbonyl content Protein carbonylation was evaluated in 200 μL of samples incubated with 800 μL of 2,4-dinitrophenylhydrazine (DNPH) or 2 M HCl in a dark place for 1 h, with mixing every 15 min (Levine et al., 1990). Samples were precipitated in 1 mL of 20% TCA and subsequently centrifuged at 10,000 ×g for 10 min at 4 °C. The same procedure was performed once with 10% TCA. The precipitate was washed with 1 mL of ethanol:ethyl acetate (1:1) 3 times and resuspended in 500 μL of 2% SDS at 37 °C. Data are expressed in nmol of carbonyl/mg protein. Catalase activity Catalase (CAT) activity was measured in the supernatant (Aebi, 1984) using hydrogen peroxide as substrate. The consumption of H2O2 was monitored spectrophotometrically (240 nm) at 37 °C. CAT activity is reported as U CAT/mg protein. Superoxide dismutase activity Superoxide dismutase (SOD) activity was measured in the supernatant (Misra and Fridovich, 1972). The auto-oxidation of adrenalin to adrenochrome at alkaline pH was monitored spectrophotometrically (480 nm) at 32 °C. SOD activity is reported as U SOD/mg protein (one unit, U = the amount of sample which inhibits the transformation of adrenaline into adrenochrome by 50%). Glutathione content Reduced glutathione (GSH) content was determined by proteins precipitation with 1.7% meta-phosphoric acid. O-phthaldialdehyde (1 mg/mL methanol) was added to the supernatant at room temperature for 15min at room temperature for 15min. Fluorescence was measured using excitation and emission wavelengths of 350 and 420 nm, respectively (Allen et al., 2000). Glutathione results are expressed as nmol/mg protein. Measurement of IL-1β cytokine IL-1β cytokine was measured in hippocampus and serum samples. Hippocampal slices were homogenized in PBS containing (in mM) 50 NaCl, 18 Na2HPO4, 83 NaH2PO4.H2O, pH 7.4, with 1 mM EGTA and 1 mM PMSF, followed by centrifugation at 1000 ×g for 5 min at 4 °C. IL-1β was measured using a commercial kit and according to the manufacturer's instructions (eBioscience, San Diego, USA). Data are expressed in pg/mg protein (tissue samples) or pg/mL (serum). S100B measurement Slices were homogenized in PBS with 1 mM EGTA and 1 mM PMSF. The S100B contents in the CSF and brain tissue were measured by ELISA, as described previously (Leite et al., 2008). Briefly, 50 μl of sample plus 50 μl of Tris buffer were incubated for 2 h on a microtiter plate that was previously coated with the anti-S100B monoclonal antibody, SH-B1. Polyclonal anti-S100 was incubated for 30 min and peroxidase-conjugated anti-rabbit antibody was then added for a further 30 min. The reaction with o-Phenylenediamine (OPD) was measured at 492 nm. Data are expressed in ng/mg protein (tissue samples) or ng/mL (CSF). GFAP measurement GFAP content was measured by ELISA, as described previously (Tramontina et al., 2007). The ELISA for GFAP was carried out by coating wells of 96-well plates with 100 μl samples containing 70 μg of protein, overnight at 4 °C. Wells were incubated with a rabbit polyclonal anti-GFAP antibody for 2 h, followed by incubation with a secondary antibody conjugated with peroxidase for 1 h, at room temperature. The color reaction with OPD was measured at 492 nm. Data are expressed in ng/mg protein. Western blotting Nitrocellulose membranes were blocked overnight at 4 °C with 2% bovine serum albumin in Tris-buffered saline (TBS), in mM: 10 Tris, 150 NaCl, pH 7.5 and 0.05% Tween 20®, and then incubated overnight at 4 °C in blocking solution containing primary antibodies (Supplementary Table 1). Subsequently, membranes were incubated for 1 h at room temperature in a solution containing horseradish peroxidase (HRP)-conjugated anti-rabbit IgG, HRP-conjugated anti-mouse IgG or HRP-conjugated anti-goat. A chemiluminescence signal was detected by a luminol substrate reaction (ECL Western Blotting System, GE Healthcare®). Immunoblots were quantified by membrane scanning in an Image4000 (GE Healthcare®), optical densities of proteins studied were determined by ImageJ software (Packard Instrument Company) and the protein/actin ratio was calculated. Protein measurement Protein was measured by Lowry's method, modified by Peterson, using bovine serum albumin as a standard (Peterson, 1977). Statistical