LDHC4 drives lung adenocarcinoma progression by inducing lactylation of RB1 at lysine 900 to disrupt the RB1–E2F1 complex Authors: Wei Peng, Zhi Lin, Cheng Sun, Lingjing Lin, Zhaolei Cui, Donghong Lin Published: 2026 ABSTRACT BackgroundThe metabolic enzyme lactate dehydrogenase C4 (LDHC4) is aberrantly expressed in cancers and linked to poor prognosis. However, its role in lung adenocarcinoma (LUAD) and the molecular mechanisms beyond glycolysis remain unclear. This study investigates whether LDHC4 promotes LUAD by modulating protein lactylation, a lactate-derived post-translational modification, focusing on the tumor suppressor retinoblastoma protein (RB1).MethodsLDHC4 expression and its correlation with clinicopathological features and survival were analyzed using public databases (UALCAN, Kaplan-Meier Plotter, LOGpc) and validated in a cohort of 90 paired LUAD tissues via immunohistochemistry. The functional impact of LDHC4 on proliferation, migration, and invasion was assessed in A549 and PC-9 cells using gain- and loss-of-function models. The global lactylation profile was analyzed using DIA-based lactylation proteomics on the Astral platform. The interaction between RB1 and E2F1 (E2F transcription factor 1) was examined through molecular dynamics simulations, co-immunoprecipitation (Co-IP), and immunofluorescence. The functional consequences of site-specific RB1 lactylation at lysine 900 (RB1–K900lac) were determined using RB1–K900R mutant constructs and cell cycle analysis.ResultsLDHC4 was significantly overexpressed in LUAD tissues, correlating with poor patient survival, and was an independent prognostic risk factor. In vitro, LDHC4 promoted LUAD cell proliferation, migration, and invasion, and its tumor-promoting role was corroborated in an LUAD xenograft model, in which derived tumors exhibited increased volume and weight compared with mock-transfected controls. Mechanistically, LDHC4 overexpression elevated global protein lactylation levels and specifically increased lactylation of RB1. Bioinformatics and molecular dynamics simulations identified K900 as a key conserved residue for RB1–E2F1 binding; its lactylation destabilized the complex by increasing structural fluctuation and weakening intermolecular interactions. Cellular experiments confirmed that the lactylation-resistant RB1–K900R mutant bound E2F1 more strongly than wild-type RB1. Functionally, cells expressing RB1–K900R exhibited suppressed malignant phenotypes and G1/S cell cycle arrest, accompanied by downregulation of CDKs/cyclins and upregulation of P21.ConclusionThis study uncovers a novel LDHC4-driven oncogenic axis in LUAD. LDHC4 facilitates RB1 lactylation at the K900 residue, which disrupts the RB1–E2F1 tumor-suppressive complex, leading to cell cycle dysregulation and tumor progression. These findings may position the “LDHC4–RB1 lactylation” axis as a promising therapeutic target for LUAD.Graphical Abstract Supplementary InformationThe online version contains supplementary material available at 10.1186/s12967-026-08070-9. Introduction Lung cancer continues to be the primary cause of cancer-related deaths worldwide [1, 2]. Among its subtypes, lung adenocarcinoma (LUAD) is the most common [3, 4]. Despite recent advancements in targeted and immunotherapies, the overall prognosis for LUAD patients, particularly those with advanced disease, remains poor [5, 6]. Metabolic reprogramming is a hallmark of cancer cells, and the Warburg effect, characterized by aerobic glycolysis, serves as a cornerstone of energy metabolism in many tumors, including LUAD [7, 8]. This metabolic shift leads to the substantial production and accumulation of lactate [9]. Lactate dehydrogenase (LDH) is the pivotal glycolytic enzyme that catalyzes the reversible conversion of pyruvate to lactate [10]. The LDH family comprises three distinct isoenzymes: LDHA, LDHB, and LDHC [11–13]. LDHA is predominantly expressed in skeletal muscle, where it plays a central role in glycolytic metabolism under conditions such as hypoxia or high energy demand [12]. LDHB is widely expressed in tissues with high oxidative capacity, notably the brain and heart, where it facilitates the conversion of lactate to pyruvate to support aerobic respiration [12]. In contrast, LDHC expression is largely restricted to the germinal epithelium of the mammalian testis and is integral to sperm function and male fertility [12, 13]. The A and B subunits can assemble into various tetrameric isoenzymes (LDH1 to LDH5) [12], whereas the C subunit typically exists exclusively as a homotetramer, LDHC4 (encoded by the LDHC gene), which is classified as a cancer-testis antigen [12]. Studies indicate that LDHC4 plays a distinct role in regulating glycolytic flux and lactate metabolism [13, 14] and is aberrantly expressed in various malignancies [12, 15]. In LUAD, LDHC4 was identified as the most consistently expressed cancer-testis antigen, suggesting its potential as a diagnostic biomarker [16, 17]. LUAD-specific research further confirmed that LDHC4 is expressed in over 80% of tumor tissues, and its expression level correlates significantly with poor patient prognosis and advanced TNM stage; in vitro and in vivo experiments have shown that LDHC4 overexpression promotes lung cancer cell proliferation and tumorigenesis by activating the PI3K/Akt/GSK-3β pathway; this results in upregulated expression of cell cycle regulators (cyclin D1, cMyc) and EMT-related factors (MMP2, MMP9, Vimentin, Twist, Slug, Snail) alongside downregulation of E-cadherin, together enhancing lactate and ATP production, tumor growth, and metastatic potential; moreover, PI3K inhibition reverses these oncogenic effects [16]. LDHC4 overexpression is associated with poor prognosis has also been reported in other cancers, such as renal cell carcinoma [18] and breast cancer [19, 20]. However, the mechanisms by which LDHC4 drives LUAD malignancy beyond its role in tumor metabolism, particularly through epigenetic or post-translational modifications, remain poorly understood. In recent years, lactate has been recognized not only as a metabolic waste product but also as an important signaling molecule and a donor for post-translational modifications [21]. Protein lysine lactylation (Klac), an emerging form of lysine modification directly mediated by lactate, can occur on both histone and non-histone proteins [22, 23]. This modification dynamically regulates protein structure, function, stability, and interaction networks, thereby influencing gene expression and cell fate, and plays a critical role in various biological processes including macrophage polarization and tumorigenesis [23–25]. Within the tumor microenvironment, high lactate concentrations favor extensive protein lactylation, making this modification a key bridge linking metabolic abnormalities to malignant tumor phenotypes [26–28]. In lung cancer, lactylation has been implicated in driving immune evasion and tumor progression under hypoxia [29, 30]. Notably, beyond its canonical role in lactate production, the glycolytic enzyme LDHC4 also functions as an upstream regulator of protein lactylation. Previous studies have identified lysine lactylation as a widespread post-translational modification in multiple cancers; in lung cancer, lactylation has been reported to modify histones (e.g., H3K18la) and several metabolic enzymes, influencing gene expression and tumor progression [29, 30]. In triple negative breast cancer (TNBC), LDHC4 has been specifically shown to catalyze lactylation at lysine 214 (K214) of acetyl-CoA acyltransferase 2 (ACAA2); this modification enhances ACAA2 enzymatic activity, accelerates fatty acid metabolism, and promotes intracellular free fatty acid accumulation, thereby inducing autophagy, stimulating cell cycle progression, and ultimately driving tumor growth [19]. However, its function and the specific lactylation targets through which it may operate in LUAD remain completely unknown. As a glycolytic enzyme responsible for lactate production, LDHC4 is upregulated in LUAD and correlates with poor prognosis [17]. These observations suggest that LDHC4 may promote tumor progression through its role as an upstream regulator of protein