High‐concentration hydrogen inhalation mitigates sepsis‐associated encephalopathy in mice by improving mitochondrial dynamics

Abstract Background Sepsis‐associated encephalopathy (SAE) is a neuronal injury with poor prognosis. Mitochondrial dysfunction is critical in SAE development, and hydrogen gas (H2) has a protective effect on septic mice. This study aimed to investigate the effect of high concentration (67%) of H2 on SAE and whether it is related to mitochondrial biogenesis and mitochondrial dynamics. Methods A mouse sepsis model was induced by cecal ligation and puncture. The mice inhalated 67% H2 for 1 h at 1 and 6 h post‐surgery, respectively. The 7‐day survival rate was recorded. Cognitive function was assessed using the Y‐maze test and Morris water maze test. Serum inflammatory factors, antioxidant enzymes, as well as mitochondrial function indexes including mitochondrial membrane potential (MMP) and ATP in the hippocampal tissue were evaluated 24 h after surgery. Mitochondrial dynamic proteins (DRP1 and MFN2) and biosynthetic proteins (PGC‐1α, NRF2, and TFAM) in the hippocampal tissue were detected. Moreover, the morphology of mitochondria was observed by transmission electron microscopy. Results Inhalation of 67% H2 improved the 7‐day survival rates and recognition memory function of septic mice, alleviated brain antioxidant enzyme activity (SOD and CAT), and reduced serum proinflammatory cytokine levels. H2 inhalation also enhanced the expression of MFN2 and mitochondrial biogenesis‐related factors (PGC‐1α, NRF2, and TFAM) and decreased the expression of fission protein (DRP1), leading to improvement in mitochondrial function, as evidenced by MMP and ATP levels. Conclusions Inhalation of high concentration (67%) of H2 in septic mice improved the survival rate and reduced neuronal injury. Its mechanism might be mediated by enhancing mitochondrial biogenesis and mitochondrial dynamics.


| INTRODUC TI ON
Sepsis is a severe condition marked by inadequate host response to infection, causing multi-organ dysfunction. 1 The latest Sepsis 3.0 guidelines underscore the importance of studying organ dysfunction in sepsis.Brain damage is a prevalent organ dysfunction associated with sepsis, which leads to neurological complications such as intensive care unit-acquired paralysis 2 and cognitive impairment, 3,4 which impairs functional status and quality of life.
Mitochondria are dynamic and multifunctional organelles that are crucial for eukaryotic cells as they facilitate biological oxidation and energy conversion through oxidative phosphorylation. 5This process generates ATP, which is essential for cellular metabolism, physiology, and organismal functions.Moreover, mitochondria regulate intracellular calcium levels, generate reactive oxygen species (ROS), and modulate cellular signaling and redox processes.Changes in the structure of mitochondria affect energy production, so the dynamics such as fusion, division, motility, and positional binding need to be controlled. 6spite mitochondrial protein depletion, sepsis survivors with multiple organ dysfunction exhibit heightened mitochondrial biosynthesis and antioxidant defense responses, preserving functional and energetic status. 7However, continuous mitochondrial damage and ensuing inflammation can cause severe outcomes such as cell death, multiple organ failure, and long-term cognitive dysfunction, [8][9][10] which are associated with sepsis prognosis and encephalopathy. 11Hydrogen gas (H 2 ), known for its properties such as scavenging ROS and modulating immune responses, has been studied as a therapeutic molecule. 12Our research team has demonstrated the efficacy of low concentration (2%) of H 2 in treating sepsis and multiorgan failure using animal and cellular models. 13Recently, high concentration of H 2 therapy devices has shown promising results in treating various diseases. 6,14However, the therapeutic effect of high-concentration of H 2 on SAE is still unclear.
Therefore, this study aimed to evaluate the effectiveness of inhaled high-concentration of H 2 in mitigating SAE and whether the mechanism was related to mitochondrial dynamics and biogenesis.