analysis Experiments with the 3PO and MCC950 inhibitors were performed at different times, and independently. Initially, data normality was evaluated using Shapiro-Wilk test and the results were considered parametric. All results were expressed as mean ± standard error mean (SEM). Multiple comparisons among groups were performed by one-way analysis of variance (ANOVA) followed by Tukey's test and two-way analysis of variance (ANOVA) followed by Bonferroni's test for rectal temperature. The level of statistical significance was set at P < 0.05. All analyses were performed using the Prism 7.0 software (GraphPad). Results 3PO reverses the LPS-induced glycolytic changes in the hippocampus To investigate whether neuroinflammation causes a shift in neurometabolism, we evaluated the activation of the glycolytic during the early LPS-induced neuroinflammation model and the potential neuroprotective role of 3PO (icv administration), an inibitor of PFKFB3. The neurometabolic shift towards increased aerobic glycolysis, in response to early neuroinflammation, has been previously reported (Vizuete et al., 2022). We sought to confirm glycolytic pathway stimulation in the LPS neuroinflammation model (Fig. 1) by measuring the glycolytic parameters; glucose uptake, HK, PFK1 and PK enzyme activities, lactate and glucose levels in CSF, and phosphorylation of the p38 MAPK enzyme. At 6 h after LPS immunochallenge, several glycolytic parameters were triggered in the hippocampus, such as elevations in glucose uptake (Fig. 1A, P = 0.0035), and in PFK1 and PK activities (Fig. 1B, P = 0.0040 and P = 0.0195, respectively). In the CSF, LPS administration increased lactate levels and reduced the glucose content (Fig. 1C, P < 0.0001 and P = 0.0001, respectively). The 3PO inhibitor reversed the alterations in glycolytic parameters that were induced by early LPS-neuroinflammation, reducing glucose uptake, and PFK1 and PK activities, without affecting metabolite contents in the CSF (lactate and glucose). Interestingly, 3PO downregulated the phosphorylation of p38 MAPK, an indicator of cellular response to stress, in LPS animals (Fig. 1D, P = 0.0307) (Supplementary Fig. 2). Moreover, to further demonstrate the enhancement of glycolytic flux, induced by early neuroinflammation, we evaluated mitochondrial parameters (Fig. 2). LPS and/or 3PO did not affect complex II activity (Fig. 2A, P = 0.2090), SDH activity (Fig. 2B, P = 0.2915), or complex IV activity (Fig. 2C, P = 0.0575). 3PO reverses the LPS-induced glycolytic changes in the hippocampus. Glucose uptake was measured by radioactivity assay. CSF glucose and lactate levels were evaluated by spectrophotometric method. Hexokinase (HK), Phosphofructokinase-1 (PFK1) and Pyruvate Kinase (PK) activities were analyzed by kinetic assay. The phosphorylation of p38 MAPK was evaluated in total homogenates by Western blotting. The 3PO inhibitor reverses the increase in glucose uptake (A), PFK1 and PK activities (B), and phosphorylation of p38 MAPK (D), induced by LPS. LPS changed energy metabolism intermediates in the CSF. Glucose levels were reduced, while lactate levels increased in the CSF, and the 3PO inhibitor did not reverse the metabolite levels (C). Values are expressed as means ± standard error. Data were analyzed by one-way ANOVA, followed by the Tukey test, assuming P < 0.05. ∗ means significant difference, when comparing Sal + Veh to LPS + Veh, # means significant difference, when comparing LPS + Veh to LPS+3PO. LPS and/or 3PO do not affect mitochondrial activity. Complex II, Succinate dehydrogenase (SDH) and Complex IV activities were measured by spectrophotometric method. LPS and the 3PO inhibitor did not affect the complex II (A), SDH (B) and complex IV (C) activities. Values are expressed as means ± standard error. Data were analyzed by one-way ANOVA, followed by the Tukey test, assuming P < 0.05. ∗ means significant difference, when comparing Sal + Veh to LPS + Veh, # means significant difference, when comparing LPS + Veh to LPS+3PO. 3PO reverses the activation of neuroinflammatory pathways and astrocyte reactivity in the hippocampus Having observed the shift in neurometabolism to glycolysis for the maintenance of early neuroinflammation, we