lactylation. The retinoblastoma protein (RB1) is a critical gatekeeper at the G1/S cell cycle checkpoint [31]. Its tumor-suppressive function is primarily mediated by the hypophosphorylated form, which binds to the E2F transcription factor family and represses their transcriptional activity to halt cell cycle progression [32, 33]. RB1 activity is finely tuned by various post-translational modifications, including phosphorylation and acetylation [34–36]. Among these, phosphorylation is the best-characterized modification and is catalyzed by cyclin-CDK complexes. This modification induces conformational changes in RB1 that modulate its interaction with binding partners such as E2F [37]. In addition to phosphorylation, methylation also contributes to RB1 regulation; specifically, SMYD2-mediated methylation of RB1 at lysine 810 enhances its phosphorylation at Ser 807/811 both in vitro and in vivo [34]. This combinatorial modification accelerates E2F transcriptional activity and promotes cell cycle progression [34]. Despite extensive investigation of RB1 post-translational modifications, lactylation of RB1 has not yet been documented in lung cancer or other malignancies. Therefore, whether RB1 undergoes lactylation and whether this modification influences its binding affinity for E2F transcription factor 1 (E2F1), thereby consequently modulating cell cycle progression, remain open questions in lung cancer biology. Based on this background and our previous data, we hypothesize that in LUAD, highly expressed LDHC4 may promote specific lactylation of the critical tumor suppressor RB1, thereby destabilizing the RB1–E2F1 complex, releasing cell cycle inhibition, and ultimately driving LUAD proliferation and metastasis. To test this hypothesis, we will first systematically analyze the expression of LDHC4 in LUAD and its relationship with patient prognosis using clinical tissues and public databases. Subsequently, in vitro functional experiments will be conducted to define the impact of LDHC4 on the malignant behavior of LUAD cells. We will then employ Lactylation DIA Proteomics by Astral to screen for key downstream target lactylated proteins of LDHC4, focusing on RB1. Finally, through molecular dynamics simulations, site-specific mutagenesis, and a series of cell and molecular biology experiments, we aim to elucidate the specific molecular mechanism by which LDHC4 promotes lactylation at specific lysine sites (e.g., K900) of RB1, disrupts RB1–E2F1 complex stability, and consequently promotes cell cycle progression. This study aims to reveal the novel role of the LDHC4–RB1 lactylation axis in LUAD progression, providing a new perspective for understanding the crosstalk between tumor metabolism and epigenetic regulation and offering potential targets for metabolic intervention in LUAD therapy. Materials and methods Public data and tools The UALCAN online database was utilized to analyze LDHC4 protein expression levels in the TCGA LUAD dataset, comparing differences with normal tissues and associations with tumor pathological grade and clinical stage [38]. The Kaplan-Meier Plotter database [39] was used to evaluate the correlation between LDHC mRNA expression levels and overall survival (OS) as well as progression-free survival (PFS) of LUAD patients. The LOGpc platform was employed for survival analysis validation using an independent dataset [40]. Immunohistochemistry (IHC) staining Based on the established clinical dataset from the HLugA180Su12 high-density tissue microarray (Shanghai Outdo Biotech, China; Ethics Approval Number: HSYJS-CP-2206002), a total of 90 LUAD patients were analyzed in this study, comprising 90 histologically confirmed cases and 0 patients lost to follow-up. Additionally, 90 matched paracancerous tissue specimens, collected between December, 2013 and July, 2015, were included. The tissue microarray provided annotated clinicopathological data, including patient age, gender, tumor size, clinical stage, and Ki-67 expression status. Clinical follow-up extended through December, 2021. After deparaffinization, rehydration, and antigen retrieval, sections were incubated overnight at 4 °C with a rabbit anti-human LDHC (subunit C) monoclonal antibody (mAb) (Proteintech Group, Inc., 1:500). The next day, horseradish peroxidase (HRP)-conjugated secondary antibody and a DAB chromogen kit were used for color development, followed by hematoxylin counterstaining. Two experienced pathologists independently scored the staining results under double blind conditions. The final expression score was calculated as the product of staining intensity and the percentage of positive cells [41]. Based on the IHC staining score for LDHC4, patients were stratified into two cohorts: a high-expression group (staining intensity from + to +++, corresponding to a quantitative expression score of 3 to 9) and a low/negative expression group (staining intensity from negative to ±, corresponding to a score of 0 to 2 [41]. Cell culture and stable cell line establishment Human LUAD cell lines A549 and PC-9 as well as normal lung epithelial cells BEAS-2B were acquired from the Cell Bank/Stem Cell Bank, Chinese Academy of Sciences. The cell lines tested negative for mycoplasma, bacterial, and fungal contamination, and their identity was confirmed by Short Tandem Repeat (STR) analysis. They were cultured in RPMI-1640 medium, supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin at 37 °C with 5% CO2. To establish the LDHC4 overexpression (LDHC4-OE) model, the full-length human LDHC cDNA was cloned into the lentiviral expression vector CV146. Lentivirus was produced by co-transfection with packaging plasmids in HEK293T cells, used to infect A549 cells, and stable clones were selected using puromycin. To establish the LDHC4 knockdown (Si-LDHC4) model, specific shRNA sequences targeting human LDHC4 were cloned into a lentiviral vector, used to infect PC-9 cells, and stable clones were selected. The following shRNA sequences were used and, for simplicity, are designated as Si1, Si2, and Si3, respectively (5'-3'): Si1, AAGGAGCAGCTAATTGAGAAGCTAA; Si2, TCGTGTAATTGGAAGTGGTTGTAAT; and Si3, CGGATTCAGATAAGGAACACTGGAA. To establish the knockdown model, the above three specific shRNA sequences targeting human LDHC4 were cloned into a lentiviral vector. Gene silencing efficiency was validated in PC-9 cells using RT-PCR and Western blotting. The Si1-LDHC4 sequence, which exhibited the most potent and stable gene knockdown, was selected for all subsequent functional assays and is hereafter uniformly designated as Si-LDHC4 group. Mock or Si-NC cells infected with empty vector virus served as controls. Interference or overexpression efficiency was verified by Western blotting. To investigate the function of RB1 K900 lactylation (RB1–K900lac), the RB1–K900R mutant plasmid was constructed by mutating the codon for lysine (K) at position 900 to arginine (R). This charge-conservative substitution was designed to preserve the electrostatic properties of the RB1 C-terminal domain while specifically abrogating the potential for lactylation at this residue. Such K→R mutations represent a standard and widely accepted methodological approach in lactylation research to dissect modification-specific effects from those attributable to the protein itself [19]. Both wild-type RB1 (RB1–WT) and RB1–K900R sequences were cloned into the lentiviral expression vector GV492 (Genechem, Shanghai). Lentiviruses were packaged by transfection in HEK293T cells and used to infect A549 and PC-9 cells. Stable expression cell lines were selected using puromycin, and expression and mutation were verified by Western blotting and DNA sequencing. Cell counting kit-8 (CCK-8) assay Logarithmic-phase A549 or PC-9 cells were seeded in 96-well plates (5 × 103 cells/well) and transfected with LDHC4 overexpression constructs (LDHC4-OE) or short hairpin RNA (Si-LDHC4), alongside Mock and negative control (Si-NC) vectors. Stable cell lines were established following puromycin selection (2 µg/mL, 96 h) and validated by Western blotting. For proliferation assessment, validated cells were reseeded in 96-well plates at 1 × 104 cells/mL (A549) or 3 × 104 cells/mL (PC-9) in 100 µL