Abstract

   Glioma is a highly aggressive brain tumor with limited treatment
   success due to its resistance to conventional therapies. Sirtuin 5
   (SIRT5) has emerged as a promising target for cancer therapy, though it
   exhibits dual roles in different cancer types. In this study, we
   investigate the role of SIRT5 in glioma and its corresponding
   mechanisms. Our findings demonstrate that SIRT5 expression is elevated
   in glioma cells both in vitro and in vivo. SIRT5 knockdown
   significantly reduced glioma cell proliferation and enhanced
   sensitivity to ferroptosis. Proteomic and metabolomic analyses
   identifies branched-chain amino acid (BCAA) metabolism as a key
   downstream pathway regulated by SIRT5 through branched-chain
   aminotransferase 1 (BCAT1). Specifically, SIRT5-mediated
   desuccinylation of BCAT1 at K39 inhibits its interaction with the E3
   ligase CHIP, thereby preventing BCAT1 degradation via the
   ubiquitin-proteasome system. Moreover, BCAT1 overexpression reverses
   the proliferation inhibition and ferroptosis sensitivity observed in
   SIRT5-knockdown cells. Clinically, we reveal a positive correlation
   between SIRT5 and BCAT1 levels in glioma samples, with higher
   expression levels predicting more advanced glioma grades and poorer
   clinical outcomes. Collectively, this study highlights the critical
   role of SIRT5 in promoting glioma progression via metabolic regulation
   and ferroptosis insensitivity, offering a potential therapeutic target
   for glioma treatment.

   graphic file with name 41419_2025_7626_Figa_HTML.jpg

   Subject terms: CNS cancer, Ubiquitylated proteins, Acetyltransferases,
   Cell death

Introduction

   Gliomas, particularly glioblastomas (GBM), are among the most
   aggressive and lethal forms of primary brain tumors. Despite
   advancements in surgical methods, radiotherapy, and chemotherapy,
   patient outcomes remain poor, with frequent recurrences and a median
   survival of less than 15 months [[36]1]. This dismal outcome is largely
   attributed to the invasive nature of glioma cells, their resistance to
   conventional therapies, and their ability to avoid programmed cell
   death mechanisms, including ferroptosis [[37]2]. Recent research has
   increasingly concentrated on the metabolic alterations of glioma cells,
   particularly in glutamine metabolism and lipid metabolism, highlighting
   its significant role in cancer progression and chemoresistance [[38]3].
   However, the molecular basis of these processes, especially the
   regulation of metabolic pathways and their link to therapeutic
   resistance, is still not well understood.

   One promising research avenue involves sirtuins, a family of
   NAD^+-dependent deacylases that are involved in various biological
   processes, including aging, metabolism and stress response [[39]4]. Of
   the seven members in the sirtuin family (SIRT1–SIRT7), SIRT5 has
   recently garnered attention for its role in controlling metabolic
   pathways [[40]5], mitochondrial function [[41]6], and oxidative stress
   [[42]7]. Unlike other Sirtuins primarily noted for their deacetylase
   activity, SIRT5 exhibits unique desuccinylase activity, highlighting
   its distinct role in post-translational modifications [[43]8]. In
   recent years, growing evidence has linked SIRT5 to cancer, suggesting
   that its function can either promote or suppress tumors, depending on
   the specific tissue and context [[44]9]. For instance, SIRT5
   desuccinylates S100 calcium-binding protein A10, promoting its
   degradation and thereby suppressing gastric cancer metastasis [[45]10].
   Lower SIRT5 expression has been observed in pancreatic ductal
   adenocarcinoma, where SIRT5 deacetylates glutamate oxaloacetate
   transaminase 1, hindering glutamine, glutathione, and pyrimidine
   metabolism, ultimately suppressing tumor growth [[46]11]. However, in
   colorectal cancer, SIRT5 promotes cell proliferation and maintains
   redox balance by desuccinylating the K280 residue of the mitochondrial
   enzyme serine hydroxymethyltransferase [[47]12]. Additionally, in
   breast cancer, SIRT5 has been shown to inhibit autophagy and mitophagy
   via regulating glutamine metabolism, which aids tumor cells in
   withstanding chemotherapy [[48]13]. Despite these advances, the
   specific role of SIRT5 in glioma, particularly in relation to
   therapeutic resistance and metabolic regulation, has not been fully
   elucidated, representing a critical gap in the current literature.

   Ferroptosis, characterized by the accumulation of lipid peroxides, is
   an iron-dependent form of cell death, distinguishing it from necrosis
   and apoptosis [[49]14]. Iron catalyzes reactive oxygen species (ROS)
   production through the Fenton reaction, triggering lipid peroxidation
   of polyunsaturated fatty acids, leading to membrane damage and cellular
   disintegration [[50]15]. GPX4, supported by GSH, is pivotal in
   counteracting ferroptosis by neutralizing lipid peroxides. GSH is
   synthesized from glutamate, which can be derived from glutamine and
   α-ketoglutarate [[51]16]. Ferroptosis-inducing agents like Erastin and
   RSL3 either inhibit cystine import or deactivate GPX4 directly,
   resulting in GSH depletion and subsequent cell death through lipid
   peroxide buildup [[52]17]. Previous research connects ferroptosis
   resistance with glioma development and poor prognosis [[53]18, [54]19].
   Sulfasalazine (SAS), an FDA-approved anti-rheumatic drug, has been
   explored for its antitumor effect in glioma through ferroptosis
   induction [[55]20, [56]21]. Nonetheless, elevated GPX4 and SLC7A11
   levels, along with reduced ferroptosis markers, such as 12-HETE and
   15-HETE, have been observed in glioma tissues and cell lines [[57]18],
   suggesting glioma’s relative insensitivity to ferroptosis.