hypothesized that the negative modulation of glycolysis by 3PO provides a neuroprotective effect by reducing the energy available for producing inflammatory mediators. As such, we evaluated several proteins that play upstream and downstream roles in neuroinflammation signaling (Fig. 3) (Supplementary Figs. 3–5). LPS upregulated the protein expressions of receptors related to inflammation activation, such as TLR4 (Fig. 3B, P = 0.0426) and RAGE (tendency towards elevation, Fig. 3B, P = 0.0661), and stimulated the signaling of key kinases cascades by raising the phosphorylation of the Akt and Erk enzymes, and the transcription factor NF-κB (Fig. 3C, P = 0.0007, P = 0.0128, and P = 0.0113, respectively). LPS induced the expressions of COX2 and Iba1, a microglial biomarker (Fig. 3D, P = 0.0296 and P = 0.0055, respectively). Moreover, LPS increased IL-1β levels in both the hippocampus and serum (Fig. 3E and F, P = 0.0016 and P = 0.0056, respectively). The 3PO inhibitor reversed the effects of LPS on inflammatory parameters, such as Erk phosphorylation, NF-κB, COX2, Iba1 and IL-1β. Interestingly, 3PO did not reduce the LPS-induced increase in serum IL-1β levels, demonstrating a selective effect on nervous tissue, since its administration was intracerebroventricular (Fig. 3F, P = 0.0056). 3PO reverses the activation of neuroinflammatory pathways in the hippocampus. Pro-inflammatory immunocontent of cytokine IL-1β was measured by ELISA. Protein expressions of TLR4, RAGE, phosphorylation of Akt and Erk, the transcription factor NF-κB, COX-2, and Iba1 were evaluated in total homogenates by Western bloting. Representative images of Western blot data (A). LPS induces early neuroinflammation and promotes elevation of TLR4 (B) and NF-κB (C), phosphorylation of Akt and Erk (C), COX2 and Iba1 protein expression (D) and increases IL-1β in the hippocampus (E) and serum (F). 3PO reverses the activation of neuroinflammation, without changes in serum IL-1β levels. Values are expressed as means ± standard error. Data were analyzed by one-way ANOVA, followed by the Tukey test, assuming P < 0.05. ∗ means significant difference, when comparing Sal + Veh to LPS + Veh, # means significant difference, when comparing LPS + Veh to LPS+3PO. Neuroinflammation activation also leads to the astrocyte reactivity (Fig. 4). LPS promoted S100B secretion as a whole in the brain, increasing levels of this protein in the CSF (Fig. 4A, P = 0.0103), without changes in the hippocampus (Fig. 4B, P = 0.0707). The 3PO inhibitor did not reverse the increase in S100B levels, but reversed the effect of LPS on GFAP, a biomarker of astrocytes (Fig. 4C, P = 0.0016). 3PO reverses LPS astrocyte reactivity induction in the hippocampus. S100B and GFAP were measured by ELISA. LPS elevates S100B protein levels in the CSF (A), without changes in protein content in the hippocampus (B). The 3PO inhibitor did not alter the change in S100B in the CSF (A). LPS increases GFAP immunocontent and the PFKFB3 inhibitor reverses the increase in GFAP (C). Values are expressed as means ± standard error. Data were analyzed by one-way ANOVA, followed by the Tukey test, assuming P < 0.05. ∗ means significant difference, when comparing Sal + Veh to LPS + Veh, # means significant difference, when comparing LPS + Veh to LPS+3PO. MCC950 reverses the LPS the activation of neuroinflammatory pathways, astrocyte reactivity and oxidative stress in the hippocampus To confirm the role of neuroinflammation in the metabolic shift to glycolysis, we looked at the contribution of the NLRP3 inflammasome to early LPS-induced neuroinflammation using MCC950, a selective inhibitor of NLRP3. MCC950 modulated neuroinflammatory signaling, downregulating NLRP3 mRNA and, consequently, the maturation of IL-1β (Fig. 5) (Supplementary Fig. 6). LPS augmented the gene expressions of NLRP3 and STAT3 (Fig. 5A, P < 0.0001 and P = 0.0052, respectively), and the protein levels of NF-κB (Fig. 5B, P = 0.0043), TLR4 (Fig. 5C, P = 0.0017) and IL-1β (Fig. 5E, P = 0.0090). MCC950 reversed the alterations in the neuroinflammation pathway that were triggered by LPS in the hippocampus and reduced the Akt phosphorylation and the expression of COX2 (Fig. 5D, P = 0.0046 and P = 0.0480, respectively), when