TIGAR maintains intestinal epithelial regeneration by stabilizing HMGCL and promoting β-catenin β-hydroxybutyrylation in burn-induced sepsis

TL;DR

TIGAR helps maintain intestinal health during burn-induced sepsis by boosting ketone production and cell regeneration.

Contribution

TIGAR's novel role in stabilizing HMGCL and promoting β-catenin β-hydroxybutyrylation is identified.

Findings

TIGAR deficiency reduces BHB production and impairs intestinal stem cell function.

TIGAR stabilizes HMGCL by blocking Park2-mediated degradation, enhancing BHB synthesis.

BHB promotes β-catenin β-hydroxybutyrylation, enhancing cell proliferation and regeneration.

Abstract

Burn-induced sepsis triggers profound intestinal injury, contributing to systemic inflammation and organ damage. Beta-hydroxybutyrate (BHB), a major ketone body, acts as a key regulator of intestinal epithelial regeneration. Its metabolic dysregulation has been implicated in impaired cell proliferation and the maintenance of intestinal stem cells (ISCs). However, the dynamic regulatory mechanisms underlying BHB fluctuation during burn sepsis-induced intestinal injury remain elusive. In this study, we demonstrate that TIGAR expression is markedly reduced in small intestinal crypts of burn sepsis mice. TIGAR deficiency substantially diminishes BHB production and compromises cell proliferation and ISC self-renewal capacity. Mechanistically, the 1-131 domain of TIGAR orchestrate dual functionality: it acts as a mitochondrial targeting signal to direct TIGAR localization and competitively binds the ketogenic enzyme HMGCL, thereby inhibiting its interaction with the E3 ubiquitin ligase Park2. This spatial interference blocks Park2-mediated K48-linked ubiquitination and proteasomal degradation of HMGCL, and stabilizing HMGCL to enhance BHB synthesis. Elevated BHB induces β-hydroxybutyrylation at lysine 335 of β-catenin, which facilitates β-catenin nuclear translocation and strengthens its interaction with TCF4, therefore driving cell proliferation and ISC self-renewal, ultimately maintaining intestinal epithelial regeneration. Collectively, this study identifies a novel role of TIGAR in maintaining intestinal barrier regeneration by promoting ketone body production. This previously unexplored mechanism of TIGAR may serve as a critical compensation for the treatment of burn-related gut barrier dysfunction.

Figures

Intestinal injury was observed in burn sepsis mice with reduced TIGAR expression and cell proliferation.A Representative images of pathological sections of the small intestine in Sham and burn sepsis groups on day 1, 3, 5 and 7. Scale bar: 100 μm. B-F Western blotting was employed to detect the expression changes of TIGAR in the small intestinal crypts of mice with burn-induced sepsis at various time points (*n* = 3 mice). G The expression of TIGAR in small intestinal tissues of mice with burn sepsis was evaluated by immunohistochemistry on post-burn day 7. Scale bar: 50 μm. H-J The mRNA expre

TIGAR promotes IEC-6 cell proliferation and intestinal organoid growth.A, B The overexpression A and knockdown B efficiency of TIGAR in IEC-6 cells were assessed by western blot analysis. C, D IEC-6 cells with TIGAR overexpression C or knockdown D were cultured in the absence or presence of LPS (100 μg/mL) for 0-72 h, the cell viability at different time points was assessed by CCK8 assay (*n* = 4 biological replicates). E–H IEC-6 cells with TIGAR overexpression E, G or knockdown F, H were cultured in the presence or absence of LPS for 5 days. Colony formation assays were subsequently performed

TIGAR facilitates the nuclear translocation of β-catenin in IEC-6 cells.A Western blot analysis was performed to detect changes in β-catenin protein levels in the intestinal crypts of mice with burn sepsis on day 7. B The alterations in the nuclear and cytoplasmic distribution of β-catenin in intestinal crypt cells of mice with burn sepsis on day 7 were assessed using nucleocytoplasmic fractionation and western blot analysis. Fibrillarin (FBL) was used as the nuclear internal control and GAPDH served as the cytoplasmic internal control. C The expression levels of β-catenin were evaluated by we

TIGAR promotes β-catenin β-hydroxybutanylation by facilitating β-hydroxybutyrate generation.A, B Volcano plot displaying metabolites differentially expressed in mice with burn sepsis on day1 A and day7 B. C–E Untargeted metabolomics revealed relative amounts of BHB from different samples: the intestinal tissues of Sham group and burn sepsis on day1, 3, 5 and 7 group mice C (*n* = 6 mice); the TIGAR knockdown HIEC-6 D (*n* = 4 biological replicates) and IEC-6 E (*n* = 5 biological replicates) cells. F The Kbhb levels of total protein in intestinal crypt of normal and burn sepsis mice were valid

BHB promotes the proliferation of IEC-6 cells and intestinal organoids growth.A–C The proliferation of IEC-6 cells following treatment with LPS (100 μg/mL) and BHB (20 mM) for 24 h was detected by CCK-8 A (*n* = 4 biological replicates) and EdU B, C assay. Scale bars: 100 μm. D, E The proliferation of IEC-6 cells was evaluated using clone formation assays after treatment with LPS and BHB over a 5-day period. F, G The nuclear translocation of β-catenin was analyzed by nuclear-cytoplasmic separation and western blot in IEC-6 cells cultured with LPS and BHB for 24 h. H LPS and BHB were co-adminis

TIGAR promotes the stability of HMGCL.A, B The protein levels of ketogenic enzymes in HIEC-6 A and IEC-6 B cells overexpressing and knockdown of TIGAR were detected by western blot. C The western blot assay was performed to validate the expression of HMGCL in *TIGAR*^*f/f*^ and *TIGAR*^*f/f*^*Vil1-Cre* mice. D IEC-6 cells were administered with 100 μg/mL LPS for 24 h, the protein level of HMGCL was verified by western blot. E The expression of HMGCL in intestinal crypts of burn sepsis on different time points (*n* = 3). F Immunofluorescence was used to detect the cellular distribution and expr

TIGAR inhibits Park2-mediated K48-linked ubiquitination of HMGCL by competitively binding to HMGCL.A Rapamycin (50 μM) was administered to IEC-6 cells for 6 h to induce autophagy, and the protein expression level of HMGCL was assessed via western blot analysis. B Stable TIGAR knockdown IEC-6 cells were treated with MG132 (20 μM) for 6 h and then HMGCL protein levels were analyzed by western blot analysis. C Using Flag-TIGAR, HA-Ub and His-HMGCL co-transfected into HEK293T cells, immunoprecipitation assay was performed to validate the ubiquitination of HMGCL. D HEK293T cells were transfected wi

Schematic representation of TIGAR’s role in sustaining intestinal cell proliferation and maintaining stem cell function.A schematic model showing that in burn sepsis mice intestinal crypt proliferative cells, including stem cells, the expression of TIGAR was significantly reduced. This reduction impaired TIGAR’s ability to competitively bind HMGCL with Park2, leading to an increased ubiquitination of HMGCL and subsequent enhanced protein degradation. Consequently, the synthesis of β-hydroxybutyrate was inhibited. The decreased β-hydroxybutyrate synthesis attenuated the β-hydroxybutyrylation mo

Full text

Introduction

Major burns represent the most severe form of trauma, posing a significant threat to human health [1]. Currently, sepsis and multiple organ dysfunction syndrome (MODS) resulting from severe burns are the primary causes of burn-related mortality, accounting for 50–60% of total deaths [2, 3]. Sepsis often results in the intestinal mucosal barrier damage [4]. This disruption may lead to enterogenous infections, promote to systemic inflammation, and exacerbate organ injury, thereby perpetuating a vicious cycle [5, 6]. Effectively preserving the integrity of the intestinal mucosa has thus emerged as one of the pivotal challenges in the management and prevention of sepsis.

The intestinal epithelium serves as the core component of the intestinal mucosa barrier, and its renewal relies on the proliferation and differentiation of LGR5^+^ intestinal stem cells (ISCs) situated at the crypt bases [7, 8]. These stem cells continuously divide, generating rapidly dividing transitional amplifying (TA) cells and proliferating progenitor cells, ensures a continuous supply of differentiated epithelial cells [9]. Among these processes, cell proliferation represents the initial and most critical step in initiating intestinal epithelial repair [10]. Modulating cell proliferation constitutes a pivotal link for maintaining intestinal homeostasis and facilitating repair. Cell proliferation relies on an adequate supply of nutrients. Numerous studies have reported that glucose and amino acids play critical roles in regulating intestinal epithelial cells, including ISCs [11–14]. In recent years, the regulatory impact of ketone bodies, which are fat metabolites, on intestinal cell proliferation and ISCs self-renewal has garnered increasing attention and has become a focal point of research in the intestinal epithelial regeneration [15].

