Phosphorylation‐mediated PI3K‐Art signalling pathway as a therapeutic mechanism in the hydrogen‐induced alleviation of brain injury in septic mice

Abstract Our previous studies illustrated that 2% H2 inhalation can protect against sepsis‐associated encephalopathy (SAE) which is characterized by high mortality and has no effective treatment. To investigate the underlying role of protein phosphorylation in SAE and H2 treatment, a mouse model of sepsis was constructed by caecal ligation and puncture (CLP), then treated with H2 (CLP + H2). Brain tissues of the mice were collected to be analysed with tandem mass tag‐based quantitative proteomics coupled with IMAC enrichment of phosphopeptides and LC–MS/MS analysis. In proteomics and phosphoproteomics analysis, 268 differentially phosphorylated proteins (DPPs) showed a change in the phosphorylated form in the CLP + H2 group (p < 0.05). Gene ontology analysis revealed that these DPPs were enriched in multiple cellular components, biological processes, and molecular functions. KEGG pathway analysis revealed that they were enriched in glutamatergic synapses, tight junctions, the PI3K‐Akt signalling pathway, the HIF‐1 signalling pathway, the cGMP‐PKG signalling pathway, the Rap1 signalling pathway, and the vascular smooth muscle contraction. The phosphorylated forms of six DPPs, including ribosomal protein S6 (Rps6), tyrosine 3‐monooxygenase/tryptophan 5‐monooxygenase activation protein gamma (Ywhag/14–3‐3), phosphatase and tensin homologue deleted on chromosome ten (Pten), membrane‐associated guanylate kinase 1 (Magi1), mTOR, and protein kinase N2 (Pkn2), were upregulated and participated in the PI3K‐Akt signalling pathway. The WB results showed that the phosphorylation levels of Rps6, Ywhag, Pten, Magi1, mTOR, and Pkn2 were increased. The DPPs and phosphorylation‐mediated molecular network alterations in H2‐treated CLP mice may elucidate the biological roles of protein phosphorylation in the therapeutic mechanism of H2 treatment against SAE.


| INTRODUC TI ON
Sepsis, which is caused by a series of pathophysiological processes, such as the excessive release of inflammatory factors and an imbalance between anti-inflammatory and proinflammatory responses, leads to multiple organ dysfunction syndrome (MODS) 1 and is the primary reason for admission into the intensive care unit. 2,3 Among the types of MODS caused by sepsis, sepsis-associated encephalopathy (SAE) is the most serious and has high morbidity 4 and mortality rates because the central nervous system is one of the most vulnerable organs. 5 As the most widespread gas in the universe, hydrogen gas (H 2 ) has been proven to alleviate many diseases and exert antiinflammatory, anti-apoptotic, and antioxidant effects. 6 Our previous studies have demonstrated that inhalation of 2% H 2 can alleviate sepsis and protect against sepsis-induced injury to various organs, such as the lungs, intestines, and brain. [7][8][9] However, the therapeutic mechanism of H 2 is still unknown.
With the widespread application and continuous development of molecular-based technologies, proteomics and phosphoproteomics analysis have been used in various studies to explore disease mechanisms and therapeutic strategies. 10 Phosphorylation and dephosphorylation are regulated by kinases and phosphatases, respectively. Approximately 2%-5% of genes in the human genome encode protein kinases and phosphatases. 11 Previously, our studies demonstrated the differences in protein levels in lung tissues and intestinal tissues from septic animals and those from normal animals. 12,13 In this study, tandem mass tag (TMT)-IMAC-LC-MS/MS was used to investigate the differences in the levels of phosphoproteins and changes in phosphorylation-mediated signalling pathways in the brain tissues of septic mice treated with H 2 compared with those of untreated septic mice to clarify the specific mechanism underlying the therapeutic effect of H 2 on sepsis.

| Animals and experiments
We purchased healthy adult male C57BL/6J mice (weighing 23-26 g and aged 8 weeks) from the Experimental Animal Center of the Academy of Military Sciences. The mice had free access to food and water and were housed in cages in a stable environment (temperature, 22-25°C; humidity, 55% ± 10%; and 12:12 h light-dark cycle).
In this study, a total of 80 mice were randomly divided into four groups: the sham, sham + H 2 , caecal ligation and puncture (CLP), and CLP + H 2 groups. The mice in the CLP and CLP + H 2 groups underwent CLP to construct a sepsis model, while mice in the sham and sham + H 2 groups underwent sham surgery. At 1 and 6 h following sham surgery or CLP, the mice in the sham + H 2 and CLP + H 2 groups were administered 2% H 2 by inhalation for 60 min. The 7-day survival rates of the mice after sham surgery or CLP were assessed (n = 20 per group). A total of 40 mice were used for brain proteomics and phosphoproteomics analysis (n = 10 mice per group), 24 mice were used for WB (n = 6 mice per group), and 16 mice were used for haematoxylin and eosin, Nissl staining, and TUNEL staining (n = 3 mice per group).

