Background
Genotoxic treatments, such as radiation and chemotherapy, are commonly employed as primary or adjuvant therapies in cancer treatment. Unlike normal cells, which respond to genotoxic insults by activating a G1 phase cell cycle checkpoint, cancer cells exhibit dysfunctional cell cycle regulation and are more sensitive to DNA damage [ 1 ]. Genotoxic drugs exert their effects through various mechanisms and can induce multiple types of DNA damage. One widely used mechanism involves the inhibition of topoisomerase enzymes, which are essential for relaxing supercoiled DNA to facilitate transcription and replication. Inhibitors of these enzymes trap topoisomerases in intermediate complexes, leading to DNA double-strand breaks in the case of topoisomerase II inhibitors (such as etoposide and doxorubicin) and single-strand breaks for topoisomerase I inhibitors [ 2 ]. The resulting DNA damage response is orchestrated by the transcription factor p53, which promotes cell cycle arrest to facilitate DNA repair or the activation of cell death pathways according to the extent of genotoxic injury [ 3 ].
We previously demonstrated the involvement of the rate-limiting enzyme of lipolysis, adipose triglyceride lipase (ATGL), and fatty acid oxidation in the proliferation of hepatocellular carcinoma (HCC) cells [ 4 , 5 ]. ATGL activity has both metabolic and signaling-related consequences within the cell [ 6 ]. Metabolically, ATGL leads to a reduction in the size and number of lipid droplets, consequently enhancing mitochondrial β-oxidation. From a signaling perspective, ATGL-mediated lipolysis results in the release of free fatty acids (FAs), which can bind and activate downstream nuclear receptors, including members of the peroxisome proliferator-activated receptor (PPAR) family [ 7 ]. In this context, ATGL activity has been specifically linked to the activation of the transcription factor PPARα in the liver [ 8 ], and our previous findings support a positive correlation between ATGL expression and PPARα activity in HCC [ 4 ]. The nature of the ligand that activates PPARα plays a crucial role in modulating its transcriptional activity, ultimately shaping the specificity of the downstream cellular response. In HCC, PPARα regulates several key pathways, including fatty acid oxidative metabolism, energy expenditure, and the control of cell proliferation and apoptosis [ 9 ].
We also observed an association between ATGL lipolytic activity and sensitivity to different therapeutic approaches in HCC. These findings are particularly relevant given that ATGL expression is lower in HCC tissues than in normal liver tissues, highlighting ATGL as a potential therapeutic target and prognostic marker. Specifically, ATGL overexpression restrains the efficacy of glycolytic inhibitors and favors the cytotoxicity of genotoxic compounds. As the effect of glycolytic inhibitors was shown to be dependent on ATGL-mediated enhancement of oxidative metabolism, which attenuates the Warburg effect [ 4 ], in the present study, we aimed to explore the role of ATGL activity in regulating the cell response to DNA-damaging agents, such as etoposide and doxorubicin, with a particular focus on the role of ATGL lipolytic activity in signaling.
Results
ATGL overexpression enhances genotoxic drug-induced DNA damage
Our previous studies on the effects of ATGL in HCC revealed different susceptibilities to common therapeutic drugs, with ATGL-overexpressing cells being more resistant to glycolysis inhibitors than to genotoxic compounds. However, the impact of ATGL levels on the DNA damage response has not been explored in depth. We investigated the extent of DNA damage upon treatment with etoposide or doxorubicin (Sigma‒Aldrich, St. Louis, MO, USA) in HepG2 cells following ATGL modulation. We demonstrated that both compounds induced increased accumulation of phosphorylated histone H2AX (γH2AX) in ATGL-overexpressing HepG2 cells, as revealed by western blotting (Fig. 1A, B ) and immunofluorescence (Fig. 1C, D ), after 24 h of treatment. This relationship between ATGL expression and genotoxic DNA damage was further confirmed by the reduction in γH2AX levels following treatment with etoposide and doxorubicin in ATGL-silenced cells (Fig. 1E, F ). Since γH2AX is an early event following genotoxic damage, a time-course experiment was performed to understand when ATGL modulates the kinetics of H2AX phosphorylation. We observed that the increased levels of DNA damage in ATGL-overexpressing cells were highly significant after 6 h of etoposide treatment (Fig. 1G, H ), which was then used as the main time point in subsequent experiments. Fig. 1 ATGL overexpression enhances genotoxic drug-induced DNA damage. HepG2 cells were transfected with empty vector (Vehicle) or ATGL-overexpressing construct (ATGL-OE) and, after 24 h, treated with 50 µM etoposide (ETO) or 2 µM doxorubicin (DOXO) for 24 h. A , B Western blot analysis of γH2AX levels was performed. Representative immunofluorescence images and quantification of γH2AX-foci per nucleus in HepG2 cells treated with etoposide ( C ) or doxorubicin ( D ). HepG2 cells were silenced for ATGL (siATGL; SCR Scramble) and, after 24 h, treated with 50 µM etoposide or 2 µM doxorubicin for 24 h. E , F Western blot analysis of γH2AX levels was performed. G , H HepG2 cells were transfected with Vehicle or ATGL-OE construct and treated with 50 µM etoposide for 2, 4, 6 or 24 h. Western blot analysis of γH2AX levels was performed. The images are representative of three independent experiments that yielded similar results. β-Actin and ATGL were used as loading and transfection controls, respectively. The data are presented as the means ± SDs from three independent experiments. Statistical significance was determined by one-way ANOVA with Tukey’s post hoc test; * p < 0.05, ** p < 0.01, *** p < 0.001 vs CTRL or as indicated by brackets.
ATGL enzymatic activity is responsible for the increased DNA damage induced by etoposide
To assess the role of ATGL’s lipolytic activity in the enhanced induction of DNA damage following etoposide treatment, HepG2 cells were treated with an inhibitor of ATGL activity, ATGListatin (ATGLi), which blocks lipolytic function without altering protein structure. Specifically, inhibition of ATGL activity resulted in reduced DNA damage induction (Fig. 2A, B ). The critical role of ATGL lipolytic activity in enhancing the genotoxic effect of etoposide was further supported by the overexpression of a catalytically inactive ATGL mutant (S47A), which harbors a mutation in the catalytic domain. Unlike overexpression of wild-type ATGL, overexpression of the S47A mutant did not lead to an increase in γH2AX levels, which remained comparable to those of the etoposide-treated control (Fig. 2C, D ). Collectively, these results demonstrated that ATGL activity is directly associated with the sensitivity of HepG2 cells to etoposide. Fig. 2 ATGL enzymatic activity is responsible for the increased DNA damage induced by etoposide. HepG2 cells were treated with 50 µM etoposide for 6 h with or without 25 µM ATGListatin (ATGLi) for 24 h. A , B Western blot analysis of γH2AX levels was performed. HepG2 cells were transfected with empty vector (Vehicle), ATGL-overexpressing (ATGL-OE) or ATGL S47A construct and treated with 50 µM etoposide for 24 h. C , D Western blot analysis of γH2AX levels was performed after 6 h of etoposide treatment. The images are representative of three independent experiments that yielded similar results. β-Actin and ATGL were used as loading and transfection controls, respectively. The data are presented as the means ± SDs from three independent experiments. Statistical significance was determined by one-way ANOVA with Tukey’s post hoc test; * p < 0.05, ** p < 0.01, *** p < 0.001 vs CTRL or as indicated by brackets.
