Inhibition of Voltage-Gated Hv1 Alleviates LPS-Induced Neuroinammation via Regulation of Microglial Metabolic Reprogramming

Neuroinammation plays an important role in the onset and advancement of cognitive loss and neurodegenerative disorders. The voltage-gated H channel (Hv1) has been reported to be involved in microglial activation and act as key drivers of neuroinammation. This study aims at evaluating the mechanism of Hv1 involvement in neuroinammation and the therapeutic potential of Hv1 inhibitor, 2-guanidinobenzimidazole (2-GBI), in a model of lipopolysaccharide (LPS)-induced neuroinammation. We investigated the inuence of Hv1 inhibitor (2-GBI) on the generation of reactive oxidative species (ROS), metabolic reprogramming, and inammatory mediators in vitro and examined the therapeutic potential of 2-GBI on microglial activation and hippocampal neuroinammation in vivo. Novel object recognition and Y-maze were employed to assess cognitive function.


Results
2-GBI reduced the LPS-induced proin ammatory response and aerobic glycolysis in microglia. HIF1α overexpression mediated aerobic glycolysis reprogramming alleviated by 2-GBI. We reported that Hv1 inhibitor exerted a protective effect on LPS-induced neuroin ammation through the ROS/HIF1α and PI3K/AKT/HIF1α pathways -mediated aerobic glycolysis. The cell death of PC12 induced by microgliamediated neuroin ammation was reversed in a transwell co-culture system by 2-GBI. Furthermore, in vivo results suggested that 2-GBI mitigated the neuroin ammatory processes and recognition injury through regulation of microglial metabolic reprogramming. Conclusion 2-GBI protects LPS-induced neuroin ammation, neuronal cell death, and subsequently reverses the hippocampus-dependent cognitive de cits through regulation of microglial metabolic reprogramming.
Taken together, these results demonstrate a key role for Hv1 in driving a pro-in ammatory microglia phenotype in neuroin ammation.

Background
Neuroin ammation is considered to be a key element to almost all neurodegenerative disorders [1].
Myeloid-derived microglia and macrophages reside within the brain, which play an indispensable role in immune responses and homeostasis maintenance in the central nervous system, as well as mediates during a neuroin ammatory process [2]. In response to immunological challenges, microglia readily becomes activated as characterized by marked morphological dynamics, and modulation of neuronal activity that impacts in ammation-mediated neuronal degeneration [3,4]. Activated microglia can secrete various cytotoxic factors and pro-in ammatory cytokines such as interleukin-1 beta (IL-1β), interleukin-6 (IL-6), and tumor necrosis factor α (TNFα) [5]. Nevertheless, exacerbated microglial activation may additionally stimulate the in ammatory responses of glial cells and lead to potentially long-lasting detrimental effects on the brain [6]. Therefore, attenuating microglia-mediated in ammatory response and neuronal injury has become an important target for improving the process of neurodegeneration.
Microglia usually change their phenotype and metabolic state in response to immune challenge [7,8].
During the classic in ammatory activation process, microglia cellular metabolism is reprogrammed from oxidative phosphorylation (OXPHOS) toward glycolysis even in the presence of oxygen, a phenomenon also known as the Warburg effect Whereas a high basal mitochondrial oxygen consumption rate (OCR) is possessed by alternatively activated macrophages [9,10]. Several reports have demonstrated that pharmacologic inhibition of glycolysis could blunt the M1 polarization in macrophages [11,12].
The voltage-gated proton channel Hv1 (also termed VSOP), encoded by Hvcn1, is mainly expressed in immune cells such as neutrophils, macrophages, B lymphocytes, and microglia [13]. Hv1 plays an essential role in alleviating the coincident intracellular acidosis via NADPH (NOX2)-dependent extrusion of hydrogen (H + ) protons [14]. Hv1 is electively expressed in microglial cells in the central nervous system [15]. The Microglial Hv1 proton canal has been suggested as an essential target that incorporates in ammation into the injured microenvironment during ischemic stroke, traumatic brain injury and spinal cord injury [16][17][18]. Hv1-de cient mice exhibited reduced in ammation and long-term neuroprotection in preclinical experimental models [19]. However, the molecular mechanisms responsible for Hv1-mediated microglial activation and neuroin ammation have not been fully elucidated. In this study, we demonstrated a novel evidence that Hv1 acts as the upstream regulator to promote metabolic reprogramming and subsequent neuroin ammation through HIF-1α. Inhibition of Hv1 by a speci c Hv1 inhibitor, 2-guanidinobenzimidazole (2-GBI), not only suppressed lipopolysaccharide (LPS)-induced microglial activation and neuroin ammation in vitro and in vivo, but also reduced cell apoptosis, and alleviated memory impairment in LPS-injected mice.

