Lipid metabolism reprogramming is a recognized hallmark of cancer cells. Identification of the underlying regulators of metabolic reprogramming in esophageal squamous cell carcinoma (ESCC) could uncover potential therapeutic targets to improve treatment. Here, we demonstrated that pre-mRNA processing factor 19 (PRP19) mediates reprogramming of lipid metabolism in ESCC. Expression of PRP19 was significantly upregulated in multiple ESCC cohorts and was correlated with poor clinical prognosis. PRP19 promoted ESCC proliferation in vitro and in vivo. Upregulation of PRP19 enhanced fatty acid synthesis through sterol regulatory element-binding protein 1 (SREBF1), a major transcription factor of lipid synthase. Moreover, PRP19 enhanced the stability of SREBF1 mRNA in an N6-methyladenosine–dependent manner. Overall, this study shows that PRP19-mediated fatty acid metabolism is crucial for ESCC progression. Targeting PRP19 is a potential therapeutic approach to reverse metabolic reprogramming in patients with ESCC.

Significance:

Upregulation of pre-mRNA processing factor 19 (PRP19) contributes to esophageal squamous cell carcinoma progression by reprogramming SREBF1-dependent fatty acid metabolism, identifying PRP19 as a potential prognostic biomarker and therapeutic target.

Esophageal cancer is a common malignancy, ranking as the sixth leading cause of cancer deaths worldwide (1). Esophageal squamous cell carcinoma (ESCC) is the main histologic subtype of esophageal cancer, accounting for over 90% of cases (2). Despite improvement in conventional therapeutic strategies, which include surgery, radiotherapy, and chemotherapy, the mortality rate of ESCC remains high (3). Therefore, it will be important to better understand the molecular mechanisms of ESCC progression and develop novel drug targets for improving ESCC prognosis.

Metabolic reprogramming is a recognized hallmark of cancer cells (4). In addition to well-understood glucose and glutamine metabolism, lipid metabolism is also increasingly being recognized (5). In contrast to normal human cells employing exogenous fatty acids, cancer cells prefer synthesizing lipids de novo to support their rapid proliferation due to their rapid lipid requirement for biological membranes and signal transduction (6). Indeed, fatty acid synthesis-related enzymes [e.g., fatty acid synthase (FASN; ref. 7), ATP-citrate lyase (ACLY; ref. 8), stearoyl-CoA desaturase (SCD1; ref. 9), and acetyl-CoA carboxylase (10)] have demonstrated pro-tumor activity in various malignancies. At the regulatory level, sterol regulatory element-binding protein 1 (SREBP1, also known as SREBF1) mainly contributes to upregulated expression of factors required for fatty acid synthesis through its transcriptional activity (11). Several studies have also shown upregulation of SREBF1 in various malignancies, including ESCC, and breast and liver cancer (12–14). Further upstream, SREBF1 is regulated by many signaling pathways, including growth factor signaling, PI3K/Akt/mTOR signaling, and cancer-specific epigenomic regulation (12, 15, 16). However, regulation of SREBF1 by mRNA modification remains unclear.

Pre-mRNA processing factor 19 (PRP19) is a multifunctional protein that participates in diverse biological processes, including the DNA damage response and pre-mRNA splicing in embryogenesis and tumor progression (17–19). In previous studies, we demonstrated that PRP19 could facilitate hepatocellular carcinoma progression by radioresistance (20), chemoresistance (20, 21), the epithelial–mesenchymal transition (22) as well as the G2–M transition (23). In addition to these widely known regulatory functions, some studies have revealed that PRP19 is involved in lipogenesis in murine preadipocyte cells and human adipose stromal cells (24–26). However, due to physiologic discrepancies in normal and tumor cells, the impact of PRP19 on tumor lipid metabolism remains unclear.

In the current study, we investigate the role of PRP19 and its underlying mechanism in ESCC. We hypothesize that PRP19 enhances ESCC progression by SREBF1-dependent reprogramming of lipid metabolism. Herein, the expression levels and functions of PRP19 in ESCC were first investigated. Next, the association between PRP19 and lipid metabolism was investigated with RNA sequencing (RNA-seq), lipidomics, and in vitro assays. Finally, the underlying mechanisms of PRP19 on SREBF1 were identified by omics techniques, and in vitro and in vivo experiments, which were further consolidated by clinical data related to ESCC prognosis.

Clinical specimens and analysis of public datasets

A total of 382 patients with ESCC from three independent cohorts were enrolled from Zhongshan Hospital of Fudan University (Shanghai, China) in this study. For Cohort 1, 30 paired snap-frozen specimens from patients with ESCC who underwent curative resection from January to December 2015, were obtained for the detection of PRP19 mRNA and protein expression. For Cohort 2, paraffin-embedded tissues were collected from 322 patients, including 102 paired tumor and nontumor tissues, 254 tumor tissues with overall survival (OS) time, and 202 tumor tissues with recurrence time. These patients underwent curative resection from January to December 2007 and were monitored until December 2012. For Cohort 3, 30 paired fresh specimens from patients with ESCC were collected for Oil Red Staining and correlation analysis between PPR19 and SREBF1. These patients underwent curative resection from January to December 2020. None of the above patients received preoperative chemotherapy or radiotherapy. This study was approved by the Human Research Ethics Committee of Zhongshan Hospital, Fudan University (no. B2020–130R/B2022–366R). Written informed consent was obtained from all participants following the guidelines of the Declaration of Helsinki.

We extracted ESCC mRNA expression files from The Cancer Genome Atlas (TCGA) using Genomic Data Commons Data Portal (https://portal.gdc.cancer.gov/). Data were downloaded in November 2021 and included RNA-seq data of 82 tissues (81 primary ESCC and 1 paired normal esophageal specimen). We first ranked ESCC samples based on the expression of PRP19 and divided the samples into two groups (high PRP19 expression: top 20 samples; low PRP19 expression: bottom 20 samples). We further obtained paired RNA-seq data of 230 ESCC specimens from the Gene Expression Omnibus (GEO) database under accession numbers GSE53624, GSE23400, GSE38129, GSE161533, and GSE20347. Gene set enrichment analysis (GSEA) was carried out using OmicShare tools (https://www.omicshare.com/tools). In addition, online visualization sites including GEPIA (http://gepia.cancer-pku.cn/), UALCAN (http://ualcan.path.uab.edu/), and ENCORI (https://starbase.sysu.edu.cn/) were used to analyze mRNA expression and common pathways of PRP19 in various tumors. Chromatin immunoprecipitation sequencing data of PRP19 in the KYSE-150 cell line was adapted from the ENCODE database (https://www.encodeproject.org/).

Cell culture

The human ESCC cell lines KYSE-150, KYSE-30, ECA-109, KYSE-180, KYSE-450, and KYSE-140 and the human immortalized esophageal cell line Het-1A were obtained from the Cell Bank of the Chinese Academy of Sciences. All cell lines were maintained in RPMI1640 medium (Invitrogen) containing 10% FBS (Invitrogen).

Quantification of neutral lipids, triglycerides, and cholesterol

Cells were seeded on coverslips and fixed with 4% paraformaldehyde for 15 minutes. Subsequently, the cells were incubated with BODIPY 493/503 (1 μg/mL, Chemegen) staining solution for 15 minutes at 37°C. Images were captured with a confocal microscope (Leica) and the fluorescence intensities of neutral lipids were quantified using ImageJ software.

Triglyceride and cholesterol contents were determined using EnzyChrom triglyceride and cholesterol kits (Bioassay Systems) according to the manufacturers’ protocols.

Fatty acid uptake assay

Fatty acid uptake capacity was determined by the Screen Quest Fluorimetric Fatty Acid Uptake Assay Kit (AAT Bioquest). Briefly, 1 × 105 cells/well were plated in 96-well plates with growth medium for 6 hours. Subsequently, the cells were transferred to serum-free medium for 1 hour. After the addition of 100 μL/well of the fatty acid dye-loading solution, cells were incubated for 1 hour and endpoint readings (bottom reading mode) were taken at Ex/Em = 485/515 nm to monitor changes in fluorescence.

mRNA stability assay

ESCC cells were transfected with indicated siRNA for 48 hours, followed by treatment with 5 μg/mL actinomycin D (ActD; MedChemExpress) for 12, 8, 4, 2, and 0 hours. Total RNAs were harvested and qPCR was performed to quantify the degradation rate and relative half-life of indicated mRNA according to the manufacturers’ instructions using GraphpadPrism 8.0 (GraphPad Inc.). The corresponding siRNA sequences and qPCR primers are listed in Supplementary Table S1 and S2, respectively.

Quantification of global N6-methyladenosine levels

The global RNA N6-methyladenosine (m6A) level was detected by the EpiQuik m6A RNA Methylation Kit (Epigentek). Briefly, total RNA was isolated from tissues or cells with TRIzol reagent, and bound to strip wells using the RNA high-binding solution. Then, the capture and detection antibodies were added to each well in turn. Finally, the relative m6A level was compared by measuring the absorbance value at a wavelength of 450 nm.

RNA-seq

Total RNA was isolated from KYSE-150 cells transfected with siRNA scramble and siPRP19_1 (n = 3) using TRIzol reagent (Invitrogen). RNA-seq detection was carried out by PANOMIX. Differentially expressed genes (DEG) between two groups were identified by limma package with R software (www.r-project.org) for Windows. The pathway analysis based on the Hallmark gene set was carried out using OmicShare tools (https://www.omicshare.com/tools). The different types of alternative splicing (AS) events were identified using rMATS software. Cutoffs of FDR and inclusion level difference (IncLevelDiff) at 0.05 were applied to screen statistically significant differential AS events. SREBF1- and ALKBH5-associated AS events were further confirmed by RT-PCR assays and the corresponding primer sequences are listed in Supplementary Table S3.

m6A-RNA immunoprecipitation assay and m6A-RNA immunoprecipitation sequencing

m6A-RNA immunoprecipitation (MeRIP) was performed to detect m6A modification of target genes using the Magna MeRIP m6A Kit (Millipore) according to the manufacturer's instructions. Briefly, 300 μg of fragmented total RNA was incubated with protein A/G magnetic beads conjugated with anti-m6A antibody (Synaptic Systems) in IP buffer at 4°C overnight. The m6A modified RNA was then isolated with elution buffer, purified by phenol/chloroform/isoamyl alcohol (25:24:1, Millipore) extraction, and subjected to qPCR assays or sequencing using Illumina NovaSeq 6000 (Illumina Inc.).

m6A-RNA immunoprecipitation sequencing (MeRIP-seq) was conducted by Cloudseq Biotech Inc. as previously described (27). Briefly, total RNA was isolated from KYSE-150 cells transfected with siRNA scramble and siPRP19_1 (n = 2) using TRIzol reagent (Invitrogen). The clean reads were aligned to the reference genome (HG19) with Hisat2 software (v2.0.4). The methylated sites on RNAs (m6A peaks) were identified by Model-based Analysis of ChIP-seq (MACS) software. Motifs were analyzed by HOMER to generate consensus sequences. The R package MetaPlotR was applied to identify metagenes of the m6A regions.

