Adjusting a Protein Modification in Ulcerative Colitis Helps Keep Mitochondria Balanced

DSS (MP Biomedicals, Lot: S2839); anti‐crotonyllysine antibody (Jingjie PTM BioLab, PTM‐502; Lot: 1037267K306; 1:1000 dilution); anti‐succinyllysine antibody (PTM‐419; Lot: 1D05032317L804; 1:1000 dilution); anti‐malonyllysine antibody (PTM‐902; Lot: 23056103K312; 1: 1000 dilution); anti‐β‐hydroxybutyryllysine antibody (PTM‐1201RM; Lot: RL053142; 1:200 dilution); anti‐lactyllysine antibody (PTM‐1401RM; Lot: RL063009; 1:300 dilution); NaCr (Shanghai yuanye Bio‐Technology, Cas No.14379‐00‐1); LEGENDplex multi‐analyte flow assay kit (BioLegend, Cat No.741043); TUBULIN antibody (Proteintech, Cat No. 11224‐1‐AP; 1:5000 dilution); GAPDH antibody (Proteintech, Cat No. 60004‐1‐Ig; 1:5000 dilution); BAX antibody (Proteintech, Cat No. 60267‐1‐Ig; 1:5000 dilution); BCL2 antibody (Proteintech, Cat No. 68103‐1‐Ig; 1:2000 dilution); CS antibody (Abcam, ab129095; 1:1000 dilution); PINK1 antibody (Abcam, ab216144; 1:1000 dilution); PARKIN antibody (Abcam, ab77924; 1:2000 dilution); LC3B antibody (Abcam, ab192890; 1:2000 dilution); P62 antibody (Abcam, ab109012; 1:10000 dilution); NLRP3 antibody (Abcam, ab263899; 1:1000 dilution); primer synthesis (Beijing Tsingke Biotech Co., Ltd).

All animal experiments were authorized by the Ethics Committee of Nanjing University of Chinese Medicine Affiliated Hospital (Approval No. 2023NL‐KS222). Female and male C57BL/6 mice were procured from Jiangsu Huachuangxinnuo Pharmaceutical Technology Co., Ltd. (Certificate No. 202332458), and underwent a 7‐day acclimatization period. Except for those in the control group, mice aged 6–8 weeks were provided with 3% DSS in their drinking water for 7 days and then sacrificed on the seventh day. In the pre‐experiments, we employed NaCr at a dosage of 10 or 20 mg kg⁻¹, with the group treated at a dosage of 20 mg kg⁻¹ demonstrating a more potent antagonistic effect on UC. Mice were gavaged with NaCr or CMC‐Na (0.5%). Body weight, fecal occult blood, and DAI were monitored daily. 24 h after the final treatment, the mice were humanely euthanized by cervical dislocation. Colon tissue and serum samples were collected for following analysis. This study adhered to ethical guidelines and rigorous scientific protocols and met the standards set by the Ethics Committee for Animal Experimentation.

Proteomic analysis was conducted by Jingjie PTM BioLab, Co., Ltd., in China. Quantitative crotonylation modification omics analysis involved a combination of TMT labeling, HPLC fractionation, crotonylation and mass spectrometry‐based quantitative proteomics. WoLF PSORT software was utilized for predicting and categorizing the sublocalization of differentially modified proteins.

Serum samples were utilized to analyze the level of IL‐6, TNF‐α, IL‐10, IL‐1β and other inflammatory factors following the instructions provided with the LEGENDplex multifactor kit (USA). Analysis was performed using Cytoflex flow cytometry.

Human normal intestinal epithelial NCM460 cell line was cultured in DMEM with 1% penicillin‐streptomycin, 10% fetal bovine serum (ABW, China) in a 5% CO₂ humidified atmosphere at 37°C. Lipopolysaccharide (LPS, 2 µg mL⁻¹, Sigma Aldrich, USA) was utilized to induce inflammation in NCM460 cells. CQ (10 µm, TargetMol, China) and Mdivi‐1 (10 µm, OriLeaf, China) were used to inhibit mitophagy. The human CS coding sequences, along with the CS K375R (substitution of AAG with AGA) and CS K375Q (substitution of AAG with CAG) mutants, were synthesized, cloned into the pGCMV/MCS/EGFP/Neo vector, and transfected into cells using Lipo8000 for 24 h under standard conditions.

