E2F1-mediated ESPL1 transcriptional activation predicts poor prognosis and promotes the proliferation of leiomyosarcoma

Datasets and bioinformatics analysis

The Cancer Genome Atlas-Sarcoma (TCGA-SARC) dataset was acquired using the University of California, Santa Cruz (UCSC) Xena Browser (https://xena.ucsc.edu/).^[11] The dataset includes various subtypes of sarcoma: LMS, dedifferentiated liposarcoma (DDLPS), myxofibrosarcoma (MFS), malignant peripheral nerve sheath tumors (MPNST), synovial sarcoma (SS), and undifferentiated pleomorphic sarcoma (UPS). Normal muscle tissue gene expression data were obtained from the genotype-tissue expression (GTEx) project. A total of 396 normal muscle samples were used for comparison with LMS tissues in the TCGA-SARC dataset. The Z scores of 1387 constituent PAthway Representation and Analysis by Direct Reference on Graphical Models (PARADIGM) pathways in LMS patients in the TCGA-SARC cohort were also obtained using the UCSC Xena browser. The correlation between ESPL1 expression and the activity of each pathway was estimated by calculating Pearson’s correlation coefficient. Only pathways with Pearson’s r > 0.80 were considered for further analysis.

The transcript expression data of ESPL1 during different cell cycle phases were obtained from the Human Protein Atlas (HPA) database (https://www.proteinatlas.org/),^[12] specifically for the U-2 OS cell line.

The Cistrome Data Browser (http://cistrome.org/db/), a comprehensive database for regulatory genomics^[13] was used to identify the chromatin regulators and transcription factors potentially binding to the ESPL1 promoter (HPRM47346; from Genecopoeia) within 1 kb of the transcription start site (TSS). The correlation between ESPL1 expression and the expression of the identified chromatin regulators and transcription factors was assessed in primary LMS samples from the TCGA-SARC. Significant correlations (Pearson’s r > 0.50) were considered for further analysis.

Kaplan-Meier survival analyses

Kaplan-Meier survival curves were generated for progression-free interval (PFI) and overall survival (OS) in the entire sarcoma cohort and the LMS subset. Patients with primary tumors were grouped based on the median expression of ESPL1.

Immunohistochemistry (IHC) analysis

IHC images of ESPL1 in normal smooth muscle tissue were obtained from the HPA database.^[12] Commercial tissue microarrays containing LMS tissues were obtained from Bioaitech (Xi’an, China) and used to examine ESPL1 protein expression with a BOND-III Automated IHC/ISH Stainer (Leica Biosystems, Germany). The tissue sections were first deparaffinized, rehydrated, and subjected to antigen retrieval using BOND Epitope Retrieval Solution 1 (Leica Biosystems). Primary antibodies specific to ESPL1 (LS-B8366; LSBio, Shirley, MA, USA) were applied, and visualization was achieved using the BOND Polymer Refine Detection system (DS9800-CN; Leica Biosystems, Germany) and a real-time digital pathology system (Aperio LV1; Leica Biosystems).

Cell culture and transfection

Human LMS representative SK-LMS-1 cells (HTB-88) were obtained from the American Type Culture Collection (Manassas, VA, USA) and were cultured as recommended. We performed short tandem repeat (STR) identification and confirmed that the STR profile of this cell line matched the records in the database, confirming its identity. In addition, we conducted mycoplasma testing, and the results indicated that the cell line was free of mycoplasma contamination. The cells were transfected with small interfering RNAs (50 nM) targeting ESPL1 (siESPL1#1, #2, #3), E2F1 (siE2F1#1, #2, #3) or nontargeting control siRNA (siNC) using Lipofectamine RNAiMAX (Thermo Fisher Scientific, USA).

Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was extracted from transfected SK-LMS-1 cells using the RNeasy Plus Mini Kit (74134; Qiagen, Germany) and reverse-transcribed into cDNA using the High-Capacity cDNA Reverse Transcription Kit (4368814; Applied Biosystems, USA). ESPL1 and E2F1 mRNA levels were quantified by qRT-PCR using SYBR Green Master Mix (Applied Biosystems) on a QuantStudio 3 Real-Time PCR System (Applied Biosystems). The following primers were used: ESPL1, forward sequence, 5’-ATCTCTGTCAGTCGGACCTGCA-3’; reverse sequence, 5’-CAGGTGGACCTTCTTCACAGAG-3’; E2F1, forward sequence, 5’- GGACCTGGAAACTGACCATCAG-3’; reverse sequence, 5’- CAGTGAGGTCTCATAGCGTGAC-3’; GAPDH (as an internal reference), forward sequence, 5’- GTCTCCTCTGACTTCAACAGCG-3’; reverse sequence, 5’- ACCACCCTGTTGCTGTAGCCAA-3’. Relative expression levels were calculated using the 2^−ΔΔCt method.