ATGL induces different post-translational regulation of p53 upon etoposide by activating the PPARα/p300 axis
By evaluating the role of ATGL in the genotoxic damage response in other HCC cell lines, we observed that the effects of ATGL in HUH7 cells mirrored those observed in HepG2 cells, in both ATGL-overexpressing (Fig. 3A, B ) and silenced cells (Supplementary Fig. 1A, B ), whereas ATGL overexpression did not lead to a similar effect in Hep3B cells, which are p53 null (Fig. 3C, D ). Thereafter, we investigated the contribution of p53, focusing on its main post-translational modifications (PTMs), which are involved in the DNA damage response. The time course experiment involving etoposide treatment revealed changes in p53 Ser-15 phosphorylation and Lys-382 acetylation after ATGL overexpression, whereas no changes in total p53 levels were observed (Fig. 3E ; Supplementary Fig. 1C–E ). Notably, the ratio of p53 acetylation to p53 phosphorylation significantly shifted in favor of acetylation, demonstrating a different response in p53 PTMs in the presence of high levels of ATGL (Fig. 3F ). The same results were obtained in HUH7 cells (Supplementary Fig. 1F–H ). First, we evaluated the activation of ATM, the main kinase contributing to p53 phosphorylation during the DNA damage response. However, no change was observed following etoposide treatment in ATGL-overexpressing cells compared with control cells (Supplementary Fig. 1I, J ). Therefore, we targeted the acetyltransferase p300 known to catalyze p53 Lys-382 acetylation in response to DNA damage [ 4 , 10 ]. Based on this, we exploited the selective inhibitor of p300, namely C646 (Sigma‒Aldrich, St. Louis, MO, USA) at a concentration of 10 µM 24 h before adding etoposide. The inhibition of p300 activity fully abolished ATGL-induced remodeling of p53 post-translational modifications and increased the level of γH2AX in both HepG2 (Fig. 3G–I ; Supplementary Fig. 1K, L ) and HUH7 cells (Supplementary Fig. 1M, N ), confirming the critical role of p53 acetylation in the response to DNA damage in the context of ATGL overexpression. Fig. 3 ATGL induces different post-translational regulation of p53 upon etoposide by activating the PPARα/p300 axis. A , B HUH7 and C , D Hep3B cells were transfected with an empty vector (Vehicle) or ATGL-overexpressing construct (ATGL-OE) for 24 h and then treated with 50 µM etoposide or 2 µM doxorubicin for 6 h. Western blot analysis of γH2AX levels was performed. E Western blot analysis of Ac-p53 Lys-382, p-p53 Ser-15 and p53 levels was performed in HepG2 cells after treatment with 50 µM etoposide for 2, 4, 6 and 24 h. F Densitometric ratios of Ac-p53 and p-p53 after treatment with 50 µM etoposide for 2, 4, 6 and 24 h. G – I HepG2 cells were transfected with empty vector (Vehicle) or ATGL-overexpressing (ATGL-OE) construct and, after 24 h, treated with 50 µM etoposide for 6 h with or without 10 µM C646 for 24 h. Western blot analysis of Ac-p53 Lys-382, p-p53 Ser-15 and p53 levels was performed. H Densitometric analysis ratios of Ac-p53 and p-p53 after treatment with 50 µM etoposide for 6 h. J – L HepG2 cells were treated with 50 µM etoposide for 6 h with or without 1 µM GW7647 for 24 h. Western blot analysis of Ac-p53 Lys-382, p-p53 Ser-15, p53 and γH2AX levels was performed. K Densitometric analysis of ratio between Ac-p53 and p-p53 expression after treatment with 50 µM etoposide for 6 h. The images are representative of three independent experiments that yielded similar results. β-Actin and ATGL were used as loading and transfection controls, respectively. The data are presented as the means ± SDs from three independent experiments. Statistical significance was determined by Student t test and one-way ANOVA with Tukey’s post hoc test; * p < 0.05, ** p < 0.01, *** p < 0.001 vs CTRL or as indicated by brackets.
The lipolytic activity of ATGL is responsible for the release of FAs, which can activate downstream transcription factors, including members of the PPAR family [ 11 ]. In particular, PPARα is known to activate p300 expression, and we observed that the use of a PPARα agonist (GW7647) upon etoposide treatment recapitulates the phenotype observed upon ATGL overexpression, both in terms of p53 PTMs and γH2AX levels in both HepG2 (Fig. 3J–L ; Supplementary Fig. 2A ) and HUH7 cells (Supplementary Fig. 2B, C ). These findings suggest that ATGL modulates p53 acetylation and the DNA damage response through activation of the PPARα/p300 axis.
ATGL levels are predictive of the commitment of p53 to apoptosis genes upon etoposide
To determine whether the increased levels of γH2AX after treatment of ATGL-overexpressing cells are due to altered DNA repair mechanisms, we assessed the recovery capacity for DNA damage after 2 h of etoposide treatment. While control cells presented reduced γH2AX levels after drug withdrawal, ATGL-overexpressing cells maintained the same extent of γH2AX. Similarly, p53 PTMs persisted in ATGL-overexpressing cells after recovery, whereas both p53 phosphorylation and acetylation were significantly reduced in control cells (Fig. 4A–D ). Fig. 4 ATGL levels are predictive of the commitment of p53 to apoptosis genes upon etoposide. A – D HepG2 cells were transfected with empty vector (Vehicle) or ATGL-overexpressing construct (ATGL-OE) and, after 24 h, treated with 50 µM etoposide for 6 h with or without recovery (Rec) with fresh medium for 2 h. Western blot analysis of Ac-p53 Lys-382, p-p53 Ser-15, p53 and γH2AX levels was performed. E , F HepG2 cells were transfected with empty vector (Vehicle) or ATGL-overexpressing construct (ATGL-OE) and, after 24 h, treated with 50 µM etoposide or 2 µM doxorubicin for 6 h. The proliferation was assayed via the Trypan blue direct counting procedure. G , H HepG2 cells were treated with 50 µM etoposide for 6 h with or without 25 µM ATGListatin (ATGLi) for 24 h. HepG2 cells were treated with 50 µM etoposide for 6 h. I – K Western blot analysis of p21 and Puma levels was performed. HepG2 cells were treated with 1 µM GW7647 for 24 h. L , M Proliferation was assayed by the Trypan blue direct counting procedure. N – P Western blot analysis of p21 and Puma levels was performed. The images are representative of three independent experiments that yielded similar results. β-Actin and ATGL were used as loading and transfection controls, respectively. The data are presented as the means ± SDs from three independent experiments. Statistical significance was determined by one-way ANOVA with Tukey’s post hoc test; * p < 0.05, ** p < 0.01, *** p < 0.001 vs CTRL or as indicated by brackets.