compared to the LPS group. MCC950 did not affect the LPS-induced increase in serum IL-1β levels, demonstrating a selective effect on nervous tissue, since its administration was intracerebroventricular (Fig. 5F, P = 0.0055). MCC950 reverses the LPS the activation of neuroinflammatory pathways in the hippocampus. Gene expressions of NLRP3 and STAT3 were evaluated by RT-PCR. Protein expressions of TLR4, RAGE, phosphorylation of Akt, the transcription factor NF-κB, COX-2, were evaluated in total homogenates by Western blotting. Pro-inflammatory immunocontent of cytokine IL-1β was measured by ELISA. MCC950 reverses the elevations in NLRP3 and STAT3 mRNA expression (A), and the increase in NF-κB protein levels (B), induced by early LPS-neuroinflammation. LPS increases TLR4 (C), phosphorylation of Akt and COX2 protein expression (D), and increases IL-1β in the hippocampus (E) and serum (F). MCC950 reverses the activation of neuroinflammation, without changing serum IL-1β levels. Values are expressed as means ± standard error. Data were analyzed by one-way ANOVA, followed by the Tukey test, assuming P < 0.05. ∗ means significant difference, when comparing Sal + Veh to LPS + Veh, # means significant difference, when comparing LPS + Veh to LPS + MCC950. We investigated the neuroprotective role of MCC950 in the oxidative stress induced by LPS-neuroinflammation in the hippocampus, since previous studies demonstrated that NLRP3 activation acts as a sensor of cellular stress, including oxidative stress, and that MCC950 can reduce oxidative stress (Abderrazak et al., 2015; Joaquim et al., 2023) (Fig. 6). MCC950 prevented LPS-induced oxidative damage and reversed the elevation of lipid peroxidation and oxidative modified protein promoted by LPS (Fig. 6A, P > 0.0001 and P = 0.0028, respectively). We did not observe any change in reactive species production between the animal groups (Fig. 6A, P = 0.5470). The inhibitor did not reverse the reduction in sulfhydryl content induced by LPS (Fig. 6A, P = 0.0005). However, MCC950 augmented oxidative defenses and reversed the reductions in CAT and SOD activities (Fig. 6B, P = 0.0051 and P = 0.0139, respectively), and the low GSH content (Fig. 6B, P = 0.0263), induced by neuroinflammation. MCC950 reverses the oxidative stress induced by LPS in the hippocampus. Total reactive oxygen species (DCF), TBARS, SH, Carbonyl and GSH contents were measured by a spectrophotometric or fluorescent method. SOD and CAT activities were analyzed by kinetic assay. LPS promotes oxidative damage by increasing TBARS and carbonyl content, and reducing SH content. MCC950 reverses the lipid peroxidation and oxidative protein modification (A). Early neuroinflammation affects the antioxidant defense by reducing the CAT and SOD activities, and the reduced GSH content. MCC950 increases both antioxidant parameters (B). Values are expressed as means ± standard error. Data were analyzed by one-way ANOVA, followed by the Tukey test, assuming P < 0.05. ∗ means significant difference, when comparing Sal + Veh to LPS + Veh, # means significant difference, when comparing LPS + Veh to LPS + MCC950. The astrocytic response followed the same pattern as described before (Fig. 7). MCC950 reversed the increase in GFAP in the hippocampus (Fig. 7C, P = 0.0064), without affecting the high S100B content in the CSF (Fig. 7A, P < 0.0001), promoted by the neuroinflammation. The astrocytic response followed the same pattern as described before (Figue 7). MCC950 reversed the increase in GFAP in the hippocampus (Fig. 7C, P = 0.0064), without affecting the high S100B content in the CSF (Fig. 7A, P < 0.0001), promoted by the neuroinflammation. MCC950 reverses LPS induction of astrocyte reactivity in the hippocampus. S100B and GFAP were measured by ELISA. LPS elevates S100B protein levels in the CSF (A), without changes in protein content in the hippocampus (B). The MCC950 inhibitor did not alter the change in S100B in the CSF (A). LPS increases GFAP immunocontent and the inhibitor reverses the increment in GFAP (C). Values are expressed as means ± standard error. Data were analyzed by one-way ANOVA, followed by the Tukey test, assuming P < 0.05. ∗ means significant difference, when comparing Sal + Veh to LPS + Veh, # means significant difference, when comparing LPS + Veh to LPS + MCC950. MCC950 partially reverses the LPS-induced glycolytic changes in the hippocampus The downregulation of neuroinflammation by MCC950 partially prevented the neuroimmunometabolic shift to glycolysis (Fig. 8), where MCC950 reversed the increase in glucose uptake (Fig. 8A, P = 0.0056) and p38 MAPK phosphorylation (Fig. 8D, P = 0.0049) (Supplementary Fig. 7). However, MCC950 did not affect the increments in PFK1 and PK activities (Fig. 8B, P = 0.0151 and P = 0.0125, respectively), or the changes in metabolite content in the CSF (lactate and glucose) (Fig. 8C, P = 0.0010 and P = 0.0136, respectively), that were associated with early LPS-induced neuroinflammation. MCC950 partially reverses the LPS-induced glycolytic changes in the hippocampus. Glucose uptake was measured by radioactivity assay. Glucose and lactate CSF levels were evaluated by spectrophotometric method. Hexokinase (HK), Phosphofructokinase-1 (PFK1) and Pyruvate Kinase (PK) activities were analyzed by kinetic assay. The phosphorylation of p38 MAPK was evaluated in total homogenates by Western blotting. The MCC950 inhibitor reverses the increase in glucose uptake (A) and the phosphorylation of p38 MAPK (D) induced by LPS, without effects on PFK1 and PK activities (B), and glucose and lactate metabolites in the CSF (C). Values are expressed as means ± standard error. Data were analyzed by one-way ANOVA, followed by the Tukey test, assuming P < 0.05. ∗ means significant difference, when comparing Sal + Veh to LPS + Veh, # means significant difference, when comparing LPS + Veh to LPS + MCC950. 3PO reverses LPS-induced sickness behavior Finally, we evaluated the effects of both inhibitors, 3PO and MCC950, on sickness behavior in animals (Fig. 9). Neither of the inhibitors reversed the elevation in rectal temperature at 3 h after LPS injection (Fig. 9A and D, P = 0.0017 and P = 0.6253, respectively). Early neuroinflammation affected exploratory behavior by reducing the distance travelled (Fig. 9B and E, P = 0.0001 and P < 0.001) and increasing immobility latency (Fig. 9C and F, P = 0.0001 and P = 0.0067). Only 3PO improved the behavior of the animals and reversed the LPS-induced sickness behavior, suggesting a cerebral action of 3PO, given that we injected the drug icv. 3PO reverses LPS-induced sickness behavior. Rectal temperature was measured by digital rectal probe at time 0 h (basal) and 3 h after saline/LPS ip and saline/3PO icv injection. The open field task was evaluated at 3 h after model induction. LPS increases the rectal temperature; neither 3PO nor MCC950 reversed this effect (A, D). LPS induced lethargy, shown by the reduced distance travelled (B, E), and elevated immobility (C, F); these effects were only reversed by the 3PO inhibitor. Values are expressed as means ± standard error. Data were analyzed by two-way ANOVA, followed by the Bonferroni test. Data were analyzed by one-way ANOVA, followed by the Tukey test, assuming P < 0.05. ∗ means significant difference, when comparing Sal + Veh to LPS + Veh, # means significant difference, when comparing LPS + Veh to LPS+3PO/LPS + MCC950. Discussion Neuroinflammation is a pathophysiological feature that is common to many neurological disorders, such as traumatic brain injuries, epilepsy, and neurodegenerative diseases. These disorders are characterized by extensive functional and morphological changes in brain cells, including glial cells (Mishra et al., 2021). The immune responses in the CNS lead to metabolic reprogramming and glycolytic pathway switching (Vizuete et al., 2022; Traxler et al., 2022), via an energetic shift similar that observed in the peripheral immune system and during tumoral metabolism responses, also known as the Warburg effect (O'Neill and Pearce, 2016). The control of this metabolic reprogramming is a potential therapeutic target for immunomodulation and, consequently, represents a neuroprotective approach for the several neurological disorders in which neuroinflammation contributes to neuronal dysfunction and death (Vizuete et al., 2022; Traxler et al., 2022). In this study, we used the LPS-induced neuroinflammation model to access the neuroprotective