Ketone bodies consist of acetoacetate, beta-hydroxybutyrate (BHB), and acetone, with BHB being the most predominant form, comprising approximately 70% [16]. The liver serves as the primary organ for ketone bodies synthesis [17]. Additionally, the intestine is recognized as a significant extrahepatic organ capable of synthesizing ketone bodies due to its possession of a complete ketone body synthesis enzyme system [18]. Currently, the role of ketone body in modulating intestinal function has garnered increasing attention. Research has shown that when mammals are deprived of nutrients, undergo prolonged fasting, or engage in exercise, fatty acid oxidative metabolism is enhanced, resulting in the production of substantial amounts of ketone bodies in the intestine [19]. BHB plays a crucial role in regulating intestinal epithelial homeostasis and promoting the regeneration of the intestinal epithelium [15, 20]. Recent studies have demonstrated that BHB serves not only as a glucose substitute but also promotes the lysine β-hydroxybutyrylation modification in multiple proteins [21, 22], thereby regulating protein functions and participating in the modulation of intestinal epithelial homeostasis [23]. The metabolic alterations of intestinal BHB following burns, as well as their association with the regulation of intestinal cell proliferation, remain unclear and warrant further investigation.

Mitochondria are the primary organelle for ketone body synthesis, and their synthesis is influenced by the cellular redox state [24, 25]. In recent years, an enzyme known as TP53-induced glycolytic and apoptotic regulator (TIGAR) has garnered increasing attention due to its ability to inhibit oxidative stress by promoting the pentose phosphate pathway (PPP) and enhancing the production of the reducing agent NADPH [26–28]. More importantly, TIGAR can translocate into mitochondria under stress conditions and function as a metabolic regulatory enzyme [29]. To date, no studies have reported the relationship between TIGAR and ketone body synthesis.

In this study, we conducted a series of investigations addressing this critical issue and identified TIGAR as a key molecule regulating intestinal ketone body synthesis. Specifically, TIGAR enhances the protein stability of the ketogenic enzyme HMGCL by inhibiting its ubiquitination, thereby increasing ketone body production and inducing β-hydroxybutyrylation of β-catenin, and ultimately enhancing cell proliferation and ISC self-renewal. Our findings reveal a previously unrecognized mechanism by which TIGAR regulates ketone body metabolism. It not only deepens our understanding of TIGAR-mediated metabolic regulation but also provides experimental evidence supporting TIGAR as a promising therapeutic target for intestinal mucosal damage.

Results

Intestinal damage in burn sepsis mice may result from decreased TIGAR expression

To investigate TIGAR dynamic expression in burn sepsis-induced intestinal injury, we established a murine burn sepsis model. Histopathological evaluation demonstrated severe disruption of mucosal architecture at critical phases (days 1-7), characterized by villus blunting, crypt dropout and a thinner basal layer (Fig. 1A). The expression of TIGAR in small intestinal crypts began to elevate at day 1 post-injury, declined at day 3, reached its nadir at day 7, and normalized by day 10 (Fig. 1B–F). Similarly, immunohistochemical analysis revealed a significant reduction in TIGAR expression in the intestinal tissues of mice with burn sepsis on day7, with its expression mainly localized to the crypts and villus (Fig. 1G). Moreover, the mRNA levels of PCNA, Ki67 and LGR5 were decreased in intestinal crypts of mice with burn sepsis on day3 and day7 compared to normal mice (Fig. 1H–J). Consistently, ISCs depletion was evidenced by LGR5^+^ cell reduction, accompanied by diminished proliferative activity (Fig. 1K–M). These results establish TIGAR downregulation as an early molecular event preceding epithelial breakdown, implicating its pathogenic role in burn sepsis-associated gut injury.Fig. 1. Intestinal injury was observed in burn sepsis mice with reduced TIGAR expression and cell proliferation.A Representative images of pathological sections of the small intestine in Sham and burn sepsis groups on day 1, 3, 5 and 7. Scale bar: 100 μm. B-F Western blotting was employed to detect the expression changes of TIGAR in the small intestinal crypts of mice with burn-induced sepsis at various time points (n = 3 mice). G The expression of TIGAR in small intestinal tissues of mice with burn sepsis was evaluated by immunohistochemistry on post-burn day 7. Scale bar: 50 μm. H-J The mRNA expression levels of Ki67 (H), PCNA (I), and LGR5 (J) in the small intestinal crypts of mice with burn sepsis on day 3 and day 7 were evaluated using qRT–PCR (n = 3 biological replicates). K–M The expression levels of Ki67 K, PCNA L, and LGR5 M in small intestinal tissues of mice with burn sepsis were assessed via immunofluorescence staining on day 3 and day 7 post-injury. Scale bars: 100 μm. Data presented are representative of three independent experiments. The data are presented as mean ± SD. Two-sided unpaired Student’s t-test was performed to assess statistical significance H–J. ** p* < 0.05, *** p* < 0.01*, *** p* < 0.001.

TIGAR deficiency exacerbates the weakened cell proliferation and loss of stem cells in the intestinal organoids induced by LPS

To simulate the inflammatory microenvironment associated with burn-induced sepsis, we subjected IEC-6 intestinal epithelial cells to LPS stimulation [30]. Using lentiviral-mediated overexpression and knockdown systems, TIGAR was shown to be significantly overexpressed (Fig. 2A), and all three shRNAs effectively suppressed TIGAR expression (Fig. 2B). The shTIGAR-1# and shTIGAR-2# constructs were selected for use in subsequent experiments. We demonstrated that TIGAR knockdown exacerbated LPS-induced proliferative defects and TIGAR overexpression alleviated the proliferation inhibition induced by LPS, evidenced by CCK8 (Fig. 2C, D), colony formation (Fig. 2E–H) and EdU assays (Fig. 2I–L). To establish in vivo relevance, we generated TIGAR^f/f^Vil1-Cre mice with intestinal specific TIGAR ablation (Fig. 2M, N). And induced organoids using the intestinal crypts of TIGAR knockout mice. LPS treatment inhibited the budding and growth of organoids (Fig. 2O). However, the knockout of TIGAR further inhibited the budding and growth of organoids (Fig. 2O). In addition, combined LPS stimulation caused the further decreased of cell proliferation and the number of LGR5^+^ cell depletion in knockout organoids compared to wild-type controls (Fig. 2P–R). These findings collectively identify the absence of TIGAR will exacerbate the weakened proliferation of intestinal epithelial cells and the reduction in the number of LGR5^+^ cells. TIGAR is a potential protector against sepsis-driven impairment of intestinal epithelial regeneration.Fig. 2TIGAR promotes IEC-6 cell proliferation and intestinal organoid growth.A, B The overexpression A and knockdown B efficiency of TIGAR in IEC-6 cells were assessed by western blot analysis. C, D IEC-6 cells with TIGAR overexpression C or knockdown D were cultured in the absence or presence of LPS (100 μg/mL) for 0-72 h, the cell viability at different time points was assessed by CCK8 assay (n = 4 biological replicates). E–H IEC-6 cells with TIGAR overexpression E, G or knockdown F, H were cultured in the presence or absence of LPS for 5 days. Colony formation assays were subsequently performed to analyze cell growth. I-L IEC-6 cells with TIGAR overexpression I, K or knockdown J, L were cultured in the presence or absence of LPS for 24 h. The cell proliferation was detected using EdU assays. Scale bars: 100 μm. M Protocol for creating intestinal-specific TIGAR knockout mice (TIGAR^f/f^Vil1-Cre). N TIGAR deletion in intestinal tissues was validated using western blot from TIGAR^f/f^Vil1-Cre and TIGAR^f/f^ mice. O Intestinal organoids derived from TIGAR^f/f^ or TIGAR^f/f^Vil1-Cre mice were cultured in the presence or absence of LPS (100 μg/mL) for 0-5 days. The variations in organoid growth were monitored and comparatively analyzed across distinct experimental groups. Scale bars: 100 μm. P–R Intestinal organoids derived from TIGAR^f/f^ or TIGAR^f/f^Vil1-Cre mice were cultured in the presence or absence of LPS (100 μg/mL) for 48 h. The number of Ki67^+^ P, Brdu^+^ Q and LGR5^+^ cells R were verified by immunofluorescence assay. Scale bars: 100 μm. Data presented are representative of three independent experiments. The data are presented as mean ± SD. Two-way ANOVA C, D and One-way ANOVA G, H, K, L and P–R were performed to assess statistical significance. ** p* < 0.05, *** p* < 0.01*, *** p* < 0.001.