BV2 Cell culture
Murine BV2 microglia and rat PC12 cells were procured from the ATCC and were subjected to culturing in Dulbecco's modi ed Eagle's medium (DMEM; Thermo Fisher Scienti c) supplemented with fetal bovine serum (FBS, 10%). Culturing of cells was carried out in a 5% CO 2 incubator at 37°C. Before the experiment, seeding of the cells was carried out into 12-or 96-well plates and incubated for 12 hrs. BV2 cells were pre-treated with 200 µM 2-GBI, a selective and state-dependent blocker of Hv1 channels for 1 hr, and then treated with LPS 50 ng/mL for 4 and 6 hrs. At various times following LPS exposure, cells were harvested and used in a metabolic assay or immunoblot analysis of neuroin ammation and glycolysis. For CoCl 2 treatment, BV2 cells were pre-treated with 200 mM CoCl 2 for 24 hrs, then treated with 200 µM 2-GBI for 9 hrs.
Preparation of primary murine microglia C57BL/6 mice of 24 hrs of age were used to establish primary microglial cells according to the previously described method [20]. Microglial cells were suspended and obtained on day 11-13, by shaking the asks for 10 min at 180 rpm and 37°C. The mature microglial cells were seeded within plates at a density of 2 × 10 5 /cm 2 and before 2-GBI (200 µM) were used for pre-treating the primary microglia 1 hr, followed by treatment with 50 ng/mL LPS in DMEM supplemented with 10% FBS for 4 hrs and 6 hrs. Following LPS exposure, cells were harvested at various times and used in metabolic assay or immunoblot analysis of neuroin ammation and glycolysis.

NO assay
The level of accumulated nitrite (NO 2 − ), a metabolite of NO is measured using the Griess assay, using the Griess reagent in the culture supernatant. BV2 cells and primary microglia were incubated overnight after being seeded in a 96-well plate at a density of 1 × 10 5 cells in each well. The cell culture supernatant was gathered followed by measurement of the concentration of NO using the Griess reagent.

Western Blotting
Western blot was conducted as elaborated in detail earlier [21]. 4-20% Sure PAGE Bis-Tris gels (GenScript, Nanjing, China) were employed to run the protein sample ( were used to combine primary antibodies and the reaction was detected with a BeyoECL Plus ECL Kit (Beyotime, China). The immunoblots were then scanned and quanti ed by the ImageJ software. The band intensity values of the target proteins were normalised to that of β-actin.

Metabolic assays
Within the culture medium, the lactate levels were ascertained by employing a Lactate Assay Kit (BioAssay Systems) following the manufacturer's instructions. For analyzing the extracellular acidi cation rates (ECAR) (103020-100, Seahorse Biosciences/Agilent Technologies, Billerica, MA, USA) and OCR (103015-100, Seahorse Biosciences/Agilent Technologies, Billerica, MA, USA), an XF96 extracellular ux analyzer (Seahorse Biosciences/Agilent Technologies, Billerica, MA, USA) was used to analyze primary microglia and BV2 cells. Individual wells containing 12,000 cells were then cultured for 12 hrs at 37°C in a 5% carbon dioxide condition in an incubator. Following LPS exposure, cells were used for metabolic assays.
Reactive oxygen species (ROS) measurement A 12-well plate was used to culture cells. 300µL of FBS-free DMEM medium supplemented with DCFH-DA was added. The 12-well plate was allowed to sit in an incubator within a 5% carbon dioxide atmosphere at 37°C for 20 minutes. After irradiation, ow cytometry (BD Accuri™ C6 Plus, BD Biosciences) was used to analyze the cells and the mean uorescence was estimated using FlowJo software (FlowJo, LLC, Ashland OR, USA).
Co-cultures of PC12 neurons and BV2 microglia As previously described [22], the transwell co-culture system was conducted by a semi-permeable 0.4-µm membrane (Costar, NY, USA). First, Cells were pre-treated with 200 µM 2-GBI for 1 hr and subsequently incubated with LPS (50 ng/mL) for the 6 h. After treatment with LPS and/or 2-GBI, BV2 microglia were seeded in the upper chamber at a density of 1×10 5 cells in 1 mL serum-free medium, and PC12 cells were plated in the lower chamber at a density of 2×10 5 cells in 2 mL serum-free medium. After co-cultivation for 24 hrs, PC12 cells were harvested for further analysis.
Cell viability assay PC12 cell viability in response to the supernatants of BV2 cells was determined by CCK-8 assay according to the manufacturer's instructions (Dojindo, Tokyo, Japan). Brie y, PC12 cells were cultivated at a density of 10,000 cells/well in 96-well plates. 10 µL of CCK-8 solution was added to each well of the plate, and the plates were incubated for 0.5 h. Assay plates were shaken on an orbital shaker for 2 min.
Data acquisition was then performed using a microplate reader.

