Interaction Between the PERK/ATF4 Branch of the Endoplasmic Reticulum Stress and Mitochondrial One-Carbon Metabolism Regulates Neuronal Survival After Intracerebral Hemorrhage

Recent investigations have revealed that oxidative stress can lead to neuronal damage and disrupt mitochondrial and endoplasmic reticulum functions after intracerebral hemorrhage (ICH). However, there is limited evidence elucidating their role in maintaining neuronal homeostasis. Metabolomics analysis, RNA sequencing, and CUT&Tag-seq were performed to investigate the mechanism underlying the interaction between the PERK/ATF4 branch of the endoplasmic reticulum stress (ERS) and mitochondrial one-carbon (1C) metabolism during neuronal resistance to oxidative stress. The association between mitochondrial 1C metabolism and the PERK/ATF4 branch of the ERS after ICH was investigated using transcription factor motif analysis and co-immunoprecipitation. The findings revealed interactions between the GRP78/PERK/ATF4 and mitochondrial 1C metabolism, which are important in preserving neuronal homeostasis after ICH. ATF4 is an upstream transcription factor that directly regulates the expression of 1C metabolism genes. Additionally, the GRP78/PERK/ATF4 forms a negative regulatory loop with MTHFD2 because of the interaction between GRP78 and MTHFD2. This study presents evidence of disrupted 1C metabolism and the occurrence of ERS in neurons post-ICH. Supplementing exogenous NADPH or interfering with the PERK/ATF4 could reduce symptoms related to neuronal injuries, suggesting new therapeutic prospects for ICH.


Supplementary figures
Figure S1.A) MAP2 expression of primary neurons, revealing a purity level of at least greater than 90%.B) CCK-8 assay detecting the cell survival rates of neurons 24 h after stimulation with different concentrations of hemin (n=3/group).C) Typical morphological characteristics of primary neurons after being treated with the indicated hemin concentrations for 24 h.Data are presented as mean ± SD. ** p < 0.01.  Figure S3.A) Expression levels of GRP78, CHOP, ATF4, the p-PERK/PERK, and p-eIF2α/eIF2α protein ratio after stimulation with different concentrations of hemin (n = 3/group).B) Western blot analysis showing the levels of ATF6 and IRE1α/XBP1s branch markers in the ERS after primary neurons were treated with varying concentrations of hemin for 24 h (n = 3/group).C) Expression levels of GRP78, CHOP, ATF4, the p-PERK/PERK, and p-eIF2α/eIF2α protein ratio in protein extracts from perihematomal mouse brain tissue at different time points after ICH (n = 3/group).D) Western blot analysis of ATF6 and IRE1α/XBP1s branch markers in the ERS in protein extracts from perihematomal mouse brain tissue at different time points after ICH (n = 3/group).E) qRT-PCR analysis of key genes involved in the ERS and 1C metabolism in RNA extracted from primary neurons treated with or without 30 μM hemin for 24 h (n = 3/group).F) Multivariate control chart illustrating targeted metabolomics data for 1C-related substances, where each point represents a sample.All samples remain within 2 standard deviations, indicating good quality control and reliable data.G) Two-dimensional score plot of PCA for the two groups, visually representing patterns between different groups.H) Two-dimensional score plot of PLS-DA for the two groups.I) Two-dimensional score plot of OPLS-DA for the two groups for selecting differential metabolites.J) Permutation test results for the OPLS-DA model for the two groups to assess the model's validity (n=6 per group.S represents control neurons; E represents hemin-treated 24-h neurons).Data are presented as mean ± SD. * p < 0.05, ** p < 0.01.with varying concentrations of GSK2606414 for 24 h using the CCK-8 kit (n = 3/group).E) Quantitative analysis of TUNEL+ apoptotic primary neurons treated with or without GSK2606414 for 24 h (n =3/group).F) Bcl2/Bax and cleaved caspase-3/caspase-3 expression levels in neurons treated with or without GSK2606414 under hemin stimulation conditions, analyzed by western blot analysis (n = 3/group).Data are presented as mean ± SD. * p < 0.05, ** p < 0.01.