| CLP model
CLP was performed as previously described. 9 Before the operation, all the animals fasted for 8 h. The mice were anaesthetized by isoflurane inhalation and placed in a prone position. The skin was sterilized, a 1-cm incision was made in the abdomen, the caecum was exposed, and 40% of it was ligated. Then, the caecum was punctured with a 21-gauge needle, and approximately 0.3 ml of the caecal content was pushed out using sterile forceps. The abdominal muscles and skin were closed with simple running sutures and metallic clips, respectively, and then the caecum was returned to the abdominal cavity. After CLP or sham surgery, the mice were subcutaneously injected with saline solution (1 ml) and administered lidocaine cream (cat# H20063466) for pain relief. The postoperative mice were housed under controlled conditions (room temperature of 20-25°C), and a heating pad was used to prevent hypothermia.

| H 2 treatment
H 2 treatment was applied as described in our previous study. 13 Briefly, the mice were placed in a closed resin box with an inlet and an outlet. H 2 was delivered using a TF-1 gas flowmeter (4 L/min; YUTAKA Engineering Corp). To monitor the H 2 concentration, a detector (HY-ALERTA Handheld Detector Model 500; H2 Scan) was used. The CO 2 expired by the mice was absorbed by Baralyme. The mice in the sham + H 2 and CLP + H 2 groups inhaled 2% H 2 gas for 60 min, 1 h, or 6 h following CLP or sham surgery, while mice in the sham and CLP groups inhaled room air.

| Haematoxylin and eosin staining
As described in a previous study, 14 24 h following sham surgery or CLP, the mice in each experimental group were deeply anaesthetized, and their brain tissues were collected for haematoxylin and eosin staining. The mice were transcardially perfused with phosphatebuffered saline (PBS) followed by 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). The brain tissues were excised, fixed with 10% formalin, embedded in paraffin, and cut into sections. The samples were stained with haematoxylin and eosin following deparaffinization and dehydration. Regional neuropathological changes were observed using a BX51 microscope (Olympus).

| Nissl staining
Brain tissues were cut into 10μm sections and stained with 1% crystal violet (Sigma) as described previously. 7 Subsequently, pyramidal neurons in the hippocampal region were observed, and Nissl bodies in the hippocampal region were counted under a microscope.

| TUNEL staining
Neuronal apoptosis was evaluated using the TUNEL assay 24 h after sham surgery or CLP. The nucleotide-labelling mix (TUNEL reagent) contained fluorescein-dUTP and fluorescein-dNTPs, both of which are used for TUNEL staining for in situ analysis of apoptosis. In addition, a nucleotide-labelling mix was used in combination with the TUNEL enzyme to prepare the TUNEL reaction mixture. Cells exhibiting DNA strand breaks were labelled with this reaction mixture, allowing us to analyse and quantify the degree of apoptotic cell death at the single-cell level among cells and in tissues. Apoptotic cells were stained green, and the nuclei of all the cells were stained blue with DAPI.

| Sample collection
First, 40 mice from all the experimental groups were anaesthetized 24 h following sham surgery or CLP. Subsequently, the heart was exposed, the right atrial appendage was cut, a cardiac puncture was performed using a sterile 18-gauge needle, and the heart was perfused via the left ventricle with precooled PBS to remove the blood. Next, the brain tissue was collected after removing the skull, after which the sample was quickly frozen using liquid nitrogen and stored at −80°C for subsequent proteomics and phosphoproteomics analysis. To control for individual differences, six samples from each group were pooled. A total of three groups of six samples including 10 biological replicates and two technical replicates were subjected to iTRAQ-based quantitative proteomics and phosphoproteomics analysis.

| Protein extraction and digestion
Two hundred micrograms of protein from each sample were mixed with 30 μl SDT buffer (4% SDS, 100 mM DTT, and 100 mM Tris-HCl [pH 7.6]). The concentration of protein was quantified with the BCA Protein Assay Kit (Bio-Rad). Trypsin protein digestion was performed according to the filter-aided sample preparation procedure described by Matthias Mann. 15 The digest peptides of each sample were desalted on C18 cartridges (Empore™ SPE Cartridges C18 [standard density], bed I.D. 7 mm, volume 3 ml, Sigma), concentrated by vacuum centrifugation and reconstituted in 40 μl of 0.1% (v/v) formic acid. The peptide content was estimated by UV light spectral density at 280 nm using an extinction coefficient of 1.1 of 0.1% (g/l) solution that was calculated based on the contents of tryptophan and tyrosine in vertebrate proteins.