In vivo xenograft model

To further identify the role of PRP19 in tumor growth, we established an ESCC subcutaneous xenograft in BALB/c nude mice. Four-week-old male BALB/c nu/nu mice were randomly assigned to groups (6 mice/group) before cell inoculation. After acclimatization for a week in the specific pathogen-free animal laboratory, 5 × 106 cells were injected into the right flank of mice. Tumor volume was monitored every 3 days and calculated as follows: tumor volume = (length × width2)/2. The mice were sacrificed at day 30, and the subcutaneous tumors were harvested and weighed. The harvested tumors were subsequently analyzed by IHC with indicated antibodies, and used for detection of intratumoral triglyceride content. The sources and working concentration of antibodies are listed in Supplementary Table S4. The above animal experiments were carried out with the approval of the Animal Care and Use Committee of Zhongshan Hospital, Fudan University (2021-057).

Statistical analysis

Statistical analyses were performed with SPSS 23.0 (IBM Corporation), GraphPad Prism 8.0 (GraphPad Inc.), and R software (www.r-project.org) for Windows. Student t test was applied to compare continuous variables and Pearson correlation coefficient was used for testing linearity degree. Survival curves were plotted with the Kaplan–Meier method and compared by log-rank tests. The χ2 test was used to compare categorical variables. Multivariate analyses were conducted with the Cox proportional hazard regression method. Data are presented as means of at least three independent replicates ± SD and P value < 0.05 was considered statistically significant. More detailed materials and methods can be found in the Supplemental Information.

Data availability

The RNA-seq data generated in this study was publicly available in Sequence Read Archive database under accession number PRJNA900516. The data analyzed in this study were obtained from TCGA database and GEO database at GSE53624, GSE23400, GSE38129, GSE161533, and GSE20347. Other data generated in this study are available upon request from the corresponding author.

PRP19 is upregulated and correlates with poor prognosis in ESCC

To explore the potential role of PRP19 in human malignancies, we compared mRNA levels in 23 tumor types based on TCGA database. The transcriptome sequencing data suggested that PRP19 expression was significantly upregulated in 13 tumor types (Fig. 1A; Supplementary Fig. S1A), indicating that PRP19 may have a common role in tumors. On the basis of availability of clinical specimens of ESCC and relatively higher expression of ESCC compared with esophageal adenocarcinoma (EACC, P = 0.013), we selected ESCC as our primary focus (Fig. 1B). To further identify the clinical significance of PRP19 expression, we investigated five independent ESCC cohorts from the GEO database (Fig. 1C; Supplementary Fig. S1B). A consistently significant upregulation of PRP19 expression was observed in ESCC samples compared with adjacent nontumorous tissues (P < 0.001).

Figure 1.

PRP19 is upregulated and correlates with poor prognosis in ESCC. A, Relative expression of PRP19 in gastrointestinal tumors including esophageal cancer, liver hepatocellular carcinoma (LIHC), cholangiocarcinoma (CHOL), stomach adenocarcinoma (STAD), colon adenocarcinoma (COAD), and rectum adenocarcinoma (READ) based on the transcriptome sequencing data of TCGA database.B, Relative expression of PRP19 in EACC and ESCC. Adapted from UALCAN database (http://ualcan.path.uab.edu/). C, Relative expression of PRP19 in three independent ESCC cohorts from GEO database (accession numbers GSE53624, GSE23400, and GSE38129). D, Relative mRNA expression of PRP19 in 30 cases of human ESCC and paired normal tissues detected by qPCR. E, Protein expression of PRP19 in 12 cases of human ESCC and paired normal tissues detected by Western blotting. Right, quantification of gray value. F, Representative IHC staining intensity of PRP19 in ESCC and normal tissues from Cohort 2. Scale bar, 200 μm (top); 50 μm (bottom). G–I, PRP19 expression was assessed by IHC in Cohort 2. G and H, PRP19 expression in ESCC specimens was examined in both total samples (G) and paired samples (H). I, PRP19 expression in normal esophagus and tumor tissues grouped into stages I to IV. J, Kaplan–Meier curves showing OS and RFS of patients with different levels of PRP19. K, Multivariable Cox analysis of clinical prognostic parameters for OS (n = 254) and RFS (n = 202). Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 1.

PRP19 is upregulated and correlates with poor prognosis in ESCC. A, Relative expression of PRP19 in gastrointestinal tumors including esophageal cancer, liver hepatocellular carcinoma (LIHC), cholangiocarcinoma (CHOL), stomach adenocarcinoma (STAD), colon adenocarcinoma (COAD), and rectum adenocarcinoma (READ) based on the transcriptome sequencing data of TCGA database.B, Relative expression of PRP19 in EACC and ESCC. Adapted from UALCAN database (http://ualcan.path.uab.edu/). C, Relative expression of PRP19 in three independent ESCC cohorts from GEO database (accession numbers GSE53624, GSE23400, and GSE38129). D, Relative mRNA expression of PRP19 in 30 cases of human ESCC and paired normal tissues detected by qPCR. E, Protein expression of PRP19 in 12 cases of human ESCC and paired normal tissues detected by Western blotting. Right, quantification of gray value. F, Representative IHC staining intensity of PRP19 in ESCC and normal tissues from Cohort 2. Scale bar, 200 μm (top); 50 μm (bottom). G–I, PRP19 expression was assessed by IHC in Cohort 2. G and H, PRP19 expression in ESCC specimens was examined in both total samples (G) and paired samples (H). I, PRP19 expression in normal esophagus and tumor tissues grouped into stages I to IV. J, Kaplan–Meier curves showing OS and RFS of patients with different levels of PRP19. K, Multivariable Cox analysis of clinical prognostic parameters for OS (n = 254) and RFS (n = 202). Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

To validate the results from public datasets, we subsequently confirmed PRP19 upregulation using qPCR in a small cohort of 30 patients with ESCC (Cohort 1). Notably, PRP19 upregulation by at least two-fold occurred in 53.33% (16/30) of patients with ESCC (Fig. 1D; P < 0.001). Next, to confirm that PRP19 upregulation in ESCC samples was also present at the protein level, Western blotting was performed in 12 paired tissues randomly selected from Cohort 1. Consistent with our data at the mRNA level, PRP19 protein expression was upregulated in 83.33% (10/12) of ESCC tissues compared with that in matched adjacent tissues (Fig. 1E; P < 0.01).

To further demonstrate the clinical significance of PRP19 expression, IHC on the tissue microarray was conducted to test PRP19 expression in an expanded cohort of 322 patients with ESCC (Cohort 2). Representative staining results are presented in Fig. 1F. Consistent with the above findings, PRP19 expression in ESCC specimens was significantly enhanced compared with that in peri-tumorous esophageal tissues in both total samples and paired samples (Fig. 1G and H; both P < 0.001). Furthermore, a significant correlation was observed between higher PRP19 level and advanced tumor–node–metastasis (TNM) stage (Fig. 1I; stage I vs. stage II, P = 0.003; stage I vs. stage III, P < 0.001; stage I vs. stage IV, P = 0.002). These results suggest that upregulation of PRP19 may contribute to malignant progression of ESCC.

All of the specimens (Cohort 2) were then divided into low and high PRP19 expression based on IHC staining scores. Detailed clinical characteristics showed that high PRP19 expression correlated with advanced stage of tumor invasion (P < 0.001), distant metastasis (P < 0.001), and clinical stage (P < 0.001; Supplementary Table S5). Patients in the high PRP19 expression group exhibited poorer OS and worse recurrence-free survival (RFS) than those in the low PRP19 expression group (Fig. 1J; both P < 0.001). Considering the significant positive correlation between PRP19 expression and TNM stage (Fig. 1I), we performed a subgroup analysis. In stage-matched patients, high PRP19 expression implied shorter OS and RFS, suggesting that the prognostic role of PRP19 is at least independent of TNM stage (Supplementary Fig. S1C and S1D). Furthermore, Cox multivariate proportional hazards model revealed that high PRP19 expression was an independent prognostic factor of poorer OS [Fig. 1K; HR, 2.295; 95% confidence interval (CI), 1.619–3.253; P < 0.001] and worse RFS (HR, 1.988; 95% CI, 1.398–2.827; P < 0.001). Taken together, these data indicate that PRP19 is significantly upregulated in ESCC and could serve as a potential predictive biomarker for patients with ESCC.

To investigate the underlying mechanism of PRP19 upregulation in ESCC, we also analyzed PRP19 promoter using the UCSC Genome Browser (https://users.soe.ucsc.edu/~cetownse/). As presented in Supplementary Fig. S1E, acetylation of H3K27 (H3K27ac) and methylation of H3K4 (H3K4me3), which could serve as positive transcription signals, were enriched in the PRP19 promoter region, suggesting PRP19 might be upregulated at the transcriptional level via histone modification.

PRP19 facilitates ESCC cell proliferation in vitro and in vivo

To identify the role of PRP19 in ESCC, we first examined PRP19 expression by qPCR and Western blotting in six ESCC cell lines. All detected ESCC cell lines upregulated PRP19 mRNA and protein expression compared with the normal esophageal epithelial cell line Het-1A (Fig. 2A and B). To further examine the functional impacts of PRP19 in ESCC progression, we used the ECA-109 and KYSE-150 cell lines, which exhibited an intermediate expression level, to establish PRP19 knockdown and overexpression cell models (Fig. 2C and D). Cell viability was initially detected by Cell Counting Kit-8 (CCK-8) assay. Endogenous knockdown of PRP19 markedly inhibited ESCC cell proliferation, whereas exogenous overexpression of PRP19 significantly facilitated proliferation (Fig. 2E). To examine long-term growth, colony formation assay revealed that knockdown of PRP19 significantly suppressed the colony forming ability of ESCC cells, whereas overexpression of PRP19 promoted colony formation (Fig. 2F and G). Importantly, the inhibition of proliferation induced by PRP19 knockdown was effectively rescued by expression of a vector encoding an siRNA-resistant form of PRP19, but not the wild-type (siRNA-sensitive) PRP19 (Supplementary Fig. S2A–S2D), indicating these results are not due to off-target effects. To determine whether PRP19 affects the proliferation of normal esophageal cells, we also performed loss- and gain-of-function assays in the Het-1A cell line. However, only a marginal effect on proliferation was observed after altered PRP19 expression, indicating a tumor-specific role of PRP19 in ESCC proliferation (Supplementary Fig. S2E–S2G). Moreover, stable knockdown of PRP19 markedly reduced the weight and volume of xenograft tumors compared with the control group, whereas exogenous overexpression of PRP19 significantly enlarged tumor volume and weight (Fig. 2HJ). Taken together, these results suggest PRP19 facilitates ESCC cell proliferation in vitro and in vivo.