TIGAR promotes the proliferation of IEC-6 cells and the growth of organoids via activating β-catenin

Intestinal epithelial homeostasis is governed by a hierarchical regulatory network, with the Wnt/β-catenin pathway serving as a master coordinator of stem cell dynamics [31, 32]. The canonical function of this pathway requires nuclear translocation of β-catenin, which interacts with TCF4 to activate proliferative transcriptional programs [33]. In burn sepsis mice, we observed preserved total β-catenin levels (Fig. 3A) but markedly reduced nuclear localization in intestinal crypts (Fig. 3B). Strikingly, while TIGAR manipulation did not alter total β-catenin abundance (Fig. 3D, E) consistent with LPS stimulation (Fig. 3C). TIGAR overexpression promoted the nuclear translocation of β-catenin (Fig. 3F, G) and restored the LPS-impaired nuclear translocation (Fig. 3H, I). Conversely, TIGAR knockdown inhibited the nuclear translocation (Fig. 3J, K). In addition, overexpression of TIGAR enhanced the interaction between β-catenin and TCF4, whereas knockdown of TIGAR diminishes this interaction (Fig. 3L). To establish the pathway specificity of TIGAR-mediated proliferation, we employed β-catenin-IN-7, a selective inhibitor disrupting β-catenin/TCF4 interaction [34]. The CCK8 assay results indicated that a concentration of 50 μM for IN7 did not cause significant cytotoxicity (Supplementary Fig. S1). Pharmacological inhibition by IN-7 exacerbated significant organoid growth arrest induced by LPS (Fig. 3M), accompanied by diminished Ki67^+^ proliferative cells (Fig. 3N) and LGR5^+^ stem cell depletion (Fig. 3O). Moreover, β-catenin inhibition abrogated TIGAR-driven proliferative advantages: co-treatment reduced EdU incorporation (Fig. 3P) and colony formation (Fig. 3Q) compared to TIGAR overexpression alone. These results demonstrate that TIGAR’s regenerative function requires intact β-catenin-TCF4 transcriptional complex activity, and TIGAR could regulate the proliferation of intestinal epithelial cells by promoting the nuclear translocation of β-catenin.Fig. 3TIGAR facilitates the nuclear translocation of β-catenin in IEC-6 cells.A Western blot analysis was performed to detect changes in β-catenin protein levels in the intestinal crypts of mice with burn sepsis on day 7. B The alterations in the nuclear and cytoplasmic distribution of β-catenin in intestinal crypt cells of mice with burn sepsis on day 7 were assessed using nucleocytoplasmic fractionation and western blot analysis. Fibrillarin (FBL) was used as the nuclear internal control and GAPDH served as the cytoplasmic internal control. C The expression levels of β-catenin were evaluated by western blot analysis in IEC-6 cells cultured with or without LPS (100 μg/mL) for 24 h. D, E After overexpression D and knockdown E of TIGAR in IEC-6 cells, the protein levels of β-catenin were verified by western blot. F After the overexpression of TIGAR in IEC-6 cells, nuclear-cytoplasmic fractionation was conducted, followed by western blot analysis to assess alterations in the subcellular distribution of β-catenin. G Immunofluorescence was employed to assess the nuclear translocation of β-catenin following TIGAR overexpression in IEC-6 cells. H, I Nuclear-cytoplasmic fractionation H and immunofluorescence I were performed to assess the nuclear translocation of β-catenin in vector and TIGAR-overexpressing IEC-6 cells cultured in the presence or absence of LPS (100 μg/mL) for 24 h. J, K Nuclear-cytoplasmic fractionation J and immunofluorescence K were performed to assess the nuclear translocation of β-catenin in shNC and TIGAR knockdown IEC-6 cells. L Immunoprecipitation analysis of the interaction of β-catenin with TCF4 by the corresponding antibodies in TIGAR overexpression (left) or knockdown (right) IEC-6 cells. M The specific inhibitor of β-catenin β-catenin-IN-7 (50 μM) and LPS (100 μg/mL) were co-administered to small intestinal organoids for 48 h, and the morphological and growth changes of the organoids were quantitatively assessed under an optical microscope. Scale bars: 100 μm. N, O β-catenin-IN-7 and LPS were co-administered to small intestinal organoids for 48 h, the number of Ki67^+^ N and LGR5^+^ O cells were acquired through confocal microscopy. Scale bars: 100 μm. P IEC-6 cells overexpressing TIGAR were treated with LPS (100 μg/mL) and β-catenin-IN-7 (50 μM) for 24 h, followed by assessment of cell proliferation using the EdU assay. Scale bars: 100 μm. Q IEC-6 cells overexpressing TIGAR were treated with LPS and β-catenin-IN-7 for 5 days, the cell growth was detected by colony formation assay. Data presented are representative of three independent experiments. The data are presented as mean ± SD. Two-sided paired Student’s t-test B, F, J and One-way ANOVA H, N–Q were performed to assess statistical significance. ** p* < 0.05, *** p* < 0.01*, *** p* < 0.001.

TIGAR promotes β-catenin β-hydroxybutanylation by facilitating beta-hydroxybutyrate generation

Emerging evidence highlights BHB as a pleiotropic regulator beyond its metabolic role, particularly through inducing β-hydroxybutyrylation (Kbhb), a post-translational modification modulating protein functionality [35]. We first analyzed the changes of metabolites in the small intestinal tissues of burn sepsis and normal mice using untargeted metabolomics, and the trend of BHB showed an initial increase followed by a decrease (Fig. 4A–C). In burn sepsis mice, we observed the dynamic of BHB dynamics in intestine: an initial surge on day 1, subsequently declined gradually on day 3 and day 5, a precipitous decline to one third of baseline by day 7 (Fig. 4C). TIGAR knockdown reducing BHB levels compare with controls in HIEC-6 and IEC-6 cells (Fig. 4D, E). Given the role of BHB in regulating post-translational modifications of proteins, we examined the Pan-Kbhb level in intestinal crypts and found that the Pan-Kbhb level was significantly decreased on day 7 in burn sepsis mice compared to sham group mice (Fig. 4F). Moreover, the overexpression of TIGAR significantly increased the level of Pan-Kbhb, whereas its level was markedly reduced following knockdown of TIGAR (Fig. 4G). Then we verified the concentration of BHB that affects protein β-hydroxybutyrylation in cells, the degree of modification was directly proportional to the concentration (Fig. 4H). We choose the concentration of 20 mM for the subsequent experiments. In addition, TIGAR knockdown decreased the β-hydroxybutyrylation level of β-catenin (Fig. 4I), while TIGAR overexpression promoted the modification level (Fig. 4J). Mass spectrometry identified six conserved lysine sites (K281, K292, K335, K435, K496, K508) on β-catenin (Supplementary Fig. S2). Site-directed mutagenesis revealed K335 as the dominant modification site (Fig. 4K). Functional studies demonstrated that K335 mutation impaired β-catenin nuclear translocation (Fig. 4L). Overall, these results demonstrate that TIGAR could activate β-catenin by facilitating the K335 β-hydroxybutyrylation.Fig. 4TIGAR promotes β-catenin β-hydroxybutanylation by facilitating β-hydroxybutyrate generation.A, B Volcano plot displaying metabolites differentially expressed in mice with burn sepsis on day1 A and day7 B. C–E Untargeted metabolomics revealed relative amounts of BHB from different samples: the intestinal tissues of Sham group and burn sepsis on day1, 3, 5 and 7 group mice C (n = 6 mice); the TIGAR knockdown HIEC-6 D (n = 4 biological replicates) and IEC-6 E (n = 5 biological replicates) cells. F The Kbhb levels of total protein in intestinal crypt of normal and burn sepsis mice were validated using western blot assay. G After overexpression (left) and knockdown (right) of TIGAR in IEC-6 cells, the levels of Kbhb in total cellular protein were assessed by western blot. H IEC-6 cells were treated with different concentrations of BHB for 24 h, and the Kbhb levels of total protein were detected by western blot. I, J The Kbhb modification level of β-catenin was evaluated by immunoprecipitation followed by western blot analysis in IEC-6 cells treated with 20 mM BHB for 24 h, with TIGAR expression modulated through either knockdown I or overexpression J. K After the transfection of wild-type and six β-catenin mutants into HEK293T cells and treatment with 20 mM BHB for 24 h, the β-hydroxybutyrylation levels of β-catenin were assessed by immunoprecipitation followed by western blot analysis. L The wild-type and K335R mutant forms of β-catenin were transfected into HEK293T cells, followed by treatment with BHB for 24 h. Subcellular fractionation and western blot analyses were subsequently conducted to assess alterations in the nuclear and cytoplasmic distribution of β-catenin. Data presented are representative of three independent experiments. The data are presented as mean ± SD. Two-sided paired Student’s t-test D, E and One-way ANOVA C were performed to assess statistical significance. *** p* < 0.01, **** p* < 0.001.