Total RNA was isolated from cells or colon tissues using a cell/tissue total RNA kit (Vazyme Biotech Co., Ltd, China). Subsequently, the RNA samples were quantified and assessed for purity using a DS‐11FX spectrophotometer (Denovix, USA). Reverse transcription was performed using a first‐strand cDNA synthesis kit (Vazyme Biotech Co., Ltd, China). qPCR analysis was carried out on a QuantStudio3 PCR detection system (Thermo Fisher Scientific, USA), and the results were analyzed using the 2^−ΔΔCT method. The primer sequences for qPCR (Table 1) were synthesized by Tsingke Biotechnology Co., Ltd., (China).

The HPIC separation was performed using a Thermo Scientific Dionex ICS‐6000 HPIC System, utilizing Dionex Ion Pac AS11‐HC (2 × 250 mm) and AG11‐HC (2 mm × 50 mm) columns. Mobile phase A consisted of 100 mm NaOH in water, while mobile phase B was ultrapure water. An additional pumping system delivered a solvent mixture (2 mm acetic acid in methanol) before entering the electrospray ionization (ESI) source at a flow rate of 0.15 mL min⁻¹. The column temperature was maintained at 30°C. The autosampler temperature was set to 4°C, with an injection volume of 5 µL.

For assay development, an AB SCIEX 6500 QTRAP⁺ triple quadrupole mass spectrometer with an ESI interface was employed. The typical ion source parameters included an ion spray voltage of −4500 V, temperature of 450°C, ion source gas 1 at 45 psi, ion source gas 2 at 45 psi, and curtain gas at 30 psi. The multiple reaction monitoring (MRM) parameters for each targeted analyte were optimized via flow injection analysis.

Transmission electron microscopy (TEM) was utilized to examine the ultrastructure of mitochondria in the colon tissue of DSS‐induced mice.

The level of MDA, SOD, and GSH in the serum of DSS‐induced mice was quantified by a colorimetric assay following the provided instructions.

Proteins from colon tissue or cells were extracted using a total protein extraction kit (Yeasen Biotechnology Co., Ltd., China), and the protein concentration was determined using the BCA method (Yeasen Biotechnology Co., Ltd., China). Subsequently, the proteins were separated by 10% SDS‐PAGE and transferred to a nitrocellulose membrane using a Bio‐Rad Trans‐Blot system. Following staining with Lichunhong dye, the bands were washed thrice with PBST. The membranes were then blocked with 5% skimmed milk and incubated overnight at 4°C with primary antibodies (TUBULIN, GAPDH, BAX, BCL2, BAD, BAK, PINK1, PARKIN, LC3, P62, NLRP3, CS). Afterward, the membranes were probed with the secondary antibody for 60 min. Finally, the target proteins were visualized using ECL reagent and quantified with ImageJ software (NIH Image, USA).

Data collected from ≥3 independent assays were presented as mean ± standard deviation (SD). Results were analyzed using a Prism 9.0 software (GraphPad, USA). Comparison between two groups were conducted using the unpaired t test (two‐tailed). Comparison among multiple groups were conducted using one‐way ANOVA and two‐way ANOVA. P < 0.05 was considered statistically significant.

To establish an acute UC mouse model, mice were treated with 3% DSS (wt/vol) for 7 days [24] (Figure 1A). Compared with those of the control group, the colons of the DSS‐induced UC group were markedly shorter and more atrophic (Figure 1B). To determine the differences in PTMs between control and DSS‐induced UC mice, we assessed the level of crotonylation, succinylation, malonylation, and lactylation in the colon tissues of the mice. The analysis revealed a significant alteration in the crotonylation level in colon tissue between the DSS‐induced UC group and control group (Figure 1C and Figure S1). Next, we used proteomics to analyze the differences in the crotonylation level between the UC group and control group. Principal component analysis (PCA), relative standard deviation (RSD) and Pearson's correlation coefficient (PCC) were used to evaluate the quantitative repeatability of the modifications (Figure 1D–F). As shown in Figure 1G, the top three sublocalizations of the differentially modified proteins were in the cytoplasm (404), nucleus (232) and mitochondria (146). The differentially modified proteins were selected for KEGG enrichment analysis, and there was statistically significant enrichment of the differentially modified proteins in the TCA cycle pathway (Figure 1H). Through proteome‐wide enrichment analysis, proteins modified by KCr, such as 2‐hydroxyglutarate dehydrogenase (DLST), CS and succinyl‐CoA ligase, were found to be sublocalized in mitochondria. Alterations of KCr in UC may be linked to changes in mitochondrial enzymes and mitochondrial homeostasis. The level of these proteins was lower in DSS‐induced UC mice than in control mice, and all of them were modified by KCr and sublocalized primarily in mitochondria (Table 2). Therefore, it is postulated that lower KCr level of enzymes in the TCA cycle is associated with UC.