Western blotting

Total protein was extracted from transfected SK-LMS-1 cells using radioimmunoprecipitation assay (RIPA) buffer supplemented with protease inhibitors (P0013; Beyotime, China). The protein concentration was determined using the bicinchoninic acid (BCA) protein assay kit (P0010; Beyotime). Protein lysates were separated by sodium dodecyl sulfate polyacrylamide gel electrophores (SDS-PAGE) (30 μg/lane) and transferred to polyvinylidene fluoride (PVDF) membranes (FFP39; Beyotime). The membranes were blocked with 5% nonfat milk in Tris Buffered Saline with Tween 20 (TBST) for 1 h at room temperature and then probed overnight at 4°C with primary antibodies specific to ESPL1 (1:1000; LS-B8366; LSBio), E2F1 (1:1000; 3742; Cell Signaling Technology), and β-actin (1:20000; 66009-1-Ig; Procell). After being washed 3 times with TBST for 10 min each, the membranes were incubated with an HRP-conjugated secondary antibody (1:5000; SA00001-2; Proteintech, China) for 1 h at room temperature. The protein bands were visualized using an ECL substrate (32209; Thermo Fisher Scientific) and detected using a chemiluminescent imaging system (Tanon 5200; Tanon Science and Technology, China). Densitometric analysis was performed using ImageJ software (NIH, USA), with β-actin serving as the loading control.

Cell growth assessment

The cell counting kit-8 (CCK-8) assay (CK04; Dojindo, Japan) was used to evaluate cell viability at 0, 24, 48, and 72 h posttransfection. The absorbance at 450 nm was measured using a microplate reader (BioTek Epoch, Agilent Technology, USA). Transfected SK-LMS-1 cells were seeded into 12-well plates and cultured for 10‒14 days. Colonies were fixed with methanol, stained with 0.5% crystal violet for 30 min, and counted. The colony formation efficiency was normalized to that of the control group.

Flow cytometry analysis of the cell cycle distribution

SK-LMS-1 cells were fixed in ice-cold ethanol, stained with propidium iodide (PI) solution containing RNase A (#4087; Cell Signaling Technology, USA), and analyzed using a BD Accuri C6 flow cytometer. The cell cycle phases were quantified using NovoExpress (v.1.5.2; Agilent Technology).

Chromatin immunoprecipitation (ChIP) and qPCR

The ChIP assay was performed using the SimpleChIP Enzymatic Chromatin IP Kit (#9002; Cell Signaling Technology, USA). SK-LMS-1 cells were fixed with formaldehyde to cross-link proteins to DNA, lysed, and sonicated to shear the DNA into 200–1000 bp fragments. Chromatin fragments were immunoprecipitated with antibodies against E2F1 or normal IgG as a control. The DNA-protein complexes were reversed, and the DNA was purified. Precipitated DNA was analyzed by qPCR using primers specifically designed for the E2F1-binding region in the ESPL1 promoter (24–125 bp), with the following primers: Forward, 5’-GTACTGGTCAGGCGGTTAAG-3’ and reverse, 5’-TTCACTTCAGCACGTACCC-3.’ qPCR was then performed as described above. The results are expressed as % input normalized to the amount of input chromatin.

Dual-luciferase reporter assay

The full ESPL1 promoter region (−1402 to +213) and a truncated promoter region (−100 to +213) were chemically synthesized, cloned, and inserted into the pGL3-basic vector (E1751; Promega, Germany) upstream of the firefly luciferase gene. SK-LMS-1 cells were cotransfected with siNC or siE2F1 (#2) and the pGL3 constructs using Lipofectamine 3000 (Thermo Fisher Scientific). At 24 h posttransfection, luciferase activity was measured using the Dual-Luciferase Reporter Assay System (E1910; Promega).

Statistical analysis

All statistical analyses were performed using GraphPad Prism 10.1.2. The log-rank test was used to assess the differences in survival between the groups with high and low ESPL1 expression. A receiver operating characteristic (ROC) curve was generated to assess the diagnostic value of ESPL1 expression for differentiating between normal muscle tissues and primary LMS. For comparisons involving multiple groups, a one-way analysis of variance followed by Tukeycee post hoc test was applied, whereas unpaired t tests were used for the comparisons between the two groups. The significance threshold was set to P < 0.05.

