The human PDAC cell line PANC-1 was obtained from the American Type Culture Collection (ATCC) and cultured in DMEM medium (Gibco, Carlsbad, CA) supplemented with 10% FBS (Gibco, Carlsbad, CA) and 1% penicillin/streptomycin (Gibco, Carlsbad, CA) in 5% CO₂ at 37°C.

Total proteins were extracted in RIPA buffer, and the protein concentration was measured by BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA). Protein was resolved in 10% Tris-SDS-PAGE gels and transferred to PVDF membranes (Millipore, Darmstadt, Germany). The membranes were incubated with anti-human SQLE antibody (Proteintech, Chicago, IL) at a dilution of 1:1000 and then probed with HRP-conjugated secondary antibody (Proteintech, Chicago, IL).

The miR-363-3p mimics (Genepharma, Shanghai, China) or negative control (NC) was transfected into PANC-1 cells using Lipofectamine 3000 Transfection Reagent (Invitrogen, Karlsruhe, Germany) according to the manufacturer’s instructions.

Cells were plated in 96-well plates at a density of 2000 cells per well. Cell viability was assessed using CCK-8 (Gaithersburg, MD). OD450 values were determined on Day 0, 1, 2, 3, and 4.

TIMER (27) and TIMER2.0 (28) were used as servers for the comprehensive analysis of SQLE expression in 33 types of human cancer, infiltration of tumor immune cells, and the expression of immune checkpoints in PAAD. EPIC (29) and McP-Counter (30) were used to validate the immune cell infiltration from SQLE expression profiles. One-way ANOVA was used to test the significant differences. Statistical significance was set at p < 0.05.

GEPIA (31) is a web tool for gene expression analyses based on The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression databases. We used GEPIA to analyze SQLE expression in 10 types of human cancers, namely, ACC, DLBC, LAML, LGG, OV, PAAD, SKCM, TGCT, THYM, and UCS. Statistical significance was set at p < 0.05. We also conducted survival analyses for SQLE, including overall survival (OS) and disease-free survival (DFS). The correlation of SQLE with ACAT2, HMGCR, HMGCS1, IDI1, and LDLR in PAAD and pan-cancer was analyzed, and the top 100 SQLE-correlated genes were identified using GEPIA.

The Starbase database (32) was first used to predict the miRNAs upstream of SQLE. PITA, RNA22, miRmap, microT, miRanda, PicTar, and TargetScan were used to identify miRNAs binding to SQLE, and miRNAs that were present in more than two programs were included in further analyses. We also used starBase to perform miRNA expression and correlation analyses for miRNA and SQLE in PAAD.

Kaplan-Meier plotter (33), a database evaluating the genes or miRNAs that are associated with survival in human cancer types, including PAAD, was used to perform survival analysis for miRNAs in PAAD. A log-rank p < 0.05 was defined as statistically significant.

cBioPortal web (34, 35) was used to analyze the alteration frequency, mutation type, and copy number alteration of SQLE in human cancers. The mutated site information of SQLE is displayed in a schematic diagram of the protein structure. The prognostic value of SQLE alterations, including OS, disease-specific survival (DSS), DFS, and progression-free survival (PFS), was determined using survival analysis. In addition, immunohistochemical images of SQLE in tumor and normal tissues were obtained. Log-rank p-values were also generated.

TISIDB is a web portal for tumor and immune system interaction (36). The relationship between SQLE expression and tumor immune biomarkers in PAAD, including lymphocytes, MHC molecules, immune inhibitors, and immunostimulators, was analyzed using TISIDB. The p-value and Spearman’s correlation coefficients (rho) were calculated automatically.

STRING website (37) was used to determine the SQLE-binding proteins network. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis was performed by “hiplot” (unpublished, https://hiplot.com.cn), which is a free and comprehensive cloud platform for scientific computation and visualization. Gene ontology (GO) analyses, including biological process (BP), cellular component (CC), and molecular function (MF), were obtained from the DAVID database (https://david.ncifcrf.gov). A two-tailed p < 0.05 was considered statistically significant.

Data are shown as the mean ± standard deviation of at least three independent experiments. Kaplan-Meier survival analysis was used to compare survival times with the log-rank test. Spearman’s correlation coefficient was used to determine the relationship between the two variables. Statistical significance was set at p < 0.05.

