1. IntroductionRadiation therapy (RT) is one of the most effective treatments for primary and secondary brain tumors in adult and pediatric patients. However, cranial irradiation induces cognitive decline and intellectual dysfunction, such as impaired learning and memory. The adverse effects are more pronounced in children, especially when the temporal lobe, where the hippocampus is located, is irradiated [1,2,3,4,5]. Due to the widespread application of RT treatment, the quality of life of an expanding number of long-term survivors is garnering increasing concern.Neural stem cells (NSCs) in the hippocampus are capable of self-renewal and differentiation into neurons, astrocytes, and oligodendrocytes [6,7]. Contrary to mature neurons, which are considered to be in an irreversible state of growth arrest, the rapidly dividing and undifferentiated NSCs are more susceptible to irradiation. Several studies have indicated that irradiation of the hippocampus induced apoptosis in the subgranular zone of the Dentate gyrus (DG) [8], diminished the proliferation of the surviving NSCs [9], and impeded the differentiation of NSCs into neurons [10]. These irradiation-induced alterations which inhibit neurogenesis have been implicated in cognitive impairment [11,12,13], and elucidating the mechanisms underlying damage to NSCs could enable the discovery of strategies to optimize cognitive brain function and lessen RT-induced adverse effects.Similarly, to various other cellular stress factors, ionizing radiation damages DNA strands by disrupting their sugar-phosphate backbone and induces overall cellular toxicity, thereby driving cells towards apoptosis, necrosis, autophagy, or senescence [14,15,16,17,18,19,20,21,22,23]. Another consequential effect of irradiation on cellular macromolecules is the generation of reactive oxygen species (ROS) and reactive nitrogen species (RNS), which are predominant sources of damage to normal tissue [24,25]. There are reports that the reduction and oxidation (redox) systems play a critical role in acute radiation syndrome and are responsible for several early and late-stage side effects [26,27]. To date, published studies have elucidated the influence and functions of free radicals in radiation-induced pressure, as well as the association between redox and mitochondrial functions. Notably, some studies applied the redox theory to discover new chemicals to enhance RT sensitivity [28,29,30,31]. However, due to the characteristics of protein modification, it is challenging for conventional omics research to profile the transcriptomic and proteomic variations in cells undergoing a redox process. The redox states of whole cellular proteins in irradiated NSCs still remain unclear.

In the present study, mouse neural stem cells were exposed to X-ray irradiation to establish the cell stress model; concurrently, fetal bovine serum (FBS) was utilized to induce differentiation. iodoTMT was employed to label-free sulfhydryl groups on cysteine residues; a proteome-wide screening was conducted, followed by a comprehensive analysis of the redox patterns. Differentially expressed proteins were identified in NSCs subjected to X-ray irradiation and induced to differentiate. From a redox-MS perspective, under sustained irradiation-induced pressure, NSCs’ natural differentiation capability could be disrupted. Furthermore, the emergence of heavily oxidized proteins in NSCs was indicative of these cells’ susceptibility to degeneration.