   In this study, we investigate the expression, clinical relevance, and
   underlying mechanisms of SIRT5 in glioma. Our findings reveal high
   SIRT5 expression in glioma both in vitro and in vivo. SIRT5 knockdown
   suppresses glioma cell proliferation, confirming its role as oncogene.
   Furthermore, SIRT5 is shown to confer resistance to SAS-induced
   ferroptosis via the GSH/GPX4 axis. Additionally, we identify BCAT1 as a
   novel substrate for SIRT5-mediated desuccinylation. SIRT5 directly
   binds to BCAT1 and desuccinylates it at K39, which prevents
   CHIP-mediated ubiquitination and subsequent degradation of BCAT1.
   Overexpression of BCAT1 rescues the proliferation and ferroptosis
   resistance deficits caused by SIRT5 knockdown. Additionally, the
   expression of SIRT5 and BCAT1 is positively correlated with advanced
   glioma grade and unfavorable prognosis. Taken together, these findings
   suggest that SIRT5 plays a crucial oncogenic role in glioma and
   positions SIRT5 as a potential therapeutic target for ferroptosis
   induction to block tumor growth.

Materials and Methods

Cell culture

   Human malignant glioblastoma multiforme U251 and U87 cell lines were
   grown in high-glucose Dulbecco’s Modified Eagle Medium (DMEM, SolarBio,
   #11995) containing 4 mM l-glutamine, and supplemented with 10% fetal
   bovine serum (FBS). Additionally, human microglial HMC3 cells were
   cultured in low-glucose DMEM supplemented with 10% FBS. The cells were
   maintained in an environment with 5% CO[2] at 37 °C and a controlled
   humidity level. These cell lines were purchased from the National
   Collection of Authenticated Cell Cultures of the Chinese Academy of
   Sciences in Shanghai, China. Authentication of the cell lines was
   confirmed through short tandem repeat (STR) analysis performed by
   Biowing Applied Biotechnology (Shanghai, China). Furthermore, all cell
   lines were verified to be free of mycoplasma contamination.

Cell transfection

   SIRT5 siRNAs (Origene, #SR323578) were transfected into HMC3 and glioma
   cells using siTran 2.0 siRNA transfection reagent (Origene, #TT320001)
   according to the manufacturer’s instructions. Plasmids were transfected
   using MegaTran 2.0 plasmid DNA transfection reagent (Origene,
   #TT210003), respectively.

Cell proliferation assay

   In the Crystal violet assay, cells were initially seeded into a 24-well
   plate (NEST Biotechnology, Wuxi, China; #702001) and subjected to
   transfection with the specified plasmids or siRNAs, or treated with
   various concentrations of MC3482 (MedChemExpress, #HY-112587).
   Afterward, fixation was carried out using 4% paraformaldehyde, followed
   by staining with 0.1% crystal violet solution. To assess absorbance,
   the stain was dissolved in 10% acetic acid, and measurements were taken
   at a wavelength of 595 nm.

   For the colony formation assay, cells were seeded into 12-well plates
   (NEST Biotechnology, #712001) and subjected to transfection with the
   specified plasmids or siRNAs. Post-transfection, cells were incubated
   for a period of 12–14 days, fixed with 4% paraformaldehyde, and stained
   with 0.1% crystal violet solution. Photographs of the stained wells
   were taken, and colony numbers were quantified.

   In the EdU staining assay, cells were cultured on glass coverslips
   placed in 12-well plates. EdU staining was carried out using the EdU
   Imaging Kit (APExBio, Shanghai, China; #K1075) following the
   manufacturer’s guidelines. The stained cells were visualized with an
   inverted fluorescence microscope (Nikon Eclipse Ti2-U, Japan).

Cell Counting Kit-8 (CCK8) assay

   Cells were plated in 96-well plates (NEST Biotechnology, #701002) at a
   density of 5 × 10³ cells per well. Following the specified treatment,
   10 µL of the Cell Counting Kit-8 reagent (AbMole BioScience, USA) was
   added to each well, and the plate was incubated at 37 °C. Absorbance at
   450 nm was measured using a microplate reader (PerkinElmer EnSpire,
   Shanghai, China).

Proteomic and metabolomic analysis

   Proteomic analysis was carried out using tandem mass tag (TMT) mass
   spectrometry, provided by Wuhan IGeneBook Biotechnology Co., Ltd
   (Wuhan, China). The raw mass spectrometry data from each sample were
   processed using the MASCOT engine (Matrix Science, London, UK; version
   2.2) within Proteome Discoverer 1.4 software for both identification
   and quantification of proteins. Differentially expressed protein
   sequences were further analyzed using the InterProScan and NCBI BLAST+
   client software to identify homologous sequences. Gene ontology (GO)
   terms were assigned, and sequences were annotated using the Blast2GO
   software. Following this, the studied proteins were mapped to pathways
   by referencing the Kyoto Encyclopedia of Genes and Genomes (KEGG)
   database using KEGG orthology identifiers.

   For metabolomic analysis, the samples were examined using an
   LC-ESI-MS/MS system (MS QTRAP® System from SCIEX and UPLC ExionLC AD),
   conducted by Wuhan IGeneBook Biotechnology Co., LTD The significant
   metabolites were identified by considering both a VIP score (VIP > 1)
   and a P value of less than 0.05 (Student’s t-test). VIP values, along
   with score and permutation plots, were extracted from the OPLS-DA
   results generated using the R package MetaboAnalystR. Data were
   log-transformed (log2) and mean-centered prior to OPLS-DA. A
   permutation test of 200 iterations was conducted to prevent
   overfitting. Identified metabolites were annotated by referencing the
   KEGG Compound database.

Immunoprecipitation (IP), GST-pull down, and western blot assay

   In the GST-pull down assay, recombinant SIRT5 protein (MedChemExpress,
   Shanghai China; #HY-[58]P74545) was incubated with GSTSep Glutathione
   Magbeads (Yeasen Biotechnology, #20562ES03) along with GST-BCAT1
   (Proteintech, #Ag4574) or GST (MedChemExpress, #HY-[59]P70270) protein
   for 2 h at 4 °C. After incubation, the beads were washed three times
   with PBS and then boiled in 1×loading buffer for 10 min at 100 °C.

   For the IP assay, cells were lysed in NP-40 lysis buffer, and the
   lysate was centrifuged at 12,000 × g for 20 min at 4 °C. The
   co-immunoprecipitation mixture, containing 30 μL of protein A/G agarose
   beads and 1 μg antibody, was incubated on a Rotator Shaker for 8 h at
   4 °C. After incubation, the beads were washed three times with lysis
   buffer and boiled in 1×loading buffer for 10 min at 100 °C.