Detection of PC12 apoptosis by TUNEL assay
After co-cultivation for 24  Four groups of the eight-week-old male mice were made (12-14 mice in each group): namely LPS (5 mg/kg, intraperitoneally), saline-treated, 2-GBI (1 mg/kg, intraperitoneally), and LPS + 2-GBI. Following 6 h of injecting LPS and 2-GBI, brain tissues were gathered and stored until further investigation at freezing temperatures (− 80°C). Following 24 h of injecting LPS and 2-GBI, behaviors analysis of the mice was carried out (described in detail below), following which they were sacri ced.

Immuno uorescence
As elaborated earlier [23], immuno uorescence analysis was carried out. Dissection of the brains was done, and the tissue processing was carried out as explained earlier [24]. Overnight incubation of the sections with the mentioned primary rabbit anti-PFKFB3 antibody (1:200, ab181861, Abcam) and Iba1 (

Y-maze
The Y-maze (Sansbio, JiangSu province) was a modi ed version of a previously described apparatus [25]. Mice were initially placed in one arm, and the sequence and number of arm entries were recorded with ANY-maze (Stoelting CO., Wood Dale, IL, USA). Each mouse was placed at the end of one arm after 24 hr of injecting LPS and 2-GBI and allowed to explore the apparatus freely for 8 minutes. The results of the test performed were recorded and the following parameters like spontaneous alteration performance (SAP), alternate arm return (AAR) and same arm return (SAR) were determined. SAP is the main criteria that tests spatial learning and memory, which was de ned as actual alternation/possible alternation. AAR and SAR in the Y-maze (AAR = alternate arm returns/total arm entries × 100, SAR = same arm return/total arm entries × 100) were used as indicators of memory impairment.

Data processing
Statistical multiple group comparisons were performed by two-way analysis of variance (ANOVA) followed by Tukey's multiple comparisons test in the GraphPad Prism software, v8.0 (CA, USA) and results were reported as mean ± standard error (SEM). A value of P < 0.05 was used to indicate signi cant difference.

2-GBI alleviates the LPS-induced in ammatory response in microglia cells
We evaluated the impact of 2-GBI, a known Hv1 inhibitor in modulation of neuroin ammatory response. For this purpose, primary microglia and BV2 cells were treated with 200 µM 2-GBI and LPS (50 ng/mL). We rst analyzed LPS-induced generation of in ammatory modulators. The results presented that 2-GBI inhibited LPS-induced NO production ( Fig. 1A and B), protein expressions of iNOS, COX2, TNFα, and IL-1β in primary microglia and BV2 cells (Fig. 1C-E).

