Introduction

Pancreatic cancer is associated with an extremely poor prognosis and is the seventh leading cause of cancer-related deaths worldwide; by 2025, it is expected to surpass breast cancer as the leading major cause of cancer death1. The dismal prognosis is primarily due to late diagnosis and poor response to cytotoxic therapy, with a 5-year survival rate of only approximately 9%2. Immunotherapy is a promising option for many types of solid tumors3; however, it only benefits just a few percent of pancreatic cancer patients4,5. Therefore, biomarkers for selecting patients for immune checkpoint inhibitors and therapeutic targets are urgently needed.

OVO-like protein 1 (OVOL1) is a member of the OVO-like proteins (OVOLs) family, a C2H2 zinc-finger transcription factor family that includes three members: OVOL1, OVOL2, and OVOL36. OVOL1 was the first OVOL protein to be identified6, and functional investigations of OVOL1 have revealed its important role in epithelial tissue development7,8, preventing epithelial-mesenchymal transition (EMT) during epithelial differentiation9. OVOL1 is aberrantly expressed in multiple cancer types10,11,12,13. OVOL1 expression is downregulated in clear cell renal cell carcinoma10 and nasopharyngeal carcinoma11, and its levels decrease as cutaneous melanoma progresses12. However, bioinformatics investigations have revealed that OVOL1 expression is elevated in some cancer types13. Furthermore, OVOL1 has been demonstrated to be associated with patient prognosis in breast cancer14, cutaneous melanoma12, and clear cell renal cell carcinoma10. Functionally, OVOL1 inhibits the malignant progression of breast cancer by enhancing TGF-β type I receptor degradation14 and reducing ZEB1 expression15. OVOL1 also suppresses ZEB1 expression in prostate cancer15 and oral squamous cell carcinoma16, impeding their progression. Additionally, OVOL1 has been implicated in the development of atopic dermatitis17. A recent study found that Ovol1 and its family member Ovol2 directly regulate inflammatory genes, and that their loss in mouse skin is sufficient to activate skin immunity18. However, neither the prognostic and immunological significance nor the function of OVOL1 in pancreatic cancer has been investigated.

This study investigated the expression of OVOL1 and its clinical implications in pancreatic cancer. Moreover, the associations between OVOL1 expression, immune cell infiltration, and the tumor mutation burden (TMB) were analyzed. Enrichment analysis was conducted to investigate the biological terms and signaling pathways associated with OVOL1. Furthermore, in vitro experiments were conducted to investigate the function and mechanism of OVOL1 in pancreatic cancer cells. Finally, the relationship between OVOL1 methylation and expression was analyzed, and the prognostic value of OVOL1 methylation was evaluated.

Results

OVOL1 expression is upregulated in pancreatic cancer and has diagnostic potential

To investigate the potential role of OVOL1 in the development and progression of pancreatic cancer, we first examined the expression of OVOL1 in The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) databases. OVOL1 expression was considerably higher in pancreatic cancer tissues than in normal tissues (Fig. 1A, p < 0.001). To further assess the upregulation of OVOL1 expression, we examined another independent pancreatic cancer cohort from the Gene Expression Omnibus (GEO) database. Consistent significant upregulation of OVOL1 expression was observed in pancreatic cancer tissues compared with that in normal tissues (Fig. 1B, p < 0.01). Furthermore, the immunohistochemistry (IHC) assay showed that pancreatic cancer tissues had higher levels of OVOL1 protein expression than normal tissues (Fig. 1C, p < 0.01). Receiver operating characteristic (ROC) curve analysis was further performed to assess the discriminative ability of OVOL1 expression levels using TCGA and GTEx data. The results showed that OVOL1 expression had an area under the curve (AUC) of 0.851 for distinguishing pancreatic cancer tissues from normal tissues (positive predictive value (PPV) = 0.879, negative predictive value (NPV) = 0.788, sensitivity = 77.10%, and specificity = 88.89%) (Fig. 1D and E). These results indicate that patients with pancreatic cancer have increased expression of OVOL1 and that OVOL1 has the potential to be a diagnostic marker.

Fig. 1
figure 1

OVOL1 expression is elevated in pancreatic cancer and has diagnostic potential. (A, B) The expression of OVOL1 in pancreatic cancer and normal tissues was investigated in the TCGA cohort (A) and GSE71729 cohort (B). (C) Immunohistochemistry analysis of OVOL1 in pancreatic cancer and adjacent normal tissues. (D, E) The diagnostic value of OVOL1 expression in the TCGA pancreatic cancer cohort was determined by ROC curve analysis (D), and the PPV, NPV, SEN, and SPE were calculated (E). PPV, positive predictive value; NPV, negative predictive value; SEN, sensitivity; and SPE, specificity. **p < 0.01 and ***p < 0.001.