| Animals
Male C57BL/6J mice, weighing 20-25 g and aged 6-8 weeks, were acquired from the Experimental Animal Centre of the Chinese Academy of Military Medical Sciences, Beijing, China.After an acclimation period, the mice were housed in a pathogen-free environment with ad libitum access to food and water (temperature of 20-22°C, humidity of 30%-70%, and 12-h cycle of light and darkness).Ethical regulations were followed throughout all experimental procedures, which received the approval of the Experimental Animal Management Committee of Tianjin Medical University (Grant No. IRB2021-DWFL-274).

| Cecal ligation and puncture (CLP)
Sepsis model was established by cecal ligation and puncture.After anesthesia with 2% isoflurane, a 1 cm incision was made in the middle of the abdomen.The cecum was ligated with surgical sutures at the end of one-half of the distance below the ileocecal flap, and a sterile 20 G needle was employed to perforate the midpoint of the ligated cecum and extrude a small amount of bowel contents.After returning both the cecum and extruded contents to the abdominal cavity, the abdominal wall incision was closed using sterile 6-0 silk in layers.Subsequently, a post-surgery subcutaneous injection of 5 mL/100 g saline was administered.

| Experimental procedures
The mice were randomly assigned into four groups (n = 62 per group): Sham group (Sham), Sham with 67% hydrogen group (Sham+H 2 ), CLP group (CLP), and CLP with 67% hydrogen group (CLP + H 2 ).The Sham group and Sham + H 2 group only laparotomy without ligation puncture cecum.The sepsis model was induced using the CLP procedure in the CLP group and the CLP + H 2 group.The Sham+H 2 group and CLP + H 2 group inhaled high concentration of H 2 for 1 h at 1 and 6 h after the operation while the remaining groups only inhaled air.
The survival rates of the mice (n = 20 per group) were monitored for 7 days post-operation.After 24 h of molding, six mice from each group were randomly selected for subsequent hippocampal damage assessment and their serum was used for the detection of serum inflammatory factors.In addition, 30 mice in each group were selected to take their hippocampus for detection of antioxidant enzyme activity (n = 6), mitochondrial membrane potential (MMP) (n = 6), ATP content (n = 6), Mitochondrial morphology (n = 6) and Western blotting (n = 6).In addition, the Y-maze test (n = 6) and the Morris water maze test (n = 6) were performed on the 7th and 8th postoperative days, respectively (Figure 1A).

| Hydrogen gas treatment
The mice in the Sham+H 2 group and CLP + H 2 group were exposed to 67% hydrogen gas and 33% oxygen mixture using an AMS-H-01 hydrogen gas and oxygen atomizer (Shanghai Asclepius Meditec Co., Shanghai, China) and placed in the sealed box.A layer of calcium Prior to the experiment, the box was charged with the hydrogenoxygen mixture for 30 min.Continuous monitoring of H 2 concentration was performed using a heat tracing gas analyzer (Thermal Fisher, Waltham, MA, USA).In the Sham+H 2 and CLP + H 2 groups, the hydrogen inhalation was exposed to the appropriate concentration for 1 h at 1 and 6 h post-operation, respectively.The Sham group and CLP group received only air.

| Y Maze test
Y-maze recognition memory experiments were conducted at 7th day after surgery to evaluate cognitive function.The Y-maze, provided by Shanghai Xinsoft Information Technology Co., consisted of three arms (A, B, and C) positioned at a 120° angle to each other.During the experiments, mice underwent a 10-min acclimation period, after which arm C was obstructed with a baffle.Mice were then free to explore arms A and B while various parameters such as distance, time, speed, entry count, latency, and resting time were tracked using a video recording system.The maze was disinfected using 75% ethanol to eliminate residual odors before removing the baffle from arm C after 1 h of training.Mice were subsequently placed at arm A and allowed to move freely for an additional five minutes, monitored by the video acquisition system.Incidences of mouse entry into each zone served as indicators of mobility.