BHB promotes the proliferation of IEC-6 cells and survival of intestinal organoid

To validate whether TIGAR promotes epithelial cell proliferation via BHB biosynthesis, we performed metabolite supplementation assays in vitro and in organoid models. In LPS-exposed IEC-6 cells, exogenous BHB effectively reversed proliferation suppression: The CCK8 activity (Fig. 5A), EdU-positive rate (Fig. 5B, C), and colony-forming capacity (Fig. 5D, E) were increased. Additionally, LPS impaired the nuclear translocation (Fig. 5F–H) and interaction with TCF4 (Fig. 5I, J) of β-catenin. Supplementation of BHB can reverse the growth inhibitory effect of LPS on intestinal organoids (Fig. 5K). Using intestinal epithelium-specific TIGAR knockout organoids, BHB supplementation rescued growth defects caused by TIGAR ablation and reconstituted the LGR5^+^ stem cell pool (Fig. 5L–O). These complementary lines of evidence establish that TIGAR-driven intestinal regeneration is mediated through sustaining BHB biosynthesis.Fig. 5BHB promotes the proliferation of IEC-6 cells and intestinal organoids growth.A–C The proliferation of IEC-6 cells following treatment with LPS (100 μg/mL) and BHB (20 mM) for 24 h was detected by CCK-8 A (n = 4 biological replicates) and EdU B, C assay. Scale bars: 100 μm. D, E The proliferation of IEC-6 cells was evaluated using clone formation assays after treatment with LPS and BHB over a 5-day period. F, G The nuclear translocation of β-catenin was analyzed by nuclear-cytoplasmic separation and western blot in IEC-6 cells cultured with LPS and BHB for 24 h. H LPS and BHB were co-administered to IEC-6 cells for 24 h, immunofluorescence analysis was conducted to examine the nuclear and cytoplasmic distribution of β-catenin in cells. Scale bars: 25 μm. I, J The interaction of β-catenin and TCF4 in IEC-6 cells administered with LPS and BHB for 24 h was validated by immunoprecipitation and western blot. K LPS and BHB were co-administered to intestinal organoids for 48 h, and the growth and morphological changes of organoids were monitored. Scale bars: 100 μm. L–O LPS and BHB were co-administered to intestinal organoids for 48 h, the number of Ki67^+^ L, M and LGR5^+^ N, O cells was verified by immunofluorescence. Scale bars: 100 μm. Data presented are representative of three independent experiments. The data are presented as mean ± SD. Two-way ANOVA A, One-way ANOVA C, E, G, J and Two-sided paried Student’s t-test M, O were performed to assess statistical significance. ** p* < 0.05, *** p* < 0.01*, *** p* < 0.001.

TIGAR maintains the stability of HMGCL to promote the nuclear translocation of β-catenin

Ketogenesis is orchestrated by sequential enzymatic reactions involving HMGCS2, HMGCL, and BDH1, with HMGCL serving as the key enzyme that cleaves HMG-CoA to generate acetoacetate [36]. Our study unveiled TIGAR as a specific regulator of HMGCL protein homeostasis. TIGAR increased the protein level of HMGCL, but had little effect on the other two enzymes (Fig. 6A, B). A decrease in HMGCL protein levels was also found in intestinal crypts of TIGAR^f/f^Vil1-Cre mice compared to TIGAR^f/f^ mice (Fig. 6C). In LPS-treated IEC-6 cells, the protein level of HMGCL decreased progressively as the concentration increases (Fig. 6D). In burn sepsis mice, HMGCL protein levels were markedly reduced in intestinal crypts (Fig. 6E, F), while its mRNA abundance remained unaltered (Fig. 6G). Similar results were also observed in cells with TIGAR knockdown and overexpression (Fig. 6H). Cycloheximide (CHX) chase assays demonstrated that TIGAR overexpression extended HMGCL protein half-life, whereas TIGAR knockdown reduced the levels (Fig. 6I). LPS treatment also inhibited the stability of HMGCL (Fig. 6J). BioGRID database prediction and co-immunoprecipitation assays confirmed direct physical interaction between TIGAR and HMGCL (Fig. 6K–M). To verify the interacting domains of TIGAR and HMGCL, we systematically generated a series of truncated mutants according to the literature (Fig. 6N) [37]. Domain mapping revealed specific binding between the 1-131 region of TIGAR and the 150-325 domain of HMGCL (Fig. 6O, P). After the knockdown of HMGCL in TIGAR overexpression cells cultured with LPS, EdU assay results demonstrated that, compared with TIGAR-overexpressing cells, HMGCL knockdown significantly suppressed cell proliferation (Fig. 6Q).Fig. 6TIGAR promotes the stability of HMGCL.A, B The protein levels of ketogenic enzymes in HIEC-6 A and IEC-6 B cells overexpressing and knockdown of TIGAR were detected by western blot. C The western blot assay was performed to validate the expression of HMGCL in TIGAR^f/f^ and TIGAR^f/f^Vil1-Cre mice. D IEC-6 cells were administered with 100 μg/mL LPS for 24 h, the protein level of HMGCL was verified by western blot. E The expression of HMGCL in intestinal crypts of burn sepsis on different time points (n = 3). F Immunofluorescence was used to detect the cellular distribution and expression of HMGCL in the small intestinal tissues of mice with burn sepsis. Scale bars: 50 μm. G, H The expression of HMGCL mRNA in intestinal crypts of burn sepsis on day7 G and overexpression and knockdown TIGAR IEC-6 cells H was assessed by qRT-PCR (n = 3 biological replicates). I, J CHX (100 μg/mL) was added to the IEC-6 cells with TIGAR knockdown and overexpression I as well as LPS (100 μg/mL) treatment J, and the cells were collected at 0, 2, 4, and 8 h respectively. The protein level of HMGCL was detected by western blot. K The Biogrid database predicts proteins that may interact with TIGAR. L Co-immunoprecipitation was used to detect the interaction between endogenous TIGAR and HMGCL in IEC-6 cells. M Using Flag-TIGAR and His-HMGCL co-transfected into HEK293T cells, the interaction between TIGAR and HMGCL was assessed by co-immunoprecipitation and western blot assays. N The schematic diagram of TIGAR and HMGCL protein truncation mutant, red: PFK domain, green: mitochondrial location reported in the literature. O Co-transfected the full-length HMGCL and the truncated TIGAR into HEK293T cells. Immunoprecipitation analysis of the interaction of TIGAR with HMGCL by anti-His. P Co-transfected the full-length TIGAR and the truncated HMGCL into HEK293T cells. Immunoprecipitation analysis of the interaction of TIGAR with HMGCL by anti-Flag. Q TIGAR overexpression plasmid and HMGCL shRNA plasmid were co-transfected into IEC-6 cells, the cell proliferation was verified by EdU assay. Scale bars: 100 μm. Data presented are representative of three independent experiments. Two-sided paired Student’s t-test A–C, G, One-way ANOVA H, Q were performed to assess statistical significance. ns no significant, ** p* < 0.05, *** p* < 0.01*, *** p* < 0.001.

Next, we investigated the impact of HMGCL on cell function. The overexpression and knockdown efficiency of HMGCL were detected in IEC6 cells (Supplementary Fig. S3A, B). ShHMGCL-3# and shHMGCL-4# were used for subsequent studies. EdU assay (Supplementary Fig. S3C–F), clone formation (Supplementary Fig. S3G–J) and CCK8 assay (Supplementary Fig. S3K) reveled that overexpression of HMGCL can reverse the cell growth inhibition caused by LPS treatment, while knockdown of HMGCL can exacerbate the LPS-induced cell growth attenuation. HMGCL did not influence the expression levels of β-catenin (Supplementary Fig. S3L, M). However, the knockdown of HMGCL decreased the nuclear translocation of β-catenin (Supplementary Fig. S3N–P), whereas the overexpression of HMGCL promoted the nuclear translocation of β-catenin (Supplementary Fig. S3Q–S). Moreover, knockdown of HMGCL exacerbated the LPS-induced decrease in Kbhb levels of β-catenin (Supplementary Fig. S3T), while overexpression of HMGCL reversed the decline in Kbhb levels of β-catenin induced by LPS (Supplementary Fig. S3U). These results suggest that TIGAR promotes the protein stability of HMGCL by interacting with it and could promote the proliferation of IEC-6 cells by stabilizing HMGCL.

Furthermore, knockdown of HMGCL in TIGAR-overexpressing cells attenuated the nuclear translocation of β-catenin induced by TIGAR overexpression (Supplementary Fig. S4A, B). Conversely, in TIGAR-knockdown cells, both overexpression of HMGCL (Supplementary Fig. S4C, D) and exogenous administration of BHB (Supplementary Fig. S4E, F) reversed the reduction in β-catenin nuclear translocation, as demonstrated by western blot and immunofluorescence assays. These results demonstrate that TIGAR promotes nuclear translocation of β-catenin through the HMGCL-BHB axis.

The 1-131 region of TIGAR has a dual function of binding to HMGCL and guiding TIGAR into mitochondria

We have confirmed that the 1-131 region of TIGAR interacts with the 150-325 region of HMGCL. HMGCL is located in the mitochondria. The prerequisite for TIGAR to bind to it is to be able to enter the mitochondria from the cytoplasm. We unexpectedly revealed that the 1-131 of TIGAR is not only the binding region with HMGCL, but also the guiding sequence that leads it into the mitochondria. This conclusion was validated through multiple approaches and extensive experimental repetitions.

It has been reported that the mitochondrial guide sequence of TIGAR is located at residues 258-261 [29]. Consequently, we carried out an in-depth investigation. We initially verified the mitochondrial localization of TIGAR and HMGCL by immunofluorescence, TIGAR is partially localized in the mitochondria (Supplementary Fig. S5A), whereas HMGCL is exclusively localized within this organelle (Supplementary Fig. S5B). By constructing multiple truncated and mutant plasmids of TIGAR, and then immunofluorescence and mitochondrial isolation were employed to assess the mitochondrial localization of these TIGAR mutants following transfection into cells. The immunofluorescence results demonstrated that the full length and 1-131 truncation could localize in mitochondria (Supplementary Fig. S5C, D). The truncated constructs (Human 132-270/Rat 132-268), containing the residues 258-261 (Human)/256-259 (Rat) which reported as the mitochondrial guide sequence, completely lost their mitochondrial targeting ability (Supplementary Fig. S5C, D). Notably, even after deletion of the 258-261/257-260 regions, TIGAR retained its mitochondrial localization efficiency (Supplementary Fig. S5C, D). By isolating the mitochondria from intestinal epithelial cells and performing western blot analysis, we confirmed the results obtained from immunofluorescence staining (Supplementary Fig. S5E, F). These findings suggest that the 1-131 domain of TIGAR may achieve mitochondrial transport via a novel guiding mechanism in intestinal epithelial cells.