Full size image

Upregulation of OVOL1 expression is correlated with clinicopathological parameters in pancreatic cancer

To assess the clinical significance of OVOL1 expression in pancreatic cancer, we analyzed the associations between OVOL1 expression and clinicopathological characteristics in the TCGA pancreatic cancer cohort. OVOL1 expression was significantly associated with T stage (Fig. 2A, p < 0.01), histologic grade (Fig. 2B, p < 0.05), history of chronic pancreatitis (Fig. 2C, p < 0.01), residual tumor status (Fig. 2D, p < 0.05), and primary therapy outcome (Fig. 2E, p < 0.01). Other clinical variables were not significantly associated with OVOL1 expression (Fig. 2F and Fig. S1A-F). These findings suggest that increased OVOL1 expression is related to the malignant progression of pancreatic cancer.

Fig. 2
figure 2

Correlation of OVOL1 expression with clinical features in patients with pancreatic cancer. (A–F) Correlation analysis of OVOL1 expression with T stage (A), histologic grade (B), history of chronic pancreatitis (C), residual tumor (D), primary therapy outcome (E), and N stage (F). ns, not significant; *p < 0.05 and **p < 0.01.

Full size image

Upregulation of OVOL1 expression predicts poor prognosis in pancreatic cancer patients

To further evaluate the clinical importance of OVOL1 expression in pancreatic cancer, we investigated the TCGA cohort to determine whether OVOL1 expression is correlated with prognosis. Kaplan–Meier analysis revealed that pancreatic cancer patients with high OVOL1 expression had shorter OS durations than those with low OVOL1 expression (Fig. 3A p = 0.004), which was confirmed in two other independent cohorts (Fig. S2A and B, p < 0.05 for both). Further analysis revealed that higher OVOL1 expression was associated with shorter DSS (Fig. 3B, p = 0.002) and PFI (Fig. 3C, p = 0.023).

Next, we conducted univariate and multivariate analyses to investigate whether OVOL1 expression could be used as an independent prognostic factor for pancreatic cancer patients. According to the univariate analysis, T stage (hazard ratio (HR) = 2.053, 95% confidence interval (CI): 1.089–3.871, p = 0.026), N stage (HR = 2.114, 95% CI: 1.259–3.549, p = 0.005), and OVOL1 expression (HR = 1.829, 95% CI: 1.206–2.774, p = 0.005) were associated with an increased risk of poor OS outcomes in pancreatic cancer patients (Table S1). These three parameters were included in the multivariate analysis, which revealed that the N stage (HR = 1.779, 95% CI: 1.029–3.075, p = 0.039) remained an independent prognostic factor for OS in pancreatic cancer patients (Table S1). Among the variables examined, OVOL1 expression demonstrated marginal significance as an independent prognostic predictor for OS outcomes in pancreatic cancer patients (HR = 1.472, 95% CI: 0.961–2.254, p = 0.075) (Table S1). Similarly, as indicated in Table S2, N stage (HR = 1.985, 95% CI: 1.038–3.797, p = 0.038) and OVOL1 expression (HR = 1.649, 95% CI: 1.017–2.674, p = 0.043) were found to be independent predictive factors for DSS outcomes in pancreatic cancer patients. None of the variables investigated were found to be independent predictive indicators for the PFI in pancreatic cancer patients (Table S3). These findings suggest that OVOL1 expression is a potential independent prognostic factor in pancreatic cancer patients.

We further created two prognostic nomograms using OVOL1 expression and N stage to estimate the probabilities of 1-, 2-, and 3-year OS and DSS in pancreatic cancer patients (Fig. 3D and E). Next, we generated calibration curves to determine the concordance between the nomogram-predicted and actual survival probability. The calibration curves showed a satisfactory match between the nomogram-predicted OS and DSS probabilities and the actual survival probability (Fig. 3F and G). Furthermore, the ROC analysis showed that combining OVOL1 and N stage (AUC = 0.641) improved predictive accuracy for OS in pancreatic cancer patients compared to N stage (AUC = 0.590) alone (Fig. S3A, p = 0.0286). Similarly, as shown in Fig. S3B, combining OVOL1 and N stage resulted in a higher AUC for predicting DSS of pancreatic cancer patients than N stage alone, though the difference was not statistically significant (p = 0.2855). Together, these results show that the constructed nomogram survival models perform well.

Fig. 3
figure 3

Prognostic value of OVOL1 expression in pancreatic cancer. (A–C) A Kaplan–Meier survival analysis was conducted to analyze the association of OVOL1 expression with OS (A), DSS (B), and PFI (C) outcomes in TCGA pancreatic cancer patients. (D, E) Prognostic nomograms were constructed to predict the probabilities of 1-, 2-, and 3-year OS (D) and DSS (E) outcomes in TCGA pancreatic cancer patients. (F, G) Calibration plots were constructed to assess the agreement between the predicted and actual probabilities of 1-, 2-, and 3-year OS (F) and DSS (G) outcomes of TCGA pancreatic cancer patients. OS, overall survival; DSS, disease-specific survival; and PFI, progression-free interval.