To investigate whether the impaired DNA damage recovery observed in ATGL-overexpressing cells translated into increased cell death upon etoposide exposure, we assessed viability after treating cells for 24 h with genotoxic drugs. We observed that ATGL-overexpressing cells exhibited increased sensitivity to both etoposide and doxorubicin (Fig. 4E, F ; Supplementary Fig. 2D ). The use of ATGLi (Fig. 4G, H ) and the overexpression of a catalytically inactive ATGL mutant (Supplementary Fig. 2E–G ) enabled us to demonstrate that this effect was dependent on ATGL lipase activity. Moreover, despite ATGL overexpression, the inhibition of ATGL activity prevented the cytotoxic outcome (Supplementary Fig. 2H ).
The results obtained thus far suggest that the state of p53 PTMs induced by ATGL overexpression induces a preferential transcriptional response toward cell death pathways. Thus, we analyzed key p53 target genes involved in the DNA damage response: p21, which promotes cell cycle arrest to allow DNA repair together with GADD45 [ 12 , 13 ], and Puma, which promotes apoptosis [ 14 ]. We found that ATGL overexpression increased Puma expression, whereas p21 levels (Fig. 4I–K ) and GADD45a nuclear foci (Supplementary Fig. 3A ) were decreased upon etoposide treatment.
In parallel, the finding that the PPARα agonist has the same effect as ATGL on etoposide sensitivity (Fig. 4L, M ), including Puma upregulation and p21 downregulation (Fig. 4N–P ), indicates that cells with increased ATGL/PPARα pathway activity are more prone to undergo apoptosis rather than cell cycle arrest for DNA repair upon etoposide treatment.
ATGL-associated transcriptional programs in HCC human specimens are linked to PPARα/p300 signaling
To validate the mechanistic evidence obtained in vitro, we interrogated transcriptomic data derived from human HCC biopsies (TCGA-LIHC dataset) revealing that PNPLA2 (ATGL) is significantly reduced in tumoral tissues compared with normal liver samples (Fig. 5A ). Consistently, Gene Set Enrichment Analysis (GSEA) revealed a significant enrichment of the Gene Ontology (GO) terms related to the positive regulation of the cell cycle in HCC samples with low PNPLA2 expression (Supplementary Fig. 3B ). No significant differences in ATGL expression were detected between TP53 wild-type and mutant tumors (Fig. 5B ), indicating that ATGL transcriptional levels are independent of p53 mutational status in HCC. Fig. 5 ATGL-associated transcriptional programs in HCC are linked to PPARα/p300 signaling and apoptotic commitment. A Boxplot showing significantly reduced PNPLA2 expression in primary HCC tissues (primary tumor; n = 371) compared with solid tumor-adjacent non-tumoral liver tissues (solid tissue normal; n = 50) samples based on TCGA data. B Boxplot of Z-score–normalized ATGL expression from TCGA-LIHC RNA-seq data based on TP53 mutation status (wild type n = 263 and mutant n = 111). C Visualization of the PPAR signaling pathway, reporting normalized enrichment score (NES) and adjusted p -value. D Bar plot showing the most significantly enriched transcription factors after a transcription factor enrichment analysis performed using TRRUST transcription factors 2019 database on differentially expressed genes (DEGs) in ATGL-high versus ATGL-low HCC samples. Adjusted p -value was reported. Scatter plot showing the correlation between E PNPLA2 and PPARα ( PPARA ); between F PNPLA2 and EP300 ; between G PPARA and EP300 ; between H PNPLA2 and Puma ( BBC3 ); between I PNPLA2 and p21 ( CDKN1A ) mRNA expression levels in HCC samples from the TCGA-LIHC cohort analyzed using GEPIA. Gene expression values are reported as log2-transformed TPM. Each dot represents an individual tumor sample.
By performing GO analysis on the top 50 genes positively correlated with PNPLA2 expression (Supplementary Fig. 3C ), apart from metabolic signatures including lipid metabolism and mitochondrial function, we highlighted the presence of the term PPAR signaling, that is also particularly enriched in samples with high ATGL levels by GSEA (Fig. 5C ; Supplementary Fig. 3D ). Notably, PPARα was the most significantly enriched transcription factor involved in the regulation of genes associated with ATGL (Fig. 5D ). The reciprocal and positive correlation between PNPLA2 and PPARA and EP300 in biopsies further supports the relevance of ATGL-PPARα–p300 signaling axis in human HCC (Fig. 5E–G ). Finally, in agreement with in vitro experiments the GSEA identified that tumors characterized by low ATGL expression were preferentially enriched in genes involved in different types of genotoxic stresses (Supplementary Fig. 3E ), a significant positive correlation was demonstrated between ATGL and the p53 target genes BBC3 (PUMA) (Fig. 5H ) and CDKN1A (p21) (Fig. 5I ).
Discussion
Given our previous characterization of the effects of increased oxidative metabolism driven by ATGL overexpression in HCC cells, we elucidated the molecular mechanism by which ATGL may increase DNA damage in response to genotoxic agents. We observed that ATGL expression levels are predictive of the response to DNA damage, as assessed by H2AX phosphorylation, in both the HepG2 and HUH7 cell lines. Etoposide-induced DNA damage increases progressively up to 6 h of treatment, with ATGL-overexpressing cells exhibiting increased levels of damage. By 24 h, the levels of DNA damage markers, including γH2AX and p53, began to decrease, which correlated with the onset of downstream effects, namely, cell death. Accordingly, a recent study also demonstrated that vimentin modulates ATGL-mediated lipolysis, thereby limiting the severity of damage [ 15 ].
The relationship between the DNA damage response and lipid metabolism has been explored across various cellular systems [ 16 , 17 ]. DNA damage has been shown to induce a rearrangement of lipid metabolism, thereby promoting inflammation [ 17 ]. Conversely, the lipid composition of cancer cells can influence genomic instability by facilitating DNA damage or modulating its repair. Notably, the number of intracellular lipid droplets has emerged as a prognostic factor for the response to genotoxic therapies in various systems, including multiple myeloma [ 18 ] and prostate cancer [ 19 ]. Furthermore, increased fatty acid oxidation has been associated with increased DNA damage, improved chemotherapeutic response, and reversal of obesity-associated chemoresistance [ 20 ]. Increased lipolytic activity impacts lipid homeostasis in distinct ways: by reducing the size of lipid droplets, thus preventing the sequestration of lipophilic chemotherapeutic agents [ 21 ]; by promoting the oxidation of FAs; and by activating downstream transcription factors via released FAs.
Interestingly, ATGL-dependent damage in HCC was opposite to that in p53-null Hep3B hepatocellular carcinoma cells, a relevant result that needs further investigation and suggests both p53-dependent and p53-independent mechanisms.
P53 is known to be involved not only in the repair response, which promotes cell cycle arrest through the induction of genes such as p21 and GADD45A but also in the proapoptotic response, which is triggered by the expression of genes such as Puma [ 22 – 24 ]. The expression of one set of genes over another directs the cell either toward DNA damage repair, when damage is limited, or toward cell death, if the extent of damage exceeds the cell’s repair capacity. In this context, phosphorylation at serine 15, a key event in the DNA damage response, is also capable of manipulating p53 transcriptional activity [ 25 , 26 ]. Indeed, this PTM primarily promotes cell cycle arrest by reducing p53 nuclear export and consequently increasing its nuclear localization [ 26 , 27 ]. Our previous report linked ATGL activity to altered p53 acetylation levels [ 4 ]. In HepG2 and HUH7 ATGL-overexpressing cells treated with etoposide, the ratio of p53 acetylation to phosphorylation was skewed toward acetylation. These findings suggest that ATGL influences the post-translational modification landscape of p53 by favoring acetylation over phosphorylation. This result is in agreement with the increased lysine acetylation of proteins in various cellular models, as FAs released via ATGL provide substrates for acetyl-CoA generation [ 28 , 29 ].