| Morris water maze (MWM) test
The MWM experiment was carried out in a round plastic tank (90 cm in diameter and 50 cm in height) containing titanium dioxide at a water temperature of 23°C.Titanium dioxide made water opaque and was used to track mice.The pool is divided into four quadrants, one of which places a circular platform 2 cm below the surface of the water.In the first 6 days of training, the mouse was randomly placed in the quadrant of the pool to allow the mice to freely find the platform.When the mouse could not find the platform within 60 s, it was led to the platform to rest for 10 s.On the seventh day, the platform was removed from the pool and the mice were placed in the water to swim freely for 60 s.The number of platform crossings and the residence time in the quadrant where the platform was located were recorded.

| Brain histopathological detection
After 24 hours of molding, six mice from each group were randomly selected for subsequent hippocampal damage assessment.The brain tissue was fixed in 4% paraformaldehyde at 4°C overnight.One part of the brain tissue was used for Nissl staining and the other part was used for HE staining.Following transcardial perfusion with 4% paraformaldehyde, the tissues were fixed in 10% formalin for 24 h and embedded in paraffin.After the 10 μm slices were dewaxed, rehydrated with xylene and ethanol, they were stained with Nissl and HE, respectively.The neuronal organization in the CA1 hippocampal region and niche morphology were scrutinized under a light microscope (Olympus, Japan).Three sections were randomly selected from each sample, and three fields of view were observed per section by two pathologists who did not know the experiment.
The normal spinal neuron count was quantified and averaged using Image pro-plus 6.0 image analysis software.

| Immunofluorescence
Mouse brain was placed in 4% paraformaldehyde at 4°C overnight.
After dehydration, OCT reagent (4593, Solarbio, China) was used for embedding.Then, the embedding block was cut into 10 μm slices.The primary antibody Drp1 (1:300, sc-271583, SantaCruz, USA) and Nrf2 (1:300, 80593-1-RR, Proteintech, China) were added to the slice separately at 4°C overnight.On the next day, fluorescent secondary antibody was added to the sections.Then the sections were observed under the fluorescence microscope (Olympus, Japan).

| Serum pro-inflammatory factors
At the 24-h post-surgery, approximately 0.6 mL blood samples were collected via the orbital vein from a random selection of six mice in each group.The samples were centrifugated at 4000 g for 10 minutes to obtain serum.ELISA analysis was used to measure serum concentrations of TNFα (MTA00B, R&D Systems, USA), IL-1β (MLB00C, R&D Systems, USA), and HMGB1 (SP14752, Saipei, China).

| Detection of antioxidant enzyme activity
At the 24-hour after operation, six mice from each group were randomly selected, and their hippocampal tissue was extracted.
The activity of Superoxide Dismutase (SOD) and Catalase (CAT) was evaluated using a 722 visible spectrophotometer (Shanghai Analytical Instrument Factory) with Cayman kits obtained from the USA.

| JC-1 assay of the mitochondrial membrane potential (MMP)
At the 24-h after operation, the hippocampus of six mice in each group was taken and mixed with mitochondrial isolation medium A for homogenization.The homogenate underwent centrifugation at 1000 g for 10 min at 4°C, and the supernatant was discarded.The collected mitochondria were obtained from the precipitate after centrifugation at 11,000 g for 10 min at 4°C.Protein concentration was determined by adding preservation medium (400 μL per gram of tissue), and MMP was measured using red and green fluorescence intensity.JC-1 staining solution (0.9 mL) was mixed with the mitochondria, and excitation wavelengths of 490 and 525 nm were measured using an EnSpire multifunctional enzyme marker (PerkinElmer, USA).Emission wavelengths of 530 and 590 nm were utilized, respectively.