Full size image

OVOL1 is correlated with immune cell infiltration in pancreatic cancer

Immune cell infiltration is a crucial factor in pancreatic cancer progression19and OVOL1 is involved in the regulation of immunity17,18. We next sought to determine whether OVOL1 expression is associated with immune cell infiltration. Single-sample gene set enrichment analysis (ssGSEA) revealed that OVOL1 expression was significantly positively correlated with the infiltration levels of Th2 cells and NK CD56bright cells (Fig. 4A, R = 0.292 and R = 0.251, respectively; p < 0.001 for both) and negatively correlated with the infiltration levels of pDCs, TFH cells, Tgd cells, mast cells, cytotoxic cells, Th17 cells, NK cells, iDCs, B cells, neutrophils, CD8 T cells, and T cells (Fig. 4A, p < 0.05 for all). Further analysis revealed that the infiltration levels of Th2 cells and NK CD56bright cells were significantly greater in the group with high OVOL1 expression than in the group with low OVOL1 expression (Fig. 4B, p < 0.001 and p < 0.01, respectively). However, the infiltration levels of pDCs, TFH cells, Tgd cells, Mast cells, cytotoxic cells, Th17 cells, NK cells, B cells, neutrophils, CD8 T cells, T cells, and Tregs were significantly lower in the group with high OVOL1 expression than in the group with low OVOL1 expression (Fig. 4B, p < 0.05 for all). Notably, OVOL1 expression was found to be negatively associated with immune score (Fig. S4A and B) and ESTIMATE score in pancreatic cancer (Fig. S4C and D). Furthermore, gene set enrichment analysis (GSEA) revealed that OVOL1 expression was negatively correlated with immune response signatures, including “GOBP_LYMPHOCYTE_MEDIATED_IMMUNITY” (Fig. 4C, normalized enrichment score (NES) = − 1.962, p = 0.001), “GOBP_ACTIVATION_OF_IMMUNE_RESPONSE” (Fig. 4D, NES = − 1.990, p = 0.001), and “GOBP_IMMUNE_RESPONSE REGULATING_SIGNALING_PATHWAY” (Fig. 4E, NES = − 1.771, p = 0.001). In support of these findings, patients with high OVOL1 levels had reduced lymphocyte infiltration signature score (Fig. S5A, p < 0.001) and TCR richness (Fig. S5B, p < 0.01). It is interesting to observe that OVOL1 was mainly expressed in pancreatic cancer cells but not immune cells (Fig. S6). These results suggest that the upregulation of OVOL1 is associated with an immunosuppressive microenvironment in pancreatic cancer.

Fig. 4
figure 4

OVOL1 expression is related to immune cell infiltration in pancreatic cancer. (A) Correlation analysis of OVOL1 expression and the infiltration levels of 24 different types of immune cells in TCGA pancreatic cancer cohort. (B) The infiltration levels of 24 different types of immune cells in OVOL1High and OVOL1Low tumors in the TCGA pancreatic cancer cohort. (C–E) GSEA plots of the enrichment of the “GOBP_LYMPHOCYTE_MEDIATED_IMMUNITY” (C), “GOBP_ACTIVATION_OF_IMMUNE_RESPONSE” (D), and “GOBP_IMMUNE_RESPONSE REGULATING_SIGNALING_PATHWAY” signatures (E) in OVOL1High versus OVOL1Low in TCGA pancreatic cancer cohort. ns, not significant; *p < 0.05, **p < 0.01, and ***p < 0.001.

Full size image

OVOL1 is related to the tumor mutation burden in pancreatic cancer

To investigate the differences in somatic mutations between the high- and low-OVOL1 groups, we compared the first 30 mutated genes in TCGA pancreatic cancer patients according to their OVOL1 expression level. Mutations in KRAS, TP53, SMAD4, CDKN2A, and RNF43 were more frequent in pancreatic tumors with high OVOL1 expression than in those with low OVOL1 expression (Fig. 5A, p < 0.05 for all). Remarkably, the high-OVOL1 pancreatic tumors had a significantly higher TMB than the low-OVOL1 tumors (Fig. 5B, p < 0.0001). Further analysis revealed that high-OVOL1 pancreatic tumors had significantly higher neoantigen loads than low-OVOL1 pancreatic tumors (Fig. 5C, p < 0.0001). Together, our findings suggest a positive relationship between OVOL1 expression and the incidence of somatic mutations.