Figure 2.

PRP19 facilitates ESCC cell proliferation in vitro and in vivo. A and B, qPCR (A) and Western blotting (B) analysis of PRP19 expression in ESCC cell lines and the normal esophageal epithelial cell line Het-1A. β-actin was used as a loading control. C and D, Transfection efficiency was confirmed in ECA-109 and KYSE-150 cells by qPCR (C) and Western blotting (D). E, CCK-8 proliferation assays were performed after PRP19 was knocked down or overexpressed in ECA-109 and KYSE-150 cells (n = 3). F, Colony formation assay was performed to examine long-term growth of ECA-109 and KYSE-150 cells with PRP19 knockdown or overexpression. G, Quantification of the colony number. HJ, Xenograft tumor growth of KYSE-150 cells with stable PRP19 knockdown or overexpression (n = 6). H, Representative images of subcutaneous xenografts. I, Growth curves of subcutaneous xenografts (n = 6). J, Quantitative analysis of xenograft weight. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 2.

PRP19 facilitates ESCC cell proliferation in vitro and in vivo. A and B, qPCR (A) and Western blotting (B) analysis of PRP19 expression in ESCC cell lines and the normal esophageal epithelial cell line Het-1A. β-actin was used as a loading control. C and D, Transfection efficiency was confirmed in ECA-109 and KYSE-150 cells by qPCR (C) and Western blotting (D). E, CCK-8 proliferation assays were performed after PRP19 was knocked down or overexpressed in ECA-109 and KYSE-150 cells (n = 3). F, Colony formation assay was performed to examine long-term growth of ECA-109 and KYSE-150 cells with PRP19 knockdown or overexpression. G, Quantification of the colony number. HJ, Xenograft tumor growth of KYSE-150 cells with stable PRP19 knockdown or overexpression (n = 6). H, Representative images of subcutaneous xenografts. I, Growth curves of subcutaneous xenografts (n = 6). J, Quantitative analysis of xenograft weight. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

PRP19 regulates fatty acid metabolism pathways

To determine potential PRP19-regulated signaling pathways, we initially performed GSEA with the Hallmark gene set and identified activated signaling pathways enriched in ESCC specimens with high PRP19 level using RNA-seq data from the TCGA_ESCC cohort and GEO datasets (accession numbers GSE20347 and GSE38129; Supplementary Fig. S3A). Among the six shared signaling pathways that overlapped in three ESCC cohorts (Fig. 3A, left), four classical pathways including MYC, mTORC1, DNA repair, and oxidative phosphorylation were expectedly enriched, consistent with our previous study (20). Interestingly, two lipid metabolism-related pathways, fatty acid metabolism and adipogenesis, were also identified (Fig. 3A, right). To validate the results from public datasets, we performed RNA-seq in KYSE-150 cells with PRP19 knockdown (Supplementary Data 1). Correspondingly, the GSEA showed that fatty acid metabolism and cholesterol homeostasis pathways were significantly downregulated in PRP19-knockdown cells (Fig. 3B).

Figure 3.

PRP19 regulates fatty acid metabolism. A, Left, overlapped signaling pathways enriched in PRP19-high samples among three public datasets of ESCC (TCGA_ESCC, GSE20347, and GSE38129). Right, individual GSEA plots of lipid metabolism pathway in TCGA_ESCC dataset. B, Left, enriched signaling pathways were identified by GSEA with the Hallmark gene set in RNA-seq data from KYSE-150 cells with PRP19 knockdown. Right, individual GSEA plots of lipid metabolism pathway in TCGA_ESCC dataset. C, Principal component analysis was used to visualize inherent clustering between control and siPRP19 groups. D, Volcano plot of LC/MS-MS-based lipidomics in KYSE-150 cells with or without PRP19 knockdown. Yellow dots, upregulated lipid ions; green dots, downregulated lipid ions; black dots, lipid ions without significant difference; red dots, lipid ions with the most significant difference (|log2FC| > 3). E, Bar plot shows cumulative differences in the relative percentage of all quantified lipid species between siPRP19 and the control group. n = 6/group. F, Relative peak area of top 10 significant TG species in the control and siPRP19 group. G, Relative peak area of all detected cholesterol ester species in the control and siPRP19 group. H and I, Neutral lipids stained with BODIPY 493/503 probe after PRP19 knockdown or overexpression in ECA-109 and KYSE-150 cells. Scale bar, 20 μm. J, Cellular content of triglycerides was detected in ECA-109 and KYSE-150 cells. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, no significance.

Figure 3.

PRP19 regulates fatty acid metabolism. A, Left, overlapped signaling pathways enriched in PRP19-high samples among three public datasets of ESCC (TCGA_ESCC, GSE20347, and GSE38129). Right, individual GSEA plots of lipid metabolism pathway in TCGA_ESCC dataset. B, Left, enriched signaling pathways were identified by GSEA with the Hallmark gene set in RNA-seq data from KYSE-150 cells with PRP19 knockdown. Right, individual GSEA plots of lipid metabolism pathway in TCGA_ESCC dataset. C, Principal component analysis was used to visualize inherent clustering between control and siPRP19 groups. D, Volcano plot of LC/MS-MS-based lipidomics in KYSE-150 cells with or without PRP19 knockdown. Yellow dots, upregulated lipid ions; green dots, downregulated lipid ions; black dots, lipid ions without significant difference; red dots, lipid ions with the most significant difference (|log2FC| > 3). E, Bar plot shows cumulative differences in the relative percentage of all quantified lipid species between siPRP19 and the control group. n = 6/group. F, Relative peak area of top 10 significant TG species in the control and siPRP19 group. G, Relative peak area of all detected cholesterol ester species in the control and siPRP19 group. H and I, Neutral lipids stained with BODIPY 493/503 probe after PRP19 knockdown or overexpression in ECA-109 and KYSE-150 cells. Scale bar, 20 μm. J, Cellular content of triglycerides was detected in ECA-109 and KYSE-150 cells. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, no significance.

Close modal

Specifically, LC/MS-MS–based lipidomics were conducted in KYSE-150 cells with PRP19 siRNA or scramble siRNA (n = 6/group). Principle component analysis was initially used to visualize inherent clustering between control and experimental classes (Fig. 3C). A noticeable separation was observed, indicating a marked difference in lipid composition between the two groups. As a result, among identified 631 lipid ions belonging to 18 classes of lipids (Supplementary Data 2), PRP19 knockdown resulted in significant downregulation of 195 and upregulation of 32 lipid ions (|log2FC| > 1, log10 (P value) < 0.05; Fig. 3D). Notably, the most significantly downregulated lipid ions were enriched in triacylglycerol (TG) class (|log2FC| > 3). For alterations of whole lipid classes, the bar plot and quantitative data showed that among the 18 classic lipids, four were downregulated, seven were unchanged, and the remaining seven were upregulated (Fig. 3E; Supplementary Fig. S3B–S3D). Among the downregulated lipid classes, the levels of most TG species and cholesteryl esters were significantly decreased in the PRP19 knockdown group compared with the control group (Fig. 3F and G), consistent with the GSEA results (Fig. 3B; Supplementary Fig. S3E). Moreover, neutral lipids labeled with the BODIPY 493/503 probe displayed diminished fluorescence in PRP19-silenced ECA-109 and KYSE-150 cells, whereas exogenous overexpression of PRP19 significantly enhanced fluorescence (Fig. 3H and I). The levels of intracellular triglyceride and cholesterol were significantly lower in PRP19-silenced cells compared with control cells, whereas overexpression of PRP19 obviously facilitated lipid content (Fig. 3J; Supplementary Fig. S3F and S3G). However, negligible effects on intracellular neutral lipids, triglycerides, and cholesterol were detected in Het-1A cells with PRP19 knockdown and overexpression (Supplementary Fig. S3H–S3J). Taken together, these results suggest that PRP19 regulates lipid metabolism in addition to lipid content in ESCC cells.

PRP19 promotes de novo lipogenesis via upregulation of lipogenic enzymes

Increased cell lipid content could be induced by accelerated lipid biosynthesis, increased fatty acid uptake, and decreased lipid catabolism. Therefore, the expression correlation between PRP19 and key molecules involved in fatty acid synthesis (SREBF1, chREBP, FASN, ACACA, ACLY, SCD1, MLYCD, ACSL1, ACSL3, ACSL4, ACSL5, ACSL6, FABP4, and FABP5), cholesterol biosynthesis (HMGCR and SREBF2), fatty acid uptake (CD36), and fatty acid oxidation (CPT1A) were initially assessed in ESCC tissues from the RNA-seq data of GSE161533 (n = 28, Fig. 4A). Pearson correlation analysis indicated significant correlations between the expression level of PRP19 and the fatty acid synthesis enzymes ACLY (Fig. 4B; r = 0.460; P = 0.014) and FASN (Supplementary Fig. S4A; r = 0.823; P < 0.001), whereas no obvious correlations were observed between PRP19 and CD36 (Fig. 4C; r = −0.218; P = 0.265), HMGCR (Fig. 4D; r = −0.164; P = 0.405), SREBF2 (Supplementary Fig. S4B; r = −0.164; P = 0.405), and CPT1A (Fig. 4E; r = −0.043; P = 0.828). These results were further validated by transcriptome data of KYSE-150 cells and other independent ESCC cohorts (Fig. 4F; Supplementary Fig. S4C and S4D). Consistently, PRP19 knockdown markedly suppressed mRNA and protein expression levels of fatty acid synthesis enzymes in ECA-109 and KYSE-150 cells, whereas the levels of key factors involved in fatty acid uptake, cholesterol biosynthesis, and fatty acid oxidation were almost unaffected. Exogenous overexpression of PRP19 in ECA-109 and KYSE-150 cells promoted the expression of lipogenic enzymes, but not fatty acid uptake, cholesterol biosynthesis, or fatty acid oxidation (Fig. 4GK; Supplementary Fig. S4E). Given the above results and the overlapped signaling pathways in ESCC tissues, we speculated that PRP19-mediated lipid upregulation in ESCC cells may be partially attributed to upregulated expression of fatty acid synthesis enzymes.