medium. After incubation at 37 °C with 5% CO₂ for 0, 24, 48, or 72 h, 10 µL of CCK-8 reagent (Beyotime Biotech, Inc., Shanghai, China) was added to each well, followed by an additional 2-h incubation. Absorbance was measured at 450 nm and 630 nm using a Multiskan spectrophotometer (Thermo Fisher Scientific, USA). All experiments were performed in triplicate. Growth curves were generated by plotting optical density values against time. Plate colony formation assay Logarithmic-phase A549 or PC-9 cells stably expressing the indicated constructs were trypsinized to single-cell suspensions. Cells were seeded in 6-well plates at 50, 100, and 200 cells per well and incubated at 37 °C with 5% CO₂ for 14 to 21 days. Colonies were fixed with methanol, stained with crystal violet, and manually counted. Each plating density was performed in triplicate. Lactate measurement Cellular lactate levels were quantified using a commercial lactate assay kit (Nanjing Jianguo Company, China). The assay is based on the enzymatic oxidation of lactate to pyruvate by lactate dehydrogenase (LDH), with concomitant reduction of NAD+ to NADH. Phenazine methosulfate (PMS) mediates the transfer of hydrogen, facilitating the reduction of nitroblue tetrazolium (NBT) to a purple formazan dye. The absorbance of the formazan product measured at 530 nm is directly proportional to the lactate concentration in the sample. All procedures were performed in accordance with the manufacturer’s instructions. Scratch wound healing assay Cell migration was assessed using a scratch wound assay. A549 or PC-9 cells in the logarithmic growth phase were cultured in 6-well plates until reaching confluence. A uniform linear wound was created in the cell monolayer using a sterile 100 µL pipette tip. Cells were then washed and incubated in serum-free medium to minimize proliferation-driven closure. Wound width was measured immediately after scratching (0 h) and at 24- or 48-h intervals using phase-contrast microscopy (Olympus IX73, Japan). Migration was quantified by determining the mean number of cells that had migrated into the denuded area across three randomly selected high-power fields per well. Transwell invasion assay Invasion capacity was evaluated using Matrigel-coated Transwell chambers. A pre-chilled mixture of serum-free medium and growth factor-reduced Matrigel (Corning, USA) at a 5:1 ratio was prepared and applied to the upper surface of Transwell inserts (100 µL per well). The Matrigel was allowed to polymerize at 37 °C for 4 to 5 h. LUAD cells were trypsinized, washed, counted, and resuspended in serum-free medium. After rehydrating the coated membranes, a cell suspension containing a defined number of cells was added to the upper chamber. The lower compartment was filled with 500 µL of medium supplemented with 20% fetal bovine serum as a chemoattractant. Following 20 to 24 h of incubation at 37 °C, non-invading cells on the upper membrane surface were removed with a cotton swab. Cells that had invaded through the Matrigel were fixed with 4% glutaraldehyde at 4 °C, stained with 0.1% crystal violet, and quantified by counting cells in at least three random microscopic fields per membrane. Establishment of subcutaneous xenograft models A549 cells were maintained in RPMI-1640 medium with 10% fetal bovine serum under standard culture conditions (37 °C, 5% CO₂). Cells in the logarithmic growth phase were trypsinized, washed with phosphate-buffered saline (PBS), and resuspended to a density of 5 × 107 cells/mL. The suspension was then mixed 1:1 with ice-cold Matrigel. Six-week-old female C-NKG mice received a subcutaneous injection of 200 μL of the cell-Matrigel mixture into the right scapular region using a needle depth of 1 cm to minimize leakage. Tumor growth was assessed weekly as previously described [19]. All animal procedures followed the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Laboratory Animal Welfare and Ethics Committee of Cyagen Biosciences (Suzhou) Co., Ltd (Ethics Approval No: TACU25SY027). Humane care was provided throughout the study. Quantitative lactylome profiling analysis Cell samples were dissolved in 200 µL of lysis buffer, boiled, and further subjected to ultrasonication. LC-MS/MS analysis was performed on an Orbitrap Astral mass spectrometer coupled with a Vanquish Neo UHPLC system (Thermo Fisher Scientific). Peptides from each sample were loaded onto a 50 cm Low-Load µPAC™ Neo HPLC Column (Thermo Scientific) at a flow rate of 2.2 µL/min. The reversed-phase HPLC mobile phase A consisted of 0.1% formic acid in water, and mobile phase B consisted of 0.1% formic acid in 80% acetonitrile. Peptides were eluted over 8 min at a flow rate of 1.25 µL/min using a linear gradient of buffer B, programmed as follows: 0–0.1 min, linear gradient from 4% to 6% B; 0.1–1.1 min, linear gradient from 6% to 12% B; 1.1–4.3 min, linear gradient from 12% to 25% B; 4.3–6.1 min, linear gradient from 25% to 45% B; 6.1–6.5 min, linear gradient from 45% to 99% B; and 6.5–8 min, held at 99% B. The eluted peptides were analyzed on an Orbitrap Astral mass spectrometer. The DIA method included a survey scan from 380 to 980 m/z at a resolution of 240,000, with an AGC target of 500% and an injection time of 5 ms. DIA MS/MS scans were acquired on the Astral from 150 to 2000 m/z using 2 m/z isolation windows, with an AGC target of 500% and an injection time of 3 ms. Normalized collision energy was set to 25%, and the cycle time was 0.6 s. Full MS and DIA scan spectra were recorded in profile and centroid mode, respectively. DIA MS data were analyzed using Spectronaut 18 (Biognosys AG, Switzerland). The MS data were searched against the UniProtKB reviewed (Swiss-Prot) database for Homo sapiens (20,427 total entries; downloaded October 2023). The technical support for the work described above was provided by Shanghai Bioprofile Technology. Co-immunoprecipitation (Co-IP) and Western blotting To detect RB1 lactylation and its interaction with E2F1, total cell protein was extracted using RIPA lysis buffer containing protease inhibitors and deacetylase inhibitors. Equal amounts of protein were incubated with anti-RB1 or anti-E2F1 antibody overnight at 4 °C, followed by incubation with Protein A/G agarose beads for 24 h. After washing the beads, bound proteins were eluted by boiling in SDS loading buffer. For immunoblotting, immunoprecipitated products and input proteins were separated by SDS-PAGE, transferred to PVDF membranes, and blocked with 5% skim milk. The membranes were then sequentially incubated with primary antibodies, followed by horseradish peroxidase (HRP)-conjugated secondary antibodies. Protein signals were visualized using an enhanced chemiluminescence (ECL) reagent. The following antibodies were used in this study: rabbit anti-human LDHC (subunit C) monoclonal antibody (mAb) (Proteintech Group, Inc., 1:1000), rabbit anti-human RB1 mAb (Proteintech Group, Inc., 1:2000), rabbit anti-human L-Lactyl Lysine mAb (PTM Bio, Inc., 1:1000), rabbit anti-human E2F1 mAb (APExBIO Technology, LLC, 1:500), rabbit anti-human Lamin B mAb (Beyotime Biotech, Inc., 1:1000), rabbit anti-human CDK1 mAb (Beyotime Biotech, Inc., 1:800), rabbit anti-human CDK2 mAb (Beyotime Biotech, Inc., 1:800), rabbit anti-human CDK4 mAb (Beyotime Biotech, Inc., 1:1000), rabbit anti-human CDK6 mAb (Beyotime Biotech, Inc., 1:1000), rabbit anti-human cyclin A2 mAb (Beyotime Biotech, Inc., 1:1000), rabbit anti-human cyclin B1 mAb (Beyotime Biotech, Inc., 1:1000), rabbit anti-human cyclin D1 mAb (Beyotime Biotech, Inc., 1:500), rabbit anti-human P21 mAb (Abmart Bio, Inc., 1:500), rabbit anti-human Chk1 mAb (Immunoway Bio, Inc., 1:2000), and rabbit anti-β-Actin monoclonal antibody (Beyotime Biotech, Inc., 1:1000). β-Actin and Lamin B served as loading controls, and protein band intensities were quantified using Image J software. Molecular dynamics simulations The structure of the human RB1 protein in complex with E2F1 was obtained from the Protein Data Bank (PDB). Models for wild type RB1 (WT), and RB1 with lactylation at the K900 site (K900lac) were constructed using the Protenix server. All molecular dynamics simulations were performed using the GROMACS software