| TMT labelling and fractionation using the high-pH reversed-phase method
A 100 μg peptide mixture from each sample was labelled using TMT reagent (Thermo Fisher Scientific) according to the manufacturer's instructions. For fractionation using the high-pH reversed-phase method, a Pierce high pH reversed-phase fractionation kit (Thermo Scientific) was used to fractionate the TMT-labelled digest samples into six fractions by increasing acetonitrile step-gradient elution according to the instructions. The peptides were loaded onto a reverse-phase trap column (Thermo Scientific Acclaim PepMap100, 100 μm*2 cm, nanoViper C18) connected to a C18 reversed-phase analytical column (Thermo Scientific Easy Column, 10 cm long, 75 μm inner diameter, and 3 μm resin) in buffer A (0.1% formic acid) and separated with a linear gradient of buffer B (84% acetonitrile and 0.1% formic acid) at a flow rate of 300 nl/min controlled by IntelliFlow technology. The mass spectrometer was operated in positive ion mode. MS data were acquired dynamically using a data-dependent top 10 method by choosing the most abundant precursor ions from the survey scan (300-1800 m/z) for HCD fragmentation. The automatic gain control (AGC) target was set to 3e6, the maximum injection time was set to 10 ms, and the dynamic exclusion duration was set to 40.0 s. Survey scans were acquired at a resolution of 70,000 at m/z 200, the resolution for HCD spectra was set to 17,500 at m/z 200, and the isolation width was set to 2 m/z. The normalized collision energy was set to 30 eV, and the underfill ratio, which specifies the minimum percentage of the target value likely to be reached at maximum fill time, was set to 0.1%. The instrument was run with peptide recognition mode enabled.

| Western Blot
Twenty-four hours after CLP or sham surgery, brain tissues were collected from the mice, weighed, and lysed using RIPA buffer (Solarbio) containing protease inhibitors (Solarbio) to extract total protein.

| Statistical analysis
Statistical analysis was performed with GraphPad Prism software (version 8.0) and spss statistical software (version 21.0). Survival rates are reported as percentages, and analysis was performed by the log-rank test. The data are presented as the mean ± standard deviation. Two-way repeated-measures anova followed Tukey's multiple comparisons test was used to analyse differences between the treatment groups. Unpaired t-test (if the values were normally distributed) or the Mann-Whitney test (if the values were nonnormally distributed) was employed to analyse differences between the two groups. For all tests, p < 0.05 was considered statistically significant.

| Survival rate
After sham surgery or CLP, several physiological and pathological changes, including unkempt hair, loose stools, a reduction in activity, and an increase in respiratory rate, were observed, consistent with a previous study. 13 All mice in the sham and sham + H 2 groups survived to 7 days, while the number of surviving mice at 7 days was significantly reduced in the sepsis and sepsis + H 2 groups (p < 0.05). Nevertheless, the survival rate in the sepsis + H 2 group was significantly improved compared with that in the sepsis group (40% vs. 60%, p < 0.05) ( Figure 1E).

| Detrimental histological changes in the hippocampi of septic mice
Haematoxylin and eosin staining showed that in the sham and sham + H 2 groups, pyramidal neurons in the hippocampal area had a clear structure without obvious abnormalities ( Figure 1A) and that there was no significant difference in pyramidal neuron structure between these two groups ( Figure 1C, p > 0.05). However, in the CLP group, pyramidal neurons were seriously damaged and disorderly arranged, the cytoplasm was deeply stained, and nuclei were pyknotic ( Figure 1A). Neuronal damage was alleviated in the CLP + H 2 group compared with the CLP group, and most of the neurons in the CLP + H 2 group showed a normal structure with slight damage. The number of normal neurons in the hippocampal area was significantly F I G U R E 1 H 2 can relieve the hippocampal histopathology damage and improve the 7-day survival rates in septic mice. At 24 h following sham or CLP operation, (A) haematoxylin and eosin staining was used to detect the hippocampal histopathology changes. Regular morphology of the pyramidal neurons in the hippocampus was observed in the sham and sham + H 2 groups. There were a large number of damaged neurons (indicated using black arrows) in the CLP group, and a significantly decreased number following H2 treatment (C).