TIGAR suppresses Park2-mediated K48-linked ubiquitination of HMGCL by competitively interacting with HMGCL

The degradation of proteins is regulated by ubiquitination or autophagy pathways. We stimulated autophagy using rapamycin and observed that the protein level of HMGCL remained unchanged (Fig. 7A). Conversely, upon treating TIGAR-knockdown cells with MG132, a marked accumulation of HMGCL protein was detected (Fig. 7B). These findings suggest that TIGAR may regulate the degradation of HMGCL protein via the ubiquitin-proteasome system. Further studies demonstrated that TIGAR specifically inhibits the ubiquitination of HMGCL (Fig. 7C, Supplementary Fig. S6A). Moreover, HMGCL existed four types of ubiquitination modifications: K6, K11, K48, and K63 (Fig. 7D). Notably, TIGAR selectively suppressed the K48-linked ubiquitination of HMGCL (Supplementary Fig. S6B). Given that TIGAR can interact directly with HMGCL, we hypothesized that TIGAR may disrupt the interaction between HMGCL and E3 ubiquitin ligase to modulate the ubiquitination status of HMGCL. Subsequently, we utilized the UbiBrowser database to predict potential E3 ubiquitin ligases for HMGCL and identified a total of eight candidate proteins (Fig. 7E). Among these candidates, MUL1 and PRKN (Park2) could localize in the mitochondria [38, 39], whereas the remaining six proteins are either not expressed in the mitochondria or are present at relatively low levels. Consequently, we prioritized functional validation of MUL1 and Park2. Our findings revealed that MUL1 neither altered the protein level of HMGCL nor its ubiquitination status (Supplementary Fig. S7A, B). In contrast, Park2 significantly decreased the protein level of HMGCL (Fig. 7F) and promoted K48-linked ubiquitination of HMGCL (Fig. 7G, Supplementary Fig. S8A). It is noteworthy that in the burn sepsis model, the expression level of Park2 in crypts did not exhibit significant changes (Supplementary Fig. S8B), and TIGAR exhibited no significant regulatory effect on the Park2 protein level (Supplementary Fig. S8C). Furthermore, the overexpression of TIGAR significantly reduced Park2-induced ubiquitination of HMGCL (Fig. 7H). Further investigations revealed that Park2 interacts with the 150-325 amino acid region of HMGCL (Fig. 7I, Supplementary Fig. S8D), which precisely overlaps with the region of HMGCL that interact with TIGAR. The overexpression of TIGAR significantly diminished the interaction between HMGCL and Park2 (Fig. 7J). These findings support our hypothesis that TIGAR may interfere with the interaction between HMGCL and the E3 ubiquitin ligase Park2 by modulating the spatial conformation of HMGCL, thereby influencing the ubiquitination and degradation processes of HMGCL.Fig. 7TIGAR inhibits Park2-mediated K48-linked ubiquitination of HMGCL by competitively binding to HMGCL.A Rapamycin (50 μM) was administered to IEC-6 cells for 6 h to induce autophagy, and the protein expression level of HMGCL was assessed via western blot analysis. B Stable TIGAR knockdown IEC-6 cells were treated with MG132 (20 μM) for 6 h and then HMGCL protein levels were analyzed by western blot analysis. C Using Flag-TIGAR, HA-Ub and His-HMGCL co-transfected into HEK293T cells, immunoprecipitation assay was performed to validate the ubiquitination of HMGCL. D HEK293T cells were transfected with the indicated plasmids. Ubiquitination assay was used to detect the ubiquitination level of His-HMGCL. E Predicting possible E3 ubiquitin ligases of HMGCL using the ubibrowse 2.0 database. F The expression of HMGCL was detected in Park2 overexpressing IEC-6 and HIEC-6 cells via western blot. G Co-transfected the Flag-Park2, His-HMGCL, HA-Ub wilt-type and mutants into HEK293T cells, immunoprecipitation assay was used to assess the ubiquitination level of His-HMGCL. H HEK293T cells were transfected with the Flag-Park2, His-HMGCL, TIGAR and HA-Ub K48 only plasmids. Immunoprecipitation assay was used to assess the ubiquitination level of His-HMGCL. I HEK293T cells were transfected with the Flag-Park2, His-HMGCL wild type and truncated plasmids. The interaction between Park2 and HMGCL was validated by co-immunoprecipitation assay. J HEK293T cells were co-transfected with Flag-Park2, His-HMGCL, and TIGAR plasmids. The interaction between Park2 and HMGCL was confirmed through co-immunoprecipitation assay. Data presented are representative of three independent experiments.

Discussion

This study reveals a new mechanism of action of the metabolic regulatory enzyme TIGAR in intestinal injury repair, which is different from the previous traditional understanding of its role in glycolytic inhibition and activation of the pentose phosphate pathway. We demonstrated that TIGAR’s mitochondrial location-dependent regulatory system for HMGCL stability, and revealed that TIGAR translocates into the mitochondria to form a complex with the ketogenic enzyme HMGCL. This interaction competitively inhibits Park2-mediated K48 ubiquitination of HMGCL, thereby significantly enhancing the stability of the HMGCL protein. This post-translational regulatory effect enhances the production of BHB, subsequently facilitating the nuclear translocation and the interaction with TCF4 of β-catenin via the β-hydroxybutyroylation on K335 of β-catenin. This process drives the renewal of intestinal epithelial cells. These findings deepened the understanding of the mechanism by which TIGAR regulates cellular metabolism and maintains the intestinal mucosal barrier.

As a metabolite of the core ketone body in the body, BHB serves as an energy supplier [25]. More importantly, it also acts as a crucial regulator of intestinal regeneration [15]. Recent studies have demonstrated that BHB maintains intestinal homeostasis via multi-faceted mechanisms: inhibiting the activation of the NLRP3 inflammasome and reducing the release of pro-inflammatory cytokines such as IL-1β [40]; remodeling the intestinal microbiota ecology to decrease the risk of pathogen translocation [41]; up-regulating the expression of tight junction proteins like claudin-3 to strengthen barrier function [18]; and regulating the fate determination of LGR5^+^ stem cells to balance their self-renewal and differentiation processes [15]. This study elucidated the pathological features of intestinal BHB metabolic disorder under burn stress. Metabolomics analysis revealed a significant decrease in intestinal mucosal BHB levels post-burn. Moreover, we demonstrated that exogenous supplementation of BHB can not only counteract the inhibition of intestinal cell proliferation induced by LPS but also restore the organoid growth disorder caused by TIGAR knockout. This series of evidence confirms the critical role of the TIGAR-BHB signaling pathway in intestinal epithelial regeneration.

TIGAR is a prominent molecule in the field of redox biology, widely recognized for its role in channeling glucose into the pentose phosphate pathway to facilitate the synthesis of NADPH. TIGAR has been shown to scavenge harmful ROS and activating AP-1, thereby supporting the proliferation of LGR5-positive intestinal stem cells [42]. This proliferative effect aligns with its canonical function in alleviating oxidative stress, a mechanism we previously demonstrated in burn injury models, where TIGAR promoted intestinal stem cell proliferation by mitigating oxidative damage [12]. However, emerging evidences indicate that TIGAR participates in biological processes beyond its classical antioxidant role. This study uncovered a novel molecular mechanism through which TIGAR modulates the Wnt/β-catenin signaling pathway. As the central signaling hub for intestinal epithelial renewal, the dynamic regulatory network of the Wnt/β-catenin pathway depends on the nuclear translocation of β-catenin. Upon entering the nucleus, β-catenin assembles a transcriptional activation complex with TCF4, thereby promoting the expression of downstream genes associated with proliferation [33]. In this study, we demonstrated that TIGAR enhances the nuclear localization of β-catenin without altering its total protein level, thereby promoting intestinal cell proliferation and self-renewal of ISCs. This finding indicates the possible presence of a novel post-translational modification regulatory mechanism for β-catenin. Mechanism analysis reveals that TIGAR-mediated BHB metabolic reprogramming is crucial in this process. Besides acting as an alternative energy substrate, BHB directly modulates protein function by inducing β-hydroxybutoylation on proteins [43]. Through mass spectrometry and site-directed mutation experiments, we demonstrated that TIGAR markedly enhances the nuclear translocation efficiency of β-catenin and strengthens its interaction with TCF4 by promoting β-catenin’s K335 site-specific β-hydroxybutyrylation modification. This discovery establishes a molecular link between ketone body metabolites and the epigenetic regulation of the Wnt signaling pathway, expanding our understanding of how TIGAR sustains intestinal regeneration.