   For Western blot analysis, proteins were separated using sodium dodecyl
   sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then
   transferred to a polyvinylidene difluoride (PVDF) membrane. Membranes
   were blocked with 5% fat-free milk and incubated overnight at 4 °C with
   the following primary antibodies: SIRT5 (Proteintech, Wuhan, China;
   #15122-1-AP), BCAT1 (Proteintech, #13640-1-AP), HA (Proteintech,
   #51064-2-AP), Flag (Proteintech, #66008-4-lg), His (Proteintech,
   #66005-1-Ig), ubiquitin (Proteintech, #10201-2-AP), GPX4 (Cell
   Signaling Technology, #52455S), FTH1 (Proteintech, #83428-1-RR), and
   GAPDH (Proteintech, #10494-1-AP). The membranes were washed three times
   with Tris-buffered saline containing 0.1% Tween-20 before being
   incubated with HRP-conjugated secondary antibodies. Protein bands were
   visualized using a chemiluminescence detection kit, and images were
   captured with a Bio-Rad ChemiDoc MP digital gel image analysis system.

Quantitative reverse transcription polymerase chain reaction (qPCR)

   Total RNA was isolated using TRIzol reagent and subsequently reverse
   transcribed into complementary DNA (cDNA) using the M5 Sprint qPCR RT
   kit with gDNA remover (Mei5bio, Beijing, China; #MF949-01).
   Quantitative real-time PCR was carried out using M5 Hiper Realtime
   Super Mix with High Rox (Mei5bio, #MF013-505). The sequences for the
   primers used in the analysis are provided in Table [60]S1.

Immunohistochemistry (IHC) staining

   Tumor microarrays were purchased from Shanghai Outdo Biotech Co., LTD.
   Sections of brain tissues (frontal lobe) were bought from Shanghai
   WeiaoBio Co., LTD. IHC staining of tumor microarrays, brain tissues,
   and xenografted tumors was carried out by Wuhan ServiceBio Technology
   Co., Ltd.

Immunofluorescence assay

   To detect the subcellular localization of SIRT5 and BCAT1, cells were
   fixed with 4% paraformaldehyde for at room temperature 20 min and
   permeabilized with phosphate-buffered saline (PBS) containing 0.5%
   Triton-X. Cells were then blocked with 5% goat serum and incubated at
   4 °C overnight with the primary antibodies. On the Next day, they were
   washed three times with PBS containing 0.5% Tween-20, followed by
   incubation with fluorescent secondary antibodies (Proteintech,
   #RGAM002, #RGAR004). Nucleus was stained with
   4′,6-diamidino-2-phenylindole (DAPI). Images were captured using an
   inverted fluorescence microscope (Nikon Eclipse Ti2-U).

   To assess reactive oxygen species (ROS) levels, cells were incubated
   with 5 μM Dihydroethidium (DHE; MedChemExpress, #HY-D0079) for 30 min
   in the dark at 37 °C, then washed twice with PBS. To visualize oxidized
   lipids, cells were incubated with 10 μM BODIPY 581/591
   C11(MedChemExpress, #HY-D1301) for 30 min in the dark at room
   temperature, then washed twice with PBS. Fluorescent images were taken
   using an inverted fluorescence microscope (Nikon Eclipse Ti2-U), and
   the fluorescence intensity was quantified through flow cytometry (BD
   FACS Canto® II).

Measurement of intracellular glutathione (GSH), glutamate, lipid peroxidation
(LPO), and malondialdehyde (MDA) levels

   The levels of intracellular GSH, glutamate, LPO, and MDA were
   quantified utilizing the Reduced Glutathione Content Assay Kit
   (Solarbio, Beijing, China; #BC1175), Glutamic Acid (Glu) Colorimetric
   Assay Kit (Elabscience, #E-BC-K903-M), the Lipid Peroxide Content Assay
   Kit (Solarbio, #BC5245), and the Malondialdehyde Content Assay Kit
   (Solarbio, #BC0025), following the protocols outlined by the
   manufacturer.

Stable cell line construction

   The full-length BCAT1 sequence was inserted into pLV-EF1α-Bsd plasmid.
   HEK-293 cells were co-transfected with the recombinant plasmid along
   with pMD2.G and psPAX2 plasmids for lentivirus packaging. SIRT5 shRNA
   lentiviral stocks were purchased from Shanghai Genechem Co., Ltd
   (#GXDL0423531_1). U251 and U87 cells were then infected with the
   lentiviral stocks and selected using blasticidin S or puromycin,
   respectively. Primer sequences are detailed in Table [61]S1.

Xenograft model

   The study was conducted under the approval of the Ethics Committee of
   Jiangxi Provincial People’s Hospital (protocol code KT078).
   Four-week-old male NOD/SCID mice were obtained from SPF Biotechnology
   Co. Ltd (Beijing, China) and were randomly assigned to groups. Each
   group of mice was housed in a Specific Pathogen-Free (SPF) animal
   facility and subcutaneously injected with 3 × 10⁶ cells. SAS (30 mg/kg)
   was administered daily via intraperitoneal injection. At the designated
   time points, the mice were sacrificed, and the xenografted tumors were
   collected. The maximal tumor size did not exceed 150 mm. The researcher
   was blinded to the group allocations throughout the experiment.

Statistical analysis

   Statistical significance between groups was determined using either a
   Student’s t-test or one-way analysis of variance (ANOVA), followed by
   Fisher’s LSD test. All data are presented as mean ± standard deviation
   (SD). A p value of less than 0.05 was considered statistically
   significant.