Given the reduced enrichment of KCr observed in DSS‐induced UC mice, we hypothesized that enhancing the general KCr level would be beneficial for alleviating UC. NaCr, a donor for KCr, has been shown to increase the level of KCr both in vivo and in vitro [25]. Also, we confirmed that NaCr promoted KCr as a donor (Figure S2). To evaluate the role of NaCr in the development of UC, we generated DSS‐induced UC mice and gavaged them with CMC‐Na (solvent control), DSS, and NaCr (DSS + NaCr) (Figure 2A). The colon, liver, spleen, thymus tissues, and serum of the mice were collected for subsequent experiments after 7 days treatment. As shown in Figure 2B–D, NaCr‐induced KCr enhancement alleviated the body weight loss, DAI scores increase and reduction in colon length induced by DSS described above. Also, KCr enhancement had a rehabilitative impact on liver, spleen and thymus indices (Figure 2E). KCr restored DSS‐induced damage to the morphology of colon villi and crypts (Figure 2F). Notably, NaCr robustly reduced the pathological score in DSS‐induced UC mice, as determined by H&E staining (Figure 2G). These results indicate that KCr alleviates the histopathological injury in UC mice. Furthermore, KCr can impede the inflammatory progression of UC when NaCr is administered after the establishment of DSS‐induced UC model without any discernible gender‐based differences or associated toxicity, and administration of NaCr at 20 mg kg⁻¹ exerted a superior anti‐inflammatory effect in UC (Figure S3–S6).

The inflammatory response in UC is orchestrated by a multitude of cytokines, including proinflammatory cytokines (such as IFN‐γ, TNF‐α, IL‐1β, IL‐2, IL‐6, IL‐9, and IL‐17A) and anti‐inflammatory cytokines (such as IL‐4 and IL‐10) [26]. Compared with those in the control group, the contents of IFN‐γ, TNF‐α, IL‐2, IL‐6, IL‐9, and IL‐17A in the DSS‐induced UC group were significantly elevated, while the contents of the anti‐inflammatory factors IL‐4 and IL‐10 were significantly reduced. Treatment with NaCr reversed the inflammatory cytokine level changes induced by DSS (Figure 2H). In addition, NCM460 cells were treated with LPS (2 µg mL⁻¹) for 24 h to establish an in vitro inflammation model [27]. We selected 80, 160, and 320 µm as treatment concentrations of NaCr according to the results of the CCK8 assay (Figure 2I). We found that NaCr reversed the upregulation of IL‐6, IL‐1β, TNF‐α, and the downregulation of IL‐4 mRNA expression induced by LPS in NCM460 cells in a concentrations‐dependent manner (Figure 2J). The above results suggest that KCr enhancement impedes the development of UC both in vivo and in vitro.

In the TCA cycle, each reaction is facilitated by a specific enzyme to ensure the smooth progression of the cycle, where the product of each step serves as the substrate for the subsequent reaction (Figure 3A). As shown in Table 2, the KCr level of enzymes involved in the TCA cycle was notably diminished in DSS‐induced UC mice. Hence, the aim of subsequent experiments was to identify the particular enzymes within the TCA cycle for which KCr is crucial for enabling the enzymes to carry out their catalytic function and drive the TCA cycle under UC conditions. Consequently, to determine the activities of CS, DLST, malate dehydrogenase (MDH), succinate dehydrogenase (SDHA), fumarate hydratase (FH), and lactate dehydrogenase (LDH), the level of their respective products was measured to evaluate the enzymatic function. The results indicated a significant reduction in the production of citrate catalyzed by CS in the DSS‐induced group, with a marked increase in citrate level observed following the induction of KCr by NaCr administration (Figure 3B). Conversely, the level of catalytic products of the other enzymes did not exhibit apparent changes (Figure 3C–G). The citrate metabolites of CS reduced in UC due to the down‐regulated KCr level of CS, increased KCr in CS enhances the production of citrate catalyzed by CS.