On the other hand, polyunsaturated FAs with 18 to 22 carbon atoms that are produced by ATGL activate PPARα [ 7 , 8 , 30 ], which is a positive regulator of the acetyltransferase p300. Importantly, the relevance of this signaling axis is supported by patient-derived transcriptomic data that revealed a significant positive association between PNPLA2 , PPARA and E P300 expression. Here, we demonstrated that p300 is required for both the increased DNA damage and the enhanced sensitivity to etoposide observed in ATGL-overexpressing cells. This finding is consistent with the recapitulation of the ATGL-induced phenotype upon treatment of HCC cells with a PPARα agonist, which alone increased DNA damage in response to etoposide and reprogrammed p53 post-translational modifications.
Under our experimental conditions, ATGL/PPARα-induced reprogramming of p53 PTMs resulted in increased levels of cell death in ATGL-overexpressing cells, as confirmed by increased expression of the proapoptotic protein Puma. Moreover, the expression of p21, a protein that mediates cell cycle arrest in response to DNA damage to facilitate repair [ 31 ], was reduced. Notably, the pharmacological activation of PPARα mimics the impact of ATGL overexpression on the same p53 downstream targets. Finally, the recovery experiment revealed a delay in DNA damage repair due to reduced p21 expression and GADD45 nuclear accumulation. This represents a critical point, as it promotes cell death. Consistently, the enrichment of pathways related to genotoxic stress in PNPLA2-low tumors suggests that HCC samples with reduced ATGL expression are transcriptionally primed to activate repair mechanisms in response to DNA damage. In contrast, higher ATGL levels are associated with attenuation of these pathways, supporting our experimental model in which ATGL overexpression enhances DNA damage and apoptotic signaling upon genotoxic stress.
Overall, our data indicates that the stimulation of ATGL activity and PPARα signaling can improve those therapeutic strategies based on DNA damaging agents that are known to be particularly unsuccessful in the treatment of HCC patients [ 32 ]. Although in vivo validation will be necessary to assess the impact of ATGL-dependent modulation in the DNA damage response, the human transcriptomic analyses support the importance of the ATGL/PPARα/p300 signaling axis in HCC.
Conclusions
This study highlights that the enzymatic activity of ATGL and the downstream activation of the PPARα/p300 axis play a role in the genotoxic stress response in HCC. These findings support the tumor-suppressive role of ATGL in HCC, not only through its previously described metabolic functions but also by influencing the cellular response to genotoxic agents. Assessing ATGL levels may be important for patient stratification, enabling more accurate prediction of therapeutic strategies dependent on p53 expression in HCC. Moreover, enhancing ATGL activity or using PPARα agonists may represent a promising approach to improve the efficacy of current treatments.
Materials and methods
Materials
Albumin (A3782), ATGListatin (5.30151), C646 (382113), crystal violet (61135), doxorubicin (DOXO) (D5220), EDTA (E6758), EGTA (E4378), etoposide (ETO) (341205), GW7647 (370698), KH 2 PO 4 (A0261677), MgCl 2 (A748033 012), Formalin solution neutral buffered 10% (HT5012), sodium deoxycholate (D6750), sodium orthovanadate (S6508), sodium pyrophosphate tetrabasic decahydrate (30411), sucrose (S0389), and Triton X-100 (T9284) were obtained from Sigma‒Aldrich, St. Louis, MO, USA. Trypan blue 0.4% solution (17-942E) was obtained from Lonza, Antwerp, Belgium. Goat anti-mouse (172-1011) and anti-rabbit (172-1019) horseradish peroxidase (H + L)-conjugated IgG were obtained from Bio-Rad Laboratories, Hercules, California. Hoechst 33342 (H1399) was purchased from Thermo Fisher Scientific. DTT (281) and protease inhibitor cocktail were obtained from AMRESCO, Radnor, Pennsylvania. Tris-base (1610716) and sodium dodecyl sulfate (SDS) (161-0300) were obtained from Bio-Rad. TEMED (110189) was obtained from PanReac AppliChem, Barcelona, Spain.
Cell lines and treatments
HepG2, HUH7 and Hep3B cell lines were purchased from American Type Culture Collection (ATCC). HepG2 and HUH7 cells were grown in RPMI 1640 (EuroClone, Milan, Italy), and Hep3B cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 1 g/L glucose (EuroClone). All media were supplemented with 10% fetal bovine serum (FBS) (EuroClone), 10 U/mL penicillin/streptomycin (EuroClone) and 2 mM L -glutamine (EuroClone). The cells were authenticated and characterized by the supplier. Mycoplasma tests were routinely carried out according to protocols from our laboratory. The cells were cultured at 37 °C in an atmosphere of 5% CO 2 in the air and plated at a density of 2 × 10 5 cells/mL for all the experiments. After 24 h of incubation, the cells were treated with etoposide (Sigma‒Aldrich, cat. number 341205) and doxorubicin (Sigma‒Aldrich, cat. number D5220) for the indicated hours.
The cells were treated with 25 μM ATGListatin (Sigma‒Aldrich, cat number 5.30151) for 24 h, 10 μM C646 (Sigma‒Aldrich, cat. number 382113) for 24 h, and 1 μM GW7647 (Sigma‒Aldrich, cat. number 370698) for 24 h.
Twenty-four h after plating, the cells were transiently transfected with pcDNA™4/HisMaxC, pcDNA™4/HisMaxC-ATGL or pcDNA™4/HisMaxC-ATGL (S47A) for 48 h with polyethylenimine (PEI) reagent according to the manufacturer’s instructions. The pcDNA™4/HisMaxC and pcDNA™4/HisMaxC-ATGL plasmids were kindly provided by Prof. Rudolf Zechner, Institute of Molecular Biosciences, Karl-Franzens-Universität Graz, Graz (Austria); the pcDNA™4/HisMaxC-ATGL (Ser47Ala) plasmid was obtained as previously described [ 33 ]. The transfection efficiency was determined by transfecting cells with pATGL-EGFP. The percentage of transfected cells was assessed as >85%. siATGL was reversely transfected using Lipofectamine® RNAiMAX Transfection Reagent (Thermo Fisher Scientific, cat. number 13778150) according to the manufacturer’s protocol.
Western blot analysis
At the end of the experiment, the cells were resuspended in lysis buffer (50 mM Tris-HCl, pH 7.4; 150 mM NaCl; 1 mM EDTA; 1% Triton X-100; 0.5% sodium deoxycholate; 0.1% SDS; 10 mM NaF; 5 mM sodium pyrophosphate; 2 mM sodium orthovanadate) supplemented with a protease inhibitor cocktail (AMRESCO). After centrifugation at 14,000× g for 15 min, the Lowry et al. [ 34 ] method was used to determine the protein concentration before electrophoresis by SDS‒PAGE and blotting onto a nitrocellulose membrane (Bio‒Rad).