Results

SIRT5 expression is upregulated in glioma and plays a significant role in
cell proliferation

   In vivo immunohistochemistry staining revealed that glioma tissues
   exhibit higher levels of SIRT5 compared to normal brain tissues (Fig.
   [62]1A). This finding was further corroborated by western blot
   analysis, which demonstrated increased SIRT5 expression in glioma cell
   lines, U251 and U87, relative to HMC3 (Fig. [63]1B). To assess whether
   SIRT5 influences cellular proliferation, crystal violet and EdU
   staining were employed. Knockdown of SIRT5 resulted in a marked
   reduction in the proliferation of U251 and U87 cells (Fig. [64]1C–F),
   whereas overexpression of SIRT5 significantly accelerated their growth
   (Fig. [65]1G–J). Moreover, the knockdown of SIRT5 impaired the colony
   formation capability of U251 cells, while its overexpression had the
   opposite outcome (Fig. [66]1K–L). Additionally, treatment with MC3482,
   a specific SIRT5 inhibitor, also inhibited the proliferation of U251
   and U87 cells (Fig. [67]S1A, B). On the Contrary, SIRT5 knockdown had
   mild inhibition on HMC3 cell proliferation, while SIRT5 overexpression
   exerted no effect, indicating that SIRT5 plays a tumor-specific role in
   glioma cells (Fig. [68]S1C, D). Collectively, these findings suggest
   that SIRT5 upregulation plays a pivotal oncogenic role in glioma
   progression.

Fig. 1. SIRT5 expression is upregulated in glioma and promotes cell
proliferation.

   [69]Fig. 1
   [70]Open in a new tab

   A SIRT5 expression in glioma tissues and normal brain tissues was
   determined by immunohistochemistry staining. Scale bar = 50 μm. B
   Western blot analysis of SIRT5 expression in HMC3, U251, and U87 cells.
   C–J Crystal violet staining and EdU staining were performed to evaluate
   the effect of SIRT5 knockdown or overexpression on the proliferative
   ability of U251 and U87 cells. Scale bar = 100 μm. K–L Colony formation
   assay was performed to evaluate the effect of SIRT5 knockdown or
   overexpression on the colony formation ability of U251 cells. Data are
   presented as mean ± SD of three independent experiments. *p < 0.01,
   **p < 0.01, ***p < 0.001.

SIRT5 knockdown enhances the sensitivity of glioma cells to ferroptosis
through GSH/GPX4 axis

   Glioma cells, due to elevated intracellular Fe²⁺ levels, exhibit
   susceptibility to ferroptosis [[71]22], making the induction of this
   form of cell death a potential therapeutic approach for glioma
   treatment, and SAS has been considered as adjuvant treatment for glioma
   by activating ferroptosis [[72]23]. Consequently, the involvement of
   SIRT5, a key component of the cellular antioxidant defense, in glioma’s
   resistance to ferroptosis was examined. In this study, the IC50 values
   of SAS were determined to be 496.5 μM and 442.2 μM in U251 and U87
   cells, respectively. Following SIRT5 knockdown, these values
   significantly dropped to 338.5 μM and 329.2 μM (Fig. [73]2A, B).
   Treatment with MC3482 also sensitized U251 and U87 cells to SAS (Fig.
   [74]S2A-B). Conversely, SIRT5 overexpression increased the IC50 of SAS
   in these cell lines (Fig. [75]2C, D). To further elucidate the
   connection between SIRT5 and ferroptosis, critical components of
   ferroptosis-related signaling pathways were analyzed. It was observed
   that SIRT5 knockdown led to a significant reduction in intracellular
   GSH levels in SAS-treated cells, concurrently increasing oxidative
   stress, whereas SIRT5 overexpression elevated GSH levels and mitigated
   oxidative stress (Fig. [76]2E–J). Given that GPX4 plays a pivotal role
   in neutralizing lipid peroxides during ferroptosis and FTH1 is
   responsible for iron storage, their expression was detected via western
   blot analysis. The results showed decreased expression of both GPX4 and
   FTH1 in SAS-treated cells, which were further downregulated following
   SIRT5 knockdown (Fig. [77]2K). Additionally, heightened levels of LPO,
   MDA, and Fe²⁺, known markers of ferroptosis, were tested. The results
   revealed that SIRT5 depletion exacerbated SAS-induced Fe²⁺
   accumulation, LPO, and MDA production, while SIRT5 overexpression
   alleviated Fe²⁺ accumulation (Fig. [78]2L–Q). Furthermore, SIRT5
   knockdown intensified SAS-induced lipid peroxidation (Fig. [79]2R).
   These findings indicate that SIRT5 facilitates ferroptosis resistance
   via the GSH/GPX4 axis.

Fig. 2. SIRT5 knockdown enhances the sensitivity of glioma cells to
ferroptosis through GSH/GPX4 axis.

   [80]Fig. 2
   [81]Open in a new tab

   A–D The IC50 values of SAS in U251 and U87 cells were determined by
   CCK8 assay following SIRT5 knockdown or SIRT5 overexpression. E–F
   Intracellular GSH content was detected in U251 and U87 cells treated
   with SAS or transfected with SIRT5 siRNA. G–H Intracellular GSH content
   was detected in U251 and U87 cells treated with SAS or transfected with
   SIRT5-Flag plasmid. I–J ROS levels in U251 cells were detected by DHE
   staining. Scale bar = 100 μm. K Western blot analysis of SLC7A11, GPX4,
   and FTH1 expression in U251 cells treated with SAS or transfected with
   SIRT5 siRNA. L–M Intracellular labile iron levels were detected by
   FerroOrange staining. Scale bar = 25 µm. N–Q Intracellular LPO and MDA
   levels were detected in U251 and U87 cells treated with SAS or
   transfected with SIRT5 siRNA. R Lipid peroxidation was detected by
   BODIPY 581/591 C11 staining. Scale bar = 50 µm. Data are presented as
   mean ± SD of three independent experiments. ns p > 0.05, *p < 0.01,
   **p < 0.01, ***p < 0.001.