package and the CHARMM36 force field in explicit solvent. Systems were equilibrated through energy minimization, NVT, and NPT steps before a 200 ns production run. The root mean square deviation (RMSD), root mean square fluctuation (RMSF), interaction energy between key residues of RB1 and E2F1, hydrogen bond number, and distance were analyzed. For a comprehensive description of the analytical methodology, readers are directed to our prior publication, which details the experimental procedures and analytical workflow [19]. Cell cycle analysis Cells were fixed overnight at −20 °C with 70% ethanol. Fixed cells were washed with PBS, incubated for 30 min at room temperature in the dark with a staining buffer containing RNase A and propidium iodide (PI). Cellular DNA content was analyzed using a flow cytometer, and the distribution of cells in different cell cycle phases (G0/G1, S, G2/M) was determined using FlowJo V10 software. Immunofluorescence staining Cells were seeded on coverslips, grown to appropriate density, fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and blocked with 5% bovine serum albumin (BSA). Subsequently, cells were incubated with primary antibodies overnight at 4 °C. The next day, cells were incubated with corresponding fluorescent dye (Cy3) labeled secondary antibodies for 1 h at room temperature in the dark. Nuclei were stained with blue fluorescent 4’,6-diamidino-2-phenylindole (DAPI). After mounting, images were observed and captured using a High-Resolution Confocal Multiphoton Microscopy System (NIKON AX RMP, Japan). Statistical analysis All experiments were independently repeated at least three times. Data are presented as mean ± standard deviation. Comparisons between two groups were performed using Student’s t-test for normally distributed data or the Mann-Whitney U test for non-normally distributed data. Comparisons among multiple groups were performed using one-way ANOVA followed by Tukey’s post hoc test. Survival analysis was performed using the Kaplan-Meier method with the log-rank test. Correlation analysis was performed using Spearman’s rank correlation test. Multivariate analysis was performed using Cox proportional hazards regression models. A p value of less than 0.05 was considered statistically significant. All statistical analyses were performed using GraphPad Prism 10.2.3, SPSS 29.0.1.0 (171), or R (4.5.2) software. The mechanism diagram was created using the BioGDP platform [42]. Results Elevated LDHC4/LDHC expression is associated with poor prognosis in LUAD To investigate the clinical significance of LDHC4/LDHC in LUAD, we first analyzed its protein expression levels using the UALCAN database. Our analysis showed a significant upregulation of LDHC4 levels in LUAD tissues compared to normal lung tissues (Fig. 1A). Further analysis indicated a positive correlation between LDHC4 expression levels and both the pathological grade (Fig. 1B) and clinical stage (Fig. 1C) of LUAD, implicating its potential involvement in tumor progression. LDHC4 expression across different lung cancer subtypes is also shown in Fig. 1D. Using the Kaplan-Meier Plotter database, we also found that high LDHC gene expression was associated with significantly shorter overall survival (OS) and progression-free survival (PFS) in LUAD patients (Fig. 1E, F). Prognostic analysis at the gene level similarly revealed that high LDHC expression was associated with poor patient outcomes (Fig. 1G). Fig. 1Elevated LDHC4/LDHC expression is associated with poor prognosis in LUAD. (A) Expression of LDHC4 protein in LUAD tissues compared with normal controls, based on UALCAN database analysis. (B, C) Association between LDHC4 expression and (B) pathological grade and (C) clinical stage of LUAD in UALCAN. (D) LDHC4 expression levels in different lung cancer subtypes. For panels A to D, an unpaired two-sample t-test was used to assess the statistical significance of expression level differences between any two sample groups. (E, F) Kaplan-Meier survival curves for (E) overall survival (OS) and (F) progression-free survival (PFS) according to LDHC expression levels, generated from the Kaplan-Meier plotter database. (G) Kaplan-Meier OS curve analyzing the relationship between LDHC expression and survival in lung cancer, using the LOGpc platform. (H) Representative immunohistochemistry (IHC) staining showing LDHC4 protein (subunit C) expression in LUAD tissue cells based on a commercial tissue-chip. (I) Quantification of LDHC4 expression on a tissue microarray comparing LUAD (n = 90) and adjacent non-tumor (n = 90) tissues; higher scores indicate greater LDHC4 expression. (J, K) Positive rate and expression level of LDHC4 in LUAD tissues versus adjacent non-cancerous tissues. (L) The correlation between LDHC4 and Ki-67 levels in tissue-chip cohorts was analyzed using R software. (M) Kaplan-Meier analysis of OS was performed with R software for LUAD tissue-chip cohorts stratified based on low and high LDHC4 expression levels. (N, O) Forest plot illustrating the independent impact of LDHC4 levels on OS in LUAD patients. ns, not significant; ***p<0.001 Elevated LDHC4/LDHC expression is associated with poor prognosis in LUAD. (A) Expression of LDHC4 protein in LUAD tissues compared with normal controls, based on UALCAN database analysis. (B, C) Association between LDHC4 expression and (B) pathological grade and (C) clinical stage of LUAD in UALCAN. (D) LDHC4 expression levels in different lung cancer subtypes. For panels A to D, an unpaired two-sample t-test was used to assess the statistical significance of expression level differences between any two sample groups. (E, F) Kaplan-Meier survival curves for (E) overall survival (OS) and (F) progression-free survival (PFS) according to LDHC expression levels, generated from the Kaplan-Meier plotter database. (G) Kaplan-Meier OS curve analyzing the relationship between LDHC expression and survival in lung cancer, using the LOGpc platform. (H) Representative immunohistochemistry (IHC) staining showing LDHC4 protein (subunit C) expression in LUAD tissue cells based on a commercial tissue-chip. (I) Quantification of LDHC4 expression on a tissue microarray comparing LUAD (n = 90) and adjacent non-tumor (n = 90) tissues; higher scores indicate greater LDHC4 expression. (J, K) Positive rate and expression level of LDHC4 in LUAD tissues versus adjacent non-cancerous tissues. (L) The correlation between LDHC4 and Ki-67 levels in tissue-chip cohorts was analyzed using R software. (M) Kaplan-Meier analysis of OS was performed with R software for LUAD tissue-chip cohorts stratified based on low and high LDHC4 expression levels. (N, O) Forest plot illustrating the independent impact of LDHC4 levels on OS in LUAD patients. ns, not significant; ***p<0.001 To corroborate these findings at the tissue level, we assessed LDHC4 expression in a tissue-chip cohort by IHC. IHC staining demonstrated relatively high LDHC4 expression in LUAD tumor cells (Fig. 1H). Quantitative scoring of a tissue microarray containing 90 paired LUAD and adjacent tissues confirmed that LDHC4 expression scores were significantly higher in tumor tissues than in matched adjacent non-tumor tissues (Fig. 1I, K). Similarly, the positive rate of LDHC4 was significantly higher in LUAD tissues compared to their adjacent counterparts (Fig. 1J). Analysis of clinicopathological characteristics showed that LDHC4 expression in LUAD tissues was significantly correlated with patient gender (χ2 = 7.571; p = 0.006), but not with histopathological grade, T stage, or lymph node metastasis (Table 1). However, LDHC4 expression levels in LUAD tissues showed a positive correlation with the proliferation marker Ki-67 (Fig. 1L), suggesting a link to active cell proliferation. Finally, survival analysis based on the tissue-chip cohort reaffirmed that high LDHC4 expression represents an independent risk factor for shortened OS in LUAD patients (Fig. 1M, N, O). Table 1Correlations between LDHC4 protein expressions and clinical characteristics in LUAD patientsClinicopathological featuresTotal case size (n = 90)LDHC4 low (n = 49)LDHC4 high (n = 41)χ² valuep valueAge (years) < 604224180.2310.631 ≥ 60482523Gender Male4518277.5710.006 