Knockout of ENO1 leads to metabolism reprogramming and tumor retardation in pancreatic cancer

The shift in glucose utilization from oxidative phosphorylation to glycolysis is the hallmark of tumor cells. The overexpression of ENO1, one of the key enzymes in the glycolysis process, has been identified in several cancers, however, its role in pancreatic cancer (PC) is yet unclear. This study identifies ENO1 as an indispensable factor in the progression of PC. Interestingly, ENO1-knockout could inhibit cell invasion and migration and prevent cell proliferation in pancreatic ductal adenocarcinoma (PDAC) cells (PANC-1 and MIA PaCa-2); meanwhile, tumor cell glucose uptake and lactate excretion also decreased significantly. Furthermore, ENO1-knockout reduced colony formation and tumorigenesis in both in vitro and in vivo tests. In total, after ENO1 knockout, 727 differentially expressed genes (DEGs) were identified in PDAC cells by RNA-seq. Gene Ontology enrichment analysis revealed that these DEGs are mainly associated with components such as the ‘extracellular matrix’ and ‘endoplasmic reticulum lumen’, and participate in the regulation of signal receptor activity. Kyoto Encyclopedia of Genes and Genomes pathway analysis revealed that the identified DEGs are associated with pathways, such as ‘fructose and mannose metabolism’, ‘pentose phosphate pathway, and ‘sugar metabolism for amino and nucleotide. Gene Set Enrichment Analysis showed that ENO1 knockout promoted the upregulation of oxidative phosphorylation and lipid metabolism pathways-related genes. Altogether, these results indicated that ENO1-knockout inhibited tumorigenesis by reducing cell glycolysis and activating other metabolic pathways by altering the expression of G6PD, ALDOC, UAP1, as well as other related metabolic genes. Concisely, ENO1, which plays a vital role in the abnormal glucose metabolism in PC, can be exploited as a target to control carcinogenesis by reducing aerobic glycolysis.


Introduction
Pancreatic cancer (PC) is a malignant tumor of the digestive system and has a high mortality rate (1). It progresses rapidly and in the lack of specific symptoms and biomarkers, its early diagnosis is challenging (2). Therefore, it is crucial to find new targets for early diagnosis and treatment of PC. The special hypoxic microenvironment involves metabolic reprogramming in pancreatic ductal adenocarcinoma (PDAC) (3). Aerobic glycolysis instead of mitochondrial oxidative phosphorylation (OXPHOS) is used for energy production, known as the "Warburg effect" (4). Glycolysis supports the vigorous growth of cancer cells by producing various substrates (5). Therefore, metabolic regulation by downregulating key glycolytic enzymes can be a novel therapy for PC.
The multifunctional glycolytic enzyme ENO1 has been shown to be commonly over-expressed in various human cancers including PC (12, 13). High expression of ENO1 is positively correlated with clinical stage, lymph node metastasis, and poor prognosis in PC. ENO1 promotes pancreatic cancer cells migration and metastasis through combining with integrins and uPAR (14). In vivo, the monoclonal antibody that block the binding of ENO1 with plasminogen inhibits metastasis formation of PDAC cells (15). In genetically engineered mice with pancreatic cancer, ENO1 DNA vaccine elicits antitumor immune responses by decreasing numbers of myeloid-derived suppressor cells and T-regulatory cells and increasing T-helper 1 and 17 responses to prolong survival (16). ENO1 may function as a promising and clinically-relevant molecular target for immunotherapeutic strategy, particularly in pancreatic cancer (15)(16)(17).
Herein, we knocked out ENO1 in pancreatic cancer cell lines and evaluated its impact on maintaining the Warburg effect and tumor growth through biochemical and functional approaches. We show that ENO1 knockout decreased PDAC cell growth, and suppressed tumorigenesis by altering the expression of metabolic pathway-related genes to trigger the metabolic patterns shift from glycolysis to OXPHOS and other metabolic pathways.

Cell culture
Two kinds of human PDAC cell lines, PANC-1 and MIA PaCa-2, were obtained from the American Type Culture Collection (ATCC). The cells were cultured in DMEM (Dulbecco's Modified Eagle Medium) added with 10% fetal bovine serum (FBS), penicillin (Thermo Fisher Scientific, USA), and 100 µg/mL of streptomycin (Thermo Fisher Scientific, USA) at 37°C and 5% CO 2 .