ESPL1 expression and prognostic significance in sarcomas and LMS

ESPL1 expression across different sarcoma subtypes was analyzed in the TCGA-SARC dataset. A significant difference was observed only between the DDLPS and UPS groups (P = 0.014) [Figure 1a]. A total of 259 primary sarcoma patients with survival data and 104 LMS patients were included in the survival analysis. K-M survival analysis revealed that higher ESPL1 expression (n = 129; top 50%) was significantly associated with worse PFI [Figure 1b; P = 0.003] and OS [Figure 1c; P = 0.02] than lower ESPL1 expression (bottom 50%) in the entire TCGA-SARC cohort.

Close

Figure 1:

ESPL1 expression and prognostic significance in sarcomas and leiomyosarcoma. (a) Violin plot showing ESPL1 expression across different sarcoma subtypes in the TCGA-SARC dataset. (b and c) Survival analysis for PFI (b) and OS (c) in primary sarcomas in the TCGASARC cohort. Patients were grouped by high (orange) or low (pink) ESPL1 expression. The median ESPL1 expression level was set as the cutoff. (d) Violin plot comparing ESPL1 expression between normal muscle tissues from the GTEx dataset (GTEx-NS, n=396) and primary LMS tissues from the TCGA-SARC cohort (n=104). (e) ROC curve illustrating the diagnostic value of ESPL1 expression for differentiating between normal muscle tissues and primary LMS tissues using the expression data in panel (d). (f and g) Survival analysis for PFI (f) and OS (g) in patients with primary LMS in the TCGA-SARC cohort. Patients were grouped by high (orange) or low (pink) ESPL1 expression. The median ESPL1 expression level was set as the cutoff. (h) Immunohistochemistry (IHC) analysis of ESPL1 protein expression. IHC staining of ESPL1 in normal smooth muscle tissue was performed using the HPA database via https://images.proteinatlas.org/73188/166539_B_5_7.jpg. A commercial tissue microarray of LMS tissues showed positive IHC staining for ESPL1 at the protein level. The insets show higher magnification images of the staining patterns. ESPL1: Extra spindle pole bodies like 1, TCGA-SARC: The cancer genome atlas sarcoma, GTEx: Genotype-tissue expression, LMS: Leiomyosarcoma, ROC: Receiver operating characteristic, DDLPS: Dedifferentiated liposarcoma; MFS: Myxofibrosarcoma; MPNST: Malignant peripheral nerve sheath tumors; SS: Synovial sarcoma; UPS: Undifferentiated pleomorphic sarcoma; PFI: Progression-free interval; OS: Overall survival; TPR: True positive rate; FPR: False positive rate; PBS: Phosphate buffered saline.

Export to PPT

Since we focused on LMS, we compared ESPL1 expression in the LMS subset with that in normal muscle tissues via GTEx. ESPL1 was significantly overexpressed in LMS tissues (n = 104) compared with normal muscle tissues (GTEx-NS, n = 396) (P < 0.001) [Figure 1d]. To assess the diagnostic value of ESPL1 expression in distinguishing between normal muscle tissues and LMS, ROC curve analysis was performed. The ROC curve demonstrated high diagnostic accuracy, with an AUC of 0.993 [Figure 1e]. Survival analysis revealed that higher ESPL1 expression in the LMS subset was significantly correlated with a worse PFI [Figure 1f, P < 0.001) and OS in the LMS subset [Figure 1g, P = 0.043].

To validate ESPL1 expression at the protein level, we retrieved IHC staining data for ESPL1 in normal smooth muscle tissue from the HPA database and performed IHC analysis on a commercial tissue microarray of LMS tissues. The results confirmed positive ESPL1 staining in LMS tissues, whereas minimal ESPL1 staining was detected in normal smooth muscle tissues [Figure 1h].

We next sought to assess whether ESPL1 expression serves as an independent prognostic biomarker in LMS. The key clinicopathological parameters of LMS patients with high and low ESPL1 expression are compared in Table 1. No significant differences were observed in OS status (P = 0.160) or age at initial pathologic diagnosis (P = 0.597) between the two groups. However, the PFI status was significantly different, with a greater proportion of patients with high ESPL1 expression experiencing disease progression (36.5% vs. 20.2%, P < 0.001). The sex distribution was similar between the two groups (P = 0.676). With respect to residual tumor status, a greater percentage of patients with high ESPL1 expression had R0 resection (34.6% vs. 33.7%, P = 0.077). In addition, margin status did not differ significantly between the groups (P = 0.206), with a slightly greater proportion of negative margins in the high-ESPL1 subgroup (41.5% vs. 31.7%) [Table 1].