We first explored SQLE expression levels in 33 types of human cancers based on TCGA dataset. As shown in Figure 1A, SQLE expression was significantly higher in tumors than in normal tissues in BLCA, BRCA, CESC, COAD, ESCA, HNSC, LIHC, LUSC, READ, STAD, and UCEC. SQLE expression was distinctly downregulated in KIRC, KIRP, PRAD, and THCA cells. Owing to an insufficient number of normal tissues as controls for several cancer types in the TCGA dataset, we verified the difference in SQLE expression between normal and tumor tissues in 10 types of human cancers by including normal tissue from the GTEx consortium of the GEPIA database. SQLE expression level was prominently increased in ACC, DLBC, OV, PAAD, THYM, and UCS and was dramatically downregulated in LAML. However, no significant differences were observed in LGG, SKCM, and TGCT (Figures 1B–K). Notably, SQLE expression in PAAD was markedly upregulated when an adequate number of normal tissues were used as controls. In summary, aberrant SQLE expression was observed in 22 types of human cancers, implying that SQLE has a tumorigenic function.

To further reveal the significance of SQLE in tumors, we performed survival analyses, including OS and DFS, in 33 types of human cancers. As shown in Figure 2, high SQLE expression predicted unfavorable OS in the following 11 cancer types: ACC (p = 0.02), BRCA (p = 0.041), CESC (p = 0.018), HNSC (p < 0.001), KIRP (p = 0.021), LUAD (p < 0.001), MESO (p = 0.016), PAAD (p = 0.0031), SARC (p < 0.001), THCA (p < 0.001), and UVM (p = 0.0026). Moreover, overexpression of SQLE was linked to poor DFS in ACC, BLCA, HNSC, LUSC, PAAD, SARC, and UVM (Supplementary Figure S1, p < 0.05). In other types of human cancers, there was no significant difference observed in SQLE expression between the high and low groups. Taken together, our results reveal that SQLE overexpression correlates with poor prognosis in patients with PAAD.

To investigate the frequency and category of SQLE mutations in human cancers, we conducted a gene alteration analysis. The highest alteration frequency of SQLE (> 25%) was observed in patients with ovarian epithelial tumors, with “amplification” as the dominant type (Figure 3A). Significantly, more than 10% of SQLE alterations (including “amplification” and “mutation”) were detected in PAAD patients (Figure 3A). Furthermore, we explored the location and number of SQLE alterations and found that the P85Lfs*25/E86* domain was detected in 4 cases, which was the most mutated location (Figure 3B). Additionally, the relationship between SQLE alterations and PAAD prognosis was demonstrated. Our results indicate that PAAD patients with SQLE alterations had worse OS (p = 2.603e-4), DSS (p = 0.0347), DFS (p = 1.021e-3), and PFS (p = 1.425e-3) than patients without SLQE alterations (Figures 3C–F). Together, SQLE alterations were frequently probed in PAAD and found associated with an unfavorable prognosis in patients with PAAD.

MicroRNAs (miRNAs) can bind to and regulate the expression of target genes. To identify the miRNAs that regulate SQLE expression, we analyzed the upstream miRNAs that could potentially target SQLE. We found 21 miRNAs that could be responsible for regulating SQLE expression in the pan-cancer analysis (Table 1). Next, we focused on these miRNAs in PAAD. As shown in Figures 4A–C, the expression of hsa-miR-194-5p, hsa-miR-363-3p, and hsa-miR-429 was different in the tumor and normal tissues, and therefore these miRNAs were confirmed as vital regulatory molecules (p < 0.05). High expression of these three miRNAs predicted favorable OS in PAAD (Figures 4D–F, p < 0.05). This phenomenon was not observed for the other 18 miRNAs in PAAD. It is well known that miRNAs negatively regulate their target genes (38). As presented in Table 1, SQLE expression showed a negative correlation with hsa-miR-363-3p but a positive correlation with hsa-miR-194-5p and hsa-miR-429 in PAAD. Thus, we hypothesized that hsa-miR-363-3p is an upstream miRNA of SQLE.

To explore the function of hsa-miR-363-3p in PAAD, we performed in vitro experiments using miR-363-3p mimic. Western blot results confirmed that both 50 nM and 100 nM mimics could effectively knock down the expression of SQLE in PANC-1 (Figure 4G). Subsequently, CCK8 results showed that the mimic could inhibit the proliferation ability of PANC-1 (Figure 4H). These results indicated that miR-363-3p could regulate the expression of SQLE and then inhibit cell proliferation in PAAD.

SQLE is a key enzyme in cholesterol metabolism and is involved in important lymphocyte functions (39, 40). Therefore, we explored the relationship between SQLE expression and immune cell infiltration in PAAD patients. The copy number of SQLE could affect the infiltration of B cells, CD8+ T cells, and CD4+ T cells (Figure 5A). In addition, SQLE expression was negatively correlated with CD4+ T cells (Figure 5B). In contrast, SQLE expression positively correlated with the infiltration of CD8+ T cells (Figure 5C) and neutrophils (Figure 5D). Our results also demonstrated that SQLE expression did not affect the infiltration of the other three types of immune cells: B cells, dendritic cells (DCs), and macrophages (Figures 5E–G).