Vector construction and ENO1-knockout stable cell line
The single guide RNA (sgRNA) sequences were obtained based on the ENO1 gene sequence and cloned into the LentiCRISPR v2 p l a s m i d u s i n g t h e B s m B I s i t e a n d p r i m e r s 5 ' -T C G C G G G A A T C C C A C T G T T G -3 ' ( f o r w a r d ) a n d 5 ' -CAACAGTGGGATTCCCGCGA-3' (reverse). For ENO1 knockout cells, we generated the recombinant lentiviruses using ViraPower Packaging Mix (Invitrogen, CA). The two knockout cell lines were obtained by infecting with lentiviruses for 48 h, followed by 2mg/L puromycin (Thermo Fisher Scientific, USA) selection for 14 days. The selected cells were subcultured to obtain the monoclonal cells. Eventually, ENO1 knockouts were confirmed by detecting ENO1 protein deficiency in corresponding cells using Western blotting.

Cell Counting Kit-8 (CCK-8) proliferation assay
Cells (5×10 3 cells/well) were seeded in 96-well plates. The cell viability was detected every 24 h for 4 days using the CCK-8 Kit (Dojindo, CK04) according to the manufacturer's instructions. Briefly, 10% CCK-8 solution was added to each well for 2 h, and then sample optical density (OD) was recorded at 485 nm using a microplate reader (BMG LABTECH) to determine the cell proliferation rate.

Colony formation assay
Cancer cells (1000 cells/well) were seeded into 6-well plates. After 2 to 3 weeks, colonies were immobilized with 4% paraformaldehyde (PFA) for 15 min. Then, the cells were stained with 0.1% crystal violet (CV) for 15 min. The stained cells were photographed with a Canon digital camera EOS M50 and counted by ImageJ software after washing. Colony formation rate is determined by counting the colonies numbers formed per 100 cells.

Transwell assay
Cells resuspended in DMEM without FBS were seeded into the upper chamber of the transwell chamber (Corning, CLS3422) at a cell density of 4×10 4 cells/well. Then, 800 mL DMEM supplemented with 20% FBS was added to the well under the chamber. The cells adhering to the upper layer of the chamber were swabbed after 24 h and then immobilized in 4% PFA for 15 min. Cells were stained with CV and then photographed and counted in four randomly selected fields under a stereomicroscope (Olympus, Japan). For the invasion assay, Matrigel (Corning, 356237) was coated on the 8 mm pores polycarbonate membrane of the transwell chamber. 4×10 4 cells were planted into each well and cultured for 48 h and the cell migration was estimated. The migration and invasion capacity was assessed by the number of cells that migrated and stained by crystal violet in a photograph. Each experiment was repeated thrice.

Glucose consumption and lactate production
Cells were seeded into 6-well plates and the DMEM medium was changed every 12 h. 500 ml of the medium was collected at 0 and 8 h of incubation to measure the initial and final concentrations of glucose and lactate, respectively, using a Silman M900 bioprocess biochemistry analyzer. Finally, the cells were digested with 0.25% trypsin solution (Thermo Fisher Scientific, USA) and counted through the countess 3 automated cell counter (Thermo Fisher Scientific, USA). The lactate and glucose levels were normalized based on cell count.

Soft agar assay
600 ml of 0.6% agar (Sigma, A1296) was added to each well of a 24-well plate. After the 0.6% agar solidified, the digested cells mixed with 0.3% agar were inoculated into the same at 600 ml/well; each well contained 1000 cells. The colonies that were >50 mm in diameter were photographed and counted under a stereomicroscope (Olympus, Japan) after three weeks.

Tumorigenicity assay in NOG mice
8-week-old male NOD-SCID IL-2 receptor gamma null (NOG) mice were used to access the tumorigenesis role of ENO1. The mice were divided into two groups (control and test groups) with 6 mice each. 2×10 6 cells were subcutaneously transplanted into the two sides of the NOG mice. The corresponding tumor sizes in mice were recorded every 5 days. The tumor volume was calculated using the formula V =L*W 2 *0.5, where L and W represent the largest and the smallest diameters, respectively. After 5-6 weeks, the mice were sacrificed for tumor collection. All animal protocols were approved by the Peking University Cancer Hospital Animal Care and Use Committee, China.