Female453114Pathological grading < 22817110.6440.422 ≥ 2623230T stage T1~29044360.0900.765 T3~41055Lymphatic metastasis Yes3415192.3490.125 No463422Distant metastasis Yes000// No904941Clinical stage I~II7037330.3200.572 ≥ III20128LUAD: lung adenocarcinoma, LDHC4: lactate dehydrogenase C4 Table 1Correlations between LDHC4 protein expressions and clinical characteristics in LUAD patientsClinicopathological featuresTotal case size (n = 90)LDHC4 low (n = 49)LDHC4 high (n = 41)χ² valuep valueAge (years) < 604224180.2310.631 ≥ 60482523Gender Male4518277.5710.006 Female453114Pathological grading < 22817110.6440.422 ≥ 2623230T stage T1~29044360.0900.765 T3~41055Lymphatic metastasis Yes3415192.3490.125 No463422Distant metastasis Yes000// No904941Clinical stage I~II7037330.3200.572 ≥ III20128LUAD: lung adenocarcinoma, LDHC4: lactate dehydrogenase C4 Correlations between LDHC4 protein expressions and clinical characteristics in LUAD patients LUAD: lung adenocarcinoma, LDHC4: lactate dehydrogenase C4 LDHC4 promotes malignant phenotypes in LUAD cells and xenograft model To elucidate the biological function of LDHC4, we performed functional experiments in LUAD cell lines. Initial screening confirmed endogenous LDHC4 expression in both A549 and PC-9 cell lines and BEAS-2B normal human bronchial epithelial cells (Fig. 2A). We selected A549 cells with relatively low endogenous LDHC4 expression for overexpression experiments and PC-9 cells with relatively high LDHC4 expression for knockdown experiments. Subsequently, we established an LDHC4 overexpression (LDHC4-OE) model in A549 cells (Fig. 2B) and an LDHC4 knockdown (Si-LDHC4) model in PC-9 cells (Fig. 2C). Fig. 2LDHC4 promotes invasion, migration, and proliferation in LUAD cells. (A) Endogenous expression levels of LDHC4 in the A549, PC-9, and BEAS-2B cell lines. (B) Lentivirus-mediated overexpression of LDHC4 in A549 cells. (C) Lentivirus-mediated knockdown (shRNA1 to shRNA3) of LDHC4 in PC-9 cells. Based on its superior silencing efficiency and stability, Si1-LDHC4 was chosen for all subsequent LDHC4 knockdown assays and uniformly renamed Si-LDHC4. (D) Cell viability measured by CCK-8 colorimetric assay. (E) Colony formation capacity. (F) Measurement of lactate levels. (G, H) Cell migration and (I, J) Invasion capabilities assessed by wound healing and Transwell invasion assays, respectively, in A549 and PC-9 cell lines stably infected with lentiviruses encoding LDHC4-OE, LDHC4-Mock, Si-LDHC4, or Si-NC vectors. (K to P) Tumor formation was monitored in C-NKG mice (n = 5 per group) that were subcutaneously inoculated with either A549 cells overexpressing LDHC4 (LDHC4-OE) or control A549 cells (-Mock), as well as PC-9 cells with LDHC4 knockdown (Si-LDHC4) or corresponding controls (Si-NC). Tumor growth curves and the weights of the excised tumors are presented. ns, not significant; ***p < 0.001 LDHC4 promotes invasion, migration, and proliferation in LUAD cells. (A) Endogenous expression levels of LDHC4 in the A549, PC-9, and BEAS-2B cell lines. (B) Lentivirus-mediated overexpression of LDHC4 in A549 cells. (C) Lentivirus-mediated knockdown (shRNA1 to shRNA3) of LDHC4 in PC-9 cells. Based on its superior silencing efficiency and stability, Si1-LDHC4 was chosen for all subsequent LDHC4 knockdown assays and uniformly renamed Si-LDHC4. (D) Cell viability measured by CCK-8 colorimetric assay. (E) Colony formation capacity. (F) Measurement of lactate levels. (G, H) Cell migration and (I, J) Invasion capabilities assessed by wound healing and Transwell invasion assays, respectively, in A549 and PC-9 cell lines stably infected with lentiviruses encoding LDHC4-OE, LDHC4-Mock, Si-LDHC4, or Si-NC vectors. (K to P) Tumor formation was monitored in C-NKG mice (n = 5 per group) that were subcutaneously inoculated with either A549 cells overexpressing LDHC4 (LDHC4-OE) or control A549 cells (-Mock), as well as PC-9 cells with LDHC4 knockdown (Si-LDHC4) or corresponding controls (Si-NC). Tumor growth curves and the weights of the excised tumors are presented. ns, not significant; ***p < 0.001 Functional assays demonstrated that LDHC4 overexpression significantly enhanced the proliferative viability (assessed by CCK-8 assay; Fig. 2D) and colony-forming capacity (Fig. 2E) of A549 cells; conversely, LDHC4 knockdown inhibited the proliferative viability (Fig. 2D) and colony formation (Fig. 2E) of PC-9 cells. In parallel experiments, lactate levels were significantly higher in cells overexpressing LDHC4 relative to controls. Conversely, knockdown of endogenous LDHC4 in PC-9 cells resulted in a significant reduction in lactate levels (Fig. 2F). Furthermore, wound healing and Transwell migration and invasion assays showed that LDHC4-OE promoted, while Si-LDHC4 inhibited, the migration and invasion abilities of LUAD cells (Fig. 2G to J). Evaluation of an LUAD xenograft model demonstrated that tumors from LDHC4-overexpressing A549 cells exhibited increased volume and weight compared with Mock-transfected controls (Fig. 2K to M), whereas knockdown of endogenous LDHC4 in PC-9 cells significantly attenuated their tumorigenic capacity (Fig. 2N to P). These results collectively indicate that LDHC4 promotes the progression of LUAD cells in vitro and in vivo. LDHC4 overexpression alters the protein lactylation landscape in LUAD cells To explore the mechanism by which LDHC4 promotes tumor progression, we employed DIA Proteomics by Astral to compare global protein lactylation modifications between LDHC4-OE A549 cells and -Mock control cells (Fig. 3A). All modification-level data were normalized to total protein quantification (Supplementary Fig. S1). Volcano plots identified a set of significantly differentially lactylated proteins: these proteins comprised 122 with upregulated lactylation (involving 338 sites) and 107 with downregulated lactylation (involving 225 sites) (Fig. 3B, C), and a heatmap depicted their expression patterns (Fig. 3D). Gene Ontology (GO) enrichment analysis revealed that these differentially lactylated proteins were significantly enriched in biological processes including the cell cycle (Fig. 3E, red dashed box). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis similarly indicated the cell cycle as one of the most significantly affected pathways (Fig. 3F, red dashed box). Gene Set Enrichment Analysis (GSEA) further supported the enrichment of lactylated proteins in lactylation-related pathways (Fig. 3G). An integrated analysis showed altered lactylation levels in multiple key proteins involved in cell cycle regulation (Fig. 3H). Detailed information regarding the lactylation modifications of all identified cell cycle-related proteins is presented in Table S1. Experimental validation showed that overall protein lactylation levels were elevated in LDHC4-OE A549 cells (Fig. 3I). Fig. 3Differential lactylation profiling in LDHC4-OE A549 LUAD cells by DIA-based proteomics on the Astral platform. (A) Schematic workflow of DIA-based lactylation proteomics analysis. (B) Volcano plot displaying differentially lactylated proteins. (C) The numbers of screened differentially lactylated proteins and modification sites. (D) Heatmap of differentially lactylated proteins. (E) Gene ontology (GO) functional classification of lactylated proteins; the red outline highlights enrichment in cell cycle-related terms. (F) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis indicating that differentially lactylated proteins are involved in processes such as cell cycle regulation (red dashed box). (G) Gene set enrichment analysis (GSEA) showing enrichment of lactylation-related pathways among the differential proteins. (H) Summary of altered proteins associated with the cell cycle pathway based on integrated lactylome analysis. Altered protein and lactylation levels in LDHC4-OE versus -Mock control are shown. (I) Detection of global lactylated proteins in LDHC4-OE A549 cells using a pan-lactylation antibody. (J, K) Increased RB1 lactylation levels in LDHC4-OE A549 cell lines, as confirmed by co-immunoprecipitation (Co-IP) and Western blotting. ***p < 0.001 Differential lactylation profiling in LDHC4-OE A549 LUAD cells by DIA-based proteomics on the Astral platform. (A) Schematic