The following primary antibodies were used: β-Actin (Cell Signaling Technology, cat. number #4970S, diluted 1:1000), γH2AX Ser-139 (Cell Signaling Technology, cat. number #9718, diluted 1:1000), ATGL (Cell Signaling Technology, cat. number 2138S, diluted 1:1000), pATM Ser-1981 (Cell Signaling Technology, cat. number #5883, diluted 1:1000), ATM (Cell Signaling Technology, cat. number #2873, diluted 1:1000), p21 (Cell Signaling Technology, cat. number #2947, diluted 1:1000), PPARα (Santa Cruz Biotechnology, cat. number sc-398394, diluted 1:1000), Puma (Cell Signaling Technology, cat. number #4976, diluted 1:1000), Ac-p53 Lys-382 (Cell Signaling Technology, cat. number #2525S, diluted 1:1000), p-p53 Ser-15 (Cell Signaling Technology, cat. number #9284S, diluted 1:1000), and p53 (Sigma-Aldrich, cat. number #P5813, diluted 1:1000). After incubation with specific horseradish peroxidase (HRP)-conjugated secondary antibodies (Bio-Rad), ImageJ software was used to perform densitometry analyses.
Cell proliferation assays
Cell proliferation was evaluated via the Trypan blue exclusion test procedure and a Crystal violet colorimetric assay (Sigma‒Aldrich cat. number 61135). For the crystal violet assay, the cells were fixed with a formalin solution containing 10% neutral buffer (containing 4% paraformaldehyde) for 10 min, washed with PBS and stained with 0.1% crystal violet for 30 min at RT. After three washes with PBS, the cells were air-dried. The elution of Crystal Violet was performed with 10% methanol, and the quantification was performed by measuring the absorbance at 595 nm via a microplate reader.
Fluorescence microscopy analysis
The medium was removed, and the cells were fixed with a neutral buffered 10% formalin solution for 10 min. After washing with PBS, they were permeabilized with PBS/0.1% Triton X-100 solution for 10 min. The cells were blocked with PBS/10% FBS solution for 1 h and incubated overnight with anti-γH2AX Ser-139 (Cell Signaling Technology, cat. number #9718) or GADD45A (Cell Signaling Technology, cat. number #4632) antibodies and then incubated for 1 h with an Alexa Fluor™ 594 donkey anti-rabbit IgG (H + L) secondary antibody; the nuclei were stained with 1 µg/mL Hoechst 33342 for 10 min. A Delta Vision Restoration Microscopy System (Applied Precision, Issaquah, WA) equipped with an Olympus IX70 fluorescence microscope (Olympus Italia, Segrate, Milano, Italy) was used to acquire fluorescence images of the cells. The fluorescence intensity was evaluated via ImageJ software.
Bioinformatic analyses
RNA-seq data of HCC and normal liver tissues were obtained from The Cancer Genome Atlas (TCGA) (TCGA-LIHC dataset). Raw RNA-seq counts of primary tumors and adjacent normal tissues were normalized using DESeq2. TP53 mutational status of TCGA-LIHC samples was obtained from cBioPortal. Statistical significance between groups was assessed using a two-tailed Wilcoxon rank-sum test. Data are presented as boxplots with individual samples shown as jittered points, and p -values ≤ 0.05 were considered statistically significant.
Gene expression correlation analysis was performed on TCGA-LIHC RNA-sequencing data using Pearson correlation coefficients to identify genes positively associated with PNPLA2 (ATGL) gene expression (LinkedOmics) [ 35 ]. The top 50 positively correlated genes were subjected to Gene Ontology over-representation analysis to identify enriched biological pathways (Webgestalt) [ 36 ]. Statistical significance was evaluated as previously described [ 35 ], and enriched terms with p -value < 0.05 were considered significant. Multiple-test correction is performed using the Benjamini and Hochberg method to generate the False Discovery Rate (FDR). LinkInterpreter was used to evaluate functional enrichment (WikiPathways database). Gene expression correlation analysis was performed using the GEPIA web server [ 37 ], which integrates RNA-sequencing data from TCGA. PNPLA2 (ATGL) and PPARA, BBC3 and P300 mRNA expression levels were extracted from the TCGA-LIHC dataset. Expression data were normalized as transcripts per million (TPM) and log2-transformed prior to analysis. Correlation between gene expression levels was assessed using correlation statistics implemented in GEPIA, and statistical significance was determined using two-tailed tests. Correlation coefficients ( R ) (Pearson test) and corresponding p -values are reported. Transcription factor enrichment analysis of genes over-expressed in ATGL-high HCC was performed using the Enrichr platform [ 38 – 40 ]. TRRUST Transcription Factors 2019 was used to identify enriched transcription factor–target gene relationships, including PPARα, while TF–PPI analysis was applied to identify associated transcriptional co-regulators. Statistical significance was evaluated by Fisher’s exact test with Benjamini–Hochberg correction.
Gene set enrichment analysis (GSEA) was performed using TCGA-LIHC RNA-seq data. Samples were ranked according to PNPLA2 expression and divided into two groups: PNPLA2 high, containing the top 25% of the ranking, and PNPLA2 low, containing the bottom 25% of the ranking. GSEA was performed using GSEA v4.4.0 software ( https://www.gsea-msigdb.org ) [ 41 ] to analyze enriched pathways in the two groups.
Data analysis
The results are expressed as the means ± SDs of data derived from at least 3 independent experiments. Student’s t test for comparisons was used for statistical analyses of only two variables, and one-way ANOVA with post hoc Tukey tests was used for multiple comparisons. GraphPad Prism 8 software was used to create graphics and perform the statistical analysis. Comparisons were considered statistically significant at p ≤ 0.05 (*), very statistically significant at p ≤ 0.01 (**) and extremely statistically significant at p ≤ 0.001 (***).
Supplementary information
Supplementary Materials
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Fabio Ciccarone, Maria Rosa Ciriolo.
Supplementary information
The online version contains supplementary material available at 10.1038/s41420-026-03048-4.
Acknowledgements
ADC and ES are enrolled in PhD Program in Cellular and Molecular Biology, Department of Biology, University of Rome Tor Vergata, Rome, Italy.
Author contributions
SC designed and managed the experiments, performed the statistical analysis, conceptualized the data and wrote the original draft; ADC and ES performed some of the Western blot experiments and provided methodological support; ED performed bioinformatic analyses; FC and MRC performed the study concept and design, edited and reviewed the paper and acquired funding. All the authors read and approved the final paper.
Funding
This work was supported in part by Italian Ministry of University and Research MUR-PNRR M4C2I1.3 PE6 project PE00000019 Heal Italia No. E83C22004670001 to MRC and in part by Italian Association for Cancer Research (AIRC) No. IG 2023 ID 29207 to MRC.
Data availability
Not applicable
Competing interests
The authors declare no competing interests.