Figure 4.

PRP19 promotes de novo lipogenesis via upregulation of lipogenic enzymes. A, Correlation analysis between expression of PRP19 and that of key molecules involved in lipid metabolism. Significantly correlated molecules (P < 0.05) are highlighted in red. BE, Pearson correlation coefficient between PRP19 and fatty acid synthesis enzyme FASN, fatty acid uptake enzyme CD36, cholesterol biosynthesis enzyme HMGCR, and fatty acid oxidation enzyme CPT1A. F, Heatmap showing dysregulation of genes involved in fatty acid synthesis (SREBF1, ACLY, SCD1, ACACA, MLYCD, ACSL1, ACSL3, ACSL5, and FABP5), fatty acid uptake (CD36), cholesterol biosynthesis (SREBF2 and HMGCR,), and fatty acid oxidation (CPT1A). G–J, Relative mRNA levels of indicated regulators of lipid metabolism following siRNA knockdown or plasmid overexpression of PRP19 in ECA-109 and KYSE-150 cells. K, Relative protein levels of indicated regulators of lipid metabolism following siRNA knockdown or plasmid overexpression of PRP19 in ECA-109 and KYSE-150 cells. L, Pearson correlation coefficient between PRP19 and SREBF1 in GSE53624 cohort (n = 119), TCGA ESCC cohort (n = 81), and GSE38129 cohort (n = 30). M, Western blot analysis of nuclear expression of SREBF1 following siRNA knockdown of PRP19 in ECA-109 and KYSE-150 cells. N, Western blot analysis of PRP19, SREBF1, FASN, ACLY, and SCD1 in ECA-109 and KYSE-150 cells as indicated. β-Actin was used as a loading control. Data are presented as mean ± SD. ns, no significance. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 4.

PRP19 promotes de novo lipogenesis via upregulation of lipogenic enzymes. A, Correlation analysis between expression of PRP19 and that of key molecules involved in lipid metabolism. Significantly correlated molecules (P < 0.05) are highlighted in red. BE, Pearson correlation coefficient between PRP19 and fatty acid synthesis enzyme FASN, fatty acid uptake enzyme CD36, cholesterol biosynthesis enzyme HMGCR, and fatty acid oxidation enzyme CPT1A. F, Heatmap showing dysregulation of genes involved in fatty acid synthesis (SREBF1, ACLY, SCD1, ACACA, MLYCD, ACSL1, ACSL3, ACSL5, and FABP5), fatty acid uptake (CD36), cholesterol biosynthesis (SREBF2 and HMGCR,), and fatty acid oxidation (CPT1A). G–J, Relative mRNA levels of indicated regulators of lipid metabolism following siRNA knockdown or plasmid overexpression of PRP19 in ECA-109 and KYSE-150 cells. K, Relative protein levels of indicated regulators of lipid metabolism following siRNA knockdown or plasmid overexpression of PRP19 in ECA-109 and KYSE-150 cells. L, Pearson correlation coefficient between PRP19 and SREBF1 in GSE53624 cohort (n = 119), TCGA ESCC cohort (n = 81), and GSE38129 cohort (n = 30). M, Western blot analysis of nuclear expression of SREBF1 following siRNA knockdown of PRP19 in ECA-109 and KYSE-150 cells. N, Western blot analysis of PRP19, SREBF1, FASN, ACLY, and SCD1 in ECA-109 and KYSE-150 cells as indicated. β-Actin was used as a loading control. Data are presented as mean ± SD. ns, no significance. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

Fatty acid synthesis enzyme genes are usually trans-activated by two master transcription factors, carbohydrate-responsive element-binding protein (chREBP) and SREBF1 (28). To elucidate the potential mechanism of PRP19-mediated upregulation of key lipogenic enzymes, we examined the correlation between PRP19 and these transcription factors. RNA-seq data from different ESCC patient cohorts demonstrated that PRP19 expression was positively correlated with SREBF1 but not chREBP (Fig. 4A and L; Supplementary Fig. S4F). These transcriptome results were further validated by qPCR and Western blotting results of ECA-109 and KYSE-150 cells (Fig. 4GK; Supplementary Fig. S4G and S4H). Notably, N-terminal transcriptionally active SREBF1 (N-SREBF1), which is cleaved from the SREBF1 precursor, exhibited an obvious downregulation when PRP19 was knocked down (Fig. 4M). Meanwhile, exogenous PRP19 was able to significantly upregulate the expression of exogenous mature N-SREBF1 when myc-PRP19 and Flag-SREBF1 plasmids were co-transfected into KYSE-150 cells, suggesting PPR19 not only promotes the mRNA expression of SREBF1, but also participates in the synthesis of transcriptionally active SREBF1 protein through proteolytic processing (Supplementary Fig. S4I).

Next, we wondered whether PRP19 facilitated the expression of FASN, ACLY, and SCD1 through SREBF1. As expected, silencing PRP19 significantly reduced the expression of FASN, ACLY, and SCD1 in ECA-109 and KYSE-150 cells, whereas these reductions could be completely or partially rescued in response to SREBF1 overexpression (Fig. 4N). Taken together, these results indicate that SREBF1 is critical for PRP19-mediated lipogenic enzyme expression.

PRP19 stabilized SREBF1 mRNA in an m6A-dependent manner

To elucidate the mechanisms through which PRP19 enhanced mRNA expression of SREBF1, we initially examined the contribution of splicing in our transcriptome sequencing data, one of the classical roles of PRP19 (29, 30). FDR < 0.05 and IncLevelDiff between the PRP19 knockdown and control group (|IncLevelDiff|) > 0.05 were applied to screen significant differential AS events. We identified 2,950 AS events (Supplementary Fig. S5A), including 1,990 skipped exons (SE), 245 alternative 3′ splice sites (A3SS), 209 alternative 5′ splice sites (A5SS), 223 retained introns, and 283 mutually exclusive exons. However, we did not identify any AS events with a significant difference from the SREBF1 gene, although 12 potentially relevant AS events were reported in unfiltered analysis results (Supplementary Data 3). To confirm the reliability of the RNA-seq data, we performed RT-PCR assays to validate these 12 PRP19-regulated AS events. Neither knockdown nor overexpression of PRP19 promoted SREBF1-associated AS events (Supplementary Fig. S5B, S5C). The above results indicated PRP19 may not promote SREBF1 expression directly through its splicing function.

Moreover, we performed Gene Ontology (GO) analysis of the transcriptome data from TCGA. Notably, the RNA modification and metabolism-related signaling pathways including “Regulation of RNA stability” were significantly enriched in PRP19-high samples (Fig. 5A; Supplementary Fig. S5D). The RNA stability assays showed that knockdown of PRP19 shortened the mRNA half-life of SREBF1, but not that of FASN/ACLY/SCD1 (Fig. 5B; Supplementary Fig. S5E). Thus, PRP19-induced upregulation of SREBF1 and lipogenic enzymes is at least partially dependent on the increased stability of SREBF1 mRNA. Moreover, we noticed that the RNA methylation pathway was also enriched in PRP19-high samples by GO analysis (Fig. 5A; Supplementary Fig. S5F). m6A, the most prevalent in RNA methylation, is regulated by specific methyltransferases (“writers”), demethylases (“erasers”), and RNA binding proteins (“readers”). At present, m6A methylation has been shown to fine-tune chemical structural features of basic RNAs, with critical roles in regulating transcript stability, translation, AS, subcellular localization, and phase separation (31). Therefore, we investigated whether PRP19 stabilized SREBF1 mRNA in an m6A-dependent manner.

Figure 5.

PRP19 stabilizes SREBF1 mRNA in an m6A-dependent manner. A, GO pathways related to RNA modification and metabolism enriched in PRP19-high samples of TCGA_ESCC cohort. B,SREBF1 mRNA half-life (t1/2) was tested at the indicated time points by qPCR in ECA-109 and KYSE-150 cells with or without PRP19 knockdown (n = 3). C, Relative global m6A level in mRNA of 30 pairs of ESCC tissues was detected using an m6A quantification kit. D, Relative global m6A level in mRNA of ECA-109 and KYSE-150 cells was detected using an m6A quantification kit. E, Motif analysis by discriminative regular expression motif elicitation (DREME) identified “GGACU” as the consensus m6A motif of KYSE-150 cells. F, Distribution of m6A peaks across the length of mRNAs in KYSE-150 cells with or without PRP19 knockdown. G, The relative abundance of m6A sites along SREBF1 mRNA in KYSE-150 cells with or without PRP19 knockdown, as detected by MeRIP-seq. The red rectangles revealed that the m6A peaks had a noticeable increased abundance. H, MeRIP-qPCR analysis of SREBF1 3′UTR m6A levels in ESCC cells with or without PRP19 knockdown. I, Treatment with a global methylation inhibitor, DAA, rescued the downregulation of SREBF1 mRNA levels caused by PRP19 knockdown in ECA-109 and KYSE-150 cells. J, Correlation analysis between the expression of PRP19 and that of the key molecules involved in m6A methylation. Significantly correlated molecules (P < 0.05) are highlighted in red. K, Relative mRNA expression of ALKBH5 in EACC and ESCC from TCGA_ESCA cohort. Adapted from UALCAN database (http://ualcan.path.uab.edu/). L, Heatmap showing dysregulation of genes encoding m6A methyltransferases (“writers”) and demethylases (“erasers”). DEGs (P < 0.05) are highlighted in red. M, Relative ALKBH5 mRNA level following siRNA knockdown of PRP19 in ECA-109 and KYSE-150 cells. N, Relative global m6A level in mRNA of ECA-109 and KYSE-150 cells with or without ALKBH5 knockdown was detected using an m6A quantification kit. O, MeRIP-qPCR analysis of SREBF1 3′UTR m6A levels in ESCC cells with or without ALKBH5 knockdown. P, Treatment with a global methylation inhibitor (DAA) rescued the downregulation of SREBF1 mRNA levels caused by ALKBH5 knockdown in ECA-109 and KYSE-150 cells. Q, Pearson correlation coefficient between ALKBH5 and SREBF1 in TCGA_ESCC cohort. Data are presented as mean ± SD. ns, no significance. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 5.