workflow of DIA-based lactylation proteomics analysis. (B) Volcano plot displaying differentially lactylated proteins. (C) The numbers of screened differentially lactylated proteins and modification sites. (D) Heatmap of differentially lactylated proteins. (E) Gene ontology (GO) functional classification of lactylated proteins; the red outline highlights enrichment in cell cycle-related terms. (F) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis indicating that differentially lactylated proteins are involved in processes such as cell cycle regulation (red dashed box). (G) Gene set enrichment analysis (GSEA) showing enrichment of lactylation-related pathways among the differential proteins. (H) Summary of altered proteins associated with the cell cycle pathway based on integrated lactylome analysis. Altered protein and lactylation levels in LDHC4-OE versus -Mock control are shown. (I) Detection of global lactylated proteins in LDHC4-OE A549 cells using a pan-lactylation antibody. (J, K) Increased RB1 lactylation levels in LDHC4-OE A549 cell lines, as confirmed by co-immunoprecipitation (Co-IP) and Western blotting. ***p < 0.001 Our previous findings demonstrated that LDHC4 is highly expressed in LUAD and is significantly associated with poor patient prognosis. In vitro cellular functional assays and lactylation sequencing further indicated that its oncogenic mechanism involves the cell cycle pathway, with RB1 showing the most pronounced differential lactylation level (3.77-fold increase) among the seven identified differentially lactylated proteins (Supplementary Table S1). RB1 is a key regulator of the cell cycle [31], which functions by binding to and inhibiting members of the E2F transcription factor family, thereby preventing the transition from the G1 to S phase and maintaining genomic stability. Loss of RB1 function not only leads to dysregulated cell cycle progression but may also promote tumor development by affecting cell differentiation, the immune microenvironment, and metabolic reprogramming [32, 33]. Additionally, RB1 activity is finely regulated by various post-translational modifications, including phosphorylation, acetylation, and ubiquitination [34–36, 43]. However, lactylation of RB1 has not been reported to date. Therefore, to investigate this further, we used co-immunoprecipitation (Co-IP) combined with Western blotting and confirmed that lactylation levels of RB1 were significantly increased in LDHC4-overexpressing A549 cells (Fig. 3J, K). The findings above indicate that LDHC4 overexpression upregulates the lactylation level of the cell cycle protein RB1. Lactylation at the K900 site of RB1 disrupts the stability of its complex with E2F1 Bioinformatics analysis indicated that lysine 900 (K900) of the RB1 protein is highly evolutionarily conserved (Fig. 4A) and that this region constitutes a critical interface for RB1 binding to the transcription factor E2F1, with a high predicted binding affinity (Fig. 4B). We therefore sought to determine how lactylation at this critical site (K900) affects the structural dynamics of RB1 and its interaction with E2F1. Molecular dynamics simulations showed that lactylation modification at the K900 site (K900lac) altered the conformational stability of RB1 itself (Fig. 4C, D). More importantly, compared to wild-type RB1 (RB1–WT), K900lac led to increased overall structural fluctuation (higher root mean square deviation [RMSD]) of the RB1–E2F1 complex (Fig. 4E), weakened key residue interactions at their interface (Fig. 4F, G), and reduced the number and stability of hydrogen bonds (Fig. 4H). Fig. 4Effect of RB1 lactylation at K900 (RB1–K900lac) on the stability of the RB1–E2F1 complex. (A) High evolutionary conservation of the K900 residue. (B) High-affinity interaction between RB1 and E2F1. (C) Dynamic comparison of root mean square deviation (RMSD) for wild-type (WT) RB1 protein versus RB1 with lactylation at K900 (K900lac). (D) Distribution of root mean square fluctuation (RMSF) for residues in RB1–WT versus RB1–K900lac. (E) Overall RMSD of the E2F1–RB1 complex over a 0–200 ns simulation period. (F) Representative conformations and (G) detailed key residue interactions of the wild-type RB1–E2F1 and RB1–K900R–E2F1 complexes. (H) Time-dependent variation in hydrogen bonds between the residue at position 900 of RB1 and the GLN260 residue of E2F1. (I) Nuclear co-localization of RB1–WT or RB1–K900R with E2F1 in A549 and PC-9 LUAD cell lines imaged using high-resolution confocal multiphoton microscopy system (NIKON AX RMP, Japan). E2F1 protein was labeled with red fluorescent Cy3, and nuclei were counterstained with blue fluorescent 4’,6-diamidino-2-phenylindole (DAPI). (J, K) Analysis of RB1–E2F1 co-immunoprecipitation in A549 and PC-9 cell lines expressing RB1–WT or RB1–K900R. (L, M) Analysis of RB1 protein lactylation levels in PC-9 cell lines expressing RB1–WT or RB1–K900R. ***p < 0.001 (vs. RB1–WT group) Effect of RB1 lactylation at K900 (RB1–K900lac) on the stability of the RB1–E2F1 complex. (A) High evolutionary conservation of the K900 residue. (B) High-affinity interaction between RB1 and E2F1. (C) Dynamic comparison of root mean square deviation (RMSD) for wild-type (WT) RB1 protein versus RB1 with lactylation at K900 (K900lac). (D) Distribution of root mean square fluctuation (RMSF) for residues in RB1–WT versus RB1–K900lac. (E) Overall RMSD of the E2F1–RB1 complex over a 0–200 ns simulation period. (F) Representative conformations and (G) detailed key residue interactions of the wild-type RB1–E2F1 and RB1–K900R–E2F1 complexes. (H) Time-dependent variation in hydrogen bonds between the residue at position 900 of RB1 and the GLN260 residue of E2F1. (I) Nuclear co-localization of RB1–WT or RB1–K900R with E2F1 in A549 and PC-9 LUAD cell lines imaged using high-resolution confocal multiphoton microscopy system (NIKON AX RMP, Japan). E2F1 protein was labeled with red fluorescent Cy3, and nuclei were counterstained with blue fluorescent 4’,6-diamidino-2-phenylindole (DAPI). (J, K) Analysis of RB1–E2F1 co-immunoprecipitation in A549 and PC-9 cell lines expressing RB1–WT or RB1–K900R. (L, M) Analysis of RB1 protein lactylation levels in PC-9 cell lines expressing RB1–WT or RB1–K900R. ***p < 0.001 (vs. RB1–WT group) Cellular experiments confirmed these simulation results. In both A549 and PC-9 cells, the mutation of K900 to arginine (RB1–K900R, mimicking a de-lactylated state) significantly enhanced the nuclear co-localization of RB1 with E2F1 compared to RB1–WT (Fig. 4I). Co-IP experiments further demonstrated that the RB1–K900R mutant exhibited a significantly stronger binding affinity for E2F1 than RB1–WT (Fig. 4J, K). Concurrently, RB1 lactylation was reduced in the RB1–K900R mutant (Fig. 4L, M). These results indicate that lactylation at the K900 site of RB1 weakens its binding to E2F1, which may release RB1-mediated transcriptional repression of E2F1. Lactylation at the RB1 K900 site promotes malignant progression of LUAD cells To validate the functional consequence of the RB1–K900 site, we constructed lentiviral expression plasmids for RB1–WT and RB1–K900R (Fig. 5A) and established stably expressing A549 and PC-9 cell lines (Fig. 5B to D). Functional assays showed that, compared to cells expressing RB1–WT, cells expressing RB1–K900R (lactylation-defective) exhibited significantly inhibited proliferative viability (CCK-8 assay; Fig. 5E), colony-forming capacity (Fig. 5F), and migration and invasion ability (Fig. 5G to J). Fig. 5Impact of RB1–K900lac on LUAD progression. (A) Construction and validation of wild-type and mutant (K900 to arginine) RB1 lentiviral expression plasmids. (B to D) Establishment of A549 and PC-9 cell lines stably expressing the RB1 constructs. (E) Analysis of cell viability by CCK-8 assay in cells expressing RB1–WT or RB1–K900R via lentiviral transduction. (F) Colony formation capacity. (G to J) Migration and invasion abilities assessed by wound healing and Transwell assays in A549 and PC-9 cell lines stably expressing RB1–WT or RB1–K900R via lentiviral vectors. Panels H and J show quantitative analyses of the wound healing assay and the Transwell invasion assay, respectively. ns, not