SIRT5 plays a pivotal role in regulating BCAA metabolism in glioma cells

   To further explore the oncogenic mechanism of SIRT5 in glioma, SIRT5
   expression was knocked down with lentiviral shRNAs, and a stable
   SIRT5-knockdown cell line was developed using SIRT5 shRNA-3 (Fig.
   [82]S3A, B). Subsequent TMT mass spectrometry was conducted to identify
   proteins differentially expressed between wild-type U251 cells and
   U251-shSIRT5 cells. Out of the 6633 proteins identified, 112 were
   upregulated, while 173 exhibited downregulation in SIRT5-knockdown
   cells (Figs. [83]3A and [84]S4A). GO enrichment analysis revealed that
   the differentially expressed proteins were predominantly involved in
   cellular processes, metabolic processes, and biological regulation
   (Fig. [85]S4B). Furthermore, KEGG pathway enrichment analysis
   highlighted significant alterations in glycosphingolipid biosynthesis,
   oxidative phosphorylation, and valine, leucine, isoleucine degradation
   upon SIRT5 knockdown (Fig. [86]3B, C). We further performed
   metabolomics analysis to elucidate the metabolic alterations associated
   with SIRT5 knockdown, which identified 474 differentially abundant
   metabolites (Fig. [87]3D). Clustering analysis results indicate that
   most differential metabolites were classified under amino acids and
   their derivates (Fig. [88]3E). KEGG analysis further confirmed that
   ß-alanine metabolism, arginine biosynthesis and tyrosine metabolism
   were notably affected by SIRT5 knockdown. Importantly, valine, leucine,
   isoleucine degradation pathways were also enriched (Fig. [89]3F), which
   correlates with findings from TMT mass spectrometry. These results
   indicate that BCAA metabolism is likely a crucial downstream pathway
   influenced by SIRT5 in glioma cells.

Fig. 3. SIRT5 regulates BCAA metabolism in glioma cells.

   [90]Fig. 3
   [91]Open in a new tab

   A Volcano plot showing proteins differentially expressed between shNC
   and shSIRT5 U251 cells (n = 3). B, C KEGG pathway analysis of the
   differentially expressed proteins. D Volcano plot depicts differential
   metabolites between shNC and shSIRT5 U251 cells (n = 6). E Heatmap
   visualizes clustering analyses of differential metabolites. F KEGG
   pathway analysis of the differential metabolism.

SIRT5 stabilizes BCAT1 through desuccinylation at K39

   BCAT1 catalyzes the initial step of BCAA metabolism and plays a pivotal
   role in glioma progression [[92]24]. Thus, we sought to determine if
   SIRT5 modulates BCAA metabolism and exerts its oncogenic functions via
   BCAT1 regulation. To assess the interaction between SIRT5 and BCAT1,
   Flag-tagged SIRT5 and HA-tagged BCAT1 were co-expressed in HEK-293T
   cells. Co-immunoprecipitation experiments demonstrated that SIRT5-Flag
   and HA-BCAT1interacted with each other (Fig. [93]4A, B), and similar
   interaction was observed between endogenous BCAT1 and SIRT5 in U251 and
   U87 cells (Fig. [94]4C). Additionally, immunofluorescence assay
   confirmed that SIRT5 and BCAT1 co-localize in U251 cells (Fig. [95]4D),
   and GST-pull down assay further verified their direct interaction in
   vitro (Fig. [96]4E).

Fig. 4. SIRT5 interacts with and stabilizes BCAT1.

   [97]Fig. 4
   [98]Open in a new tab

   A, B The interaction between Flag-tagged SIRT5 and HA-tagged BCAT1 was
   detected by co-IP and western blot in HEK-293T cells. C The interaction
   between endogenous SIRT5 and BCAT1 in U251 and U87 cells was detected
   by co-IP and western blot. D Immunofluorescence assay of SIRT5 and
   BCAT1 cellular localization in U251 cells. Scale bar = 10 μm. E
   GST-pull down assay of the interaction between SIRT5 and BCAT1. F–G
   Western blot analysis of SIRT5 and BCAT1 expression in U251 and U87
   cells transfected with scramble siRNA or SIRT5 siRNA. H, I Western blot
   analysis of SIRT5 and BCAT1 expression in U251 and U87 cells
   transfected with vector or SIRT5-Flag plasmid. J, K Western blot
   analysis of BCAT1 expression in U251 cells treated with 50 μg/mL
   cycloheximide. Data are presented as mean ± SD of three independent
   experiments. *p < 0.01, **p < 0.01, ***p < 0.001.

   We then examined whether SIRT5 influences BCAT1 expression at the
   protein or mRNA level. Western blot analysis revealed that SIRT5
   knockdown significantly reduced BCAT1 protein levels in both U251 and
   U87 cells (Fig. [99]4F, G), while SIRT5 overexpression elevated BCAT1
   protein levels (Fig. [100]4H, I). Notably, these changes in BCAT1
   expression occurred at the protein level, with no significant
   alterations observed in BCAT1 mRNA levels. (Fig. [101]S5A–D).
   Furthermore, cycloheximide chase assays demonstrated that SIRT5
   overexpression delayed BCAT1 protein turnover, whereas SIRT5 depletion
   accelerated its degradation (Fig. [102]4J, K), suggesting that SIRT5
   stabilizes BCAT1 in glioma cells.

   Considering SIRT5’s enzymatic role as both deacetylase and
   dessuccinylase [[103]25], we assessed its effect on BCAT1 acetylation
   and succinylation. Our results indicated that BCAT1 is subjected to
   both modifications (Figs. [104]5A and [105]S6A). However, BCAT1
   succinylation was lower in U251 and U87 cells compared to HMC3 cells,
   while its acetylation levels remained consistent across the cell lines.
   (Figs. [106]5B and [107]S6B). Interestingly, SIRT5 knockdown increased
   BCAT1 succinylation (Fig. [108]5C, D), while ectopic expression of
   wild-type SIRT5, but not its enzyme-deficient mutant, lowered BCAT1
   succinylation (Fig. [109]5E, F). Moreover, neither of these
   interventions affected BCAT1 acetylation (Fig. [110]S6C, D). To
   pinpoint the key succinylation site, five candidate lysine residues
   (K26, K28, K39, K176, K305) were predicted using LMSuccSite Platform
   (Table. [111]S2), and each was mutated to arginine to mimic
   desuccinylation state. As shown in Fig. [112]5G, H, the BCAT1 K39R
   mutant exhibited reduced succinylation and a longer half-life compared
   to wild-type BCAT1, suggesting that K39 is the primary succinylation
   site on BCAT1. We then explored whether SIRT5 regulated BCAT1
   expression through desuccinylation at K39. The results revealed that
   SIRT5 overexpression had no impact on the succinylation level or
   half-life of BCAT1 K39R mutant (Fig. [113]5I, J). And SIRT5 showed a
   weaker interaction with the K39R mutant than with wild-type BCAT1 (Fig.
   [114]5K), indicating that SIRT5 stabilizes BCAT1 via K39
   desuccinylation.