| Measuring ATP content
After 24-h post-surgery, hippocampal tissue was extracted from six randomly selected mice in each group.The tissue was added to the ATP detection lysate (Biyuntian Institute of Biotechnology) and homogenized 10 times with a precooled glass homogenizer.The resulting homogenate was centrifuged at 12,000 g for 5 min at 4°C, and the supernatant was discarded.Subsequently, 20 μL of tissue samples were mixed with an EnSpire multifunctional enzyme marker (PerkinElmer, USA) to determine relative para-optical unit values in the assay wells.Finally, the ATP content was calculated using the resultant standard curve.

| Western blotting
Hippocampal tissues were retrieved from an ultralow temperature refrigerator, weighed, and placed in precooled PBS buffer with protease inhibitor.After a 5-min homogenization using an ultrasonic cell crusher, the samples were centrifuged at 15,000 g for 10 min at 4°C.The resulting supernatant underwent BCA protein quantification, followed by the addition of protein loading buffer, boiling, denaturation, and storage at −80°C.Following electrophoretic separation and membrane transfer, the membranes were blocked with skimmed milk and incubated overnight at 4°C with primary antibodies against Drp1 (1:1000, Abcam, USA), Mfn2 (1:1000, Abcam, USA), PGC-1α (1:1000, Abcam, USA), NRF2 (1:1000, Abcam, USA), TFAM (1:1000, Abcam, USA), and GAPDH (1:10,000, Abcam, USA).Subsequently, the membranes underwent five minutes washes with TBST, followed by incubation with goat anti-rabbit secondary antibody (1:5000, Affinity, Australia) at room temperature for one hour.Expression levels were determined by analyzing the gray values using Image J software, and the ratio of the target protein's gray value to that of the reference GAPDH band was used as a reflection of expression level.

| Transmission electron microscope observation
The hippocampal tissue was cut into small pieces and put them in 2.5% glutaraldehyde at 4°C overnight.After fixation in 1% OsO4 and alcohol dehydration, tissues were embedded with Epon-Araldite resin.Then, the embedding block was cut into 50 μm ultrathin sections and stained sequentially with 2% uranyl acetate and lead solution.Finally, the sections were observed with the transmission electron microscope (Hitachi, Ltd., Tokyo, Japan).

| Statistical analysis
The statistical analysis was performed using SPSS 19.0 software.Shapiro-Wilk test was used to test the positive pacificity of the data.Normally distributed data were presented as mean ± standard deviation (x ± s), and survival rates were expressed as percentages.
One-way anova was used to analyze measurement data, while the survival rate was analyzed by Kaplan-Meier method and log-rank test.Statistical significance was determined at p < 0.05.

| High concentration of H 2 inhalation improved 7-day survival rate and MSS score of sepsis mice
As shown in Figure 1B, the Sham and Sham + H 2 groups maintained 100% survival over the 7-day period.Compared with the Sham group, septic mice exhibited a 50% 7-day survival rate (p < 0.01), which significantly improved to 90% after 67% H 2 inhalation (p < 0.01).Inhalation of 67% H 2 notably enhanced the 7day survival rate of septic mice.In addition, MSS scores were also performed on sepsis mice (Figure 1C).There was no significant difference between the Sham and the Sham+H 2 groups (p > 0.05), while the MSS score of the CLP group was significantly increased (p < 0.001), and the MSS score decreased in the CLP + H 2 group (p < 0.001).

| High concentration of H 2 inhalation ameliorated cognitive dysfunction
Recognition memory of the mice was assessed using morris water maze (MWM) test and Y-maze test.As the number of training increases, the time for mice to find a platform gradually decreased.
Notably, mice in the CLP group took longer to find a platform than in the Sham group during the same training time (p < 0.001).However, the time to find the platform was reduced after H 2 therapy (p < 0.01).
After removing the platform on the seventh day, the proportion of mice staying on the platform (Figure 2C) and the number of times they crossed the platform (Figure 2D) were reduced compared with the Sham group (p < 0.001), while the CLP + H 2 group significantly increased both (p < 0.001).In addition, the Y maze results showed that the CLP group demonstrated significantly reduced C-arm dwell times compared to the Sham group (p < 0.001), while the Sham + H 2 group showed no significant differences compared to the Sham group (p > 0.05).The CLP + H 2 group exhibited a significant increase in C-arm dwell times relative to the CLP group (p < 0.001) (Figure 2E).