We subsequently conducted univariate and multivariate analyses for OS [Table 2] and PFI [Table 3] in patients with LMS. According to the univariate analysis, high ESPL1 expression was associated with poorer OS (Hazard ratio [HR] = 1.467; 95% confidence interval [CI]: 1.102–1.952; P = 0.009). High ESPL1 expression remained an independent prognostic factor for poor OS (HR = 1.837; 95% CI: 1.249–2.704; P = 0.002) after adjustment for margin status in the multivariate analysis [Table 2].

According to the univariate analysis, higher ESPL1 expression was associated with a significantly shorter PFI (HR = 1.543; 95% CI: 1.236–1.925; P < 0.001). This relationship remained strong in the multivariate analysis, where ESPL1 expression independently predicted a worse PFI (HR = 1.683; 95% CI: 1.270–2.231; P < 0.001) [Table 3]. Patients with R2 residual tumors had a significantly worse PFI (HR = 4.996; 95% CI: 1.764–14.155; P = 0.002). However, in the multivariate analysis, the effect of residual tumor status was no longer significant (HR = 3.752; 95% CI: 0.651–21.615; P = 0.139) [Table 3].

Knockdown of ESPL1 induces G2/M arrest in SKLMS-1 cells

To explore the molecular mechanisms associated with ESPL1 in LMS, we analyzed the correlation between ESPL1 expression and the activity of 1387 constituent PARADIGM pathways in LMS cases from the TCGA-SARC dataset [Supplementary Table 1]. The heatmap [Figure 2a] shows the most significant correlations (Pearson’s r > 0.80) between ESPL1 expression and biological pathways. Higher ESPL1 expression levels were associated with increased activity in pathways such as the following: Mitotic_ prometaphase; kinesins; E2F_transcription_factor_network; resolution_of_Sister_Chromatid_cohesion; Aurora_B_ signaling; PLK1_signaling_events; FOXM1_transcription_ factor_network; E2F_mediated_regulation_of_DNA_ replication; G1/S-Specific_Transcription; mitotic_metaphase/ anaphase_transition; and separation_of_Sister_chromatids [Figure 2a]. These pathways are generally related to mitotic processes, transcription factor networks, and chromatid cohesion, suggesting a role for ESPL1 in cell cycle regulation and genomic stability.

Close

Figure 2:

Knockdown of ESPL1 induces G2/M arrest in SK-LMS-1 cells. (a) Heatmap showing ESPL1 expression and the activity of constituent PARADIGM pathways in LMS patients from the TCGA-SARC dataset. The rows represent individual LMS cases (n=104), and the columns represent various biological pathways. Pathways with Pearson’s r > 0.80 are shown. High expression and pathway activity are indicated in red, whereas low expression and pathway activity are indicated in green. (b) Box plot depicting the variation in normalized ESPL1 transcript expression (log2[TPM]) during different phases of the cell cycle (G1, S-tr, and S&G2) in U-2 OS cells. Images were obtained from the HPA: https://www.proteinatlas.org/ENSG00000135476-ESPL1/subcellular#cell_cycle. (c) Scatter plot showing the continuous variation in ESPL1 transcript expression over a 24-h cell cycle period in U-2 OS cells. (d) qRT-PCR validation of ESPL1 knockdown in SK-LMS-1 cells transfected with siNC or siESPL1 (#1, #2, #3). (e and f) ESPL1 protein levels in SK-LMS-1 cells 24 h after transfection with siNC or siESPL1 (#1, #2, #3). (g) 24 h after the transfection of siNC or siESPL1 (#2, #3), the cell viability at the indicated time points was measured and compared. # indicates a comparison between siESPL1#2 and siNC. ^✶ indicates a comparison between siESPL1#3 and siNC. (h and i) Effect of ESPL1 knockdown on the clonogenic capacity of SK-LMS-1 cells. Representative images (h) of crystal violet-stained colonies and quantification (i) of relative colony formation are shown. (j) Cell cycle distribution of SK-LMS-1 cells transfected with siNC or siESPL1 (#2 and #3). PI staining was used to quantify the cell cycle phases. (k) Quantification of the cell cycle distribution from the flow cytometry data. The data are shown as means ± SD from three independent experiments. ^✶✶ and ^##P<0.01, ^✶✶✶P<0.001. ESPL1: Extra spindle pole bodies like 1, LMS: Leiomyosarcoma, OS: Overall survival, TCGA-SARC: The Cancer Genome atlas sarcoma, qRT-PCR: Quantitative real-time polymerase chain reaction, HPA: Human protein atlas, PLK1: Polo-like kinase 1, FOXM1: Forkhead Box M1, TPM: Transcripts per million., DJF: Doublet-J discriminant function, RMS: Root mean square, si-NC: Small interfering RNA-negative control, OD: Optical density.