References
1. Larsen BD Benada J Yung PYK Bell RAV Pappas G Urban V Cancer cells use self-inflicted DNA breaks to evade growth limits imposed by genotoxic stress Science 2022 376 476 83 10.1126/science.abi6378 35482866 Larsen BD, Benada J, Yung PYK, Bell RAV, Pappas G, Urban V, et al. Cancer cells use self-inflicted DNA breaks to evade growth limits imposed by genotoxic stress. Science. 2022;376:476–83. 35482866 10.1126/science.abi6378
2. Swift LH Golsteyn RM Genotoxic anti-cancer agents and their relationship to DNA damage, mitosis, and checkpoint adaptation in proliferating cancer cells Int J Mol Sci 2014 15 3403 31 10.3390/ijms15033403 24573252 PMC3975345 Swift LH, Golsteyn RM. Genotoxic anti-cancer agents and their relationship to DNA damage, mitosis, and checkpoint adaptation in proliferating cancer cells. Int J Mol Sci. 2014;15:3403–31. 24573252 10.3390/ijms15033403 PMC3975345
3. Jamil S Lam I Majd M Tsai SH Duronio V Etoposide induces cell death via mitochondrial-dependent actions of p53 Cancer Cell Int 2015 15 79 10.1186/s12935-015-0231-z 26251638 PMC4527242 Jamil S, Lam I, Majd M, Tsai SH, Duronio V. Etoposide induces cell death via mitochondrial-dependent actions of p53. Cancer Cell Int. 2015;15:79. 26251638 10.1186/s12935-015-0231-z PMC4527242
4. Di Leo L Vegliante R Ciccarone F Salvatori I Scimeca M Bonanno E Forcing ATGL expression in hepatocarcinoma cells imposes glycolytic rewiring through PPAR-α/p300-mediated acetylation of p53 Oncogene 2019 38 1860 75 10.1038/s41388-018-0545-0 30367149 PMC6756110 Di Leo L, Vegliante R, Ciccarone F, Salvatori I, Scimeca M, Bonanno E, et al. Forcing ATGL expression in hepatocarcinoma cells imposes glycolytic rewiring through PPAR-α/p300-mediated acetylation of p53. Oncogene. 2019;38:1860–75. 30367149 10.1038/s41388-018-0545-0 PMC6756110
5. Castelli S Ciccarone F De Falco P Ciriolo MR Adaptive antioxidant response to mitochondrial fatty acid oxidation determines the proliferative outcome of cancer cells Cancer Lett 2023 554 216010 10.1016/j.canlet.2022.216010 36402229 Castelli S, Ciccarone F, De Falco P, Ciriolo MR. Adaptive antioxidant response to mitochondrial fatty acid oxidation determines the proliferative outcome of cancer cells. Cancer Lett. 2023;554:216010. 36402229 10.1016/j.canlet.2022.216010
6. Vegliante R Di Leo L Ciccarone F Ciriolo MR Hints on ATGL implications in cancer: beyond bioenergetic clues Cell Death Dis 2018 9 1 10 10.1038/s41419-018-0345-z 29472527 PMC5833653 Vegliante R, Di Leo L, Ciccarone F, Ciriolo MR. Hints on ATGL implications in cancer: beyond bioenergetic clues. Cell Death Dis. 2018;9:1–10. 29472527 10.1038/s41419-018-0345-z PMC5833653
7. Pawlak M Lefebvre P Staels B Molecular mechanism of PPARα action and its impact on lipid metabolism, inflammation and fibrosis in non-alcoholic fatty liver disease J Hepatol 2015 62 720 33 10.1016/j.jhep.2014.10.039 25450203 Pawlak M, Lefebvre P, Staels B. Molecular mechanism of PPARα action and its impact on lipid metabolism, inflammation and fibrosis in non-alcoholic fatty liver disease. J Hepatol. 2015;62:720–33. 25450203 10.1016/j.jhep.2014.10.039
8. Sapiro JM Mashek MT Greenberg AS Mashek DG Hepatic triacylglycerol hydrolysis regulates peroxisome proliferator-activated receptor α activity J Lipid Res 2009 50 1621 9 10.1194/jlr.M800614-JLR200 19304987 PMC2724052 Sapiro JM, Mashek MT, Greenberg AS, Mashek DG. Hepatic triacylglycerol hydrolysis regulates peroxisome proliferator-activated receptor α activity. J Lipid Res. 2009;50:1621–9. 19304987 10.1194/jlr.M800614-JLR200 PMC2724052
9. Pan Y Li Y Fan H Cui H Chen Z Wang Y Roles of the peroxisome proliferator-activated receptors (PPARs) in the pathogenesis of hepatocellular carcinoma (HCC) Biomed Pharmacother 2024 177 117089 10.1016/j.biopha.2024.117089 38972148 Pan Y, Li Y, Fan H, Cui H, Chen Z, Wang Y, et al. Roles of the peroxisome proliferator-activated receptors (PPARs) in the pathogenesis of hepatocellular carcinoma (HCC). Biomed Pharmacother. 2024;177:117089. 38972148 10.1016/j.biopha.2024.117089
10. Sakaguchi K Herrera JE Saito S Miki T Bustin M Vassilev A DNA damage activates p53 through a phosphorylation-acetylation cascade Genes Dev 1998 12 2831 41 10.1101/gad.12.18.2831 9744860 PMC317174 Sakaguchi K, Herrera JE, Saito S, Miki T, Bustin M, Vassilev A, et al. DNA damage activates p53 through a phosphorylation-acetylation cascade. Genes Dev. 1998;12:2831–41. 9744860 10.1101/gad.12.18.2831 PMC317174
11. Fougerat A Bruse J Polizzi A Montagner A Guillou H Wahli W Lipid sensing by PPARα: Role in controlling hepatocyte gene regulatory networks and the metabolic response to fasting Prog Lipid Res 2024 96 101303 10.1016/j.plipres.2024.101303 39521352 Fougerat A, Bruse J, Polizzi A, Montagner A, Guillou H, Wahli W. Lipid sensing by PPARα: Role in controlling hepatocyte gene regulatory networks and the metabolic response to fasting. Prog Lipid Res. 2024;96:101303. 39521352 10.1016/j.plipres.2024.101303
12. Tamamori-Adachi M Koga A Susa T Fujii H Tsuchiya M Okinaga H DNA damage response induced by Etoposide promotes steroidogenesis via GADD45A in cultured adrenal cells Sci Rep 2018 8 9636 10.1038/s41598-018-27938-5 29941883 PMC6018231 Tamamori-Adachi M, Koga A, Susa T, Fujii H, Tsuchiya M, Okinaga H, et al. DNA damage response induced by Etoposide promotes steroidogenesis via GADD45A in cultured adrenal cells. Sci Rep. 2018;8:9636. 29941883 10.1038/s41598-018-27938-5 PMC6018231
13. Schäfer A Gadd45 proteins: key players of repair-mediated DNA demethylation Adv Exp Med Biol 2013 793 35 50 10.1007/978-1-4614-8289-5_3 24104472 Schäfer A. Gadd45 proteins: key players of repair-mediated DNA demethylation. Adv Exp Med Biol. 2013;793:35–50. 24104472 10.1007/978-1-4614-8289-5_3