| Changes in protein phosphorylation in the brain tissues of septic mice
To determine if the changes in protein phosphorylation identified by proteomics analysis are associated with the therapeutic effects of H 2 on brain damage in septic mice, we collected brain tissues from and tyrosine (114) phosphorylation sites (accounting for 12.36% and 0.72% of total phosphopeptides, respectively) ( Figure 2C). In addition, most of the phosphoproteins (57.85%) contained more than two phosphorylation sites; for example, the Q8BTI8 protein had 293 phosphorylation sites ( Figure 2D). The mean phosphorylation frequency of all quantified proteins was as high as 0.71 per hundred amino acids (Figure 2 E). These findings suggest that a prominent change in protein phosphorylation, mainly serine phosphorylation, is critically important for the therapeutic effect of H 2 on brain damage in septic mice.

| Combined proteomics and phosphoproteomics analysis
Proteins are the functional executors of life activities, and their regulatory function is mainly affected by their expression levels and posttranslational modifications (PTMs). They ultimately affect the phenotypic changes of organisms. Previous studies have shown that H 2 treatment leads to changes in protein levels and modification in the brain tissues of septic mice. 16 Therefore, we performed a combined analysis of proteins and modified proteins to further assess the ability of H 2 to alter protein function in mice with brain injury.
As shown in Figure 3A-C, the levels of 6883 proteins were quantified and six showed differential abundance. Moreover, the levels of  Table 2.

| Pathway analysis of the DPPs
To further elucidate the biological roles of the DPPs (proteins for which only the phosphorylated form showed a change in abundance) and the signalling events they regulate, we performed pathway analysis of the 268 DPPs (p < 0.05, CLP + H 2 group vs. CLP group, one-sample t-test) using the KAAS. KEGG pathway analysis of the differentially abundant proteins and DPPs was performed.
The results are shown in a single image because they can be intuitively compared to identify synergistic relationships between protein expression and modification and differential regulation of these processes from the perspective of pathways ( Figure 5A).
We defined significant enrichment as an FDR <0.05. The top 20 pathways are shown in Figure 5B.  In this way, the PI3K-Akt signalling pathway can regulate a wide variety of biological processes, including inflammation, cellular proliferation, autophagy, and apoptosis. 18 Therefore, the PI3K-Akt signalling pathway and up-and downstream signalling molecules may play a key role in the therapeutic effect of H 2 on brain damage in septic mice.

| Verification of PI3K-Akt signalling pathwayrelated DPPs in CLP + H 2 group mice compared to CLP group mice
Among the 268 DPPs in CLP + H 2 group mice compared to CLP group mice identified by quantitative phosphoproteomics, we chose to validate the changes in the levels of ribosomal protein S6 The protein-protein interaction network and associated pathways of these six proteins involved in the PI3K-Akt signalling pathway are shown in Figure 6A. Immunoblotting showed that the total protein levels of Ywhag, Magi1, mTOR, Rps6, Pten, and Pkn2 were not different between the CLP + H 2 group and CLP group (p > 0.05, Figure 6B,C). Mass spectrometry was then used to identify the differentially phosphorylated sites in Ywhag, Magi1, and mTOR, with the results are shown in Figure 6C. Western blotting (WB) showed that the levels of phosphorylated Rps6 (S240 + S244), Pten (S370), and Pkn2 (T816) were increased in septic mice treated with H 2 compared with CLP model mice, while the total levels of those proteins were not changed ( Figure 6D). These results are consistent with the results of MS/MS-based quantitative phosphoproteomics in our study. In addition, consistent with a previous study, 19 WB showed that the level of phosphorylated mTOR (S2448) was decreased in the CLP + H 2 group compared with the CLP group. This was because H 2 could regulate the mTOR-autophagy signalling pathway to attenuate sepsis-induced neuroinflammation.
In summary, in septic mice, H 2 might regulate the phosphorylation of proteins (Rps6, Ywhag, Pten, Pkn2, Magi1, and mTOR) in the PI3K-Akt signalling pathway to regulate cell metabolism, growth, and proliferation.