Previous studies highlighted the close association of KCr with mitochondria, next, TEM was employed to visualize the mitochondrial structure. As shown in Figure 4A, while mitochondria in the control group displayed typical mitochondrial morphology, characterized by a smooth outer membrane, an inner membrane connected to vesicular cristae and an electron‐dense, granular internal matrix, the mitochondria in the DSS‐induced group exhibited a perinuclear clustering, accompanied by the presence of damaged or absent cristae, fragmentation, and irregularities in the outer membrane, and mitochondrial damage was reduced after NaCr treatment. Damaged mitochondria was observed alongside mitophagosomes and autolysosomes [28]. The abundance of these structures was observed to increase after NaCr administration compared to DSS‐induced group (Figures 4A and S7). The equilibrium between mitochondrial function and oxidative stress is crucial for preserving normal cellular metabolism and preventing diseases. Oxidative stress is triggered when mitochondrial dysfunction leads to excessive ROS production or decreased antioxidant defenses [29]. Therefore, we assessed markers of oxidative stress. Compared with those in the control group, the level of MDA, SOD, and GSH in the serum of the DSS‐induced group exhibited notable alterations, while the SOD activity and GSH content were significantly increased and the content of MDA was significantly reduced in the NaCr group (Figure 4B). It is widely acknowledged that inflammation and oxidative stress have the potential to trigger apoptosis [30]. Thus, the expression level of apoptosis‐related proteins was assessed using Western blot and qPCR. In comparison to the control group, the expression of BAX, BAD, and BAK, pro‐apoptotic markers, were increased in colon tissue of the DSS‐induced UC group, whereas the increase in BAX, BAD, and BAK expression was attenuated in the NaCr group. Additionally, the expression of the anti‐apoptotic marker BCL2 was decreased in UC group, and elevated KCr significantly reversed the downregulation of BCL2 expression (Figure 4C–D). Next, we constructed a LPS‐induced UC model in NCM460 cell lines in vitro. Similarly, we observed a substantial increase in the down‐regulated expression of BCL2 following KCr enhancement (Figure 4E,F). These findings suggest that enhanced KCr preserves mitochondrial homeostasis, reduces oxidative stress and impedes apoptosis.

Mitophagy is a critical process for maintaining mitochondrial homeostasis. We postulated that KCr could maintain the mitochondrial homeostasis and enhance mitochondrial function through mitophagy. The level of LC3 and P62, markers of autophagy, was evaluated by Western blot, qPCR and immunofluorescence co‐localization. Compared with control group, elevated level of both LC3II and P62 was observed in the DSS‐induced UC group, indicating a blockade in autophagic flux where the initiation of autophagic processes occurred but efficient substrate degradation was impeded, likely due to compromised autophagic clearance mechanism. Increased KCr level suppressed the elevation of P62 induced by DSS, indicating KCr triggered an unimpeded autophagic flow. However, LC3II expression was also reduced in the NaCr groups, the reason may be due to an excessive activation of autophagic flux (Figure 5A–D). We also observed that compared with control group, the expression level of LC3II and P62 was elevated in the LPS‐induced group. The upregulation of LC3II and P62 in LPS‐induced inflammation was attenuated by KCr, indicating the induction of autophagy by KCr (Figure 5E–G). Given that the majority of KCr was produced in mitochondria, while the mitochondria initiate mitophagy to preserve mitochondrial homeostasis. Additionally, the canonical PINK1/PARKIN‐mediated mitophagy pathway was examined. PINK1 and PARKIN are pivotal regulators involved in maintaining mitochondrial quality and are crucial for mitophagy. The level of PINK1 and PARKIN was down‐regulated in the colon of UC mice in comparison to the control group; conversely, elevated level of KCr resulted in the significant upregulation of PINK1 and PARKIN expression, indicating a positive effect of KCr in restoring mitophagy (Figure 5A–C). Also, the level of PINK1 and PARKIN was decreased in the LPS‐induced group compared to the control group, while the expression level of PINK1 and PARKIN was significantly elevated by KCr (Figure 5E–G). These observations suggest that mitophagy is induced by KCr through the activation of the PINK1/PARKIN pathway, both in DSS‐induced UC mice and in LPS‐induced inflammation.