PRP19 stabilizes SREBF1 mRNA in an m6A-dependent manner. A, GO pathways related to RNA modification and metabolism enriched in PRP19-high samples of TCGA_ESCC cohort. B,SREBF1 mRNA half-life (t1/2) was tested at the indicated time points by qPCR in ECA-109 and KYSE-150 cells with or without PRP19 knockdown (n = 3). C, Relative global m6A level in mRNA of 30 pairs of ESCC tissues was detected using an m6A quantification kit. D, Relative global m6A level in mRNA of ECA-109 and KYSE-150 cells was detected using an m6A quantification kit. E, Motif analysis by discriminative regular expression motif elicitation (DREME) identified “GGACU” as the consensus m6A motif of KYSE-150 cells. F, Distribution of m6A peaks across the length of mRNAs in KYSE-150 cells with or without PRP19 knockdown. G, The relative abundance of m6A sites along SREBF1 mRNA in KYSE-150 cells with or without PRP19 knockdown, as detected by MeRIP-seq. The red rectangles revealed that the m6A peaks had a noticeable increased abundance. H, MeRIP-qPCR analysis of SREBF1 3′UTR m6A levels in ESCC cells with or without PRP19 knockdown. I, Treatment with a global methylation inhibitor, DAA, rescued the downregulation of SREBF1 mRNA levels caused by PRP19 knockdown in ECA-109 and KYSE-150 cells. J, Correlation analysis between the expression of PRP19 and that of the key molecules involved in m6A methylation. Significantly correlated molecules (P < 0.05) are highlighted in red. K, Relative mRNA expression of ALKBH5 in EACC and ESCC from TCGA_ESCA cohort. Adapted from UALCAN database (http://ualcan.path.uab.edu/). L, Heatmap showing dysregulation of genes encoding m6A methyltransferases (“writers”) and demethylases (“erasers”). DEGs (P < 0.05) are highlighted in red. M, Relative ALKBH5 mRNA level following siRNA knockdown of PRP19 in ECA-109 and KYSE-150 cells. N, Relative global m6A level in mRNA of ECA-109 and KYSE-150 cells with or without ALKBH5 knockdown was detected using an m6A quantification kit. O, MeRIP-qPCR analysis of SREBF1 3′UTR m6A levels in ESCC cells with or without ALKBH5 knockdown. P, Treatment with a global methylation inhibitor (DAA) rescued the downregulation of SREBF1 mRNA levels caused by ALKBH5 knockdown in ECA-109 and KYSE-150 cells. Q, Pearson correlation coefficient between ALKBH5 and SREBF1 in TCGA_ESCC cohort. Data are presented as mean ± SD. ns, no significance. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

Initially, we detected the relative global m6A level in ESCC tissues, and found that ESCC tissues showed lower levels of RNA m6A compared with the counterparts (Fig. 5C), implying m6A may be involved in the pathogenesis of ESCC. The capacity of PRP19 to downregulate global m6A was also verified in ECA-109 and KYSE-150 cells, as measured by EpiQuik m6A quantification assay (Fig. 5D). Next, transcriptome-wide MeRIP-seq was performed in KYSE-150 cells. The m6A consensus motif of GGACU was presented with high enrichment in KYSE-150 cells, enriched in the vicinity of the coding sequence and 3′ untranslated region (3′UTR) of the mRNA (Fig. 5E and F). Peak calling analysis revealed that the m6A peak enrichment in the 3′-UTR of SREBF1 mRNA was upregulated upon PRP19 knockdown according to the MeRIP-seq data (Fig. 5G); these results were further confirmed by MeRIP-qPCR (Fig. 5H), indicating that the low m6A level of SREBF1 in ESCC cells is at least partially dependent on PRP19 regulation. Moreover, administration of a global methylation inhibitor, 3-deazaadenosine (DAA), could significantly rescue the downregulated expression of SREBF1 caused by PRP19 inhibition in KYSE-150 and ECA-109 cells (Fig. 5I), indicating PRP19 may stabilize SREBF1 mRNA in an m6A-dependent manner.

Among the major regulatory elements of m6A modification, the “writers” or “erasers” directly determine the m6A methylation level of target transcripts. To identify the potential molecules that perform the m6A modification function of PRP19, we carried out correlation analysis between PRP19 and m6A-related gene expression profiles in the TCGA_ESCC cohort (Fig. 5J). Among the reported methyltransferases and demethylases, only expression of AlkB homolog H5 (ALKBH5), a demethylase highly expressed in ESCC (Fig. 5K), was found to correlate with PRP19 levels (Supplementary Fig. S5G). Our transcriptome sequencing of KYSE-150 cells also suggested that knocking down PRP19 had no significant effect on the major “writers” and “erasers” except ALKBH5 (Fig. 5L); this was further verified by qPCR analysis (Fig. 5M). In terms of the mechanism of PRP19 regulation of ALKBH5 expression, although splicing analysis of transcriptome data and RT-PCR data suggested that the splicing function of PRP19 does not play a direct role in ALKBH5 expression (Supplementary Fig. S5H; Supplementary Data 3), the exact mechanism remains to be elucidated. Moreover, potential binding between ALKBH5 and SREBF1 mRNA was indicated in the m6A2Target database (Supplementary Fig. S5I). ALKBH5 knockdown significantly upregulated the global m6A level in ESCC cells and m6A enrichment of SREBF1 mRNA (Fig. 5N and O), consistent with its canonical role in tumors (32, 33). As expected, downregulated ALKBH5 significantly suppressed SREBF1 levels in ECA-109 and KYSE-150 cells, whereas the methylation inhibitor DAA completely rescued suppression of SREBF1 mRNA levels (Fig. 5P). The strong expression correlation between ALKBH5 and SREBF1 mRNA was also observed in clinical tissues of the TCGA cohort (Fig. 5Q; r = 0.449; P < 0.001). These results indicated that PRP19 stabilization of SREBF1 mRNA depends on ALKBH5-mediated demethylation.

Subsequently, we identified the potential m6A reader that recognizes and binds m6A-modified SREBF1 mRNA. The YT521-B homology (YTH) domain-containing proteins (YTHDF1/2/3, YTHDC1/2) are well-established m6A readers in mammalian cells; these proteins are involved in mRNA decay, with the exception of YTHDC1, which is responsible for RNA splicing and translocation (34, 35). qPCR data showed that knockdown of YTHDC2, but not YTHDF1/2/3 and YTHDC1, significantly enhanced the mRNA level of SREBF1 in ECA-109 and KYSE-150 cells (Fig. 6A). In fact, significant downregulation of YTHDC2 mRNA expression was observed in ESCC samples compared with adjacent nontumorous tissues (Fig. 6B), consistent with its suppressed role in lung cancer (36). We demonstrated that YTHDC2 could directly bind SREBF1 mRNA; this was also verified by cross-linking–immunoprecipitation and high-throughput sequencing (CLIP-seq) data from the m6A2Target database (Fig. 6C; Supplementary Fig. S6A). Silencing YTHDC2 obviously facilitated SREBF1 mRNA stability, thereby partially rescuing the mRNA degradation caused by PRP19 knockdown (Fig. 6D). A similar role was observed at the protein level (Supplementary Fig. S6B). Moreover, decreased ALKBH5 expression in ECA-109 and KYSE-150 cells impaired SREBF1 mRNA stabilization induced by YTHDC2 knockdown (Fig. 6E). Taken together, these results indicate YTHDC2 may be responsible for recognizing PRP19-mediated demethylation of SREBF1 mRNA.

Figure 6.

YTHDC2 may recognize PRP19-mediated demethylation of SREBF1 mRNA. A, Relative mRNA levels of SREBF1 in ECA-109 and KYSE-150 cells following siRNA knockdown of YT521-B homology domain family members. B, Relative mRNA expression of YTHDC2 in two independent ESCC cohorts from the GEO database (accession numbers GSE23400 and GSE38129). C, RIP-qPCR revealing binding enrichment of YTHDC2 to SREBF1 mRNA in ECA-109 and KYSE-150 cells. D and E,SREBF1 mRNA half-life (t1/2) was tested at the indicated time points by qPCR in ECA-109 and KYSE-150 cells transfected with indicated siRNAs (n = 3). Data are presented as mean ± SD. ns, no significance. ***, P < 0.001.

Figure 6.

YTHDC2 may recognize PRP19-mediated demethylation of SREBF1 mRNA. A, Relative mRNA levels of SREBF1 in ECA-109 and KYSE-150 cells following siRNA knockdown of YT521-B homology domain family members. B, Relative mRNA expression of YTHDC2 in two independent ESCC cohorts from the GEO database (accession numbers GSE23400 and GSE38129). C, RIP-qPCR revealing binding enrichment of YTHDC2 to SREBF1 mRNA in ECA-109 and KYSE-150 cells. D and E,SREBF1 mRNA half-life (t1/2) was tested at the indicated time points by qPCR in ECA-109 and KYSE-150 cells transfected with indicated siRNAs (n = 3). Data are presented as mean ± SD. ns, no significance. ***, P < 0.001.

Close modal

SREBF1 is a functionally important target gene of PRP19 in ESCC

Rescue assays were next carried out to identify whether SREBF1 participates in the biological function of PRP19 in ESCC. BODIPY 493/503 staining and triglyceride quantification assays demonstrated that silencing SREBF1 could reverse the increased lipid content of KYSE-150 cells seen with PRP19 overexpression (Fig. 7A and B), whereas overexpressing SREBF1 rescued the decreased lipid content of KYSE-150 cells seen with PRP19 knockdown (Supplementary Fig. S7A and S7B). CCK-8 and colony formation assays revealed that SREBF1 knockdown significantly suppressed PRP19-induced cell proliferation and colony formation (Fig. 7C and D), whereas overexpressing SREBF1 rescued the decreased proliferation caused by PRP19 knockdown in KYSE-150 cells (Supplementary Fig. S7C and S7D). Importantly, the proliferation inhibition induced by SREBF1 knockdown was effectively rescued upon expression of a vector encoding an siRNA-resistant form of SREBF1, but not the wild-type (siRNA-sensitive) SREBF1 (Supplementary Fig. S7E–S7H), indicating these results are not due to off-target effects. Moreover, the subcutaneous xenograft model showed that SREBF1 suppression could markedly downregulate tumor growth seen with PRP19 overexpression (Fig. 7EG). The enhanced triglyceride content was also reversed by SREBF1 downregulation in tumors with stable overexpression of PRP19 (Fig. 7H). IHC staining and qPCR assays showed increased expression of ALKBH5, SREBF1, FASN, ACLY, and SCD1 in PRP19-overexpressed subcutaneous tumor tissues (Fig. 7I; Supplementary Fig. S7I and S7J), consistent with our results in vitro. The above effect could be counteracted when cells were cotransfected with lentiviruses overexpressing PRP19 and silencing SREBF1. Thus, SREBF1 mediates the regulatory function of PRP19 in ESCC cells.