3. DiscussionIonizing radiation of the developing or adult brain is acknowledged as a potential cause of cognitive impairment and neurodegeneration, particularly when neural stem cells are affected [39,40,41,42]. Elucidating the mechanism underlying irradiation-mediated NSC injury would contribute to alleviating the side effects of radiotherapy and demystifying the induction of neural inflammation, brain development, and even neurodegeneration-associated mechanisms [43,44,45]. The substitution models employed in preclinical radiation research vary from cultured cells to small or large animals [46,47]; the majority of these models have been established according to the linear quadratic (LQ) model [48]. Currently, the emergence of 3D tissue models and organoids has been beneficial in understanding radiation-induced tissue response and in precision medicine [49,50].With the advent of sequencing technology, neural stem cells have been investigated from a system-wide perspective, including transcriptomics, proteomics, and metabolomics, shedding new light on their complex regulatory mechanism [51,52,53,54]. Taking into consideration the properties and limitations of radiobiology models, as well as the complexity of the neural stem cell microenvironment in the brain, we sought to determine how NSCs respond to X-ray irradiation stress in the absence of cellular interactions. Therefore, in the present study, we designed a neural stem cell in vitro radiation model and integrated expression and redox proteomic techniques to analyze global protein expression in differentiated neural stem cells following X-ray irradiation. The proteomic expression profile demonstrated that irradiation impaired NSC proliferation, the cell cycle, and differentiation; in particular, the oxidation of those upregulated proteins posed an extremely high risk of neurodegeneration.Proliferating neural stem cells or progenitor cells are tremendously sensitive to ionizing radiation-induced DNA damage and apoptosis [55,56]. This phenomenon was also reflected in our BrdU assay. When DNA damage is induced, the replication checkpoint initiates the DNA repair response and delays the cell cycle progress. In neural stem cells, the cell cycle is also associated with cell differentiation: prolonged G1 and upregulated p57 enable cells to respond to signals rapidly and differentiate properly [56,57,58,59]. The manipulation of the G1 phase by CDKs could regulate the NSCs’ fate, proliferation, or differentiation [60,61,62]. After irradiation, the G1 phase was shortened in differentiating NSCs, suggesting that irradiation disrupted the conditions for normal NSC differentiation. Nestin, an intermediate filament protein, is universally considered a marker of neural stem/progenitor cells [63]. Upregulated nestin expression was detected in stem/progenitor cells during the early development stage in which cells are engaged in active proliferation. Once these cells ceased dividing and initiated differentiation, nestin expression became downregulated [64]. Nestin expression is representative of NSCs’ pluripotential. It has been reported that irradiation significantly reduced the nestin-positive cells in the mouse brain’s dentate gyrus [65]. When co-cultured with irradiated vascular endothelial cells, nestin-positive NSCs exhibited a marked decline [66]. In this study, nestin mRNA expression was similarly downregulated when NSCs were subjected to X-ray irradiation, indicating a deleterious effect of irradiation on NSCs’ stemness. Under pathological conditions, nestin should be re-expressed for the repair process to be initiated [67]. However, it is difficult to determine whether those nestin-deficient NSCs are capable of completing the repair task in irradiation-induced brain injury.The predominant cytotoxic effects of irradiation are DNA damage and cell cycle arrest [68,69]; another adverse effect of irradiation that could cause cognitive impairment is reduced neurogenesis. Irradiation induces apoptosis in dividing cells, reduces the pool of mitotic NSCs, hampers the generation of new neurons [3], affects the microenvironment of the targeted brain tissue site, and alters the NSC niche [70,71]. The expression of neurogenesis-related proteins in our proteomic datasets also reflected the detrimental effects of irradiation. Cyclin-dependent kinase 5 regulatory subunit-associated protein2 (CDK5RAP2) has been implicated in the proliferation of neuronal progenitors in the developing neocortex [72] and was also shown to cause Seckel syndrome [73]. Justin Miron et al. reported that CDK5RAP2 was prevalent in the hippocampus of brains that develop Alzheimer’s disease (AD). Notably, we also detected increased CDK5RAP2 expression in irradiated NSCs. Similar characteristics seem to occur for other neurological disease-related genes. Appb1, which interacts with amyloid precursor protein in Alzheimer’s disease, was downregulated in irradiated NSCs. Appb1 deletion was discovered to increase