| High concentration of H 2 inhalation significantly reduced brain injury
As Shown in Figure 3A, the CA1 area of the hippocampus in the Sham and Sham + H 2 groups displayed well-organized neuronal cells with neat arrangements and numerous, large, purple-blue Nissier vesicles, while the CLP group showed sparse neuronal cell structure with significantly reduced normal neurons, irregular arrangements, and reduced Nissier vesicle counts.Moreover, mice in the CLP + H 2 group exhibited slightly improved neuronal arrangements with reduced pathological changes and significantly lower apoptotic neuron counts compared to the CLP group (p < 0.001) (Figure 3B).In HE staining (Figure 3C), the hippocampal pyramidal neurons in the Sham

| High concentration of H 2 alleviated mitochondrial dysfunction
MMP is essential for ATP generation.In the present study (Figure 4A,B), MMP and ATP levels were significantly reduced in the CLP group compared to the Sham group (p < 0.001), indicating that mitochondrial function was impaired due to sepsis.However, the CLP + H 2 group exhibited a significant increase in MMP and ATP compared to the CLP group (p < 0.001).At the same time, the activities of antioxidant enzymes SOD and CAT in the hippocampus were also detected (Figure 4C,D).We observed that a significant decrease in SOD and CAT activity (p < 0.001) in the hippocampus tissue than those of the Sham group, which was reversed by inhalation of 67% H 2 (p < 0.001).Moreover, transmission electron microscopy was used to observe the morphology of mitochondria directly.
As shown in Figure 4E, Mitochondrial morphology was abnormal in the CLP group, including mitochondrial swelling, ridge collapse and mitochondrial membrane rupture.After H 2 therapy, mitochondrial damage was significantly reduced.

| High concentration of H 2 improved mitochondrial dynamics
GTPases play a predominant role in regulating mitochondrial fusion, while Drp1 and Mfn1/Mfn2 proteins are essential for both mitochondrial fission and fusion.In the CLP group, there was a significant increase in DRP1 expression (p < 0.001), whereas Mfn2 expression showed a significant decrease (p < 0.001).In contrast, the CLP + H 2 group displayed a significant decrease in DRP1 protein expression (p < 0.001) and a notable increase in MFN2 protein expression (p < 0.001) compared to the CLP group (Figure 5A-C).In addition, Drp1 was labeled with green fluorescence and directly observed by immunofluorescence (Figure 5D).Compared with the Sham group, the expression of Drp1 in the CLP group was significantly increased, while, the expression of Drp1 decreased significantly after H 2 treatment.Therefore, CLP led to a substantial imbalance between mitochondrial fusion and fission in hippocampal tissue, which was improved by exposure to high levels of H 2 .

| High concentration of H 2 improved mitochondrial biogenesis
Mitochondrial biogenesis regulates the generation of new mitochondria.Relevant proteins in hippocampus tissue were measured 24 h after casting.As shown in Figure 6A-D Inhalation of 67% H 2 further up-regulated the expression levels of these proteins compared to the CLP group (p < 0.001).Moreover, NRF2 was labeled with red fluorescence and directly observed by immunofluorescence (Figure 6E).Compared with the Sham group, the expression of NRF2 in the CLP group was significantly increased, while the expression of NRF2 increased further after H 2 treatment.These results indicated that high concentrations of H 2 enhanced hippocampal mitochondrial function and alleviated sepsis-induced brain injury in mice.