Export to PPT

The variation in ESPL1 transcript expression during different phases of the cell cycle was investigated in U-2 OS cells based on subcellular data from the HPA. Box plot analysis [Figure 2b] revealed that ESPL1 expression varied during the cell cycle, with higher expression observed in the S-tr and S&G2 phases than in the G1 phase. A continuous scatter plot analysis over a 24-h cell cycle period [Figure 2c] further confirmed that ESPL1 expression fluctuated, with peak levels occurring in the late S and G2/M phases. To assess the functional impact of ESPL1 knockdown, LMS representative SK-LMS-1 cells were transfected with siRNAs targeting ESPL1 (siESPL1#1, #2, #3) or nontargeting control siRNA (siNC). qRT-PCR confirmed the significant decrease in ESPL1 mRNA levels in the siESPL1 groups compared with the siNC group [Figure 2d]. Western blot analysis further validated the reduction in ESPL1 protein levels 24 h post-transfection [Figure 2e and f].

A cell proliferation assay (CCK-8) demonstrated that ESPL1 knockdown significantly reduced the viability of SK-LMS-1 cells over a 72-h period [Figure 2g]. Additionally, the colony formation assay revealed a significant decrease in clonogenic capacity upon ESPL1 knockdown [Figure 2h and i]. To determine how ESPL1 knockdown influences the cell cycle, we performed flow cytometry analyses of the cell cycle distribution using PI staining. ESPL1 knockdown resulted in an evident increase in the G2/M population, with a corresponding decrease in the G1 and S phase populations [Figure 2j and k].

Bioinformatics analysis identified E2F1 as a transcription factor with potential regulatory effects on ESPL1 expression in LMS

To identify potential transcriptional regulators of ESPL1 in LMS, we conducted a bioinformatics analysis using expression data from the TCGA-SARC dataset. We first identified 200 chromatin regulators and transcription factors that potentially bind to the ESPL1 promoter [Supplementary Material 1] within 1 kb of the TSS from the Cistrome DB [Supplementary Table 2]. Then, their expression correlation with ESPL1 in LMS patients in the TCGASARC cohort was visualized [Figure 3a] and calculated. Pearson’s correlation analysis revealed the top transcription factors and chromatin regulators with potential regulatory effects on ESPL1 expression, including RAD51 recombinase (RAD51) (r = 0.71), RAD21 cohesin complex component (RAD21) (r = 0.53), E2F1 (r = 0.49), TAR DNA binding protein (TARDBP) (r = 0.49) and structural maintenance of chromosomes 1A (SMC1A) (r = 0.48) [Figure 3b and c].

Close

Figure 3:

Bioinformatics analysis identified E2F1 as a transcription factor with potential regulatory effects on ESPL1 expression in LMS. (a) Heatmap showing the expression correlation of ESPL1 (left) and 200 chromatin regulators and transcription factors with potential binding to the ESPL1 promoter (right) (within 1 kb of the TSS site) in primary LMS cases from the TCGA-SARC dataset (n=104). (b) Table listing the top transcription factors and chromatin regulators with significant Pearson correlation coefficients with ESPL1 expression in LMS patients. (c) Scatter plot depicting the correlation between ESPL1 and E2F1 expression in LMS patients, with a Pearson correlation coefficient of 0.49 (n=104). (d) Visualization of E2F1 binding motifs (MA0024.2 and MA0024.1) predicted in the ESPL1 promoter region. Binding motifs are indicated with green and pink bars showing the predicted locations of the MA0024.2 and MA0024.1 binding sites, respectively. ESPL1: Extra spindle pole bodies like 1, LMS: Leiomyosarcoma, TCGA-SARC: The cancer genome atlas sarcoma, TSS: Transcription start site.