the risk of AD [74]. Likewise, Appb1 knockout in mice resulted in impaired learning and memory [75]. SOD1, a superoxide scavenger, is frequently upregulated during redox reactions [76]. We found that SOD1 was upregulated in irradiated NSCs, and chiefly attributed this to the IR-induced ROS. SOD1 was also upregulated in amyotrophic lateral sclerosis (ALS) patients [77], which indicates potential connections between irradiation and neurodegenerative disorders. Nrcam, a cell adhesion molecule, has been associated with autism spectrum disorders (ASD) [78]. Nrcam-knockout mice demonstrated autism-related behaviors, such as impaired sociability, cognitive rigidity, and repetitive behavior [79]. In the present study, Nrcam expression was also decreased in irradiated NSCs. Numerous neurogenesis-associated proteins altered by IR could not all be listed here. Nevertheless, IR’s impact on NSCs is considerably more complex than appreciated, especially the potential risk for neurodegeneration.Beta tubulin III, also known as Tuj-1, a class III member of the beta tubulin protein family, is regarded as a neuron-specific marker to detect progenitor cell differentiation. Consistent with Hyeon Soo Eom et al.’s study [80], we observed upregulated Tuj-1 in irradiated NSCs. MAP2, another neuron marker, was upregulated in our MS/MS detection; however, Anggraeini Puspitasari et al. demonstrated that MAP2 expression was upregulated during the early stage of irradiation (4 days) and progressively diminished in the subsequent 20 days [81]. Recent research suggested that the two markers belong to two distinct types of neurons: Tuj1 are from pan-neurons, and MAP2 are from mature neurons [82]. The inconsistency between results for Tuj-1 and MAP2 expressions indicated that during NSCs differentiation, IR’s effects on neurons might vary depending on cell types; nonetheless, the specific mechanisms warrant further investigation.Reactive oxygen species (ROS), a group of aerobic respiration metabolic byproducts, are responsible for cellular redox homeostasis. During exposure to ionizing radiation, abundant quantities of ROS and reactive nitrogen species (RNS) are generated by extracellular water radiolysis and mitochondrial membrane destruction [83,84]. ROS and RNS are the principal sources of oxidative damage to normal tissues. Concurrently, excessive ROS or RNS causes the oxidation of lipids, DNA, and proteins [85,86,87]. The oxidation of protein cysteine by ROS or RNS has been recognized as a prominent class of protein posttranslational modifications, which are heavily associated with aging and multiple diseases [88,89,90]. Two kinds of protein oxidative modifications exist irreversible oxidation and reversible oxidation. Irreversible oxidation results in protein dysfunction. In comparison, reversible oxidation, primarily of cysteine residues, could regulate the activity, the redox balance, and signaling cascades [91,92]. In the present study, we utilized cysteine-reactive tandem mass tags (iodo TMT) to detect reversible oxidation. The LC-MS/MS data could provide a proteome-wide protein oxidation profile beneficial for the analysis of the adverse effects of IR-induced oxidative stress.Ionizing radiation significantly elevated the protein oxidation level in differentiating NSCs. The proteins with a dual increase in expression and oxidation levels, especially Sdha, Atp5a1, and Ndufab1, have been documented in studies of neurodegenerative diseases [93,94,95,96,97]. Nevertheless, the oxidation of these disease-marker proteins received scant attention. IR-induced proteome-wide protein oxidation could be associated with an increased risk of neurodegeneration, whereas limiting the oxidation of certain risk proteins would provide an auxiliary strategy for alleviating radiotherapy-induced brain injury.In recent years, the majority of patients worldwide have turned toward photon therapy, and the utilization of charged particle therapies, including proton and carbon ion therapy, has substantially expanded [98]. Particle therapy treatment could substantially diminish the exposure of healthy tissue to radiation and long-term side effects [99,100], particularly among pediatric patients, in whom exposure of healthy organs to radiation doses can induce long-term detrimental effects [99]. We also have been conducting a collaborative Boron neutron capture therapy (BNCT) research project with the institute of high energy physics of the Chinese Academic of Sciences (CAS). Referring to economic considerations and indications such as meningiomas, ionizing radiation still has clinical utility. Investigations of radiation-induced injury could enable a deeper understanding of our coping mechanism when subjected to stressful radioactive rays and the progression of senescence. It is anticipated that the survival rates of cancer patients will continuously improve due to the constant evolution of modern radiotherapy. 4. Materials and Methods 4.1. Cells and X-ray Irradiation