significant; **p < 0.01, ***p < 0.001 Impact of RB1–K900lac on LUAD progression. (A) Construction and validation of wild-type and mutant (K900 to arginine) RB1 lentiviral expression plasmids. (B to D) Establishment of A549 and PC-9 cell lines stably expressing the RB1 constructs. (E) Analysis of cell viability by CCK-8 assay in cells expressing RB1–WT or RB1–K900R via lentiviral transduction. (F) Colony formation capacity. (G to J) Migration and invasion abilities assessed by wound healing and Transwell assays in A549 and PC-9 cell lines stably expressing RB1–WT or RB1–K900R via lentiviral vectors. Panels H and J show quantitative analyses of the wound healing assay and the Transwell invasion assay, respectively. ns, not significant; **p < 0.01, ***p < 0.001 RB1–K900 lactylation promotes tumor progression by affecting the cell cycle pathway To elucidate the downstream mechanism, cell cycle analysis by flow cytometry revealed that, compared to the RB1–WT group, a significantly higher proportion of cells in the RB1–K900R group were arrested in the G1/S phase (Fig. 6A to C). Immunofluorescence preliminarily confirmed reduced nuclear expression of CDK1 in RB1–K900R cells (Fig. 6D). Western blot analysis further showed downregulated expression of cyclin-dependent kinases CDK1, CDK2, CDK4, and CDK6 in RB1–K900R cells (Fig. 6E, F). Correspondingly, expression of the cyclins A2, B1, and D1 was reduced (Fig. 6G, H), while expression of the cell cycle negative regulators P21 and Chk1 was upregulated in RB1–K900R cells (Fig. 6I, J). Fig. 6Influence of RB1–K900lac on the cell cycle pathway. (A to C) Cell cycle analysis by flow cytometry in A549 and PC-9 cell lines stably expressing RB1–WT or RB1–K900R via lentiviral vectors. (D) Representative immunofluorescence images showing the distribution of CDK1 in A549 and PC-9 cells. CDK1 protein was labeled with red fluorescent Cy3, and nuclei were counterstained with blue fluorescent DAPI. Images were acquired using a high-resolution confocal multiphoton microscopy system (NIKON AX RMP, Japan). (E, F) Western blot analysis of cell cycle-related CDK molecule expression. (G, H) Expression of cell cycle-related cyclin molecules. (I, J) Expression of P21 and Chk1 molecules. (K) Schematic diagram illustrating how LDHC4 promotes the cell cycle by inducing RB1 lactylation. **p < 0.01, ***p < 0.001 Influence of RB1–K900lac on the cell cycle pathway. (A to C) Cell cycle analysis by flow cytometry in A549 and PC-9 cell lines stably expressing RB1–WT or RB1–K900R via lentiviral vectors. (D) Representative immunofluorescence images showing the distribution of CDK1 in A549 and PC-9 cells. CDK1 protein was labeled with red fluorescent Cy3, and nuclei were counterstained with blue fluorescent DAPI. Images were acquired using a high-resolution confocal multiphoton microscopy system (NIKON AX RMP, Japan). (E, F) Western blot analysis of cell cycle-related CDK molecule expression. (G, H) Expression of cell cycle-related cyclin molecules. (I, J) Expression of P21 and Chk1 molecules. (K) Schematic diagram illustrating how LDHC4 promotes the cell cycle by inducing RB1 lactylation. **p < 0.01, ***p < 0.001 In summary, our findings reveal a mechanism by which high expression of LDHC4 in LUAD induces lactylation modification at the K900 site of the RB1 protein, thereby destabilizing the RB1–E2F1 complex, which ultimately releases the inhibition on cell cycle progression and promotes LUAD advancement (Fig. 6K). Discussion The progression of lung adenocarcinoma (LUAD) involves complex metabolic reprogramming and epigenetic dysregulation [6, 9, 29, 30]. By integrating bioinformatics, clinical sample analysis, functional experiments, quantitative lactylome profiling, and computational biology, this study systematically reveals a novel molecular mechanism through which LDHC4 drives LUAD malignancy. We demonstrate that LDHC4 induces lactylation at the K900 site of the tumor suppressor RB1, leading to disruption of the RB1–E2F1 complex. This discovery not only deepens the understanding of LDHC4's oncogenic function but also establishes a direct link between tumor metabolism and core cell cycle regulation via the emerging post-translational modification of protein lactylation. Our findings provide the first functional confirmation of the oncogenic role of LDHC4 in LUAD and establish its clinical significance as an independent biomarker of poor prognosis. Previous work has demonstrated that LDHC, as a member of the cancer/germline antigen family, is expressed in a substantial proportion of non-small cell lung cancer (NSCLC) specimens and, given its immunogenic properties, may serve as a promising target for multi-antigen cancer vaccines [44]. Building upon this foundation, our study refines the clinical relevance of LDHC4 specifically in the LUAD subtype, providing functional evidence for its tumor-promoting role and highlighting its potential not only as a prognostic biomarker but also as a candidate for targeted therapeutic strategies. Both the TCGA database and our cohort of 90 matched clinical samples consistently show high LDHC4 expression in LUAD tissues, which positively correlates with tumor pathological grade and clinical stage. More importantly, survival analysis using the Kaplan-Meier Plotter database indicated that high LDHC4 expression is associated with significantly shorter OS and PFS, a finding validated in the independent LOGpc dataset [40]. The robustness of this bioinformatic analysis is supported by the database’s use of real-time calculations, a manually curated clinical database, and correction for multiple hypothesis testing. Our subsequent multivariate Cox analysis based on the tissue-chip cohort confirms that high LDHC4 expression is an independent risk factor affecting LUAD patient prognosis, further solidifying its candidacy as a molecular indicator for clinical prognosis assessment. At the functional level, we confirmed that LDHC4 significantly promotes the proliferation, colony formation, migration, and invasion of LUAD cells. This aligns with prior findings that LDHC4 promotes lung cancer cell proliferation via PI3K/Akt pathway activation [16], but our study extends its oncogenic function to include the critical metastasis-related phenotypes of migration and invasion. The significant positive correlation between LDHC4 expression and the proliferation marker Ki-67 further corroborates its role in promoting tumor growth and provides the rationale for investigating its impact on cell cycle regulation. Traditionally, protein lysine lactylation (Klac) is a finely regulated process involving the coordinated activity of multiple members of the LDH family. Emerging evidence indicates that LDHA-mediated lactate production is critical for histone lactylation in multiple pathophysiological processes [45–47], while LDHB expression modulates the availability of lactate as a substrate for protein lactylation [48]. As a key isozyme within the LDH family, LDHC4 primarily catalyzes the interconversion between pyruvate and lactate [12, 13, 16]. In the present study, we demonstrate that LDHC4 significantly promotes lactate production in LUAD cell lines. More importantly, lactylome analysis (LDHC4-OE vs. -Mock) revealed that, after excluding the confounding effects of endogenous LDHA- and LDHB-mediated lactylation, the net effect of LDHC4 upregulation was a marked increase in global protein lactylation levels in A549 cells, identifying LDHC4 as a key regulator of the protein lactylation network in this context. These findings suggest that distinct LDH family members may collaboratively orchestrate tumor metabolic reprogramming through the integrated regulation of lactate metabolism and protein lactylation. Future studies are warranted to elucidate the specific roles of LDHA and LDHB within the LDHC4-centered lactylation regulatory network and their potential interplay. In addition to its canonical enzymatic function, our study identifies and validates a non-canonical role for LDHC4: remodeling the global landscape of intracellular lactylation. Using DIA-based lactylation proteomics on the Astral platform, we observed that LDHC4 overexpression elevates overall protein lactylation