RNA-sequencing
The cells were lysed with TRIzol reagent (Invitrogen,15596018); the library construction and RNA-sequencing were conducted by Novogene Co., Ltd (Tianjin, China). The raw sequencing data have been deposited in the Genome Sequence Archive (GSA) database under accession number HRA001089. All of the data are also available as the sequence read archive (SRA) format in the National Center for Biotechnology Information (NCBI) with the accession number of SRP414959. The gene expression differences were analyzed using the DEseq2 software (1.20.0). Data with p-value <0.05 after adjustment were considered significantly different. The Gene Ontology (GO) and KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway enrichment analyses of differentially expressed genes (DEGs) were accomplished by clusterProfiler (3.4.4); GO terms with a p-value <0.05 were considered significantly enriched. Gene set enrichment analysis (GSEA) results were visualized by Omicshare Online tools.

TIMER and GEPIA based gene expression and prognostic value analysis of ENO1 in human cancers
The TIMER (https://cistrome.shinyapps.io/timer/) and GEPIA databases (http://gepia2.cancer-pku.cn/#analysis) were used to analyze the difference in ENO1 expression levels between the human cancers and paired normal tissue. ENO1 expression levels (Log 2 TPM) are displayed using box plots, with statistical significance of differential expression evaluated using the Wilcoxon test in the TIMER 2.0 version. We use log2 (TPM + 1) for log-scale and match TCGA normal and GTEx data in GEPIA databases analysis. Furthermore, The Kaplan-Meier Plotter database (http://kmplot. com/analysis/) was used to find the prognostic value of ENO1 expression level in human cancers. We choose mRNA (RNA-seq) to start KM Plotter for pan-cancer. Patients are splited by selecting best cutoff value automatically.

Statistical analysis
Data were analyzed by GraphPad Prism 8. The data differences were assessed by t-test and those with p-value <0.05 were considered statistically significant.

ENO1 is significantly upregulated in human pancreatic cancer and other cancer tissues
The results from the TIMER database showed that ENO1 expression was significantly higher in most human cancers tissue compared to adjacent normal tissue, such as in BLCA (bladder urothelial carcinoma), BRCA (breast invasive carcinoma), CHOL (cholangiocarcinoma), COAD (colon adenocarcinoma), ESCA (esophageal carcinoma), HNSC (head and neck cancer), KIRC (kidney renal papillary cell carcinoma), KIRP (kidney renal papillary carcinoma), LIHC (liver hepatocellular carcinoma), LUAD (lung adenocarcinoma), LUSC (lung squamous cell carcinoma), P R A D ( p r o s t a t e a d e n o c a r c i n o m a ) , R E A D ( r e c t u m adenocarcinoma), STAD (stomach adenocarcinoma), THCA (thyroid carcinoma) and UCEC (uterine corpus endometrial carcinoma). Notably, ENO1 expression was significantly lower only in KICH (kidney chromophobe) ( Figure 1A). Similarly, GEPIA data also showed that ENO1 expression was high in most cancer types, which was consistent with TIMER analysis. The expression of ENO1 is significantly greater in PAAD (pancreatic adenocarcinoma) tumor tissues (T) than in normal tissues (N) with the highest Log 2 FC value ( Figure 1B and Supplementary Figure 1A). Altogether, these results suggested that ENO1 may promote the occurrence of various cancers including pancreatic cancer.

ENO1 knockout significantly reduced the growth and colony formation rate of PDAC cells
To examine the biological function(s) of ENO1 in PC, we knocked out ENO1 in PANC-1 and MIA PaCa-2 cells ( Figure 3A). Both the ENO1-knockout and corresponding control cells were cultured and tested for cell viability using the CCK-8 assay. The cell growth curves (reflected by OD values) revealed that ENO1 knockout markedly inhibited the cell proliferation in both the PDAC cells ( Figure 3B). The cells were cultured for 10-15 days, the colonies were immobilized and stained, and then photographed and counted. We found that the colony numbers of the ENO1 knockout cells were sharply decreased compared to the corresponding control cells. ENO1 knockout decreased the proliferation abilities of PANC-1 and MIA PaCa-2 cells significantly compared with the control cells ( Figures 3C, D). Thus, knockout of ENO1 could inhibit the proliferation of PDAC cells.