GFP-transfected C57BL/6 mouse neural stem cells (NSCs), derived from 12.5 dpc embryos, were purchased from Cyagen Biosciences (MUBNF-01101, Guangzhou, China). The NSCs were maintained in a humidified incubator with 5% CO2 at 37 °C in Cyagen recommended medium (OriCellTM Neural Stem Cell Growth Medium, MUCMX-90011). The medium was changed every 2 days. Oricell Neural stem cell growth medium was replaced by 10% fetal bovine serum (FBS)/DMEM-F12K (Gibco) for differentiation. For X-ray irradiation (IR) treatment, the cells were irradiated at 1 Gy or 5 Gy with an Xstrahl X-ray system, Model CIX2 (Xstrahl, Walsall, West Midlands, UK). The follow-up procedures are described in subsequent sections.

4.2. qRT-PCR AnalysisThe experiment was conducted for six groups, namely: ctrl, differentiation group (NSCs treated with FBS), irradiation group (cells exposed to X-ray,1 Gy or 5 Gy), differentiation after IR group (after 1 Gy or 5 Gy irradiation, the culture medium was immediately changed to FBS/DMEM-F12K). Total RNA extraction was performed at 24 h post-X-ray irradiation or cell differentiation using TRIzol Plus RNA kit (Invitrogen, Carlsbad, CA, USA). cDNA was prepared using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA). The RT-PCR reaction was performed using Universal SYBR Green Supermix (Bio-Rad, Hercules, CA, USA) according to the manufacturer’s instructions. Information on the primers is listed in Supplementary Table S1. The statistical analysis was performed using GraphPad Prism 8.0 Software. The results were presented as the Mean ± standard error of the mean. Student’s t-test was used to compare values between the two groups. Differences were considered statistically significant when p values were 4.3. Cell Cycle Analysis

The experimental groups and study design were consistent with the statements mentioned above. 24 h after each specific treatment, cells were collected via trypsinization. Furthermore, supernatants and PBS used during wash steps were kept ensuring the collection of both adherent and detached cells. After collection, the cells were fixed in ice-cold 70% ethanol at 4 °C overnight. Subsequently, the cells were stained with PI solution (50 μL PI and 50 μL RNase A in 10 mL PBS) for 30 min at room temperature before measurement. The data were obtained using a flow cytometer (Beckman Coulter, Brea, CA, USA) and analyzed using the ModFitLT software (Version 5.0; Verity Software House, Topsham, ME, USA).

4.4. BrdU Assay

NSCs were seeded in 96-well plates and subjected to specific stimulation (mentioned above). 24 h after treatment, the NSCs’ proliferation in each well was evaluated using a Cell Proliferation ELISA BrdU Kit (Roche, Mannheim, Germany) according to the manufacturer’s protocol. The absorbance, which represents BrdU incorporation during DNA synthesis, was measured at 450 nm using a microplate spectrophotometer (Thermo, Swedesboro, NJ, USA)