The generation of ketone bodies is primarily regulated by three key enzymes: HMGCS2, HMGCL, and BHD1. This study systematically elucidated the molecular landscape of the TIGAR-HMGCL axis in regulating ketone body production and signal transduction. In the classical ketogenic enzyme system, HMGCL functions as a critical enzyme in the regulation of ketone body metabolism that catalyzes the cleavage of HMG-CoA to produce acetoacetate [44]. However, this study demonstrated that TIGAR specifically modulates the protein stability of HMGCL without influencing the expression levels of HMGCS2 and BHD1. Moreover, we identified the TIGAR-HMGCL-BHB-β-catenin signaling pathway. HMGCL facilitates the nuclear translocation and interaction of β-catenin with TCF4 by enhancing β-catenin β-hydroxybutyrylation. This finding refines the previously reported mechanism of HMGCL promoting β-catenin nuclear entry [45], and addresses a critical gap in the regulatory mechanism of this process. At the level of HMGCL protein stability regulation, we investigated the mechanism by which TIGAR inhibits HMGCL degradation. Literature indicates that HMGCL protein stability is regulated through ubiquitination [46]. In this study, via systematic screening of E3 ubiquitin ligases, we identified Park2 as a novel E3 ubiquitin ligase for HMGCL. Park2 is an E3 ubiquitin ligase that can be localized to the cytoplasm, nucleus, Golgi complex, and mitochondria [47–49]. Its primary function involves participating in the degradation of intracellular proteins and regulating mitochondrial autophagy [50]. In this study, we found that Park2 catalyzes the formation of K48-linked polyubiquitination chains by binding to the HMGCL domain (residues 150-325), thereby targeting it for proteasomal degradation. TIGAR interacts with HMGCL 150-325 regions, generating a steric hindrance effect and competitively inhibiting the assembly of the Park2-HMGCL complex. This “molecular shield” mechanism effectively inhibits the ubiquitination-mediated degradation pathway of HMGCL. These findings uncovered novel functions of Park2 in non-autophagy pathways, extended its biological relevance in regulating mitochondrial protein homeostasis. TIGAR may dynamically compete with Park2 to regulate the stability of HMGCL.

Additionally, this study serendipitously revealed a novel insight into the mitochondrial localization signal of TIGAR. Using a systematic truncation mutagenesis approach, we identified that the region responsible for TIGAR’s interaction with HMGCL lies within amino acids 1–131. Surprisingly, this same region also contains the sequence that directs TIGAR into mitochondria, a finding distinct from the previously reported 258–261 site [29]. To confirm this accidental discovery, we repeated the experiments in multiple cell lines and found that the 132–275 truncation mutant, which includes residues 258–261, completely lost its ability to localize to mitochondria. In contrast, the 1–131 fragment and mutants lacking the 258–261 sequence retained efficient mitochondrial targeting. The heterogeneity in this subcellular localization pattern may be influenced by several factors, including differences in mitochondrial membrane transport systems across various cell types and dynamic reprogramming of molecular chaperone networks under stress conditions [51–53]. Our study indicates that the 1–131 fragment not only mediates TIGAR’s interaction with HMGCL but also contributes to its mitochondrial transport through a potentially novel mechanism in intestinal epithelial cells. Given that the focus of this study was to investigate the interaction between TIGAR and HMGCL and its downstream effects, we did not further explore this unexpected finding to stay focused on the original goal. Nonetheless, this finding provides valuable experimental evidence for our future studies on the mitochondrial translocation mechanism of TIGAR.

In conclusion, we demonstrated that TIGAR regulates ketone body metabolism via a mitochondrial location-dependent metabolic reprogramming mechanism. Specifically, TIGAR translocates into the mitochondria and forms a complex with the ketogenic enzyme HMGCL, thereby inhibiting Park2-mediated ubiquitination and degradation of HMGCL through a spatial competition mechanism. This interaction ultimately promotes β-hydroxybutyrate-mediated post-translational modification of β-catenin, which contributes to the maintenance of intestinal epithelial homeostasis by promoting cell proliferation and survival, as well as sustaining intestinal stem cell function (Fig. 8). These findings reveal a novel biological function of TIGAR, highlight its role as a metabolic regulatory hub beyond antioxidant stress, and delineate the potential targeted therapeutic axis of TIGAR-HMGCl-β-catenin for intestinal barrier injury in burn-induced sepsis.Fig. 8. Schematic representation of TIGAR’s role in sustaining intestinal cell proliferation and maintaining stem cell function.A schematic model showing that in burn sepsis mice intestinal crypt proliferative cells, including stem cells, the expression of TIGAR was significantly reduced. This reduction impaired TIGAR’s ability to competitively bind HMGCL with Park2, leading to an increased ubiquitination of HMGCL and subsequent enhanced protein degradation. Consequently, the synthesis of β-hydroxybutyrate was inhibited. The decreased β-hydroxybutyrate synthesis attenuated the β-hydroxybutyrylation modification of lysine 335 (K335) on the β-catenin protein. This attenuation disrupted the nuclear translocation of β-catenin, diminished its interaction with the transcription factor TCF4, and ultimately suppressed the proliferation of intestinal cells as well as the self-renewal capacity of LGR5^+^ stem cells.

Materials and methods

Antibodies and reagents

Anti-TIGAR (sc-166291) was purchased from Santa Cruz; Anti-TIGAR (22136-1-AP), Anti-HMGCL (16898-1-AP), anti-Park2 (14060-1-AP), anti-Tubulin (66031-1-Ig), anti-ubiquitin (10201-2-AP), anti-β-catenin (51067-2-AP), anti-TCF4 (22337-1-AP), anti-His-Tag (66005-1-Ig), anti-Flag-Tag (20543-1-AP, 66008-4-Ig), anti-HA-Tag (51064-2-AP), anti-PCNA (10205-2-AP), anti-FBL (16021-1-AP), anti-BrdU (66241-1-Ig), anti-Tom20 (11802-1-AP) were purchased from Proteintech (Wuhan, China); HRP-linked second Antibody (A0216, A0208) and anti-LGR5 (AF0165) was from Beyotime Biotechnology (Shanghai, China); anti-Ki67 (MA5-14520) was purchased from Invitrogen (California, USA); Anti-β-Hydroxybutyryllysine (PTM-1201RM) was purchased from PTM bio (Hangzhou, China); VeriBlot for IP Detection Reagent (HRP) (ab131366), Alexa Fluor® 488 Goat Anti-Rabbit IgG (ab150077) and Alexa Fluor® 488 Goat Anti-Mouse IgG (ab150113) were purchased from Abcam (UK); Cy3 Goat Anti-Rabbit IgG (SDI0012) and Cy3 Goat Anti-Mouse IgG (SDI0013) were purchased from SIDBIO (Chongqing, China); LPS (L9143) and CHX (239765) were purchased from Sigma-Aldrich (Saint Louis, Germany); (R)-3-Hydroxybutanoic acid sodium (HY-W015851), MG132 (HY-13259) and rapamycin (HY-10219) were purchased from MedChemExpress (New Jersey, USA); IntestiCult Organoid Growth Medium (Mouse, 06005) was purchased from STEMCELL Technologies (Vancouver, Canada).

Cell lines and culture

The intestinal epithelial cell line HIEC-6 and IEC-6, and HEK293T cells were obtained from the American Type Culture Collection (ATCC) (Manassas, USA). HIEC-6, IEC-6, and HEK293T cells confirmed mycoplasma-free by ATCC. The HEK293T and IEC-6 cells were cultured in RPMI 1640 medium with 10% fetal bovine serum, 1 × streptomycin, and penicillin at 37 °C in 5% CO_2_. The HIEC-6 cells were cultured in Opti-MEM (Gibco, 31985) with 20 mM HEPES, 10 mM GlutaMAX (Gibco, 35050), 10 ng/mL Epidermal Growth Factor, 4% fetal bovine serum (Gibco), 1 × streptomycin, and penicillin at 37 °C in 5% CO_2_.

Animals and experimental models

Healthy male BALB/c mice (22–25 g, 8-10 weeks old) were purchased from the Experimental Animal Center of the Third Military Medical University. Under specific pathogen-free conditions, animals are raised in separately ventilated cages, with a light/dark cycle of 12 h, freely obtaining food and water. Before the experiment, the mice were fed a standardized diet for a week. All procedures comply with the institutional guidelines and regulations. The experimental protocol was approved and implemented by the Laboratory Animal Welfare and Ethics Committee of the Third Military Medical University (License No. : AMUWEC20210636).