Export to PPT

While RAD51 and RAD21 (two chromatin regulator genes) are strongly correlated with ESPL1 expression, their roles are more related to DNA repair and homologous recombination than to direct transcriptional regulation.^[6,14] E2F1 is a well-established transcription factor that directly regulates the expression of genes involved in critical cellular processes, including cell cycle progression, DNA damage response, and apoptosis.^[15] Given the relevance of these pathways in the pathogenesis of LMS, we focused on E2F1 in this study.

To further validate this potential regulatory relationship, we examined the E2F1 binding motifs (MA0024.2 and MA0024.1) within the ESPL1 promoter region using the Scan module in JASPAR 2024 [Figure 3d].^[16] Only the predicted binding sites with relative scores >0.85 were considered for further analysis [Table 4]. The ESPL1 promoter region spans chr12: 53266897–53268512, with the TSS at position 53268299. Fourteen potential E2F1 binding sites (green and pink bars) were observed in this region [Figure 3d].

Validation of the effects of E2F1 transcriptional activation on the ESPL1 promoter in LMS

E2F1 expression was also significantly greater in LMS tissues (TCGA-SARC, n = 104) than in normal muscle tissues (GTEx-NS, n = 396) [Figure 4a]. Next, we investigated the impact of E2F1 knockdown on ESPL1 expression in SKLMS-1 cells. ESPL1 expression following transfection with E2F1 siRNAs was significantly lower than that following transfection with the nontargeting control (siNC) [Figure 4b-f].

Close

Figure 4:

Validation of the effects of E2F1 transcriptional activation on the ESPL1 promoter in LMS. (a) Violin plot comparing ESPL1 expression between normal muscle tissues from the GTEx dataset (GTEx-NS, n=396) and primary LMS tissues from the TCGA-SARC dataset (n=104). (b and c) The relative mRNA expression of E2F1 (b) and ESPL1 (c) in SK-LMS-1 cells 24 h after transfection with siNC or siE2F1 (#1, #2, #3). (d-f) Western blotting images (d) and quantification (e and f) of E2F1 and ESPL1 protein levels in SK-LMS-1 cells 24 h after transfection with siNC or siE2F1 (#1, #2, #3). (g) A schematic diagram illustrating the ESPL1 promoter region and the location of the targeting region in the ChIP-qPCR assay. (h) Luciferase reporter assay to measure the transcriptional activity of the ESPL1 promoter in SK-LMS-1 cells. The cells were cotransfected with siNC or siE2F1 (#2) and luciferase reporter constructs containing either the full ESPL1 promoter region (−1402 to +213) or the truncated promoter region (−100 to +213). The data are presented as the fold change in relative luciferase activity normalized to that of siNC. (i) ChIP-qPCR analysis of E2F1 binding to the ESPL1 promoter in SK-LMS-1 cells with or without E2F1 knockdown. IgG was used as a negative control. The enrichment of E2F1 binding is shown as the % input. ^✶✶✶P<0.001. ESPL1: Extra spindle pole bodies like 1, ChIP: Chromatin Immunoprecipitation, GTEx: Genotype-tissue expression, LMS: Leiomyosarcoma, OS: Overall survival, IgG: Immunoglobulin G, qPCR: Quantitative polymerase chain reaction.

Export to PPT

To test whether E2F1 directly regulates ESPL1 promoter activity, we performed a luciferase reporter assay. SKLMS-1 cells were cotransfected with siNC or siE2F1 (#2) and luciferase reporter constructs based on pGL3-basic containing either the full ESPL1 promoter region (pGL3/−1402~213) or a truncated promoter region (pGL3/−100~213) [Figure 4g]. The results demonstrated that both pGL3/−1402~213 and pGL3/−100~213 had significantly greater luciferase activity than the pGL3-basic control [Figure 4h]. In addition, E2F1 knockdown significantly reduced the luciferase activity of these two plasmids but had a limited influence on the activity of the pGL3-basic control [Figure 4h].

To validate the direct binding of E2F1 to the ESPL1 promoter, we designed a pair of primers covering two predicted E2F1 binding sites in the promoter region (24~125) for ChIP-qPCR analysis [Figure 4g]. ChIP-qPCR analysis was then performed in SK-LMS-1 cells with or without E2F1 knockdown. The results revealed significant enrichment of E2F1 binding at the ESPL1 promoter in control cells, which was reduced upon E2F1 knockdown [Figure 4i].