2-GBI attenuates the LPS-induced aerobic glycolysis in microglial cells
Glycolytic reprogramming plays an important role in polarization of in ammatory cells. In 2-GBI-treated cells, the LPS-induced glycolysis-related proteins of HK2 and PFKFB3 were decreased ( Fig. 2A-B). Therefore, we examined whether Hv1-mediated aerobic glycolysis and enhanced in ammatory responses were dependent on HIF1α expression. As shown in Fig. 2A-B, protein expression of HIF1α were elevated in response to LPS stimulation and mitigated in response to 2-GBI. Next, 2-GBI-treated BV2 cells and primary microglia unveiled the lactate levels for LPS-activated status undergoing a marked decline (Fig. 2C). In addition, as an indirect indicator of lactate generation and improved glycolytic metabolism ECAR was seen to be elevated in LPS-treated primary microglia and BV2 cells, and 2-GBI signi cantly inhibited LPSinduced glycolytic metabolism in microglial cells following 4 hrs of LPS treatment (Fig. 2D). As shown in Fig. 2D, addition of 2-GBI was found to have pronounced inhibitory effect on the basal OCR, ATP-linked OCR, maximal respiration and SRC of microglial cells and no increasing effect on ECAR. ATP-linked OCR and SRC were noticeably increased by LPS (50 ng/mL) exposure, while 2-GBI treatment effectively rescued this induction. The OCR was signi cantly lower in 2-GBI/LPS group than the LPS group in microglial cells following 4 hrs of LPS treatment. It is clearly evident from the data that mitochondrial respiration and aerobic glycolysis in microglial cells was inhibited by Hv1.
The anti-in ammatory effect of Hv1 depends on HIF1α To further explore linkage between decreased expression of HIF1a and aerobic glycolysis due to Hv1 inhibitor, we assessed PFKFB3 and HK2 levels in BV2 cells induced HIF1α by CoCl 2 treatment (Fig. 3). As shown in Fig. 3A-D, the protein level of HIF1α, HK2 and PFKFB3 in microglia were signi cantly increased after CoCl 2 treatment, while this upregulation effects of CoCl 2 were relieved by 2-GBI ( Fig. 3A-D).
Effects of 2-GBI on PI3K/AKT/mTOR and ERK1/2 activation in microglia The PI3K/AKT and ERK1/2 pathways are involved in the regulation of HIF1α. We investigated whether 2-GBI could affect PI3K/AKT and ERK1/2 phosphorylation in microglia. As shown in Fig. 4A-C, 2-GBI signi cantly decreased the phosphorylation of PI3K/AKT/mTOR and ERK1/2 expression in BV2 cells and primary microglia after 4 hours of LPS exposure. These results suggest that the PI3K/AKT and ERK1/2 pathway have a role in LPS-induced microglial activation, and that 2-GBI reduces LPS-induced in ammatory responses by PI3K/AKT/HIF1α and ERK1/2/HIF1αsignaling. An increasing body of evidence shows that activated microglial cells generated intracellular ROS and play a signi cant role in HIF1α-mediated enhancement of the in ammatory response [26]. Hence, the impact of Hv1 on the production of intracellular ROS was determined using ROS-sensitive indicators DCFH-DA. 2-GBI-treated BV2 cells manifested a decrease in intracellular ROS for LPS-activated status (Fig. 4D).

2-GBI Regulates the Expression of In ammatory Mediators in Neurons/Glial Cells Co-cultures Submitted to LPS Stimulus
To study the neuroprotective effects of 2-GBI, we constructed a noncontact co-culture system of BV2 and PC12 cells by the transwell system to mimic the growth environment of microglia and dopaminergic neurons. First, we measured the viability of PC12 cells at 24 hr after co-culture with LPS-activated BV2 using CCK-8. The results showed that the viability of PC12 cells after co-culture with LPS-activated BV2 decreased signi cantly at 24 hours (Fig. 5A). Next, we observed that 2-GBI signi cantly reversed PC12 cell death (Fig. 5A). Furthermore, western blot results showed that LPS-induced BV2 upregulated cleavedcaspase-3 levels and downregulated synaptophysin levels in PC12 cells, resulting in a signi cant in ammation-mediated neurotoxicity in PC12 cells. In contrast, 2-GBI signi cantly protect against in ammation by improving the levels of cleaved-caspase-3 and synaptophysin in PC12 cells (Fig. 5B-F).