Fig. 5. K39 is the primary site in SIRT5-mediated BCAT1 desuccinylation.

   [115]Fig. 5
   [116]Open in a new tab

   A IP and western blot analysis of BCAT1 succinylation levels in
   HEK-293T cells transfected with HA-BCAT1 plasmid. B IP and western blot
   analysis of endogenous BCAT1 succinylation levels in HMC3, U87, and
   U251 cells. C, D IP and western blot analysis of BCAT1 succinylation
   levels in U87 and U251 cells transfected with scramble siRNA or SIRT5
   siRNA. E, F IP and western blot analysis of BCAT1 succinylation levels
   in U87 and U251 cells transfected with vector or SIRT5-Flag plasmid. G
   IP and western blot analysis of the succinylation levels of wild-type
   BCAT1 and BCAT1 mutant in U251 cells. H Western blot analysis of BCAT1
   expression in U251 cells treated with 50 μg/mL cycloheximide. I IP and
   western blot analysis of the succinylation levels of BCAT1 K39R mutant
   in U251 cells. J Western blot analysis of BCAT1 expression in U251
   cells treated with 50 μg/mL cycloheximide. K co-IP and western blot
   analysis of the interaction between SIRT5-Flag and wild-type BCAT1 or
   BCAT1 K39R mutant. Data are presented as mean ± SD of three independent
   experiments. ns p > 0.05, *p < 0.01, ***p < 0.001.

SIRT5-mediated BCAT1 desuccinylation suppresses its ubiquitin-proteasome
degradation

   Our previous study has shown that E3 ligase CHIP mediates BCAT1
   ubiquitination, promoting its degradation [[117]26]. We sought to
   investigate whether the stabilization of BCAT1 by SIRT5 is associated
   with its ubiquitination status. Western blot analysis revealed that the
   BCAT1 K39R mutant exhibited lower levels of ubiquitination compared to
   wild-type BCAT1 (Fig. [118]6A). Additionally, SIRT5 overexpression was
   found to decrease both wild-type and K48-linkage BCAT1 ubiquitination
   (Fig. [119]6B, C). In contrast, depletion of SIRT5 increased these
   forms of BCAT1 ubiquitination (Fig. [120]6D, E). To further explore the
   interplay between BCAT1 succinylation and ubiquitination, Flag-tagged
   CHIP was co-transfected with wild-type BCAT1 or BCAT1 K39R mutant.
   Western blot analysis indicated that CHIP specifically increased the
   ubiquitination of wild-type BCAT1, while having no significant effect
   on K39R mutant (Fig. [121]6F). Moreover, CHIP interacted less with the
   BCAT1 K39R mutant compared to wild-type BCAT1, and SIRT5 overexpression
   reduced the interaction between CHIP and BCAT1 (Fig. [122]6G, H).
   Notably, CHIP overexpression did not significantly affect the half-life
   of the BCAT1 K39R mutant (Fig. [123]6I). In summary, these findings
   suggest that SIRT5-mediated desuccinylation of BCAT1 at K39 inhibits
   its interaction with CHIP, thereby preventing ubiquitin-proteasome
   degradation of BCAT1.

Fig. 6. SIRT5-mediated BCAT1 desuccinylation suppresses its
ubiquitin-proteasome degradation.

   [124]Fig. 6
   [125]Open in a new tab

   A IP and western blot analysis of the ubiquitination levels of
   wild-type BCAT1 and BCAT1 mutant. B, C IP, and western blot analysis of
   wild-type and K48-linkage BCAT1 ubiquitination levels in U251 cells
   transfected with vector or SIRT5-Flag plasmid. D, E IP and western blot
   analysis of wild-type and K48-linkage BCAT1 ubiquitination levels in
   U251 cells transfected with scramble siRNA or SIRT5 siRNA. F IP and
   western blot analysis of the ubiquitination levels of wild-type BCAT1
   and BCAT1 K39R mutant in U251 cells transfected with vector or
   Flag-CHIP plasmid. G, H co-IP and western blot analysis of the
   interaction between CHIP and BCAT1 in U251 cells. I Western blot
   analysis of BCAT1 expression in U251 cells treated with 50 μg/mL
   cycloheximide. Data are presented as mean ± SD of three independent
   experiments. ns p > 0.05.

SIRT5 stabilizes BCAT1 to promote cell proliferation and tumor growth

   To further validate the functional role of SIRT5/BCAT1 axis in glioma,
   stable cell lines were generated in which SIRT5 was knocked down, and
   BCAT1 was overexpressed in U87 glioma cells (Fig. [126]7A). Cell
   proliferation assay and EdU staining revealed that compared to
   wild-type U87 cells, those with SIRT5 knockdown (U87-shSIRT5) exhibited
   a significantly reduced proliferation rate, which was restored upon
   BCAT1 overexpression (U87-shSIRT5 + BCAT1) (Fig. [127]7B, C).
   Similarly, in a xenograft mouse model, mice injected with U87-shSIRT5
   cells showed notably smaller tumor volumes and reduced tumor weights
   compared to those xenografted with wild-type U87 cells. However, tumor
   growth was restored in mice xenografted with U87-shSIRT5 + BCAT1 cells
   (Fig. [128]7D–F). IHC staining for Ki67, a marker of cell
   proliferation, confirmed lower expression in the shSIRT5 group but
   elevated levels in the shSIRT5 + BCAT1 group (Fig. [129]7G). Together,
   these results clearly indicate that SIRT5 promotes glioma progression
   by regulating BCAT1.

Fig. 7. SIRT5 stabilizes BCAT1 to promotes cell proliferation and tumor
growth in vitro and in vivo.