14. Sperka T Song Z Morita Y Nalapareddy K Guachalla LM Lechel A Puma and p21 represent cooperating checkpoints limiting self-renewal and chromosomal instability of somatic stem cells in response to telomere dysfunction Nat Cell Biol 2011 14 73 9 10.1038/ncb2388 22138576 Sperka T, Song Z, Morita Y, Nalapareddy K, Guachalla LM, Lechel A, et al. Puma and p21 represent cooperating checkpoints limiting self-renewal and chromosomal instability of somatic stem cells in response to telomere dysfunction. Nat Cell Biol. 2011;14:73–9. 22138576 10.1038/ncb2388
15. Wang F Rong M Zhang L Solomon AD Gui W Li J Vimentin intermediate filaments orchestrate DNA nonhomologous end joining repair and lipolysis after DNA damage Oncogene 2025 44 3025 36 10.1038/s41388-025-03465-2 40550847 Wang F, Rong M, Zhang L, Solomon AD, Gui W, Li J, et al. Vimentin intermediate filaments orchestrate DNA nonhomologous end joining repair and lipolysis after DNA damage. Oncogene. 2025;44:3025–36. 40550847 10.1038/s41388-025-03465-2
16. Xu J Fei P Simon DW Morowitz MJ Mehta PA Du W Crosstalk between DNA damage repair and metabolic regulation in hematopoietic stem cells Cells 2024 13 733 10.3390/cells13090733 38727270 PMC11083014 Xu J, Fei P, Simon DW, Morowitz MJ, Mehta PA, Du W. Crosstalk between DNA damage repair and metabolic regulation in hematopoietic stem cells. Cells. 2024;13:733. 38727270 10.3390/cells13090733 PMC11083014
17. Hamsanathan S Anthonymuthu T Han S Shinglot H Siefken E Sims A Integrated -omics approach reveals persistent DNA damage rewires lipid metabolism and histone hyperacetylation via MYS-1/Tip60 Sci Adv 2022 8 eabl6083. 10.1126/sciadv.abl6083 35171671 PMC8849393 Hamsanathan S, Anthonymuthu T, Han S, Shinglot H, Siefken E, Sims A, et al. Integrated -omics approach reveals persistent DNA damage rewires lipid metabolism and histone hyperacetylation via MYS-1/Tip60. Sci Adv. 2022;8:eabl6083. 35171671 10.1126/sciadv.abl6083 PMC8849393
18. Garcia De Paco E Soulet C Requirand G Robert N Cartron G Vincent L Lipid-mediated modulation of DNA damage signaling as a prognostic and therapeutic strategy against multiple myeloma Blood 2023 142 3322 10.1182/blood-2023-181039 Garcia De Paco E, Soulet C, Requirand G, Robert N, Cartron G, Vincent L, et al. Lipid-mediated modulation of DNA damage signaling as a prognostic and therapeutic strategy against multiple myeloma. Blood. 2023;142:3322.
19. Ribeiro CF Rodrigues S Bastos DC Fanelli GN Pakula H Foiani M Blocking lipid synthesis induces DNA damage in prostate cancer and increases cell death caused by PARP inhibition Sci Signal 2024 17 eadh1922 10.1126/scisignal.adh1922 38593154 PMC11161871 Ribeiro CF, Rodrigues S, Bastos DC, Fanelli GN, Pakula H, Foiani M, et al. Blocking lipid synthesis induces DNA damage in prostate cancer and increases cell death caused by PARP inhibition. Sci Signal. 2024;17:eadh1922. 38593154 10.1126/scisignal.adh1922 PMC11161871
20. Hwang S Yang S Park K Kim B Kim M Shin S Induction of fatty acid oxidation underlies DNA damage-induced cell death and ameliorates obesity-driven chemoresistance Adv Sci (Weinh) 2024 11 e2304702 10.1002/advs.202304702 38145969 PMC10933680 Hwang S, Yang S, Park K, Kim B, Kim M, Shin S, et al. Induction of fatty acid oxidation underlies DNA damage-induced cell death and ameliorates obesity-driven chemoresistance. Adv Sci (Weinh). 2024;11:e2304702. 38145969 10.1002/advs.202304702 PMC10933680
21. Antunes P Cruz A Barbosa J Bonifácio VDB Pinto SN Lipid droplets in cancer: from composition and role to imaging and therapeutics Molecules 2022 27 991 10.3390/molecules27030991 35164256 PMC8840564 Antunes P, Cruz A, Barbosa J, Bonifácio VDB, Pinto SN. Lipid droplets in cancer: from composition and role to imaging and therapeutics. Molecules. 2022;27:991. 35164256 10.3390/molecules27030991 PMC8840564
22. Williams AB Schumacher B p53 in the DNA-damage-repair process Cold Spring Harb Perspect Med 2016 6 a026070 10.1101/cshperspect.a026070 27048304 PMC4852800 Williams AB, Schumacher B. p53 in the DNA-damage-repair process. Cold Spring Harb Perspect Med. 2016;6:a026070. 27048304 10.1101/cshperspect.a026070 PMC4852800
23. Zhang H Xu J Long Y Maimaitijiang A Su Z Li W Unraveling the Guardian: p53’s multifaceted role in the DNA damage response and tumor treatment strategies Int J Mol Sci 2024 25 12928 10.3390/ijms252312928 39684639 PMC11641486 Zhang H, Xu J, Long Y, Maimaitijiang A, Su Z, Li W, et al. Unraveling the Guardian: p53’s multifaceted role in the DNA damage response and tumor treatment strategies. Int J Mol Sci. 2024;25:12928. 39684639 10.3390/ijms252312928 PMC11641486
24. Aubrey BJ Kelly GL Janic A Herold MJ Strasser A How does p53 induce apoptosis and how does this relate to p53-mediated tumour suppression? Cell Death Differ 2018 25 104 13 10.1038/cdd.2017.169 29149101 PMC5729529 Aubrey BJ, Kelly GL, Janic A, Herold MJ, Strasser A. How does p53 induce apoptosis and how does this relate to p53-mediated tumour suppression? Cell Death Differ. 2018;25:104–13. 29149101 10.1038/cdd.2017.169 PMC5729529
25. Abuetabh Y Wu HH Chai C Al Yousef H Persad S Sergi CM DNA damage response revisited: the p53 family and its regulators provide endless cancer therapy opportunities Exp Mol Med 2022 54 1658 69 10.1038/s12276-022-00863-4 36207426 PMC9636249 Abuetabh Y, Wu HH, Chai C, Al Yousef H, Persad S, Sergi CM, et al. DNA damage response revisited: the p53 family and its regulators provide endless cancer therapy opportunities. Exp Mol Med. 2022;54:1658–69. 36207426 10.1038/s12276-022-00863-4 PMC9636249
26. Wen J Wang D Deciphering the PTM codes of the tumor suppressor p53 J Mol Cell Biol 2021 13 774 85 10.1093/jmcb/mjab047 PMC8782589 34289043 Wen J, Wang D. Deciphering the PTM codes of the tumor suppressor p53. J Mol Cell Biol. 2021;13:774–85. 10.1093/jmcb/mjab047 PMC8782589 34289043