| DISCUSS ION
In this study, we found that 67% H 2 could improve the survival rate of septic mice, alleviate the pathological damage of brain tissue as well as cognitive dysfunction, and reduce the inflammatory factors as well as redox imbalances.This beneficial effect was related to the improvement of mitochondria dynamics and biogenesis.
In sepsis, the brain tissue is one of the most vulnerable target organs.Clinically, SAE has a high fatality rate, and survivors often present with chronic autonomic dysfunction, delirium or cognitive impairment of varying degrees. 15The mortality rate of CLP model mice established in this study was as high as 50% and the cognitive function of mice was impaired, which was consistent with previous studies. 16Inhalation of 67% H 2 could effectively improve the survival rate and had a protective effect on cognitive function.
Studies have found that cognitive dysfunction in sepsis patients is closely related to hippocampal inflammation and pathological damage. 17,18Proinflammatory cytokines play a key role in the pathophysiology of SAE in critically ill patients. 19,20Elevated levels of proinflammatory factors lead to neuronal apoptosis, necrosis, 21 axon damage, brain edema, 22 inhibition of neurotransmitter transporters, and breakdown of the blood-brain barrier (BBB). 23In this study, elevated hippocampal inflammatory factors such as TNFα, IL-1β, and HGMB1 were observed in mice with sepsis, as well as neuropathological changes in the CA1 region of the hippocampus, including disordered neuronal arrangement, reduced structural density, and reduced number of normal neurons.Chen et al. shown that H 2 alleviates hippocampal damage in a variety of neurological diseases, primarily by regulating inflammatory responses and alleviating histological damage. 17This is consistent with our findings that 67% H 2 treatment improves hippocampal inflammatory cytokine levels and pathological tissue.
Brain tissue has lower antioxidant defense level and higher oxygen consumption compared with other organs.Therefore, it is more susceptible to oxidative damage in sepsis. 24The imbalance of oxidative stress can disrupt cellular respiration and lead to abnormal metabolism, and the free radicals produced can induce inflammatory mediators, causing the breakdown of the blood-brain barrier and secondary brain damage. 25Experimental and clinical studies have shown that the increased production of ROS in sepsis is associated with the consumption of antioxidants. 24CAT and SOD are crucial antioxidant enzymes that reflect the body's antioxidant levels and damage capacity. 26,27In our study, SOD and CAT were significantly decreased in the hippocampus of CLP mice, while inhalation of 67% H 2 significantly improved antioxidant capacity.Furthermore, ROS can impair ATP synthesis by inducing mitochondrial structural changes that lead to loss of mitochondrial enzyme activity, which is associated with acute brain dysfunction in SAE patients. 28At the same time, the mitochondrial permeable pore (mPTP) opens abnormally due to the imbalance of cellular oxidative stress.Excessive water and electrolyte enter the mitochondrial matrix, resulting in mitochondrial swelling, mitochondrial outer membrane rupture, MMP depletion, and a large amount of oxygen free radicals are produced. 29Excessive ROS causes damage to mitochondrial proteins, and ROS clearance is impaired, which may cause a vicious cycle and further lead to neuronal injury. 29Our previous studies [30][31][32][33] have shown that 2% H 2 has antioxidant and anti-inflammatory effects.In addition to protecting multiple organs in sepsis, it can also improve mitochondrial function. 31In this study, we found that 67% H 2 also had the same function, improving MMP and ATP production.
Mitochondria play a vital role in neuronal function.Once changes in the mitochondrial dynamics occur, including fission and fusion, harmful effects may occur. 34These processes are regulated by key proteins, such as Drp1, Mfn1, and Mfn2.Drp1 and mitochondrial fission protein 1 are associated with mitochondrial fission, while mitofusin, Mfn1, Mfn2, and optic atrophy protein 1 (OPA1) are involved in mitochondrial fusion. 35In the SAE model, Haileselassie et al. 36  Consistent with references, Drp1 increased and Mfn2 decreased in the hippocampus of sepsis mice.However, 67% H 2 treatment can improve the expression of the above proteins, suggesting that 67% H 2 has an improved effect on mitochondrial dynamics.Mitochondrial biogenesis, described as the growth and division of mitochondria, improves survival in patients with sepsis by producing new mitochondria, while inhibition of mitochondrial biogenesis worsens prognosis. 38,39It is currently widely believed that damaged mitochondrial clearance in sepsis organs can be compensated by increased mitochondrial biogenesis to produce new mitochondria, which may be a key reason for improving survival rates and reversing organ damage in sepsis patients. 40Mitochondrial biogenesis is regulated by nuclear and mitochondrial factors such as PGC-1α, NRF1, NRF2, and TFAM.Specifically, PGC-1α activates NRF1 and NRF2, which regulate mitochondrial biogenesis by overseeing the expression of nuclear-encoded proteins involved in mitochondrial respiratory function and translation. 41Moreover, PGC-1α activation triggers TFAM translocation from the cytosol to the mitochondria, where it binds to mtDNA and governs its transcription and replication. 42Up-regulation of PGC-1α has been shown to reduce the production of and mitochondrial ROS, as well as levels of intracellular inflammatory factors. 43,44Zhao et al. observed that the biogenesis of mitochondria in cerebral cortical astrocytes was strengthened to cope with the increased energy demand of astrocytes under sepsis conditions, and finally the ultrastructure of mitochondria was restored under mild injury. 45This is consistent with our findings that the expression levels of mitochondrial biosynthetic proteins such as PGC-1α, NRF2, and TFAM were significantly elevated in mice with sepsis, and the expression of biosynthetic proteins was higher after 67% H 2 treatment.These results suggest that 67% of H 2 improves SAE by promoting mitochondrial biogenesis.In 2007, Ohsawa et al. 46 first reported that hydrogen (H 2 ), as a therapeutic antioxidant, can effectively remove the most cytotoxic oxygen free radicals and has a protective effect on cerebral ischemia-reperfusion injury.Our previous studies have also shown that 2% H 2 has a protective effect on multiple organs of sepsis, including intestinal injury, 32 lung injury, 31 liver injury, 33 and brain injury. 16though 2% H 2 has a protective effect on multiple organs in sepsis, this does not mean that 2% H 2 is the optimal therapeutic concentration, and the optimal concentration is still unknown.
In the past, there was a view that high concentration of hydrogen was easy to explode and its safety was poor.However, with the development of advanced high concentration H 2 therapeutic devices, 66.7% hydrogen and 33.3% oxygen can be produced by electrolyzing water, and the safety of use is ensured through safety mechanisms such as sealing loop and concentration monitoring feedback.
Using these devices, high concentration H 2 (67%) has been used to treat a variety of diseases, leading to favorable outcomes. 6,14,47haling high concentrations of H 2 boosts its antioxidant effect by theoretically elevating its concentration in tissues, cells, and organelles. 14Jun et al.'s study showed that 67% of H 2 inhalation can be maintained in tissues for a better duration than 4% H 2 and was more protective against acute kidney injury. 47In addition, Xie et al.'s study showed that inhalation of 67% H 2 could better improve 7-day survival rate and sepsis induced liver, kidney and lung injury in mice than 2% H 2 , which may be related to the more activation of Nrf2 signaling pathway. 14This suggested that the therapeutic effect of high concentration H 2 inhalation is better than that of low concentration H 2 inhalation.Therefore, we investigated whether high levels of H 2 have a protective effect on septic brain injury.
There are limitations in our study.First, we only use a dose of 67% H 2 , so the dose effect and optimal dose are unknown.Second, the current treatment of inhalation of high concentrations of H 2 is relatively mature, but H 2 still has the risk of explosion.Finally, we only explored the protective effects of high concentration H 2 on the brain, liver and kidney, and whether it has protective effects on other organs is worth further exploration.