   [130]Fig. 7
   [131]Open in a new tab

   A Western blot analysis of BCAT1 and SIRT5-Flag expression in wild-type
   U87 cells, U87-shSIRT5 cells, and U87-shSIRT5 + BCAT1 cells. B, C
   Crystal violet staining and EdU staining were performed to evaluate the
   proliferative ability of wild-type U87 cells, U87-shSIRT5 cells, and
   U87-shSIRT5 + BCAT1 cells. Scale bar = 100 μm. D–F Photographs, volume,
   and weight of the xenografted tumors. G IHC staining of the xenografted
   tumor sections in each group. Scale bar = 50 μm. Data are presented as
   mean ± SD (n = 6). **p < 0.01, ***p < 0.001. H, I Representative IHC
   images of glioma microarray showing SIRT5 and BCAT1 expression. J–K
   Correlation analysis of SIRT5 and BCAT1 expression with glioma grades.
   L, M) Kaplan–Meier survival curve of glioma patients based on SIRT5 and
   BCAT1 expression. N Correlation between SIRT5 and BCAT1 expression in
   glioma tissue microarrays.

High SIRT5 and BCAT1 expression correlates with poor prognosis of glioma
patients

   To better understand the clinical relevance of SIRT5 and BCAT1 in
   glioma, tissue microarrays were analyzed for their expression levels in
   glioma patient samples (Fig. [132]7H, I). The results demonstrated that
   both SIRT5 and BCAT1 were significantly upregulated in patients with
   advanced tumor grades (Fig. [133]7J, K). Survival analysis indicated
   that patients with higher levels of either SIRT5 or BCAT1 expression
   had notably shorter overall survival (Fig. [134]7L, M). Furthermore, a
   strong positive correlation was identified between SIRT5 and BCAT1
   expression within the tissue microarray samples (Fig. [135]7N). These
   findings suggest that the elevated expression of SIRT5 and BCAT1
   contributes to glioma progression and may serve as valuable prognostic
   markers for glioma patients.

SIRT5-mediated BCAT1 stabilization facilitates ferroptosis resistance

   We further examined the impact of SIRT5/BCAT1 axis on ferroptosis
   resistance in glioma cells. U87-shSIRT5 cells exhibited enhanced
   sensitivity to SAS, and ectopic expression of BCAT1 restored resistance
   to SAS in SIRT5-knockdown U87 cells (Fig. [136]8A). Upon SAS treatment,
   U87-shSIRT5 cells exhibited increased ROS levels and reduced GSH and
   glutamate levels in comparison to wild-type U87 and U87-shSIRT5 + BCAT1
   cells (Fig. [137]8B–D). In addition, the levels of LPO, MDA, and Fe²⁺,
   all markers of ferroptosis, were significantly elevated in U87-shSIRT5
   cells, which were ameliorated by ectopic expression of BCAT1 (Fig.
   [138]8E–G). Moreover, the expression of GPX4 and FTH1 was markedly
   decreased in U87-shSIRT5 cells (Fig. [139]8H). In in vivo experiments,
   mice xenografted with wild-type U87 cells or U87-shSIRT5 cells were
   intraperitoneally injected with SAS every other day. Results
   demonstrated that SIRT5 depletion potentiated SAS’s anti-tumor efficacy
   (Fig. [140]8I–K). IHC staining also revealed decreased expression of
   GPX4 and FTH1in the tumors from U87-shSIRT5 cells administrated with
   SAS (Fig. [141]8L). In sum, these data suggest that SIRT5/BCAT1 axis
   effectively enhances ferroptosis resistance in glioma.

Fig. 8. SIRT5-mediated BCAT1 stabilization facilitates ferroptosis
resistance.

   [142]Fig. 8
   [143]Open in a new tab

   A CCK8 assay was performed to evaluate the IC50 values for SAS in
   wild-type U87 cells, U87-shSIRT5 cells, and U87-shSIRT5 + BCAT1 cells.
   B DHE staining was performed to assess ROS levels in wild-type U87
   cells, U87-shSIRT5 cells, and U87-shSIRT5 + BCAT1 cells treated with
   SAS. Scale bar = 100 μm. C–F Intracellular GSH, glutamate, LPO, and MDA
   levels were detected in wild-type U87 cells, U87-shSIRT5 cells, and
   U87-shSIRT5 + BCAT1 cells treated with SAS. G Intracellular labile iron
   levels were detected by FerroOrange staining. Scale bar = 25 µm. H
   Western blot analysis of SLC7A11, GPX4, and FTH1 expression wild-type
   U87 cells, U87-shSIRT5 cells, and U87-shSIRT5 + BCAT1 cells treated
   with SAS. I–K Photographs, volume, and weight of the xenografted
   tumors. L IHC staining of the xenografted tumor sections in each group.
   Scale bar = 50 μm. Data are presented as mean ± SD (n = 6). *p < 0.01,
   **p < 0.01, ***p < 0.001.

Discussion

   SIRT5 is emerging as a promising target for cancer therapy due to its
   critical role in various metabolic pathways [[144]25]. Interestingly,
   SIRT5 demonstrates a dual function in the context of cancers, acting as
   a tumor suppressor in renal cell carcinoma and pancreatic cancer, while
   exhibiting oncogenic properties in breast cancer and non-small cell
   lung cancer [[145]27]. The expression and performance of SIRT5 in
   glioma, however, remain ambiguous and disputed. While certain studies
   report that SIRT5 expression is elevated in glioma tissue compared to
   adjacent non-cancerous tissue [[146]28], Tang et al., in contrast,
   associate higher SIRT5 expression with improved survival rates in
   glioma patients [[147]29]. In our research, we observe increased SIRT5
   expression in U251 and U87 cells compared to HMC3 cells, and in glioma
   tissues compared to normal brain tissues. Through microarray analysis,
   we establish that SIRT5 expression positively correlates with glioma
   grade and unfavorable prognosis. Functional experiments reveal that
   overexpressing SIRT5 enhances glioma cell proliferation, whereas SIRT5
   knockdown impairs cell proliferation and tumor growth both in vitro and
   in vivo. Overall, our findings indicate that SIRT5 functions as an
   oncogene in glioma.