ENO1 knockout decreased the cell migration, invasion, and glycolysis in PDAC cells
The effect of ENO1 knockout on cell migration and invasion in PDAC cells was evaluated by transwell assay. We found that fewer cells migrated through the polycarbonate membrane in the ENO1 knockout groups than in the control groups ( Figures 4A, B). Consistently, when the polycarbonate membrane of the chamber was coated with Matrigel, the number of cells invading through the polycarbonate membrane was lower in the ENO1 knockout groups than in the control groups ( Figures 4C, D). These results revealed that ENO1 knockout significantly decreased the migration and invasion of PDAC cells. To examine the effect of ENO1 knockout on glycolysis, we detected the possible change in glucose and lactate concentrations in the two PDAC cells. The corresponding cell culture media were collected at 0 and 8 h to estimate the amounts of glucose and lactate, reflecting any possible change in glucose uptake and lactate secretion after ENO1 knockout. We found that the glucose consumption and lactate production in both the ENO1 knockout PDAC cells were significantly lower compared to that in corresponding control cells ( Figures 4E, F). This indicated that ENO1 knockout reduces glucose metabolism levels in PDAC cells.

ENO1 knockout inhibited the tumorigenicity of PDAC cells
Cells were seeded into 24-well plates and cultured in 0.3% agar for 2-3 weeks. We found that the colony numbers were significantly lower in the ENO1 knockout groups than in the control groups ( Figures 5A, B). This indicated that ENO1 knockout markedly reduced the tumorigenicity of PDAC cells in vitro. For in vivo assay, 2×10 6 control or ENO1 knockout cells were subcutaneously inoculated into NOG mice. The tumor size was recorded every 5 days to obtain the growth curves. The results showed that tumor sizes were significantly smaller in ENO1 knockout mice than in control mice, indicating that ENO1 knockout reduced the tumorigenicity of PDAC cells (Figures 5C, D).

ENO1 knockout induced compensatory upregulation of other metabolic pathways in PC
After examining the biological function of ENO1 in PC, we tried to understand its underlying molecular mechanism in abnormal cell metabolism in PC. Accordingly, RNA-seq was conducted to find the DEGs after ENO1 knockout in PANC-1 cells (Supplementary Table 1). The heatmap from the transcriptome sequencing data revealed 727 DEGs (p-value <0.05), including 370 upregulated and 357 downregulated genes ( Figure 6A). These DEGs were also analyzed by a volcano map, which showed the overall changes in gene expression after ENO1 knockout. The red and green dots represent the upregulated and downregulated DEGs, respectively ( Figure 6B). Furthermore, the identified DEGs were subjected to GO and KEGG pathway enrichment analysis. GO terms are defined as biological processes (BP), molecular functions (MF), and cellular components (CC). Based on the GO enrichment analysis, the top 20 GO terms of upregulated and downregulated genes were selected based on the -log 10 (P-value) ( Supplementary Figures 2A-F). The top 10 GO terms indicated that the main BP terms enriching the DEGs were heart development and the regulation of signaling receptor activity; the enriched CC terms were extracellular matrix and dendrites; the enriched MF terms were receptor ligand and regulator activity ( Figure 6C). KEGG pathway analysis was used to annotate the DEGs related physiological and biochemical reaction pathways; the top 20 pathways were selected based on -log 10 level (Pvalue). Many pathways showed substantial changes between the knockout and control cells ( Figure 6D). These pathways include the cancer pathways, Rap1 signaling pathway, Wnt signaling pathway, and many metabolic pathways. The number and percentages of the DEGs belonging to these pathways are listed in Figure 6D.
To understand how ENO1 participates in the abnormal metabolism of PDAC cells, we mainly focused on the identified cell metabolism-associated pathways, including the pentose phosphate pathway and metabolism of fructose, mannose, glutathione, amino, and nucleotide sugar. The DEGs involved in these four metabolic pathways and their relative expression levels (sgENO1 vs control) are listed in Supplementary Table 2. These DEGs may be the regulatory targets of ENO1. We randomly selected 7 DEGs to validate the RNAseq data by qRT-PCR, including 5 upregulated and 2 downregulated genes. We found that RNA-seq data were consistent with qRT-PCR ( Figure 6E).
In addition, we performed GSEA to explore the potential downstream pathways of ENO1. GSEA highlighted that ENO1