Furthermore, these results were validated in 178 patients with PAAD from the TCGA cohort. Patients were divided into two groups according to the median expression level of SQLE. McP-Counter and EPIC methods were performed to validate the immune cell infiltration in 2 groups. The McP-Counter results showed that there were significant differences in T cell, CD8+ T cell, B cell, NK cell, DC, and endothelial cell (Figure 5H), which was consistent with EPIC results (Figure 5I). In conclusion, SQLE expression has a complex regulatory effect on immune cell infiltration in PAAD.

To further investigate the relationship between SQLE and tumor immune biomarkers, we used the GEPIA and TISIDB databases. Our results showed that SQLE expression was significantly negatively correlated with PDCD1 (Figure 6A), LAG3 (Figure 6C), cytotoxic T-lymphocyte associated protein 4 (CTLA4, Figure 6E), and CD160 (Figure 6G)-all checkpoint inhibitors in the GEPIA database analysis. Our TISIDB database analysis confirmed these findings. SQLE expression was related to PDCD1 (Figure 6B, rho = -0.347, p < 0.01), LAG3 (Figure 6D, rho = -0.334, p < 0.01), CTLA4 (Figure 6F, rho = -0.241, p < 0.01), and CD160 (Figure 6H, rho = -0.447, p < 0.01). We performed further analyses to reveal the correlation among SQLE expression, copy number, methylation, and tumor immune features in PAAD, including lymphocytes (Supplementary Figure S2-A), immuno-inhibitors (Supplementary Figure S2-B), MHC molecules (Supplementary Figure S2-C), and immunostimulators (Supplementary Figure S2-D). Remarkably, the relationship between SQLE expression and tumor immune features was always contrary to the results of SQLE methylation (Supplementary Figure S2). Our results indicate that SQLE might function as a regulator of the immune microenvironment in PAAD.

SQLE is a pivotal gene regulating cholesterol biosynthesis. Therefore, we performed an enrichment analysis of SQLE-related partners. A list of 50 SQLE-binding proteins was obtained from the STRING database. We constructed a network of 20 proteins that were most strongly associated (Figure 7A), and most of these proteins were involved in cholesterol metabolism. The top 100 genes related to SQLE expression pan-cancer were selected from the GEPIA2 database analysis. As presented in Figures 7B–F, SQLE expression level was positively correlated with ACAT2 (R = 0.53), HMGCR (R = 0.5), HMGCS1 (R = 0.56), IDI1 (R = 0.51), and LDLR (R = 0.49) genes (all p < 0.001). We obtained similar results for PAAD (Figures 7G–K, all p < 0.001). A combined analysis of the two datasets suggested three common molecules, namely, DHCR7, NSDHL, and MSMO1 (Figure 7L). Subsequently, we conducted KEGG and GO enrichment analyses. The results of the former showed that “metabolic pathways” and “steroid biosynthesis” were involved in the function of SQLE in carcinogenesis (Figure 7M). The results of the latter implied that these genes were related to oxidation-reduction, cholesterol biosynthesis, iron ion binding, and oxidoreductase activity, among other reactions (Figure 7N). The annotations of the X-axis in Figure 7N are presented in Supplementary Table S1.

TMB is a favorable predictor of immunotherapy. Our results suggest that SQLE alteration correlated with high TMB pan-cancer and in PAAD (Figure 8A, p < 0.001). We also analyzed the results of two clinical trials of anti-PD1 treatment in melanoma and found that SQLE expression was inversely associated with cytotoxic lymphocyte levels (CTLs), OS, and PFS in melanoma patients (Figure 8B, p < 0.05). Subsequently, we analyzed SQLE expression and its association with biomarkers of MHC (B2M, HLA-B, HLA-C, TAP1, and TAP2), dendritic cells (BATF3), macrophages (CD68 and IL1A), type-I anti-tumor responses (CD8A and GZMB), and cell proliferation (MKI67) in 178 PAAD tissues (Figure 8C). SQLE was positively associated with MHC molecules (B2M, r = 0.221; HLA-B, r = 0.146; HLA-C, r = 0.143; TAP1, r = 0.187; and TAP2, r = 0.240), macrophages (CD68, r = 0.196 and IL1A, r = 0.275), and cell proliferation (MKI67, r = 0.380) but negatively associated with dendritic cells (BATF3, r = -0.181) and type-I anti-tumor responses (CD8A, r = -0.173 and GZMB, r = -0.171). Furthermore, our results showed that high SQLE expression indicated low CTL infiltration and poor OS in PAAD patients (Figure 8D, p < 0.05). These results suggest that high SQLE expression predicted depletion of cytotoxic lymphocytes and loss of anti-tumor ability, leading to unfavorable responses to immunotherapy.