Figure 7.

SREBF1 is a functionally important target gene of PRP19 in ESCC. A, Detection of neutral lipids by BODIPY 493/503 staining in KYSE-150 cells with indicated treatment. Right, quantification of fluorescence intensity with ImageJ. Scale bar, 20 μm. B, Cellular content of triglycerides was detected in KYSE-150 cells. C, CCK-8 proliferation assays were performed in KYSE-150 cells with PRP19 overexpression or SRBF1 knockdown (n = 3). D, Colony formation assay was performed in KYSE-150 cells with PRP19 overexpression or SREBF1 knockdown. Right, quantification of colony number. EG, Xenograft tumor growth of KYSE-150 cells with stable PRP19 overexpression or SREBF1 knockdown (n = 6). E, Representative images of subcutaneous xenografts. F, Growth curves of subcutaneous xenografts. G, Quantitative analysis of xenograft weight. H, Detection of triglyceride content in harvested xenograft tumors. I, Representative IHC staining of PRP19, ALKBH5, YTHDC2, SREBF1, FASN, ACLY, and SCD1 in xenograft tumors of different groups. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, no significance.

Figure 7.

SREBF1 is a functionally important target gene of PRP19 in ESCC. A, Detection of neutral lipids by BODIPY 493/503 staining in KYSE-150 cells with indicated treatment. Right, quantification of fluorescence intensity with ImageJ. Scale bar, 20 μm. B, Cellular content of triglycerides was detected in KYSE-150 cells. C, CCK-8 proliferation assays were performed in KYSE-150 cells with PRP19 overexpression or SRBF1 knockdown (n = 3). D, Colony formation assay was performed in KYSE-150 cells with PRP19 overexpression or SREBF1 knockdown. Right, quantification of colony number. EG, Xenograft tumor growth of KYSE-150 cells with stable PRP19 overexpression or SREBF1 knockdown (n = 6). E, Representative images of subcutaneous xenografts. F, Growth curves of subcutaneous xenografts. G, Quantitative analysis of xenograft weight. H, Detection of triglyceride content in harvested xenograft tumors. I, Representative IHC staining of PRP19, ALKBH5, YTHDC2, SREBF1, FASN, ACLY, and SCD1 in xenograft tumors of different groups. Data are presented as mean ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, no significance.

Close modal

Coexpression of PRP19 and SREBF1 was associated with poorer prognosis in patients with ESCC

Although the clinical correlation between PRP19 and SREBF1 has been preliminarily validated in the public dataset (Fig. 4), we further collected clinical specimens from our institution for direct evidence. Firstly, 30 ESCC specimens from Cohort 3 were subjected to qPCR (Fig. 8A). A marked positive correlation between mRNA expression of PRP19 and SREBF1 was observed (r = 0.528; P = 0.003), which was also observed at the protein level (Fig. 8B and C). Moreover, an expanded cohort of 322 patients (Cohort 2) was employed to identify a clinical correlation by IHC. We performed IHC staining of SREBF1. SREBF1 located in the nucleus represents transcriptional activity that promotes downstream FASN production. We thus focused on SREBF1 staining in the nucleus rather than in the cytoplasm. Strikingly, ESCC tissues had a higher ratio of high nuclear SREBF1 expression compared with nontumoral esophageal tissues (Fig. 8DF; all ESCC tissues: 123/322; paired ESCC tissues: 37/102; nontumoral tissues: 2/102). Consistent with qPCR and Western blotting results, Pearson correlation analysis based on IHC scores indicated that the level of PRP19 was positively associated with that of SREBF1 (Fig. 8G; r = 0.435; P < 0.001). Furthermore, Kaplan–Meier analysis revealed that patients with two highly expressed markers PRP19/SREBF1 exhibited the poorest OS (Fig. 8H; median survival: PRP19highSREBF1low: 33 months; PRP19lowSREBF1high: 32 months; PRP19highSREBF1high: 24 months) and RFS (median survival: PRP19highSREBF1low: 31 months; PRP19lowSREBF1high: 49.5 months; PRP19highSREBF1high: 18 months), indicating that the combination of these two biomarkers may be more useful for evaluating prognosis in patients with ESCC.

Figure 8.

Coexpression of PRP19 and SREBF1 was associated with poorer prognosis in patients with ESCC. A, Pearson correlation coefficient between mRNA expression of PRP19 and that of SREBF1 in Cohort 3 from our institution. B, Protein expression of PRP19 and SREBF1 in 24 pairs of ESCC tissues from Cohort 3. C, Pearson correlation coefficient of the gray value quantification of PRP19 and SREBF1. D, Representative IHC staining image of SREBF1 in ESCC and normal tissues from Cohort 2. Scale bar, 200 μm (top); 50 μm (bottom). E and F, SREBF1 nuclear expression in ESCC specimens was examined in both total samples (E) and paired samples (F). G, Pearson correlation coefficient between the IHC scores of PRP19 and that of SREBF1 in Cohort 2 from our institution. H, Kaplan–Meier curves exhibiting OS and RFS of patients with different levels of PRP19 or SREBF1. I, Frozen sections from patients with ESCC showing lipid content, and PRP19 and SREBF1 expression. Scale bar, 50 μm. J, Paraffin sections from patients with ESCC showing expression of PRP19, ALKBH5, YTHDC2, SREBF1, FASN, ACLY, and SCD1 by IHC staining. Scale bar, 50 μm. K, Schematic diagram illustrating the proposed relationship among PRP19, m6A modification, and fatty acid metabolism in ESCC progression. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 8.

Coexpression of PRP19 and SREBF1 was associated with poorer prognosis in patients with ESCC. A, Pearson correlation coefficient between mRNA expression of PRP19 and that of SREBF1 in Cohort 3 from our institution. B, Protein expression of PRP19 and SREBF1 in 24 pairs of ESCC tissues from Cohort 3. C, Pearson correlation coefficient of the gray value quantification of PRP19 and SREBF1. D, Representative IHC staining image of SREBF1 in ESCC and normal tissues from Cohort 2. Scale bar, 200 μm (top); 50 μm (bottom). E and F, SREBF1 nuclear expression in ESCC specimens was examined in both total samples (E) and paired samples (F). G, Pearson correlation coefficient between the IHC scores of PRP19 and that of SREBF1 in Cohort 2 from our institution. H, Kaplan–Meier curves exhibiting OS and RFS of patients with different levels of PRP19 or SREBF1. I, Frozen sections from patients with ESCC showing lipid content, and PRP19 and SREBF1 expression. Scale bar, 50 μm. J, Paraffin sections from patients with ESCC showing expression of PRP19, ALKBH5, YTHDC2, SREBF1, FASN, ACLY, and SCD1 by IHC staining. Scale bar, 50 μm. K, Schematic diagram illustrating the proposed relationship among PRP19, m6A modification, and fatty acid metabolism in ESCC progression. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

Furthermore, we validated PRP19-related signaling pathways in clinical specimens. Frozen sections from patients with ESCC that exhibited high PRP19 expression showed increased lipid accumulation and high SREBF1 expression in ESCC lesions (Fig. 8I), suggesting a potential association between PRP19 and lipid metabolism. Consecutive paraffin sections from patients with ESCC exhibited high expression levels of target proteins (ALKBH5, SREBF1, FASN, ACLY, and SCD1) and low levels of YTHDC2 (Fig. 8J). These results further suggest RRP19 may promote expression of fatty acid synthesis enzymes via SREBF1 in ESCC.

According to the data presented in Figs. 1 and S1, PRP19 expression is significantly upregulated in various tumors. To preliminarily examine whether PRP19 also promotes progression of other tumors via SREBF1-dependent fatty acid metabolism, we analyzed RNA-seq data from TCGA with ENCORI database (https://starbase.sysu.edu.cn/). Among seven shared signaling pathways based on all samples from TCGA, five signaling pathways (fatty acid metabolism, adipogenesis, MYC, mTORC1, and oxidative phosphorylation) overlapped with those from ESCC, suggesting PRP19 shares biological functions in tumors (Supplementary Fig. S8A). Consistent with ESCC data, expression correlations between PRP19 and SREBF1, ALKBH5, and YTHDC2 were also observed in breast invasive carcinoma, head and neck squamous cell carcinoma, lung adenocarcinoma, acute myeloid leukemia, and kidney renal clear cell carcinoma (Supplementary Fig. S8B–S8D). Thus, we have shown that PRP19 may function as an oncogene by stabilizing SREBF1 mRNA expression in tumor progression.

The current study investigated the impact and mechanism of PRP19 on ESCC lipid metabolism. We demonstrated that the expression of PRP19 was significantly upregulated in multiple ESCC cohorts and was correlated with poor clinical prognosis. The combination of RNA-seq, lipidomics, and cell biological analyses showed that upregulation of PRP19 enhanced fatty acid synthesis. Mechanistically, we revealed that PRP19 enhanced the stability of SREBF1 mRNA, a major transcription factor of lipid synthase, in an m6A dependent manner (Fig. 8K). To the best of our knowledge, this study is the first to document that PRP19-mediated fatty acid metabolism is crucial for ESCC progression.

PRP19 was originally studied for its functions in the DNA damage response and pre-mRNA splicing. PRP19 is becoming increasingly recognized as a widely upregulated biomarker in malignancies including hepatocellular carcinoma (20), acute leukemia (37), tongue cancer (38), and neuroblastoma (39), and is correlated with poor clinical prognosis in patients. Whether a similar role of PRP19 exists in ESCC is yet to be determined. Here, we showed upregulation of PRP19 in eight independent cohorts totaling 694 patients with postoperative ESCC, and identified its association with advanced tumor stage, significantly worse survival, and more frequent tumor recurrence, suggesting PRP19 is likely to play a role in ESCC progression. Although many studies have reported that PRP19 expression is significantly enhanced in tumor tissues, there are few reports on the mechanism of its upregulation. Herein, we showed that histone modifications including H3K27ac and H3K4me3, active transcription signals, were enriched in the PRP19 promoter region in KYSE-150 cells, suggesting that PRP19 might be upregulated at the transcriptional level. Nevertheless, the upstream molecular mechanisms that mediate these histone modifications remain unclear and require further investigation.