levels in LUAD cells and specifically alters the lactylation status of a distinct subset of proteins. Functional enrichment analyses further revealed that these differentially lactylated proteins are significantly enriched in pathways governing the cell cycle, directly linking this metabolism-associated enzyme to core biological processes regulating cell proliferation and tumorigenesis. The importance of protein lactylation as a key bridge connecting the glycolytic tumor microenvironment with cellular function has become increasingly recognized [8, 21–23, 26, 30]. Previous studies in gastrointestinal malignancies and gynecological/breast cancers have shown lactylation promotes cancer progression by modifying specific proteins [23, 49, 50]. This study is the first to establish LDHC4 as a key regulator of the protein lactylation profile in LUAD, thereby linking a specific metabolic enzyme with a global epigenetic-like modification and providing a new paradigm for understanding how oncogenes remodel cellular signaling by modulating metabolite-driven modifications. Among the numerous differentially lactylated proteins, we focused on the core cell cycle inhibitor RB1, whose inactivation is a common pathway in cancer development, traditionally believed to be primarily regulated by phosphorylation [36, 51]. This study is the first to report and detail lactylation modification at the K900 site of RB1 (RB1–K900lac), demonstrating it to be a novel mechanism directly leading to the loss of its tumor-suppressive function. Our molecular dynamics simulations reveal the disruptive impact of this modification at the atomic level. The RB1 tumor suppressor governs the G1-to-S phase transition primarily through its interaction with the E2F family of transcription factors, with the RB1–E2F1 complex serving as a key regulator of this checkpoint [52, 53]. The K900 site, located at the critical RB1–E2F1 binding interface, is highly evolutionarily conserved. Simulations show that lactylation at K900 not only alters RB1's conformational stability but, more critically, leads to significantly increased overall structural fluctuation (RMSD) of the RB1–E2F1 complex, weakened interface interactions, and decreased hydrogen bond stability. These computational predictions were validated in cellular experiments: compared to wild-type RB1, the K900R mutant (mimicking a de-lactylated state) showed significantly enhanced binding and tighter co-localization with E2F1. This evidence strongly indicates that the high lactate environment fostered by LDHC4 induces RB1 K900 lactylation, which physically impairs the RB1–E2F1 interaction, thereby alleviating RB1-mediated inhibition of E2F1 transcriptional activity. This discovery expands the “modification map” regulating RB1 function by adding lactylation, a metabolism-sensitive modification, to the known repertoire that includes phosphorylation and acetylation [36, 51]. It also establishes a clear signaling axis from metabolic enzyme (LDHC4) to metabolite (lactate) to post-translational modification (lactylation) to dissociation of a core tumor suppressor complex (RB1–E2F1) and, ultimately, to cell cycle dysregulation and tumor progression. To confirm the functional consequences of RB1–K900lac, we performed rescue experiments. The results showed that LUAD cells expressing the lactylation-defective RB1–K900R mutant exhibited significantly inhibited proliferation, colony formation, migration, and invasion capabilities compared to cells expressing wild-type RB1. This directly demonstrates that lactylation at the K900 site is the key molecular event leading to RB1 functional inactivation and driving LUAD malignant phenotypes. Mechanistically downstream, the RB1–K900R mutation caused G1 phase cell cycle arrest, accompanied by characteristic changes: downregulated expression of CDK/cyclin complexes and upregulated expression of the cycle inhibitor P21. These findings are consistent with RB1's role as a brake at the G1/S checkpoint [31]. When RB1 is lactylated at the K900 site, this brake is released, freeing E2F1 to activate downstream genes that drive S-phase entry, leading to uncontrolled proliferation. Our work is the first to precisely map lactylation to a specific amino acid site on RB1 and systematically delineate the resulting signal transduction and phenotypic output chain. Specifically, as a member of the lactate dehydrogenase family, LDHC4 primarily catalyzes the interconversion between pyruvate and lactate [12, 13]. Our data suggest that LDHC4 may indirectly promote RB1 lactylation by elevating the intracellular lactate pool, which serves as the substrate for lactylation. Although LDHC4 has not been shown to possess intrinsic lactyltransferase activity, its role in supplying the metabolic precursor for lactylation is crucial. Whether specific acyltransferases (e.g., p300) directly mediate the transfer of the lactyl group derived from LDHC4 to the K900 residue of RB1 remains to be experimentally validated. This study also has certain limitations. First, while we have demonstrated the necessity of the K900 lactylation site using a K-to-R loss-of-function mutation, the precise mechanism by which LDHC4 regulates RB1 lactylation requires further clarification. As a key enzyme in lactate production, LDHC4 may primarily promote this modification indirectly by elevating the intracellular lactate pool; however, the possibility of more direct regulation, such as through protein-protein interactions or a non-canonical enzymatic function, cannot be excluded and warrants future investigation. Second, although the K-to-R mutation effectively blocks lactylation, future studies utilizing a gain-of-function model would provide complementary and more definitive evidence. While the lysine-to-glutamine (K-to-Q) mutation is commonly employed to mimic acylated lysines, it may not fully recapitulate the unique structural properties of a lactyl group. Therefore, the application of advanced chemical biology tools, such as genetic code expansion for the site-specific incorporation of lactylated lysine, will be essential to precisely dissect the gain-of-function effects of RB1 lactylation in LUAD. Such an approach would further substantiate our finding that RB1–K900lac is a direct trigger of cell cycle dysregulation. Furthermore, while we focused on RB1, other differentially lactylated proteins revealed by our lactylome analysis, particularly those enriched in the cell cycle pathway, may also play important synergistic roles. Besides, an important focus of our subsequent work will be to determine how RB1 lactylation influences its broader protein interaction network. Specifically, we plan to investigate the impact of this modification on RB1's binding with various predicted partners beyond E2F1, which will be essential for understanding the full spectrum of RB1 regulation by lactylation. Finally, although we confirmed this mechanism in cell models, validating the relationship between LDHC4 expression, RB1–K900lac levels, therapeutic response, and prognosis in larger, prospective LUAD clinical cohorts is a crucial next step for clinical translation. Conclusions In summary, this study systematically reveals LDHC4 as an oncogenic driver and poor prognostic biomarker in LUAD and elucidates the complete mechanism by which it upregulates protein lactylation to specifically induce RB1–K900 lactylation. This modification destabilizes the RB1–E2F1 complex, leading to cell cycle dysregulation and tumor progression. The discovery of this “LDHC4-lactate-RB1 lactylation” axis innovatively integrates tumor metabolic abnormality, protein post-translational modification, and core cell cycle regulation. This not only deepens the understanding of LUAD pathogenesis but also provides a novel theoretical foundation and potential intervention targets for developing targeted therapies, such as inhibiting LDHC4 enzymatic activity, RB1 lactylation, or the function of the released E2F1. Electronic supplementary material Below is the link to the electronic supplementary material. supplementary material 1 supplementary material 1 supplementary material 2 supplementary material 2 supplementary material 3 supplementary material 3