effect of glycolysis downregulation in the hippocampal structure using the 3PO inhibitor. We also employed MCC950, an inhibitor of NLRP3 inflammasome assembly, to reinforce the role of the Warburg effect in early neuroinflammation signaling, and as a potential strategy to control LPS-induced neuroinflammation. The 3PO inhibitor reduces glycolytic flux by inhibiting PFKFB3 activity in neoplastic cells, macrophages and in the hippocampus structure (Vizuete et al., 2022; Clem et al., 2008; Finucane et al., 2019), and is a promising neuroprotective drug for reducing cerebral ischemia-reperfusion injury (Burmistrova et al., 2019), stroke (Yan et al., 2022), LPS-induced neuroinflammation (Vizuete et al., 2022), and neurodegenerative disease (Dempsey et al., 2017). In contrast, the MCC950 inhibitor maintains the NLRP3 inflammasome inactive, thereby reducing oxidative stress and neuroinflammation. This inhibitor has been reported to prevent reactive gliosis in the sepsis model (Danielski et al., 2020), in the LPS-neuroinflammation model (Dolunay et al., 2016), during ischemia (Joaquim et al., 2023), and Alzheimer's disease (Dempsey et al., 2017). In the CNS, astrocytes are predominantly glycolytic and demonstrate high expression and activity of the PFKFB3 (Bolaños, 2016). However, during the neuroimmune response, upregulation of PFKFB3 occurs in both reactive astrocytes and microglia, in association with aerobic glycolysis (Backes et al., 2016). We observed a shift in glycolytic flux that was induced by the LPS model. Early neuroinflammation (6 h after LPS administration) increased glycolytic parameters, such as glucose uptake, PFK1 and PK enzyme activities, and phosphorylation of p38 MAPK, in the hippocampus. We also observed an increase in glucose consumption as the energetic fuel in the brain, as shown by the high lactate levels and the low glucose content in CSF. The increase in glycolytic demand promotes and maintains neuroimmunometabolism in response to the LPS immunological challenge, and was independent of oxidative metabolism, since we did not observe any change in mitochondrial parameters. In fact, LPS leads to phenotype reactive changes in macrophages, microglia and astrocytes, and induces metabolic activity shifts, directing bioenergetics to the glycolysis flux (Vizuete et al., 2022; Yan et al., 2022; Palsson-Mcdermott et al., 2015). Although PFKFB3 inhibition with 3PO did not affect the lactate and glucose contents in the CSF, it downregulated the glycolysis flux in the hippocampus. The inhibitor reversed the effect of LPS on the neuroimmunometabolic shift to glycolysis by reducing glucose uptake, PFK1 and PK activities, and the phosphorylation of p38 MAPK. In fact, the mechanism of action of 3PO occurs indirectly from its PFK1 activity inhibition (due to a reduction in the intermediary effector, Fru-2,6-BP), and the downregulation of PK observed is a consequence of the reduction of Fru-1,6-BP (Alves-Filho and Pålsson-McDermott, 2016). In contrast, the downmodulation of neuroinflammation by the MCC950 inhibitor partially reversed the increase in glycolytic flux, since the inhibitor reduced the glucose uptake and the phosphorylation of p38 MAPK, without affecting glycolytic enzymes, lactate and glucose CSF levels. Brain energy metabolism dysfunction is associated with numerous neurological disorders (Zhang et al., 2021), and brain injury leads elevates glycolytic flux in glial cells (Vizuete et al., 2022). The 3PO inhibitor controls the glycolysis shift and diminishes neuroimmunometabolism more efficiently than MCC950, since it inhibits PFK1 activity. However, it is notable that the inhibition of neuroinflammation has a certain impact on brain energy metabolism. NLRP3 is a sensor of stress cellular environment, a key metabolic driver of several conditions and responsible for the initiation and progression of metaflammation (i.e., metabolically-induced inflammation), and related diseases (Sokolova et al., 2020). It has been suggested that the NLRP3 inflammasome can sense changes in glycolytic flux (Olona et al., 2022); however, the effect of glycolytic flux on NLRP3 inflammasome activity needs to be clarified (Watanabe et al., 2020). For example, the inhibition of hexokinase