It has been well documented that ESCC cell proliferation requires lipids for providing energy, biological membranes, and signal transduction (40, 41). However, the potential mechanisms for upregulation of lipid metabolism in ESCC have not yet been fully investigated. PRP19 is indispensable for lipogenesis and may promote lipid synthesis by upregulation of PPAR-γ expression in human adipose stromal cells (26). In the current study, we explored the correlation between PRP19 and lipid metabolism in three independent ESCC RNA-seq datasets. We found that the overlapped fatty acid metabolism pathways were significantly enriched in PRP19-high samples, which was verified in RNA-seq data from KYSE-150 cells with PRP19 knockdown. Moreover, LC/MS-MS–based lipidomics revealed that PRP19 knockdown markedly reduced TG content. These results indicate that upregulation of PRP19 is involved in fatty acid metabolism, consistent with the conclusions of previous studies (24–26). However, we revealed a negative correlation between PRP19 and PPAR-γ expression in ESCC tissues. This may be partly attributed to the fact that although PPAR-γ promotes lipogenesis in normal cells, it also inhibits cell proliferation in multiple tumors, including ESCC, colorectal cancer, and hepatocellular carcinoma (42–45), which is inconsistent with the role of PRP19 in tumor progression. Therefore, PPR19 may not promote tumor lipid synthesis through PPAR-γ expression. Here, we focused on two other transcription factors responsible for lipid synthesis, SREBF1 and chREBP. We found that the expression levels of PRP19 and SREBF1 were significantly correlated in multiple ESCC cohorts, and experiments in vitro suggested PRP19 promoted mRNA and protein expression of SREBF1. However, no expression and regulatory correlation was observed between PRP19 and chREBP. Therefore, PRP19 may promote lipid content and ESCC progression by enhancing the expression of SREBF1, which was further validated by rescue assays in vitro and in vivo. In addition to fatty acid synthesis, our transcriptome and lipidomics data suggested that PRP19 was involved in cholesterol homeostasis. Although PRP19 could not regulate the expression levels of some master cholesterol biosynthesis genes (HMGCR and SREBF2), PRP19 may be directly involved in cholesterol biosynthesis by promoting SREBF1 expression (46). In addition, PRP19 may also be involved in cholesterol homeostasis through regulating cholesterol uptake, efflux, and esterification. The underlying mechanism remains to be further investigated.

m6A methylation is a reversible RNA modification that ubiquitously occurs in tumor cells. Emerging evidence has indicated that dysfunction of m6A methylation is involved in tumorigenesis and progression of ESCC (47), although it remains unclear whether m6A methylation participates in ESCC progression by regulating fatty acid metabolism. In the current study, we revealed the involvement of PRP19 in RNA methylation based on GSEA results from TCGA. Given that m6A is the most abundant RNA modification, we speculated that PRP19 might enhance SREBF1 expression through m6A methylation. Consistent with a previous study (48), we showed that m6A modification was downregulated in 30 ESCC tissues, whereas PRP19 knockdown promoted the global m6A level in ESCC cells, suggesting PRP19 may be involved in tumor progression by inhibiting m6A methylation in ESCC. However, a recent study reported that increased m6A modification were observed in eight ESCC tissues compared with adjacent normal tissues (49). The discrepancy may be attributed to the small sample size, or indicates that m6A methylation plays a more complicated role in ESCC progression than we currently understand. Furthermore, m6A-seq and MeRIP-qPCR confirmed that PRP19 could downregulate the m6A modification of SREBF1 mRNA, and the demethylase ALKBH5, which has been reported to facilitate ESCC cell proliferation (50), may be responsible for PRP19-mediated demethylation. On the basis of TCGA data, we revealed that ALKBH5 was upregulated in ESCC. However, several other studies have reported decreased ALKBH5 expression in ESCC (51, 52). The controversial role of ALKBH5 combined with the global m6A methylation levels suggests that m6A methylation plays a more complicated role in ESCC progression, which will require further investigation. The m6A “readers” are responsible for recognizing and binding m6A sites, leading to different destinies of target transcripts (53). Herein, we demonstrated SREBF1 expression could be regulated by YTHDC2, but not other members of the YTH domain family. In addition, RNA immunoprecipitation (RIP) analysis and CLIP-seq data showed YTHDC2 could directly bind SREBF1 mRNA. Silencing YTHDC2 improved SREBF1 mRNA stability, partially rescuing mRNA degradation caused by PRP19 or ALKBH5 knockdown. These results indicate YTHDC2 recognizes PRP19-mediated m6A modification of SREBF1 mRNA.

Currently, there is no appropriate biomarker to evaluate the clinical prognosis of patients with ESCC, which has hindered the development of personalized treatment. Herein, we revealed a positive correlation between PRP19 and SREBF1 expression in a large ESCC cohort from our institution, and the combination of PRP19 and SREBF1 was more valuable than the individual biomarkers for estimating prognosis of patients with ESCC. Moreover, pan-cancer analysis based on TCGA database indicated PRP19 may be involved in fatty acid metabolism in multiple tumors via stabilizing SREBF1 mRNA, similar to its role in ESCC. These results suggest PRP19 may be a promising biomarker for the diagnosis and treatment of pan-cancer, but will require further experimental validation in the future.

Overall, our study has revealed the clinical significance of enhanced PRP19 expression in patients with ESCC in multiple independent clinical cohorts. Upregulation of PRP19 contributes to tumor progression by reprogramming SREBF1-dependent fatty acid metabolism. PRP19 may therefore be a promising prognostic biomarker in patients with ESCC, as well as being a novel therapeutic target.

No disclosures were reported.

G.C. Zhang: Data curation, writing–original draft. X.N. Yu: Validation, investigation. H.Y. Guo: Visualization, methodology. J.L. Sun: Validation, investigation, visualization. Z.Y. Liu: Formal analysis, validation, methodology. J.M. Zhu: Supervision, funding acquisition, project administration. T.T. Liu: Conceptualization, resources, funding acquisition. L. Dong: Resources, project administration. X.Z. Shen: Resources, supervision, funding acquisition, project administration. J. Yin: Conceptualization, supervision, writing–review and editing.

This study was supported by Shanghai Sailing Program (No. 21YF1407200), China Postdoctoral Science Foundation (No. 2021M700831), Postdoctoral Science Foundation of Zhongshan Hospital Fudan University (No. LCBSHZX006), and National Natural Science Foundation of China (No. 81970505).