27. Hammond EM Denko NC Dorie MJ Abraham RT Giaccia AJ Hypoxia links ATR and p53 through replication arrest Mol Cell Biol 2002 22 1834 43 10.1128/MCB.22.6.1834-1843.2002 11865061 PMC135616 Hammond EM, Denko NC, Dorie MJ, Abraham RT, Giaccia AJ. Hypoxia links ATR and p53 through replication arrest. Mol Cell Biol. 2002;22:1834–43. 11865061 10.1128/MCB.22.6.1834-1843.2002 PMC135616
28. Shang S Liu J Hua F Protein acylation: mechanisms, biological functions and therapeutic targets Sig Transduct Target Ther 2022 7 396 10.1038/s41392-022-01245-y PMC9797573 36577755 Shang S, Liu J, Hua F. Protein acylation: mechanisms, biological functions and therapeutic targets. Sig Transduct Target Ther. 2022;7:396. 10.1038/s41392-022-01245-y PMC9797573 36577755
29. McDonnell E Crown SB Fox DB Kitir B Ilkayeva OR Olsen CA Lipids reprogram metabolism to become a major carbon source for histone acetylation Cell Rep 2016 17 1463 72 10.1016/j.celrep.2016.10.012 27806287 PMC5123807 McDonnell E, Crown SB, Fox DB, Kitir B, Ilkayeva OR, Olsen CA, et al. Lipids reprogram metabolism to become a major carbon source for histone acetylation. Cell Rep. 2016;17:1463–72. 27806287 10.1016/j.celrep.2016.10.012 PMC5123807
30. Ong KT Mashek MT Davidson NO Mashek DG Hepatic ATGL mediates PPAR-α signaling and fatty acid channeling through an L-FABP independent mechanism J Lipid Res 2014 55 808 15 10.1194/jlr.M039867 24610891 PMC3995459 Ong KT, Mashek MT, Davidson NO, Mashek DG. Hepatic ATGL mediates PPAR-α signaling and fatty acid channeling through an L-FABP independent mechanism. J Lipid Res. 2014;55:808–15. 24610891 10.1194/jlr.M039867 PMC3995459
31. Ticli G Cazzalini O Stivala LA Prosperi E Revisiting the function of p21CDKN1A in DNA repair: the influence of protein interactions and stability Int J Mol Sci 2022 23 7058 10.3390/ijms23137058 35806061 PMC9267019 Ticli G, Cazzalini O, Stivala LA, Prosperi E. Revisiting the function of p21CDKN1A in DNA repair: the influence of protein interactions and stability. Int J Mol Sci. 2022;23:7058. 35806061 10.3390/ijms23137058 PMC9267019
32. Yuan JN Chao Y Lee WP Li CP Lee RC Chang FY Chemotherapy with etoposide, doxorubicin, cisplatin, 5-fluorouracil, and leucovorin for patients with advanced hepatocellular carcinoma Med Oncol 2008 25 201 6 10.1007/s12032-007-9013-3 18488159 Yuan JN, Chao Y, Lee WP, Li CP, Lee RC, Chang FY, et al. Chemotherapy with etoposide, doxorubicin, cisplatin, 5-fluorouracil, and leucovorin for patients with advanced hepatocellular carcinoma. Med Oncol. 2008;25:201–6. 18488159 10.1007/s12032-007-9013-3
33. Castelli S Ciccarone F Tavian D Ciriolo MR ROS-dependent HIF1α activation under forced lipid catabolism entails glycolysis and mitophagy as mediators of higher proliferation rate in cervical cancer cells J Exp Clin Cancer Res 2021 40 94 10.1186/s13046-021-01887-w 33706793 PMC7948341 Castelli S, Ciccarone F, Tavian D, Ciriolo MR. ROS-dependent HIF1α activation under forced lipid catabolism entails glycolysis and mitophagy as mediators of higher proliferation rate in cervical cancer cells. J Exp Clin Cancer Res. 2021;40:94. 33706793 10.1186/s13046-021-01887-w PMC7948341
34. Lowry OH Rosebrough NJ Farr AL Randall RJ Protein measurement with the Folin phenol reagent J Biol Chem 1951 193 265 75 10.1016/S0021-9258(19)52451-6 14907713 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265–75. 14907713
35. Vasaikar SV Straub P Wang J Zhang B LinkedOmics: analyzing multi-omics data within and across 32 cancer types Nucleic Acids Res 2018 46 D956 63 10.1093/nar/gkx1090 29136207 PMC5753188 Vasaikar SV, Straub P, Wang J, Zhang B. LinkedOmics: analyzing multi-omics data within and across 32 cancer types. Nucleic Acids Res. 2018;46:D956–63. 29136207 10.1093/nar/gkx1090 PMC5753188
36. Wang J Vasaikar S Shi Z Greer M Zhang B WebGestalt 2017: a more comprehensive, powerful, flexible and interactive gene set enrichment analysis toolkit Nucleic Acids Res 2017 45 W130 7 10.1093/nar/gkx356 28472511 PMC5570149 Wang J, Vasaikar S, Shi Z, Greer M, Zhang B. WebGestalt 2017: a more comprehensive, powerful, flexible and interactive gene set enrichment analysis toolkit. Nucleic Acids Res. 2017;45:W130–7. 28472511 10.1093/nar/gkx356 PMC5570149
37. Tang Z Kang B Li C Chen T Zhang Z GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis Nucleic Acids Res 2019 47 W556 60 10.1093/nar/gkz430 31114875 PMC6602440 Tang Z, Kang B, Li C, Chen T, Zhang Z. GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res. 2019;47:W556–60. 31114875 10.1093/nar/gkz430 PMC6602440
38. Xie Z Bailey A Kuleshov MV Clarke DJB Evangelista JE Jenkins SL Gene set knowledge discovery with Enrichr Curr Protoc 2021 1 e90 10.1002/cpz1.90 33780170 PMC8152575 Xie Z, Bailey A, Kuleshov MV, Clarke DJB, Evangelista JE, Jenkins SL, et al. Gene set knowledge discovery with Enrichr. Curr Protoc. 2021;1:e90. 33780170 10.1002/cpz1.90 PMC8152575
39. Chen EY Tan CM Kou Y Duan Q Wang Z Meirelles GV Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool BMC Bioinforma 2013 14 128 10.1186/1471-2105-14-128 PMC3637064 23586463 Chen EY, Tan CM, Kou Y, Duan Q, Wang Z, Meirelles GV, et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinforma. 2013;14:128. 10.1186/1471-2105-14-128 PMC3637064 23586463
40. Kuleshov MV Jones MR Rouillard AD Fernandez NF Duan Q Wang Z Enrichr: a comprehensive gene set enrichment analysis web server 2016 update Nucleic Acids Res 2016 44 W90 97 10.1093/nar/gkw377 27141961 PMC4987924 Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 2016;44:W90–97. 27141961 10.1093/nar/gkw377 PMC4987924
41. Subramanian A Tamayo P Mootha VK Mukherjee S Ebert BL Gillette MA Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles Proc Natl Acad Sci USA 2005 102 15545 50 10.1073/pnas.0506580102 16199517 PMC1239896 Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA. 2005;102:15545–50. 16199517 10.1073/pnas.0506580102 PMC1239896