| DISCUSS ION
Sepsis is a systemic inflammatory syndrome that can lead to neurological impairment, including behavioural alterations and neurological impairment. 7,12 The CLP model, the gold standard model of sepsis, was applied in our present study. 20 In our previous studies, we found that 2% H 2 inhalation can alleviate sepsis and sepsis-induced damage to organs, including the lung, liver, kidney, and brain. [7][8][9]14,21 In this study, inhalation of 2% H 2 effectively increased the 7-day survival rate of septic mice. In our previous studies, we showed that the mechanism underlying the TA B L E 2 Top 20 correlated proteins differential expressed only in phosphoproteins in the brain of CLP + H 2 mice compared to CLP mice  apoptosis and autophagy in neurons, as previously reported. 19,24 KEGG pathway analysis showed that the DPPs were significantly enriched in 20 signalling pathways, including autophagy, endocytosis, the PI3K-Akt signalling pathway, the HIF-1 signalling pathway, the cGMP-PKG signalling pathway, and the Rap1 signalling pathway, which are related to signal transduction, transport, and catabolism. Thus, the therapeutic effects of H 2 may be associated with cell growth and metabolism.
It has been proven that sepsis regulates the PI3K-Akt signalling pathway. The PI3K-AktmTOR signalling pathway is a major regulator of cell growth and metabolism, promoting anabolic processes such as ribosome biogenesis and the synthesis of proteins, nucleotides, fatty acids, and lipids and inhibiting catabolic processes such as autophagy. 25 Evidence has shown that dysregulation of mTOR signalling is linked to many human diseases, including diabetes, neurodegenerative diseases, and cancer. 26  Then, PRKs phosphorylate pyrin, leading to the recruitment of 14-

3-3 proteins and inhibition of inflammasome activation. Genetic
alterations that lead to dysregulation of PI3K signalling, most commonly PI3K mutations, HER2 amplification, and events that inactivate PTEN, are prevalent in cancer. 34 Translational regulation of PTEN is a major process by which physiological PI3K signalling is homeostatically regulated and plays a role in reducing pathway activation by oncogenic PIK3CA mutants and the antitumour F I G U R E 6 Differentially expressed phosphoproteins (DPPs) (Rps6, Pten, Ywhag, mTOR, Magi1, and Pkn2) participate in the PI3K-AKT signalling pathway were significantly different in the therapeutic mechanism of hydrogen in CLP-induced mice. (A) The western blot were used to identify the differentially expressed phosphoproteins and the total proteins. The results showed there were no changes in the total level of Ywhag, mTOR, and Magi1 quantified by the ratio of band density to those of β-actin, respectively (B) and the MS/MS spectrums from Phosphorylated peptides. (C) (p-Ywhag assigned to TSADGNEKK; p-Magi1 assigned to AENEVPSPASSHHSSNQPASLTEEK; p-mTOR assigned to IMLRMAPDYDHLTLMQKVEVFEHAVNNTAGDDLAK). The b and y ions including loss of ammonia and water were considered when we calculated the PTM score.) (D) The expression of Rps6, p-Rps6, Pten, p-Pten, mTOR, p-mTOR, Pkn2, and p-Pkn2 in CLP + H2 were detected by western blot. Quantitative analysis of p-Rps6, p-Pten, p-mTOR, and p-Pkn2 is shown as the ratio of band density to that of Rps6, Pten, mTOR, and Pkn2, respectively. NS, no significance; *p < 0.05; **p < 0.01; ***p < 0.001. Magi1, membrane-associated guanylate kinase1; mTOR, mammalian target of rapamycin; Pkn2, protein kinase N2; Pten, phosphatase and tensin homologue deleted on chromosome ten; Rps6, Ribosomal protein S6; Ywhag/14-3-3, tyrosine 3-Monooxygenase/tryptophan 5-monooxygenase activation protein gamma activity of PI3K pathway inhibitors. 35 As an indirect downstream effector of mTOR, rpS6 not only participates in glucose homeostasis but is also essential for regulating the size of at least some cell types. 36 In addition, it was recently shown that specific small inhibitory RNA-mediated knockdown of RPS6 disrupts 40S ribosomal biogenesis, showing that RPS6 is a critical regulator of 5′ terminal oligopyrimidine (5′ TOP) translation. 37 In our research, the activation of proteins related to the PI3K-Akt signalling pathway (Rps6, Ywhag, Pten, Pkn2, Magi1, and mTOR) was closely associated with the therapeutic effect of H 2 on brain damage in septic mice.

| CON CLUS ION
In the present study, we quantitatively assessed the changes in protein phosphorylation in the H 2 -treated CLP model mice and identified functions and signalling cascades associated with the phosphoproteins that exhibited significant changes in abundance.
Importantly, we revealed the critical involvement of the PI3K-Akt signalling pathway and that related phosphoproteins (p-Rps6, p-Ywhag, p-Pten, p-Magi1, p-mTOR, and p-Pkn2) in the ability of H 2 to relieve brain damage in septic mice. Taken together, these findings provide novel insights into the mechanisms underlying the thera- Yu yonghao: Funding acquisition (supporting); writing -review and editing (supporting).

ACK N OWLED G EM ENTS
This work was supported by the National Natural Science

CO N FLI C T O F I NTE R E S T
The authors declare that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the fndings of this study are available from the corresponding author upon request.