Method S1
Non-targeted metabolites extraction and analysis

Sample preparation
The sample was added with 200 μL of water, swirled for 30 s, and thawed three times with liquid nitrogen.
They were then treated with ultrasound in an ice bath for 10 minutes.Homogenate 50 μL for protein quantification.Homogenate 150 μL and add 600 μL(-40℃) of pre-cooled extract containing isotopically labeled internal standard mixture (methanol: acetonitrile = 1:1).Swirl for 30 s, ultrasound in ice water bath for 10 min, incubate at -40℃ for 1 h, centrifuge at 12000 rpm at 4℃ for 15 min.The resulting supernatant was transferred to a fresh glass vial for LC/MS analysis.Quality control (QC) samples are prepared by mixing the supernatant of all samples in equal parts.

LC-MS/MS analysis
The Vanquish (Thermo Fisher Scientific) ultra-high performance liquid chromatograph was used in this project.The target compounds were separated by Waters ACQUITY UPLC BEH Amide (2.1 mm x 100 mm, 1.7μm) liquid chromatography column.Phase A was aqueous, containing 25 mmol/L ammonium acetate and 25 mmol/L ammonia hydroxide in water (pH=9.75),and phase B was acetonitrile.The auto-sampler temperature was 4 ℃, and the injection volume was 2μL.The Thermo Q Exactive HFX mass spectrometer is capable of primary and secondary mass spectrometry data acquisition under the control of the control software (Xcalibur, Thermo).Detailed parameters are as follows: sheath gas flow rate as 30 Arb, Aux gas flow rate as 25 Arb, capillary temperature 350 ℃, full MS resolution as 120000, MS/MS resolution as 7500, collision energy as 10/30/60 in NCE mode, spray Voltage as 3.6 kV (positive) or -3.2 kV (negative), respectively.

Data processing and analysis
After the original data is converted into mzXML format by ProteoWizard software, the selfcompiled R program package (kernel is XCMS) is used for peak identification, extraction, alignment and integration, and then matched with BiotreeDB (V2.1) self-built secondary mass spectrometry database for material annotation.The cutoff value for the algorithm score is set to 0.3.

Method S2
Targeted metabolites extraction and analysis

Sample preparation
The mass spectrometric pure methanol, acetonitrile and formic acid used in this experiment were purchased from Thermo-Fisher Scientific (FairLawn, NJ, USA).The target substance standard and AQC (6aminoquinoline-n-hydroxysuccinimate) were purchased from Sigma Corporation (St. Louis, MO, USA).The standard product is dissolved in 50% methanol and prepared into a mother liquor with a concentration of 1.0mg/ml, and then diluted into a series of concentration standard samples to obtain the standard curve.
AQC derivative reagent (1.5mg/ml): Accurately weigh 1.5mg AQC powder into a clean and dry glass vial, add 1000uL dehydrated acetonitrile, and fully dissolve at 55℃ (not more than 15min).To reduce degradation, the sample is thawed under an ice bath.Each sample was added with 100μL frozen methanol solution containing inner target, and then centrifuged at 4 °C at a speed of 14000g for 20min after ultrasonic breaking.
The 50μL supernatant was transferred to the 96-well plate and affixed with a plastic sealing film for the detection of leaf acids.For amino acids, another 10μL supernatant was taken, and then 70μL boric acid buffer and 20μL AQC derivative reagent were added in turn.After the reaction at 55℃ for 10 minutes, 900μL deionized water was added to dilute it.After mixing, take 100μL supernatant and dilute it 10 times before testing on the machine.

Data processing and analysis
MassLynx software (v4.1, Waters, Milford, MA, USA) was used to process the raw data file generated by UPLC-MS/MS to integrate, calibrate, and quantify the peaks of each metabolite.The self-developed iMAP software (V1.0,MetaboProfile, Shanghai, China) was used for subsequent statistical analysis, such as PCA, OPLS-DA and unidimensional statistical test.

Figure S2 Figure
Figure S2

Figure S4 Figure
Figure S4

Figure S5 Figure
Figure S5

Figure S6 Figure
Figure S6

Figure S7 Figure
Figure S7
). E) KEGG pathway classification of differentially expressed genes compared with the control group after knockdown of SHMT2, MTHFD2, or ALDH1L2.