   Although certain ferroptosis inducers show therapeutic promise in
   various cancer types, sulfasalazine remains the only reagent that has
   been under evaluation for glioma treatment in clinical trials [[148]30,
   [149]31]. However, several unsolved obstacles remain that limit its
   clinical use. For instance, glioma cells have a higher level of GSH
   than normal brain cells [[150]32], making it insensitive to ferroptosis
   inducers. Indeed, in vitro studies have shown that SAS only becomes
   gliomatoxic at concentrations exceeding 200 μM [[151]21]. Therefore,
   overcoming ferroptosis insensitivity remains a key research focus.
   SIRT5 has been shown to enhance cellular antioxidant defenses by
   promoting IDH2 desuccinylation and G6PD deglutarylation, as well as by
   desuccinylating SOD1, thus mitigating oxidative stress [[152]33,
   [153]34]. These mechanisms indicate that SIRT5 may prevent ferroptosis.
   Recent findings suggest that SIRT5 regulates ferroptosis through the
   Nrf2/HO-1 pathway in ischemic stroke models [[154]35]. SIRT5 also
   regulates glutamine metabolism [[155]13], which is a key source of
   glutamate and GSH. However, no evidence has linked SIRT5 to ferroptosis
   insensitivity in cancer. Our study demonstrates that high SIRT5
   expression correlates ferroptosis insensitivity in glioma cells.
   Specifically, SIRT5 knockdown decreases the IC50 of SAS to in glioma
   cells. Mechanistically, GPX4 and FTH1 expression are markedly
   downregulated following SAS treatment in SIRT5-knockdown cells,
   resulting in Fe^2+ accumulation, ROS production, and lipid
   peroxidation. These findings provide strong evidence that SIRT5 acts as
   a ferroptosis repressor in glioma.

   Enhanced BCAA metabolism is a hallmark of glioma, and its inhibition
   triggers apoptosis in glioma cells both in vitro and in vivo [[156]36].
   Our proteomic and metabolomic analysis indicates a relationship between
   SIRT5 and BCAA metabolism. BCAT1, which catalyzes the initial step in
   BCAA metabolism, is overexpressed in most cancer types, including
   glioma, where its high expression correlates with unfavorable patient
   prognosis [[157]37]. It has been reported that increased BCAT1
   expression boosts glutamate levels, enhancing GSH production and
   supporting GPX4 activity, which protects against lipid peroxidation and
   ferroptosis [[158]38]. Post-translational modification plays crucial
   role in the regulation of BCAT1. Phosphorylation and palmitoylation
   regulate BCAT1’s activity and intracellular localization [[159]39].
   Additionally, CHIP-mediated ubiquitination of BCAT1 leads to its
   proteasomal degradation, reducing cell proliferation and resistance to
   temozolomide in glioma [[160]26]. In this study, we reveal for the
   first time that BCAT1 undergoes both acetylation and succinylation in
   glioma cells. Glioma cells exhibit reduced succinylation of BCAT1,
   which stabilizes the protein. We further identify BCAT1 as a novel
   desuccinylation substrate of SIRT5, with SIRT5 directly binding to
   BCAT1 and desuccinylating it at K39. This desuccinylation suppresses
   the interaction between BCAT1 and the E3 ligase CHIP, preventing BCAT1
   degradation. These findings reveal a novel post-modification cross-talk
   on BCAT1. Furthermore, BCAT1 overexpression rescues the proliferation
   inhibition observed in SIRT5 knockdown cells. Following SAS treatment,
   BCAT1 overexpression restores intracellular GSH levels and mitigates
   oxidative stress, thereby conferring resistance to SAS-induced
   ferroptosis. To further study the role of SIRT5 and BCAT1 in clinical,
   our data shows a correlation between SIRT5 and BCAT1 expression in
   glioma microarray, with higher expression levels predicting higher
   glioma grade and worse clinical outcome.

   In conclusion, our study proposes a hierarchical model in which
   elevated SIRT5 expression stabilizes BCAT1 through desuccinylation,
   promoting BCAA metabolism and glioma cell proliferation. Additionally,
   the SIRT5/BCAT1 axis increases intracellular GSH levels, alleviates ROS
   production, and inhibits lipid peroxidation, contributing to
   ferroptosis insensitivity in glioma cells. Thus, our study is the first
   to highlight the crucial role of SIRT5 in the ferroptosis resistance,
   emphasizing its therapeutic potential for glioma patients.

Supplementary information

   [161]Supplemental Figures^ (384.5KB, pdf)
   [162]Table S1^ (15.3KB, docx)
   [163]Table S2^ (15.2KB, docx)
   [164]Raw WB Data^ (4.5MB, pdf)

Author contributions

   Conceptualization: TW, ZL; Methodology: TW; Investigation: TW, XHH,
   JJC, XW, ZZ; Visualization: ZL; Formal Analysis: ZL; Writing-original
   draft: TW; Writing-review & editing: XJH, ZL.

Funding

   This work was supported by National Natural Science Foundation of China
   (82303229) and Jiangxi Provincial Natural Science Foundation
   (20232BAB216078, 20242BAB25508).

Data availability

   The data that support the findings of this study are available from the
   corresponding author upon reasonable request.

Competing interests

   The authors declare no competing interests.

Ethics approval and consent to participate

   This study was approved by the Ethics Committee of Jiangxi Provincial
   People’s Hospital (protocol code KT078), and the maximal tumor size
   permitted is 150 mm. All experiments were performed in accordance with
   the approved protocol and guidelines. Written informed consent for
   using clinical information and tissue samples was obtained from all
   patients.

Footnotes

   Edited by Jean-Ehrland Ricci

   Publisher’s note Springer Nature remains neutral with regard to
   jurisdictional claims in published maps and institutional affiliations.

Contributor Information

   Tao Wang, Email: wangtaoalepellis@gmail.com.

   Zhuo Lu, Email: ndyfy10004@ncu.edu.cn.

Supplementary information

   The online version contains supplementary material available at
   10.1038/s41419-025-07626-9.

References