This study utilized the classical burn sepsis model [54], specifically the subeschar bacterial injection model of burns, which is also referred to as the invasive wound infection model in burn research. The specific grouping and experimental procedures are as follows: Based on the previously described slightly modified burn-sepsis model, 40 mice were randomly allocated into two groups: the sham operation group (Sham group, n = 20) and the burn sepsis group (BS group, n = 20). Under general anesthesia induced by intraperitoneal injection of ketamine (80 mg/kg) and xylazine (10 mg/kg), the dorsal hair of mice in each group was carefully removed using electric shavers. Anesthetized mice were positioned in a template to expose approximately 20% of their body surface area (calculated using the Meeh formula). The shaved backs of the mice were then immersed in water maintained at 90 ± 0.2 °C for 8 s, resulting in scald injuries covering approximately 20% of the total body surface area. Then, 1.5 mL of 1% 37 °C lactate Ringer’s solution per kg body weight was intraperitoneally injected for fluid resuscitation, and 100 μL of buprenorphine (0.3 mg/mL) was administered subcutaneously for analgesia. The burned mice were housed individually in sterile cages and provided with sterile water and food ad libitum. The sham-operated mice underwent the same experimental procedure, except that they were exposed to water at 37°C instead of hot water. For burn wound infection, 100 μL of a 10 mM MgSO4 solution containing 5 × 10^5^ Pseudomonas aeruginosa should be subcutaneously injected following the induction of burn wound injury. On days 1, 3, 5, 7, and 10 post-infection, euthanasia was conducted via cervical dislocation. Subsequently, an aseptic midline laparotomy was performed to incise the abdomen, followed by the aseptic resection of the small intestine. Small intestinal crypts were then extracted for subsequent experiments.

TIGAR^f/f^ mice have two lox sites on both sides of exon 3 in the TIGAR locus. Floxed mice were crossed with Vil1-Cre mice to obtain tissue-specific knockout of TIGAR.

No animals were excluded from this study. Animals were allocated to experimental groups using a random number generator. Experiments were performed in a blinded manner; mice were coded and randomized by investigators not involved in experimental performance. Sample sizes as well as the definitions of statistical methods andmeasures for each experiment are provided in the corresponding figure legend.

Crypt isolation and culture

After euthanizing the mice, the small intestines are excised and immediately transferred to a 10 cm dish containing cold PBS. The serosa, blood vessels, and adipose tissue surrounding the intestine were carefully removed. Using a syringe, PBS was injected into one end of the small intestine and flushed repeatedly until all contents were thoroughly cleared. The intestine was then placed in a new dish, longitudinally incised, and rinsed six times with fresh PBS. Subsequently, the cleaned small intestinal tissue was segmented into approximately 5 mm pieces and transferred to a 50 mL centrifuge tube containing 40 mL of ice-cold PBS supplemented with 5 mM EDTA. The tissue was digested by gentle rotation at 4 °C for 30 min. Following digestion, the supernatant was discarded, and the tissue fragments were resuspended in 30 mL of PBS containing 10% FBS. The centrifuge tube was vigorously up and down for 10 s, filtered through a 70 μm cell strainer, and the filtrate was collected in a clean 50 mL tube. This filtration step was repeated twice, and the filtrates were pooled. Finally, the pooled filtrate was centrifuged at 300×g for 10 min. The supernatant was discarded, leaving the pellet, which contained the crypts.

For organoid culture, the crypts are resuspended in 1 ml organoid medium. Mix the crypts with Matrigel on ice, ensuring that Matrigel constitutes more than 70% of the mixture and the crypts density is maintained at 8–20/μl. Inoculate the mixed suspension into the center of the bottom of a preheated 48-well plate, using approximately 30 μl per well. Avoid contact between the suspension and the side walls of the well. After inoculation, place the culture plates in a 37 °C incubator for approximately 15 min to allow Matrigel to solidify. Subsequently, add 200 μl of preheated mouse intestinal organoid complete medium (stem cell culture medium with 1× B27 supplement, 1× N2 supplement, 1 × GlutaMAX, 1% Pen/Strep and 10 μM Y27632) to each well, ensuring that the Matrigel droplets are fully immersed in the medium. Transfer the 48-well plate to a 37 °C incubator with 5% CO_2_ for continuous observation, and replace the culture medium every 2–3 days.

Stable knock down and overexpressing cell lines

Stable knockdown cell lines of TIGAR and HMGCL were established using shRNA technology. The shRNAs were purchased from Tsingke Biotech (Beijing, China). The sequence was cloned into the pLKO.1 vector and co-transfected with packaging plasmids into HEK293T cells. After 48 h, the viral supernatant was collected, and target cells were infected in the presence of polybrene. Infected cells were then selected with a specific concentration of puromycin for 7 days to generate stable knockdown cell lines. The detailed shRNA sequence information is provided in Supplementary Table S1. The expression plasmids and their mutants, including Flag-TIGAR, Flag-Park2, Flag-Mul1, His-HMGCL, and HA-Ub, were constructed using standard molecular biology techniques. The wild-type (WT) or mutant sequences were cloned into the pLV4ltr-Puro-CMV vector. The method for constructing stable overexpression cell lines was consistent with that used for shRNA-mediated knockdown.

RNA isolation and real-time PCR

The total RNA was extracted from tissues and cells using Trizol reagent according to the manufacturer’s protocol. Subsequently, 1000 ng of the extracted RNA was reverse transcribed into cDNA using the PrimeScript RT Reagent Kit (Takara, Shiga, Japan). The relative expression levels of mRNA were quantified by qRT-PCR with SYBR Premix Ex TaqTM (Takara, Shiga, Japan). The thermal cycling conditions were as follows: initial denaturation at 95°C for 3 min, followed by 40 cycles of 95°C for 5 s, 58°C for 34 s, and a final extension step at 72°C for 60 s. All experiments were independently repeated at least three times to ensure reproducibility. The relative mRNA expression levels were analyzed using the 2^-ΔΔCt^ method and normalized to the internal reference gene β-actin. The primer sequences: Mouse HMGCL-F 5’-ACCCACCCCAGTGAAGATC-3’, R 5’-TCAGACACAGCACCGAAGA-3’, Rat HMGCL-F 5’-GTTTCCCGGCATCAACTACC-3’, R 5’-CAACTTCTTGGCGACCTCAG-3’, Mouse LGR5-F 5’-GCAGACTACGCCTTTGGAAA-3’, R 5’-GCCTGGGGAAGAGATGAGAT-3’, Mouse Ki67-F 5’-CCCAGGCAGTCCATAAAACA-3’, R 5’-CAGGTAGGCCAGAGCAAG TC-3’, Mouse PCNA-F 5’-TGTGGATAAAGAAGAGGAGGCG-3’, R 5’-ACTGTAGGA GACAGTGGAGTGG-3’, Mouse Actin-F 5’-GCTCAGTAACAGTCCGCCTAG-3’, R 5’-AGTG TGACGTTGACATCCG-3’, Rat Actin-F 5’-GAGAGGGAAATCGTGCGTGA-3’, R 5’-TGGAA GGTGGACAGTGAGGC-3’, Rat TIGAR primer1-F 5’-AGGAGAGACAGTAGAGCAGG-3’, R 5’-GTCACTCAGGAAATAACCAA-3’, Rat TIGAR primer2-F 5’-CGAGGACTAAGCAGA CCATA-3’, R 5’-CCCAGAATCAACTGACAAAT-3’.

Cytoplasmic and nuclear fractionation

Cytoplasmic and nuclear fractionation was performed using the nuclear protein and cytoplasmic protein extraction kit (Beyotime, Shanghai, China). Briefly, after washing the cells with PBS, carefully remove the PBS, add 200 μL of Cell Separation Buffer A, gently scrape the cells using a cell scraper, incubate on ice for 15 min, and subsequently add 10 μL of Cytoplasmic Protein Extraction Reagent B. Maximum speed vortexing for 5 s, followed by an ice bath for 1 min. Perform another vortexing at maximum speed for 5 s, then centrifuge at 12,000–16,000 g at 4 °C for 5 min. Immediately transfer the supernatant to a pre-cooled EP tube, this constitutes the extracted cytoplasmic protein. For the precipitate, completely aspirate the remaining supernatant, wash with PBS and then discarding PBS, and subsequently add 50 μL of nuclear protein extraction reagent containing protease inhibitors. Perform the fastest and most intense vortexing for 15 to 30 s to fully suspend and disperse the cell pellet. Place the sample back into the ice bath and vortex vigorously at high speed for 15 to 30 s every 1 to 2 min, repeating this process for a total of 30 min. Centrifuge the sample at 12,000–16,000 g at 4 °C for 10 min. Immediately transfer the supernatant to a pre-cooled EP tube to obtain the extracted nuclear protein.

Cytoplasmic and mitochondria fractionation

Cytoplasmic and mitochondria fractionation was performed using the Mitochondria Isolation Kit for Cultured Cell (Thermo, USA). Briefly, Centrifuge the suspensions contain 2 × 10^7^ cells in 2-mL centrifuge tubes at 850 × g for 2 min. Carefully aspirate and discard the supernatant. Add 400 μL of Mitochondrial Separation Reagent A, vortex at medium speed for 5 s, and incubate on ice for 2 min. Transfer the cell suspension to a glass homogenizer and thoroughly grind the cells on ice to ensure efficient lysis. Transfer the lysed cells back to the original tube, add 400 μL of mitochondrial separation reagent C, rinse the homogenizer with 200 μL of mitochondrial separation reagent A, and combine all components into the tube containing the sample. Invert several times to ensure thorough mixing. Centrifuge at 700 × g for 10 min at 4 °C, then carefully transfer the supernatant to a new 2 mL tube. Centrifuge again at 12,000 × g for 15 min at 4 °C. To purify mitochondrial components and minimize contamination from lysosomal and peroxidized substances, centrifuge the resulting pellet at 3000 × g for 15 min at 4 °C. Carefully transfer the supernatant (cytosolic fraction) to a new tube, while the precipitate will contain the isolated mitochondria. Add 500 microliters of mitochondrial separation reagent C to the precipitate for washing. Subsequently, add the protein lysis buffer and fully lyse the mitochondria on ice. Centrifuge the mixture at 12,000 g for 5 min, and collect the supernatant, which contains the mitochondrial protein.