| CON CLUS ION
Inhalation of high concentration (67%) of H 2 in septic mice improved survival and reduced neuronal injury.Its mechanism might be mediated by enhanced mitochondrial biogenesis and dynamics.
biogenesis, mitochondrial dynamics, sepsis-associated encephalopathy | 3 of 11 CUI et al. lime was placed at the bottom of the box to absorb exhaled CO 2 .
group and the Sham+H 2 group were well-structured and tightly arranged.In the CLP group, the nuclear nucleus of hippocampal pyramidal neurons was condensed, the cytoplasm was hyperchromatic, and the arrangement was disordered.After treatment with H 2 , the F I G U R E 2 67% H 2 inhalation improved cognitive dysfunction in mice with sepsis.(A) Trajectory plot of the seventh day after training.(B) Escape latency during training in mice of each group.***p < 0.01 vs. Sham group, ## p < 0.01, ### p < 0.001 vs.CLP group.(C) Time ratio of mice in the quadrant where the platform is located on 7th day after training.(D) The number of times the mice crossed the platform on 7th day after training.(E) Times in the novel arms in the Y-maze test.All the data were expressed as means ± SD (n = 6 per group).***p < 0.001 compared with two groups.morphology and arrangement of hippocampal pyramidal neurons improved.For the detection of inflammatory factors (Figure 3D-F), TNFα, (p < 0.001), IL-1β (p < 0.001) and HMGB1 (p < 0.001) were significantly increased in the CLP group compared with the Sham group, while these inflammatory factors were significantly reduced in the CLP + H 2 group (p < 0.001).
, the protein expression levels of PGC-1α, NRF2, and TFAM significantly increased F I G U R E 3 67% H 2 inhalation ameliorated brain injury in septic mice.(A) Nissl staining of hippocampus (×200).(B) The number of abnormal neurons was calculated based on Nissl staining.(C) HE staining of brain tissue.(D-F) Serum TNFα, IL-1β, and HMGB1.All the data were expressed as means ± SD (n = 6 per group).The red arrows represent the obvious differences.***p < 0.01 compared with two groups.in the CLP group compared to the Sham group (p < 0.001).This indicates that septic stimulation induced an active phase and increased biosynthesis of hippocampal tissue mitochondria.