2-GBI suppresses LPS-induced brain in ammation in mouse
To study the effect of Hv1 inhibitor on LPS-induced memory impairment and neurotoxicity in vivo, adult C57BL/6J mice were treated with 5 mg/kg LPS intraperitoneally with or without adminsstration of 2-GBI (1 mg/kg, intraperitoneally). The levels of TNFα, IL-1β and iNOS underwent a marked increase in the mouse hippocampus 6hrs following the LPS challenge. Nevertheless, the elevation of these cytokines was averted by 2-GBI ( Fig. 6A and B). Furthermore, increased PFKFB3 positive microglia were detected in the CA1 and DG regions of the LPS-treated mice, and 2-GBI alleviated LPS-induced glycolysis in the hippocampus (Fig. 6C-E).
We carried out a Y-maze behavioral test to assess spatial working memory affected by the Hv1 inhibitor. Alternation of arm entries in a Y-maze is driven by an instinct to visit a novel place and requires the animal to remember which arms it entered in its immediately previous exploration. During the 8 min test session in the Y-maze, LPS treatment signi cantly reduced the probability of SAP to 52% compared to 62% in vehicle-treated control mice (Fig. 7A). 2-GBI prevented the decrease of SAP caused by LPS administration (Fig. 7A). Correspondingly, LPS signi cantly increased the probability of SAR, while 2-GBI reversed the cognition de cits (Fig. 7B). AAR did not produce a difference among groups in the Y-maze test (Fig. S1). These results suggest that 2-GBI reverses the spatial working memory impairment caused by LPS-induced neuroin ammation.

Discussion
The current work reveals evidence that microglial Hv1 acts as an immunometabolic regulator that controls in ammatory cytokine production. In both primary and BV2 microglia, Hv1 inhibitor impairs LPSinduced pro-in ammatory mediator synthesis and aerobic glycolysis. Mechanistic studies identi ed that PI3K/AKT/HIF1α and ROS/HIF1α signaling pathways were involved in Hv1 mediated aerobic glycolysis.
We further demonstrated that Hv1 inhibitor was able to suppress the in ammatory responses and the altered metabolic processes caused by LPS insult in both vitro vivo. Systematic administration of Hv1 inhibitor also improved LPS-induced de cits in recognition memory in mice.
Hv1, having a functional expression within microglia, aids in NOX-dependent generation of ROS and controls intracellular pH [14]. Recent studies have revealed that a de ciency of Hv1 was able to weaken the disruption of white matter integrity induced by bilateral common carotid artery stenosis [27], alleviate neuronal apoptosis and neuronal pyroptosis following SCI [28], and reduce LPC-mediated myelin damage [29]. Our group has previously found that Hv1 upregulation in the aged brain exaggerates postoperative neuroin ammatory responses after peripheral tibia fracture surgery [15]. In the current work, we further con rmed previous nding that Hv1 plays a critical role in microglia-mediated neuroin ammation and emerges as a potential therapeutic target for neuroprotection.
Evidence suggests an essential involvement of metabolic reprogramming in regulating the inherent in ammatory response [30]. In response to immune challenge, cells choose to use glycolysis instead of mitochondrial catabolic pathways for the conservation and generation of metabolic resources. As a consequence of this switch, there is an increase in lactate production and glucose uptake with activation of the pentose phosphate pathway (PPP) and a simultaneous decrease in oxygen consumption by mitochondria [31]. Succinate, the Krebs cycle intermediate, regulates HIF1α in M1 macrophages to drive a sustained production of IL-1β [32]. The addition of 2-DG, a glycolysis inhibitor, blocked the immunometabolic reprogramming of microglia resulting from an LPS-induced rise in aerobic glycolysis [12]. A high or moderate dose of LPS-induced cells became more dependent on glycolysis than mitochondrial respiration [33]. The molecular mechanisms behind this response is not known, but it has been proposed that the energy depletion elicits mitochondrial damage. Without LPS treatment, impaired Hv1 activity by 2-GBI results in signi cantly reduced OCR. The current study showed that Hv1 was found to have pronounced down-regulatory effect on the OCR of microglia and no increasing effect on ECAR. Nevertheless, LPS pushed aerobic glycolysis and generation of the proin ammatory cytokine in microglial cells, and 2-GBI effectively attenuated LPS-induced glycolysis in microglial cells, thereby implying that Hv1 play an important role in control of microglial activation via metabolic reprogramming.
The switch to glycolysis is promoted by the transcription factor HIF1α so that these cells can continue to generate ATP. HIF1α aids in this metabolic switch by bonding with the hypoxia response elements within target genes, such as glycolytic enzymes and the glucose transporter GLUT1 [34]. HIF1α also increments the expression of glycolysis-related genes PFKFB3, a major driver of glycolysis by its ability to synthesize fructose-2,6-bisphosphate [35]. When HIF1α is absent, the cellular ATP levels are greatly reduced [36]. In the current study, we detected an elevation of HIF1α expression in LPS-stimulated microglial cells. Inhibition of HIF1α expression in microglia attenuated LPS-induced aerobic glycolysis as well as the in ammatory response, thereby implying that the regulating microglial metabolic reprogramming and subsequent in ammatory responses by Hv1 were dependent upon the expression of HIF1α.
In microglia, ROS are generated primarily by NOX2 [37]. It has been suggested that Hv1 channels are an essential regulator of ROS production in microglia by counteracting the charge imbalance caused by the activation of NADPH oxidase [38]. In the current study, we found that the Hv1 inhibition suppressed ROS expression and abolished enhancement of the in ammatory response in LPS stimulated microglial cells.
There is evidence that ROS act as second messengers to propagate microglial immune activation by in uencing multiple key signaling pathways, including PI3K/AKT/mTOR and MAPKs [38]. Phosphorylated mTOR and ERK1/2 were suggested to increase the expression of HIF1a [39][40]. We found that 2-GBI inhibited PI3K/AKT/mTOR and ERK1/2 phosphorylation. Thus, these data indicated that Hv1 regulates HIF1α expression and metabolic reprogramming through the PI3K/AKT/HIF1α and ROS/HIF1α mediated pathway.
This work investigates the effect of a prototypical Hv1 inhibitor (2-GBI) on LPS induced neuroin ammation both in vitro and in vivo. 2-GBI inhibits Hv1 proton conduction by binding to the VSD from its intracellular side [41]. We found that 2-GBI effectively ameliorated neuroin ammation through inhibition of HIFα-mediated aerobic glycolysis in cultured microglia. Systematic administration of 2-GBI facilitated the recovery of recognition memory in mice challenged with LPS. The 2-GBI associated cognition recovery from LPS stimulation was paralleled by reduced levels of IL-1β, TNFα and iNOS in the hippocampus. Our nding suggested that Hv1 inhibitors can be considered as potential pharmacological treatments for diseases caused by Hv1 hyperactivity in future.