The NLRP3 inflammasome is recognized as a key regulator in the pathogenesis of various inflammation‐related diseases, including UC [31]. Next, the modulatory effect of NaCr on NLRP3 was assessed through Western blot and qPCR. The results indicated that the expression of NLRP3 was increased in the DSS‐induced UC group compared to the control group, whereas enhanced KCr led to a decrease in NLRP3 expression (Figure 6A–C). Also, compared with control group, NLRP3 expression was elevated in LPS‐induced group, and the expression of NLRP3 was decreased by KCr (Figure 6D–F). Given the ability of NaCr to induce mitophagy, CQ was used to inhibit mitophagy [32]. Subsequently, the reduced of NLRP3 by KCr was reversed by CQ due to the mitophagy inhibition (Figure 6G–I). The results showed that inhibition of mitophagy by Mdivi‐1 was also effective in reversing the downregulation of NLRP3 by KCr (Figure S8). The NLRP3 inflammasome plays a crucial role in the inflammatory cascade, with its activation triggering a cascade of inflammatory responses that impact the generation and secretion of diverse inflammatory cytokines such as IL‐1β and TNF‐α [33]. The mRNA expression level of IL‐1β and TNF‐α in NCM460 cells was assessed by qPCR. Elevated level of KCr reduced inflammatory factor level, while treatment with CQ increased the release of inflammatory cytokines, indicating that increased KCr level induced mitophagy to decrease the production of inflammatory factors (Figure 6I). Therefore, these results suggest that the enhancement of KCr protects against NLRP3 inflammasome activation through the induction of mitophagy.

CS is a rate‐limiting enzyme in the TCA cycle. Biological processes that catalyze the condensation of oxaloacetic acid and acetyl‐CoA to produce citrate play important roles in the TCA cycle and are also involved in regulating mitochondrial homeostasis [34]. Proteomic analysis revealed a decreased level of CS in the colon tissue from DSS‐induced mice (Figure 7A). Subsequently, Western blot and qPCR analyses were employed to assess the expression level of CS. The results showed that the expression of CS was lower in DSS‐induced UC mice than in control mice, whereas enhancement of KCr led to an increase in CS expression (Figure 7B–D). Similarly, CS expression was reduced during LPS‐induced inflammation in comparison to the control group, while KCr elevated CS expression (Figure 7E–G). Additionally, a significant reduction in K375 crotonylation in CS was observed in DSS‐induced mice compared with the control group according to proteomic analysis (Figure 7H). To investigate the correlation between CS K375 crotonylation and mitophagy, site‐specific mutations were introduced in CS. Plasmids were utilized to transfect NCM460 cells with CS mutants in which K375 was substituted with arginine (K375R), a mutant that cannot undergo KCr at the K375 site, or glutamine (K375Q), a KCr‐mimicking mutant. Western blot was conducted to evaluate the impact of alterations in the K375 site in CS on mitophagy and NLRP3. The overexpression of wild‐type (WT) CS had a minimal effect on mitophagy in NCM460 cells. Conversely, mitophagy was significantly increased in NCM460 cells overexpressing the CS K375Q mutant compared to cells expressing the CS WT, while mitophagy was notably reduced in cells expressing the CS K375R mutant. However, compared with cells transfected with the CS WT mutant or GFP controls, cells transfected with the CS K375R mutant exhibited substantially impeded mitophagy following LPS stimulation (Figure 7I,J). By using NaCr treatment or rescuing the phenotype with the K375Q mutant in K375R‐transfected cells, we demonstrated that K375 is the crucial crotonylation site for the induction of mitophagy and inhibition of NLRP3 inflammasome activation (Figure S9). Therefore, it can be inferred that KCr induces mitophagy and inhibits NLRP3 inflammasome activation via the continuous activation of K375 site in CS.