Discussion
Identifying the key regulators of PC is an important research challenge for the timely diagnosis and treatment of PC (19). In PC, tumor cells undergo reprogrammed metabolism to meet energy requirements and support malignant behaviors. Notably, dysregulation of ENO1 has been associated with several cancers (14,20,21). Multivariate analyses have shown that overexpression of ENO1 can be a predictor of tumor progression (11,13). These studies suggest ENO1 is a promising therapeutic and diagnostic target in human tumors.
Here, we show that ENO1 is widely expressed in normal human tissues and is significantly upregulated in most TCGA (The Cancer Genome Atlas) cancer patients. Furthermore, OS analysis found that high expression of ENO1 was significantly associated with poor prognosis of BLCA, BRCA, CESC, ESCC, HNSC, LIHC, LUAD, PAAD, and SARC cancers. Concerning the RFS, high expression of ENO1 was significantly related to poor prognosis of BRCA, LIHC, SARC, PAAD, and TGCT cancers. These findings are consistent with prior studies showing a correlation of high ENO1 expression and poorer survival in cancer patients (22,23). These data highlighted ENO1 as an oncogene in various cancers.
Recent study have shown that ENO1 overexpression promoted proliferation, invasion and migration of SKCM cells; and increased pyruvate and lactate production (24). Subsequently, we explored the hypothesis that ENO1 is one of the leading regulators of the Warburg effect and thus plays a major role in carcinogenesis and tumor maintenance. We found that ENO1 knockout markedly reduced the cell proliferation, migration, and invasion of PDAC cells; meanwhile, glycolysis levels and tumorigenicity were also reduced. RNA-seq analysis identified key 727 DEGs after ENO1 knockout, which are mainly distributed in the extracellular matrix and endoplasmic reticulum, and participate in the regulation of signal receptor activity. KEGG pathway enrichment analysis revealed that the DEGs were related to key cellular metabolism pathways, including the pentose phosphate pathway, and the metabolism of fructose, mannose, glutathione, and others. In addition, RAS oncogene family members RAP1B, RHOH and Wnt family member WNT9A expression level were significantly down-regulated in ENO1 knockout group. Interestingly, we found that the expression levels of stem cell marker CD24 and CD34 were significantly decreased after knockout of ENO1. This is consistent with the report that ENO1 regulates stem cell-like properties in cancer cells (25).
In this study, we found that ENO1 knockout in PDAC cells affected the expression of abundant metabolism-related genes, confirming that the inhibition of ENO1 decreases proliferation and also suppresses tumor growth in vivo and in vitro. G6PD in the pentose phosphate pathway was upregulated after ENO1 knockout, suggesting a shift of glucose metabolism to the pentose phosphate pathway in PDAC cells. It seems that ENO1 silencing can redistribute excessive glucose to the pentose phosphate pathway, decreasing lactate levels. Metabolic pattern changes can promote autophagy and fatty acid oxidation, reducing the growth of cancer cells (26). ENO1 knockout downregulated ALDOC levels in fructose and mannose metabolic pathways. The aberrant expression of aldolase family members has been demonstrated to promote tumor progression; ALDOC is upregulated in various cancers and acts as a regulator of Wnt signaling (27). The knockdown of ALDOC reduces cell growth, glucose uptake, and glycolysis in cancer cells (28). Silencing of UAP1 inhibited the growth and colony formation of cancer cells (29). Our RNA-seq data showed that UAP1 levels decreased affecting the amino and nucleotide sugar metabolism pathway. These genes may be indispensable to ENO1-mediated regulation of cell metabolism in PC, however, the biological functions of other DEGs in PC must be examined in the future. Meanwhile, GSEA revealed that genes encoding for mitochondrial functions and lipid metabolism were significantly upregulated in ENO1 knockout cells, indicating a shift in metabolism patterns in cancer cells.
In summary, our findings show the metabolic analysis following ENO1 knockout. ENO1 is a potential oncogene and its knockout may suppress the tumorigenicity of PDAC cells by triggering the metabolic pattern change from glycolysis to other metabolic pathways such as pentose phosphate pathway, mitochondrial OXPHOS and lipid metabolism. ENO1 can be exploited as a therapeutical target for reducing aerobic glycolysis in PC. Inhibition of ENO1 alone or in combination with other pathways activated by ENO1 knockout, opens novel avenues for future cancer therapeutic approaches. Nonetheless, this study had certain limitations. We mainly validated the biological functions of ENO1 in PDAC cells by examining "loss-of-function", while "gain-of-function" studies can further uncover the mechanism of ENO1 activating intracellular and extracellular signals in cancer cells.

Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: The raw sequencing data have been deposited in the Genome Sequence Archive (GSA) database under accession number HRA001089. All of the data are also available as the sequence read archive (SRA) format in the National Center for Biotechnology Information (NCBI) with the accession number of SRP414959.

Ethics statement
The animal study was reviewed and approved by Peking University Cancer Hospital Animal Care and Use Committee.