Conclusions
In summary, the present study provides the evidence showing that Hv1 modulates microglial activation via regulation of metabolic reprogramming. HIF1α is essential for Hv1-mediated in ammation, and activation of the PI3K/AKT/HIF1α as well as ROS/HIF1α pathway are involved in Hv1-mediated glycolytic reprogramming. The observations that pharmacological targeting of microglial Hv1 channels can affect hippocampal related cognitive function and neuroin ammation in mice are particularly interesting and interfering with this pathway may provide a promising therapeutic option for neuroin ammatory diseases.  exposure. (E) Band intensity was quanti ed by ImageJ software, and the values of target protein were normalised to that of β-actin. All the data are expressed as means ± SEM and were analyzed by two-way ANOVA with the Tukey's post hoc test. N = 6 independent measurements. *P < 0.05; **P < 0.01 for comparisons shown. LPS-induced increased ECAR dependent glycolysis, glycolytic capacity and glycolytic reserve in cultured primary microglia and BV2 cells. OCR and ECAR measured are expressed in bar graph format as the mean ± SD, n = 6. *, P < 0.05 compared to Ctrl. # , P < 0.05 compared to LPS group.  (C) Band intensity in BV2 cells was quanti ed by ImageJ software, and the values of target protein were normalised to that of β-actin. All the data are expressed as means ± SEM and were analyzed by two-way ANOVA with the Tukey's post hoc test. N = 6 independent measurements. *P < 0.05; **P < 0.01 for comparisons shown. (D) The BV2 cells with or without 2-GBI were incubated with LPS (50 ng/mL) for 4 hr and intracellular ROS levels were measured by DCFH-DA. Fluorescence assay were detected in BV2. Data are presented as mean ± SEM for at least 6 independent experiments..