by N-acetylglucosamine promotes NLRP3 inflammasome activation (Wolf et al., 2016), but the 2-deoxyglucose inhibitor of hexokinase has contrasting effects on NLRP3 activity (Olona et al., 2022). Moreover, NLRP3 inflammasome activation upregulates glycolysis flux by increasing PFKFB3, and consequently PFK1 activity (Finucane et al., 2019), and the PKM2 isozyme promotes NLRP3 inflammasome activation (Xie et al., 2016). LPS upregulated several biochemical hallmarks of neuroinflammation signaling in the hippocampus. We observed an increase in TLR4 expression, leading to downstream activation and amplification of the intracellular signaling cascade. This occurs via phosphorylation of p38 MAPK, Akt and Erk, which are key enzymes in the immune cellular response (Cianciulli et al., 2016), and the consequent elevation of the transcription factors, NLRP3, STAT3 and NF-κB, and the expressions of COX2 and the IL-1β. Furthermore, LPS augmented the glial response by increasing microglial and astrocytic biomarkers, Iba1 and GFAP, and the secretion of S100B into the CSF. Our data provide further evidence that the LPS/TLR4 signaling pathway stimulates the innate immune response in glial cells (Pålsson-McDermott and O'Neill, 2004). The control of cerebral innate immune pathways has been suggested as a potential target to improve the nervous tissue in neurodegenerative disorders where inflammation is a common pathological event (Ellwardt and Zipp, 2014). We focused on the strategy of regulating NLRP3 inflammasome assembly using the MCC950 inhibitor due to its role as a sensor of glycolysis (Olona et al., 2022), its global brain expression in neurons, microglia and astrocytes (Liu et al., 2013), its activation by the TLR4 downstream pathway (Dolunay et al., 2016) and its importance in the innate imune response (Abderrazak et al., 2015). This approach resulted in the downregulation of NLRP3 inflammasome assembly in the hippocampus, thereby decreasing the TLR4/Akt/p38 MAPK/NF-κB/STAT3 signaling, COX2 and IL-1β expression, and astrocyte reactivity (decreased GFAP) that were induced by early neuroinflammation. Moreover, the inhibitor prevented the oxidative stress associated with LPS-induced neuroinflammation by improving antioxidant defenses (CAT and SOD enzyme activities, GSH content), and reducing oxidative damage in lipids (TBARS) and proteins (carbonyl content), according to previous studies (Joaquim et al., 2023; Danielski et al., 2020). We demonstrated that the downregulation of metabolic reprogramming by 3PO in glial cells affects the molecular cascade of LPS-induced neuroinflammation. The 3PO inhibitor does not directly regulate any mediator of the inflammation pathway, but specifically acts on the PFKFB3 enzyme and reduces the glycolytic flux. 3PO abrogated p38 MAPK and Erk phosphorylation, which resulted in inhibition of the NF-κB pathway, therefore suppressing COX2 and IL-1β levels. These alterations occurred exclusively in the nervous tissue, since its administration was intracerebroventricularly, and there was no reversal effect on serum IL-1β levels. Moreover, the inhibitor affected microglia and astrocyte responses and reduced the classical biomarkers, Iba1 and GFAP, respectively. These findings reinforce the importance of energy metabolism reprogramming to aerobic glycolysis in the neuroimmune response and highlights a potential strategy to decrease neuroinflammation signaling, since 3PO decreases glycolytic parameters (such as glucose uptake, PFK1 and PK activities) and consequently reduces the energy available for the synthesis and release of inflammatory mediators. Moreover, 3PO improved the behavior of animals, reducing the lethargy induced by LPS in the open field task. Some limitations of this work should be mentioned. Firstly, our evaluation of sickness behavior was limited to body temperature control, in which there was no improvement with both inhibitors, and the open field test, in which 3PO reversed the effects of LPS on exploratory behavior more efficiently than the NRLP3-inhibition strategy. In fact, different behavioral tasks, such as the tail suspension test and forced swim test could expand on the observations found (Mazuco et al., 2019; Chen et al., 2013; Tastan et al., 2021). Secondly, we used only male rats, despite knowledge regarding the sex dependence of the neuroinflammatory response (e.g.