F
I G U R E 4 67% H 2 inhalation alleviated mitochondrial dysfunction of septic mice.(A) The JC-1 aggregate/monomer ratio was used to assess the MMP.(B) The ATP content was detected based on fluorescein.(C) Hippocampal CAT.(D) Hippocampal CAT.(E) The morphology of mitochondria was observed by transmission electron microscopy.The red arrows represent the obvious differences.All the data were expressed as means ± SD (n = 6 per group).***p < 0.01 compared with two groups.
observed that the disruption of the cell respiratory chain under LPS stimulation, F I G U R E 5 67% H 2 improved the mitochondrial dynamics in septic mice.(A-C) The expression of DRP1 and MFN2 was detected by western blotting.(D) Drp1 was observed by immunofluorescence staining.All the data were expressed as means ± SD (n = 6 per group).***p < 0.01 compared with two groups.leading to the loss of MMP, resulting in an increase in Drp1 and the recruitment of p53 to the mitochondrial outer membrane, followed by the initiation of the cell death pathway.At the same time, Li et al. have shown that HO-1 can reduce sepsis-induced renal injury by regulating the expression of Drp1 and up-regulating the expression of Mfn2. 37

F I G U R E 6
67% H 2 improved the mitochondrial biogenesis in septic mice.(A-D) The expression of mitochondrial biogenesis-related proteins (PGC-1α, NRF2, and TFAM) was assayed by western blotting.(E) NRF2 was observed by immunofluorescence staining.All the data were expressed as means ± SD (n = 6 per group).***p < 0.01 compared with two groups.