4.5. Immunofluorescence StainingNeurospheres were trypsin-digested into a single-cell suspension and cultured on 0.01% poly-L-lysine (Sigma-Aldrich, St. Louis, MO, USA) pre-coated coverslips in a 24-well plate. The cells were induced to differentiate following 0 Gy, 1 Gy, or 5 Gy irradiation. After 5 days of differentiation, the cells were fixed with 4% Paraformaldehyde (PFA), followed by PBS washing thrice and blocking for 1 h with 0.5% bovine serum albumin (BSA) and 0.1%Triton X-100. The blocking solution was also used for antibody dilution: Rabbit anti-GFAP (1:1000, Abcam, Cambridge Biomedical Campus, Cambridge, UK), Mouse anti-O4 (1:1000, R&D systems, Minneapolis, MN, USA), and Mouse anti-beta 3 tubulin (1:1000, Sigma), and the primary antibodies were incubated at 4 °C overnight. After several washes with TBS, the corresponding secondary antibodies were added for 2 h at room temperature. The utilized secondary antibodies are as follows: Donkey anti-mouse IgM Alexa 555 and Donkey anti-rabbit IgM Alexa 633 (Thermo, Waltham, MA, USA). The cell climbing slices were mounted on glass slides with an antifade reagent mounting medium (BOSTER Biological Tech, Wuhan, China). All the stained fluorescent markers were captured using an LSM 700 laser scanning confocal microscope (Axio-observer Z1; Carl Zeiss, Oberkochen, Germany) and analyzed using the software ZEN lite (Zeiss, https://www.zeiss.com/microscopy/en/products/software/zeiss-zen-lite.html/, accessed on 18 March 2020) 4.6. Protein and LC MS/MS and TMT Label

Protein sample preparation. 36 h after corresponding treatments, all NSC samples were lysed in RIPA buffer with PMSF (Abcam), then centrifuged at 12,000× g for 10 min at 4 °C; the supernatants containing total proteins were collected. The protein concentration per sample was determined using Pierce BCA Protein Assay Kit (Thermo scientific, Rockford, IL, USA) according to the manufacturer’s protocol. Aliquots of 50 μg proteins were used for proteomics analysis. Proteins’ disulfide bonds were reduced with 10 mM Dithiothreitol for 45 min at 55 °C, then alkylated with 25 mM iodoacetic acid for 30 min in the dark, followed by overnight acetone precipitation. The obtained precipitants were dissolved in EPPS (Thermo Fisher Scientific, Rockford, IL, USA) and re-dissociated with Trypsin overnight at 37 °C. Peptide and the sulfhydryls of cysteine-containing peptides labeling were performed using TMT10-plex and iodoTMT Mass Tag Labelling Kit (Thermo Fisher Scientific, Rockford, IL, USA) following the manufacturer’s protocol. The labeled samples were acidified with trifluoroacetic acid, followed by a desalination procedure using a C18 Sep-pak column, and then vacuum dried.

LC-MS/MS analysis. The peptide samples were dissolved in 0.1% formic acid, then preconcentrated and desalted using PepMap C18 nanotrap column (Thermo Fisher Scientific, Rockford, IL, USA) A reversed-phase analytical column (EASY-Spray C18, Thermo Fisher Scientific, Rockford, IL, USA) was utilized for peptide separation in a binary solvent system. Gradient conditions were: 4–26% solvent B for 120 min and 26–95% B for 10 min. The peptides were analyzed using a data-dependent acquisition method at a resolution of 120,000, a scan range of 375–1500 m/z, and at a resolution of 60,000 with a target value of 2 × 105 ions and a maximum injection time of 120 ms. The fixed first m/z was 100, and the isolation window was 1.2 m/z units. The raw data files were processed using the Andromeda search engine in MaxQuant 1.5.6.5 software (https://www.maxquant.org/, accessed on 11 November 2019) 4.7. Bioinformatic AnalysisAll statistics of protein expression data were computed using Excel software (Microsoft Excel, version 2013), and the differentially expressed (DE) proteins were screened via the t-test (p |0.5|); related expression volcano plots were generated using GraphPad Prism V 7.0. The clustered heatmap profile of protein expression among each group was conducted using the “pheatmap” package (version 1.0.12, https://cran.rstudio.com/web/packages/pheatmap/index.html/, accessed on 15 February 2022) in R.The principal component analysis of protein expression patterns among groups was performed using the “FactoMineR” package (version 2.4, https://cran.r-project.org/web/packages/FactoMineR/index.html/, accessed on 15 February 2022) in R, and the output data were plotted using GraphPad Prism. For correlation analysis, the normalized protein expression values of particular experimental groups were transformed on a Log2 scale, then analyzed and visualized with GraphPad Prism.