(Vázquez-Martínez et al., 2018)). Thirdly, our goal was to observe acute changes, but examining other timepoints in this preclinical model will allow valuable observations for defining a post-translational approach and, for this, we will need to use a less invasive route of administration, such as the intranasal route. Despite these limitations, the model used and the results obtained were highly satisfactory. However, future studies should evaluate the effects of treatments with 3PO and MCC950 on the cognitive impairment and anxious-type behavior that is induced by neuroinflammation. Additionally they should investigate gender response and determine whether acute intervention in the neuroinflammatory process triggers a long-term neuroprotective response in animals. Conclusions In conclusion, taken together, our findings show the induction neuroinflammation by LPS, where negative glycolysis modulation presents an interesting target for reducing neuroinflammation. The 3PO inhibitor reversed inflammation and glial parameters, as well as the glycolysis shift induced by LPS in the hippocampus, and improved LPS-induced sickness behavior. The NLRP3 inhibition strategy also downregulated neuroinflammation, oxidative stress and glucose uptake. Our data provide further evidence of the role of the Warburg effect in the early neuroinflammation response, and suggest new strategic pharmacological approaches to mitigate neuroinflammation at a critical moment that can trigger future neurodegenerative diseases. CRediT authorship contribution statement Adriana Fernanda K. Vizuete: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Fernanda Fróes: Methodology, Formal analysis. Marina Seady: Methodology, Formal analysis. Aline Castro Caurio: Methodology, Formal analysis. Osmar Vieira Ramires Junior: Methodology, Formal analysis. Ana Karla Oliveira Leite: Methodology, Formal analysis. Clarissa Penha Farias: Methodology, Formal analysis. Angela T.S. Wyse: Resources, Methodology. Carlos-Alberto Gonçalves: Writing – review & editing, Supervision, Resources, Project administration. Disclosures All authors declare that there is no duality of interest associated with their contribution to this manuscript. All authors read and approved the final manuscript. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments The authors gratefully acknowledge the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul (FAPERGS), National Institute of Science and Technology for Excitotoxicity and Neuroprotection (INCTEN/CNPq), INNT(CNPq: E-465346/2014-6), Fundação de Amparo à Pesquisa do Rio de Janeiro (FAPERJ: E-26/010/000080/2018). Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.bbih.2024.100901. Supplementary data The following are the Supplementary data to this article: Multimedia component 1 Experimental design. Animals were submitted to stereotactic surgery (PN25) for implantation of cannulas in the ventricular region. After recovery from surgery, the animals were subjected to the in vivo LPS neuroinflammation model, whereby rats were treated with LPS (1 mg/kg, i.p.) or saline (0.9% NaCl, i.p.). After 3 h, bilateral infusions (2 μl, icv) of vehicle (DMSO 0.01%/Kg), the 3PO inhibitor (20 μM/kg) or the MCC950 inhibitor (10 μM/kg) were administered. Animals were divided into the following groups: 1) Sham + vehicle (saline i.p., vehicle icv); 2) sham + inhibitor (saline i.p., 3PO/ MCC950 icv); 3) LPS + vehicle (LPS i.p., vehicle icv); and 4) LPS + inhibitor (LPS i.p., 3PO/ MCC950 icv). Representative membrane analysis by Western blotting from Fig. 1. Representative membrane analysis by Western blotting from Fig. 3A. Representative membrane analysis by Western blotting from Fig. 3B. Representative membrane analysis by Western blotting from Fig. 3C. Representative membrane analysis by Western blotting from Fig. 5. Representative membrane analysis by Western blotting from Fig. 8. Data availability Data will be made available on request. References Associated Data Supplementary Materials Multimedia component 1 Data Availability Statement Data will be made available on request.