Pan antibody-based PTM enrichment and LC-MS/MS analysis

Collect the small intestinal crypts from mice and resuspend them in IP buffer (1 mM EDTA, 50 mM Tris-HCl, 100 mM NaCl, 0.5% NP-40, pH 8.0). Subsequently, incubate the suspension with pre-washed Anti-β-Hydroxybutyryl lysine antibody conjugated agarose beads (PTM Bio, TMT-1204) at 4°C for 12 h. The bound peptides are eluted with 0.1% trifluoroacetic acid (TFA). After vacuum drying, the eluted peptides are desalted using Millipore C18 ZipTips and subsequently subjected to LC-MS/MS analysis. The LC-MS/MS analysis is supported by Jingjie PTM BioLabs (Hangzhou, China). The specific peptides and β-hydroxybutyrylation modification sites that were identified are presented in Supplementary Table S2.

Untargeted metabolomics analysis

For cell samples, add 1 mL of pre-cooled (-80 °C) 80% (v/v) mass spectrometry-grade methanol to a culture dish. Scrape the cells off the dish using a cell scraper. Lyse the cells at -20 °C by treating them with a low-temperature ultrasonic disruptor for three cycles of 60 s each, followed by a 30-s pause between cycles. For the mouse small intestinal tissue, 25 mg of the tissue was accurately weighed and mixed with 750 μL of pre-cooled 80% methanol. The tissue was then homogenized using a low-temperature tissue grinder at -20°C for three cycles, each consisting of 60 s of homogenization followed by a 30-s pause. The homogenates of the transferred cells and the small intestine were incubated at -20 °C for 60 min. Subsequently, the samples were centrifuged at 18,410 × g for 10 min at 4 °C. After centrifugation, collect the supernatant, dry it under a stream of nitrogen, re-dissolve the precipitate in 50% acetonitrile, centrifuge again, and retain the supernatant for subsequent analysis. To prepare the quality control (QC) mixture for mass spectrometry analysis, withdraw 20 μL from each sample. The concentration of the standard β-hydroxybutyric acid solution is 10 μM.

The hydrophilic metabolites extracted from cells and small intestinal tissues were profiled using the Orbitrap Exploris 120 mass spectrometer (Thermo Fisher Scientific) via a targeted analytical approach. The column temperature was set at 45°C. In the negative ion mode, a mixed gradient elution was performed on a BEH Amide HILIC chromatographic column (2.1 × 150 mm, 1.7 μm, Waters) using mobile phase A (5% ACN, 10 mM ammonium bicarbonate, pH 9) and mobile phase B (95% ACN, 10 mM ammonium bicarbonate, pH 9). The linear gradient is as follows: 0 min, 99% B; 0.1 min, 99% B; 6 min, 30% B; 6.5 min, 99% B; 10 min, 99% B. In the negative ion mode, the spray voltage is set to 2.5 kV, the capillary temperature is maintained at 350 °C, and the mass range (m/z) is from 70 to 1050.

The original files of non-targeted metabolomics were processed using Compound Discoverer 3.3 (Thermo Fisher Scientific), leveraging the mzCloud and mzVault databases for compound identification. The mass tolerance was set to 10 ppm for precursor ions and 15 ppm for fragment ions. For quantitative analysis, a maximum retention time deviation of 0.25 min was permitted.

Ubiquitination assay

The specified construction plasmid was co-transfected into HEK293T or hiec-6 cells along with the HA-Ub plasmid. Forty-eight hours post-transfection (cells were treated with MG132 (20 µM) for 4 h prior to harvest), the cells were washed with PBS and subsequently lysed in 460 µl of IP lysis buffer (1% SDS, 50 mM Tris-HCl, pH 7.5, 0.5 mM EDTA, 1 mM DTT) on ice. The lysate was then centrifuged at 13,000 rpm for 10 min. From the obtained supernatant, mix 60 µl with 15 µl of 5x SDS buffer, heat at 100 °C for 5 min to prepare the input sample. The remaining supernatant was incubated overnight with the anti-His antibody. Subsequent steps should be performed according to the procedure outlined in the immunoprecipitation section.

CCK-8 cell proliferation and viability assays

Cell viability and proliferation rate were assessed using Solarbio’s Cell Counting Kit-8 (CCK-8). For the cell proliferation assay, cells were seeded at a density of 2 × 10^3^ cells per well in 96-well plates. Following incubation at 37°C for the designated time periods, the original medium was replaced with 100 µL of fresh medium containing 10 µL of CCK-8 solution, and the cells were further incubated for 2 h. The absorbance at 450 nm was measured using a microplate reader (Thermo) to quantify the number of viable cells. For the cell viability assay, 8 × 10^3^ cells per well were seeded into a 96-well plate. Following incubation at 37°C for 24 h and treatment with the respective drugs, the viable cells were assessed using the aforementioned CCK-8 method.

EdU cell proliferation assays

EdU detection was carried out using the BeyoClickTM EDU-555 Cell Proliferation Kit (Beyotime) following the manufacturer’s instructions. Specifically, cells from each group were seeded into a 24-well plate and treated as indicated, followed by incubation with EdU (10 µM) for 2 h. After fixation and permeabilization, the cells were incubated with the click reaction cocktail solution. Subsequently, nuclear staining was performed using Hoechst 33342. Six random fields per well were captured using a Leica microscope to quantify the percentage of EdU-positive cells.

Colony formation assays

500 cells were seeded into 6-well plates and cultured in complete medium for 10 days until visible colonies formed. Subsequently, the colonies were fixed with 4% paraformaldehyde, stained with crystal violet, and then washed with PBS. The colony numbers were quantified using ImageJ software.

Immunofluorescence

The cells were fixed with 4% paraformaldehyde for 15 min and permeabilized with 0.5% Triton X-100 at room temperature for 15 min. Subsequently, the cells were blocked with a solution of 5% normal serum at room temperature for 1 h. After removing the blocking solution, an appropriate dilution of primary antibody was added to each well and incubated overnight at 4 °C. The cells were then washed three times with PBST solution (3 min per wash). Next, the diluted fluorescent secondary antibody was applied and incubated at 37 °C for 1 h. Following this step, DAPI staining was performed on the specimens in darkness for 10 min. Finally, to prevent fluorescence quenching, a mounting medium containing an anti-fade reagent was applied before imaging under a fluorescence microscope.

Western blot

The protein from the cell or tissue sample was extracted and its concentration was quantified using the BCA method. The protein samples were then separated by SDS-PAGE electrophoresis and transferred onto a PVDF membrane. The membrane was blocked with 5% skim milk powder and incubated overnight at 4°C with the primary antibody working solution at the appropriate dilution. Afterward, the membrane was washed three times with TBST buffer for 10 min each and subsequently incubated with the secondary antibody working solution at room temperature for 90 min. Finally, the membrane was washed three times with TBST buffer and visualized using the ECL luminescent solution.

Co‑immunoprecipitation (Co‑IP)

The co-immunoprecipitation assay was performed using relevant antibodies in conjunction with the Pierce™ Classic Magnetic IP/Co-IP Kit (Thermo Fisher Scientific). Briefly, cells were lysed on ice in an IP buffer supplemented with protease inhibitors, followed by centrifugation at 14,000 rpm for 20 min at 4 °C. The supernatant was collected and incubated overnight at 4°C with 5 μg of the respective antibody while shaking. Subsequently, 25 μl of pre-washed Protein A/G magnetic beads were added, and the mixture was shaken for 4 h at 4°C. The magnetic beads were then collected, washed sequentially with IP buffer and wash solution buffer, and finally eluted with 100 μl of elution buffer to obtain proteins for western blot analysis.

Statistical analysis

Statistical analysis was conducted using GraphPad Prism 8.0. No animals were excluded from this study. Animals were allocated to experimental groups using a random number generator. Experiments were performed in a blinded manner; mice were coded and randomized by investigators not involved in experimental performance. Sample sizes as well as the definitions of statistical methods andmeasures for each experiment are provided in the corresponding figure legend. The Shapiro–Wilk normality test was performed to determine the data distribution. All data were presented as mean values ± standard deviation (SD) from at least three independent experiments. Differences between two groups were evaluated using the Student’s t-test. Multi-group comparisons were analyzed through one-way ANOVAs or two-way ANOVAs. Statistical significance was defined as P < 0.05. In all statistical analyses, ** p* < 0.05, *** p* < 0.01, and **** p* < 0.001 were used to denote levels of significance.

Supplementary information

Supplementary Table S2 Uncropped western blots Supplementary figures and legends