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

1.
Sung
H
,
Ferlay
J
,
Siegel
RL
,
Laversanne
M
,
Soerjomataram
I
,
Jemal
A
, et al
.
Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries
.
CA Cancer J Clin
2021
;
71
:
209
49
.
2.
Arnold
M
,
Soerjomataram
I
,
Ferlay
J
,
Forman
D
.
Global incidence of esophageal cancer by histological subtype in 2012
.
Gut
2015
;
64
:
381
7
.
3.
Lagergren
J
,
Smyth
E
,
Cunningham
D
,
Lagergren
P
.
Esophageal cancer
.
Lancet
2017
;
390
:
2383
96
.
4.
Hanahan
D
.
Hallmarks of cancer: new dimensions
.
Cancer Discov
2022
;
12
:
31
46
.
5.
Currie
E
,
Schulze
A
,
Zechner
R
,
Walther
TC
,
Farese
RV
Jr
.
Cellular fatty acid metabolism and cancer
.
Cell Metab
2013
;
18
:
153
61
.
6.
Medes
G
,
Thomas
A
,
Weinhouse
S
.
Metabolism of neoplastic tissue. IV. A study of lipid synthesis in neoplastic tissue slices in vitro
.
Cancer Res
1953
;
13
:
27
9
.
7.
Menendez
JA
,
Lupu
R
.
Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis
.
Nat Rev Cancer
2007
;
7
:
763
77
.
8.
Carrer
A
,
Trefely
S
,
Zhao
S
,
Campbell
SL
,
Norgard
RJ
,
Schultz
KC
, et al
.
Acetyl-CoA metabolism supports multistep pancreatic tumorigenesis
.
Cancer Discov
2019
;
9
:
416
35
.
9.
Lien
EC
,
Westermark
AM
,
Zhang
Y
,
Yuan
C
,
Li
Z
,
Lau
AN
, et al
.
Low glycaemic diets alter lipid metabolism to influence tumor growth
.
Nature
2021
;
599
:
302
7
.
10.
Svensson
RU
,
Parker
SJ
,
Eichner
LJ
,
Kolar
MJ
,
Wallace
M
,
Brun
SN
, et al
.
Inhibition of acetyl-CoA carboxylase suppresses fatty acid synthesis and tumor growth of non–small cell lung cancer in preclinical models
.
Nat Med
2016
;
22
:
1108
19
.
11.
Horton
JD
,
Goldstein
JL
,
Brown
MS
.
SREBPs: activators of the complete program of cholesterol and fatty acid synthesis in the liver
.
J Clin Invest
2002
;
109
:
1125
31
.
12.
Li
LY
,
Yang
Q
,
Jiang
YY
,
Yang
W
,
Jiang
Y
,
Li
X
, et al
.
Interplay and cooperation between SREBF1 and master transcription factors regulate lipid metabolism and tumor-promoting pathways in squamous cancer
.
Nat Commun
2021
;
12
:
4362
.
13.
Zhang
N
,
Zhang
H
,
Liu
Y
,
Su
P
,
Zhang
J
,
Wang
X
, et al
.
SREBP1, targeted by miR-18a-5p, modulates epithelial–mesenchymal transition in breast cancer via forming a co-repressor complex with Snail and HDAC1/2
.
Cell Death Differ
2019
;
26
:
843
59
.
14.
Wang
C
,
Tong
Y
,
Wen
Y
,
Cai
J
,
Guo
H
,
Huang
L
, et al
.
Hepatocellular carcinoma-associated protein TD26 interacts and enhances sterol regulatory element-binding protein 1 activity to promote tumor cell proliferation and growth
.
Hepatology
2018
;
68
:
1833
50
.
15.
Shao
W
,
Espenshade
PJ
.
Expanding roles for SREBP in metabolism
.
Cell Metab
2012
;
16
:
414
9
.
16.
Peterson
TR
,
Sengupta
SS
,
Harris
TE
,
Carmack
AE
,
Kang
SA
,
Balderas
E
, et al
.
mTOR complex 1 regulates lipin 1 localization to control the SREBP pathway
.
Cell
2011
;
146
:
408
20
.
17.
Chan
SP
,
Kao
DI
,
Tsai
WY
,
Cheng
SC
.
The Prp19p-associated complex in spliceosome activation
.
Science
2003
;
302
:
279
82
.
18.
Mahajan
K
.
hPso4/hPrp19: a critical component of DNA repair and DNA damage checkpoint complexes
.
Oncogene
2016
;
35
:
2279
86
.
19.
Fortschegger
K
,
Wagner
B
,
Voglauer
R
,
Katinger
H
,
Sibilia
M
,
Grillari
J
.
Early embryonic lethality of mice lacking the essential protein SNEV
.
Mol Cell Biol
2007
;
27
:
3123
30
.
20.
Yu
X-N
,
Zhang
G-C
,
Liu
H-N
,
Zhu
J-M
,
Liu
T-T
,
Song
G-Q
, et al
.
Pre-mRNA processing factor 19 functions in DNA damage repair and radioresistance by modulating cyclin D1 in hepatocellular carcinoma
.
Mol Ther Nucleic Acids
2022
;
27
:
390
403
.
21.
Yin
J
,
Zhang
YA
,
Liu
TT
,
Zhu
JM
,
Shen
XZ
.
DNA damage induces downregulation of Prp19 via impairing Prp19 stability in hepatocellular carcinoma cells
.
PLoS One
2014
;
9
:
e89976
.
22.
Yin
J
,
Wang
L
,
Zhu
JM
,
Yu
Q
,
Xue
RY
,
Fang
Y
, et al
.
Prp19 facilitates invasion of hepatocellular carcinoma via p38 mitogen-activated protein kinase/twist1 pathway
.
Oncotarget
2016
;
7
:
21939
51
.
23.
Huang
R
,
Xue
R
,
Qu
D
,
Yin
J
,
Shen
XZ
.
Prp19 arrests cell cycle via Cdc5L in hepatocellular carcinoma cells
.
Int J Mol Sci
2017
;
18
.
24.
Cho
SY
,
Shin
ES
,
Park
PJ
,
Shin
DW
,
Chang
HK
,
Kim
D
, et al
.
Identification of mouse Prp19p as a lipid droplet-associated protein and its possible involvement in the biogenesis of lipid droplets
.
J Biol Chem
2007
;
282
:
2456
65
.
25.
Cho
SY
,
Park
PJ
,
Lee
JH
,
Kim
JJ
,
Lee
TR
.
Identification of the domains required for the localization of Prp19p to lipid droplets or the nucleus
.
Biochem Biophys Res Commun
2007
;
364
:
844
9
.
26.
Khan
A
,
Dellago
H
,
Terlecki-Zaniewicz
L
,
Karbiener
M
,
Weilner
S
,
Hildner
F
, et al
.
SNEV(hPrp19/hPso4) regulates adipogenesis of human adipose stromal cells
.
Stem Cell Reports
2017
;
8
:
21
9
.
27.
Hao
W
,
Dian
M
,
Zhou
Y
,
Zhong
Q
,
Pang
W
,
Li
Z
, et al
.
Autophagy induction promoted by m(6)A reader YTHDF3 through translation upregulation of FOXO3 mRNA
.
Nat Commun
2022
;
13
:
5845
.
28.
Röhrig
F
,
Schulze
A
.
The multifaceted roles of fatty acid synthesis in cancer
.
Nat Rev Cancer
2016
;
16
:
732
49
.
29.
Srivastava
A
,
Ambrósio
DL
,
Tasak
M
,
Gosavi
U
,
Günzl
A
.
A distinct complex of PRP19-related and trypanosomatid-specific proteins is required for pre-mRNA splicing in trypanosomes
.
Nucleic Acids Res
2021
;
49
:
12929
42
.
30.
de Moura
TR
,
Mozaffari-Jovin
S
,
Szabó
CZK
,
Schmitzová
J
,
Dybkov
O
,
Cretu
C
, et al
.
Prp19/Pso4 is an autoinhibited ubiquitin ligase activated by stepwise assembly of three splicing factors
.
Mol Cell
2018
;
69
:
979
92
.
31.
Fu
Y
,
Dominissini
D
,
Rechavi
G
,
He
C
.
Gene expression regulation mediated through reversible m⁶A RNA methylation
.
Nat Rev Genet
2014
;
15
:
293
306
.
32.
Guo
X
,
Li
K
,
Jiang
W
,
Hu
Y
,
Xiao
W
,
Huang
Y
, et al
.
RNA demethylase ALKBH5 prevents pancreatic cancer progression by posttranscriptional activation of PER1 in an m6A-YTHDF2-dependent manner
.
Mol Cancer
2020
;
19
:
91
.
33.
Zhang
S
,
Zhao
BS
,
Zhou
A
,
Lin
K
,
Zheng
S
,
Lu
Z
, et al
.
m(6)A demethylase ALKBH5 maintains tumorigenicity of glioblastoma stem-like cells by sustaining FOXM1 expression and cell proliferation program
.
Cancer Cell
2017
;
31
:
591
606
.
34.
Lee
Y
,
Choe
J
,
Park
OH
,
Kim
YK
.
Molecular mechanisms driving mRNA degradation by m(6)A modification
.
Trends Genet
2020
;
36
:
177
88
.
35.
Deng
X
,
Su
R
,
Weng
H
,
Huang
H
,
Li
Z
,
Chen
J
.
RNA N6-methyladenosine modification in cancers: current status and perspectives
.
Cell Res
2018
;
28
:
507
17
.
36.
Ma
L
,
Chen
T
,
Zhang
X
,
Miao
Y
,
Tian
X
,
Yu
K
, et al
.
The m(6)A reader YTHDC2 inhibits lung adenocarcinoma tumorigenesis by suppressing SLC7A11-dependent antioxidant function
.
Redox Biol
2021
;
38
:
101801
.
37.
Zerkalenkova
E
,
Lebedeva
S
,
Borkovskaia
A
,
Soldatkina
O
,
Plekhanova
O
,
Tsaur
G
, et al
.
BTK, NUTM2A, and PRPF19 are novel KMT2A partner genes in childhood acute leukemia
.
Biomedicines
2021
;
9
:
924
.
38.
He
Y
,
Huang
C
,
Cai
K
,
Liu
P
,
Chen
X
,
Xu
YI
, et al
.
PRPF19 promotes tongue cancer growth and chemoradiotherapy resistance
.
Acta Biochim Biophy Sin
2021
;
53
:
893
902
.
39.
Cai
Y
,
Chen
K
,
Cheng
C
,
Xu
Y
,
Cheng
Q
,
Xu
G
, et al
.
Prp19 is an independent prognostic marker and promotes neuroblastoma metastasis by regulating the Hippo-YAP signaling pathway
.
Front Oncol
2020
;
10
:
575366
.
40.
Yuan
Y
,
Zhao
Z
,
Xue
L
,
Wang
G
,
Song
H
,
Pang
R
, et al
.
Identification of diagnostic markers and lipid dysregulation in esophageal squamous cell carcinoma through lipidomic analysis and machine learning
.
Br J Cancer
2021
;
125
:
351
7
.
41.
Tao
M
,
Luo
J
,
Gu
T
,
Yu
X
,
Song
Z
,
Jun
Y
, et al
.
LPCAT1 reprogramming cholesterol metabolism promotes the progression of esophageal squamous cell carcinoma
.
Cell Death Dis
2021
;
12
:
845
.
42.
Tontonoz
P
,
Spiegelman
BM
.
Fat and beyond: the diverse biology of PPARγ
.
Annu Rev Biochem
2008
;
77
:
289
312
.
43.
Sawayama
H
,
Ishimoto
T
,
Watanabe
M
,
Yoshida
N
,
Sugihara
H
,
Kurashige
J
, et al
.
Small molecule agonists of PPAR-gamma exert therapeutic effects in esophageal cancer
.
Cancer Res
2014
;
74
:
575
85
.
44.
Yu
J
,
Shen
B
,
Chu
ES
,
Teoh
N
,
Cheung
KF
,
Wu
CW
, et al
.
Inhibitory role of peroxisome proliferator-activated receptor gamma in hepatocarcinogenesis in mice and in vitro
.
Hepatology
2010
;
51
:
2008
19
.
45.
Ogino
S
,
Shima
K
,
Baba
Y
,
Nosho
K
,
Irahara
N
,
Kure
S
, et al
.
Colorectal cancer expression of peroxisome proliferator-activated receptor gamma (PPARG, PPARgamma) is associated with good prognosis
.
Gastroenterology
2009
;
136
:
1242
50
.
46.
Shimano
H
,
Sato
R
.
SREBP-regulated lipid metabolism: convergent physiology - divergent pathophysiology
.
Nat Rev Endocrinol
2017
;
13
:
710
30
.
47.
Zhang
X
,
Lu
N
,
Wang
L
,
Wang
Y
,
Li
M
,
Zhou
Y
, et al
.
Recent advances of m(6)A methylation modification in esophageal squamous cell carcinoma
.
Cancer Cell Int
2021
;
21
:
421
.
48.
Cui
Y
,
Zhang
C
,
Ma
S
,
Li
Z
,
Wang
W
,
Li
Y
, et al
.
RNA m6A demethylase FTO-mediated epigenetic upregulation of LINC00022 promotes tumorigenesis in esophageal squamous cell carcinoma
.
J Exp Clin Cancer Res
2021
;
40
:
294
.
49.
Li
R
,
Zeng
L
,
Zhao
H
,
Deng
J
,
Pan
L
,
Zhang
S
, et al
.
ATXN2-mediated translation of TNFR1 promotes esophageal squamous cell carcinoma via m(6)A-dependent manner
.
Mol Ther
2022
;
30
:
1089
103
.
50.
Nagaki
Y
,
Motoyama
S
,
Yamaguchi
T
,
Hoshizaki
M
,
Sato
Y
,
Sato
T
, et al
.
m(6) A demethylase ALKBH5 promotes proliferation of esophageal squamous cell carcinoma associated with poor prognosis
.
Genes Cells
2020
;
25
:
547
61
.
51.
Xiao
D
,
Fang
TX
,
Lei
Y
,
Xiao
SJ
,
Xia
JW
,
Lin
TY
, et al
.
m(6)A demethylase ALKBH5 suppression contributes to esophageal squamous cell carcinoma progression
.
Aging
2021
;
13
:
21497
512
.
52.
Chen
P
,
Li
S
,
Zhang
K
,
Zhao
R
,
Cui
J
,
Zhou
W
, et al
.
N6-methyladenosine demethylase ALKBH5 suppresses malignancy of esophageal cancer by regulating microRNA biogenesis and RAI1 expression
.
Oncogene
2021
;
40
:
5600
12
.
53.
He
L
,
Li
H
,
Wu
A
,
Peng
Y
,
Shu
G
,
Yin
G
.
Functions of N6-methyladenosine and its role in cancer
.
Mol Cancer
2019
;
18
:
176
.