Fig. 5
figure 5

OVOL1 is associated with mutation profile, TMB, and neoantigen loads in pancreatic cancer. (A) The frequency of the top 30 mutated genes in high- and low-OVOL1 TCGA pancreatic tumors. (B) The relationship between OVOL1 expression and TMB in TCGA pancreatic tumors. (C) The relationship between OVOL1 expression and neoantigen loads in TCGA pancreatic tumors. TMB, tumor mutation burden. ****p < 0.0001.

Full size image

OVOL1 is correlated with various biological terms and pathways in pancreatic cancer

To investigate the potential roles of OVOL1 in pancreatic cancer patients, DEGs (log2FC > 1, p adj < 0.05) in TCGA pancreatic cancer patients with OVOL1High versus OVOL1Low were obtained (Fig. S7 and Table S4) and subjected to gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses. The GO biological process annotations revealed that the DEGs were involved in biological processes including “Epithelial cell differentiation,” “Positive regulation of gene expression,” and “Regulation of ERK1 and ERK2 cascade” (Fig. 6A). The molecular function annotations revealed that these DEGs were associated with molecular functions including “Sequence-specific double-stranded DNA binding,” “Structural constituent of epidermis,” and “Sequence-specific DNA binding” (Fig. 6A). The cellular component annotations demonstrated that these genes were enriched in cellular components such as “Extracellular region,” “Chromatin,” and “Keratin filament” (Fig. 6A). KEGG analysis revealed that these genes were enriched in pathways including “Retinol metabolism,” “Metabolic pathways,” and “MAPK signaling pathway” (Fig. 6B).

Fig. 6
figure 6

OVOL1 is associated with multiple biological terms and signaling pathways in pancreatic cancer. (A) GO analysis of the DEGs (log2FC > 1, p adj < 0.05) in TCGA pancreatic cancer patients with OVOL1High versus OVOL1Low. (B) KEGG enrichment analysis of the DEGs (log2FC > 1, p adj < 0.05) in TCGA pancreatic cancer patients with OVOL1High versus OVOL1Low.

Full size image

Upregulation of OVOL1 is associated with proliferation and the ERK/JNK/p38 MAPK signaling pathway in pancreatic cancer

To investigate the putative biological activities and underlying mechanisms of OVOL1 in pancreatic cancer, we used GSEA to identify the signaling pathways associated with OVOL1 in the TCGA pancreatic cancer cohort. The GSEA results showed that OVOL1 expression was positively associated with the “BENPORATH_PROLIFERATION” (Fig. 7A, normalized enrichment score (NES) = 3.083, p = 0.025) and “REACTOME_RAF_INDEPENDENT_MAPK1_3_ACTIVATION” signaling pathways (Fig. 7B, NES = 1.669, p = 0.004), which was consistent with the findings revealed by the GO and KEGG analyses that OVOL1 expression was positively associated with the “Regulation of ERK1 and ERK2 cascade” (Fig. 6A) and “MAPK signaling pathway” (Fig. 6B). Inspired by these findings, we employed a CCK-8 assay to investigate the effects of OVOL1 on pancreatic cancer cells. The results showed that knocking down OVOL1 significantly decreased the proliferative activity of MIAPaCa-2 cells (Fig. 7C, p < 0.01 for siOVOL1 #1 and p < 0.05 for siOVOL1 #2; Fig. S8) and AsPC-1 cells (Fig. 7D, p < 0.001 for both siOVOL1 #1 and siOVOL1 #2; Fig. S9). Notably, OVOL1 knockdown had a minor effect on the motility of MIAPaCa-2 cells (Fig. 7E), despite reaching statistical significance. Consistent with the GSEA results, the results of the western blot analysis revealed that OVOL1 knockdown substantially inhibited mitogen-activated protein kinases (MAPKs), including p-ERK, p-JNK, JNK, and p-p38 in MIAPaCa-2 cells (Fig. 7F). Further analysis revealed a positive correlation between OVOL1 expression and Ki67, a proliferative marker20in TCGA pancreatic cancer (Fig. 7G, R = 0.545, p < 0.01). Tumors with high OVOL1 expressionhad higher Ki67 expression than those with low OVOL1 expression (Fig. 7H, p < 0.001). Thus, these data suggest that OVOL1 may influence cell proliferation by activating the ERK/JNK/p38 MAPK signaling pathway.

Fig. 7
figure 7

Knockdown of OVOL1 suppresses proliferation and impairs the ERK/JNK/p38 signaling pathway in pancreatic cancer cells. (A, B) GSEA plots of enrichment of “BENPORATH_PROLIFERATION” (A) and “REACTOME_RAF_INDEPENDENT_MAPK1_3_ACTIVATION” (B) signaling pathways in OVOL1High versus OVOL1Low in TCGA pancreatic cancer cohort. (C, D) CCK-8 assay of MIAPaCa-2 (C) cells and AsPC-1 (D) transfected with siOVOL1#1, siOVOL1#2, or siCtrl. (E) Wound healing assay of MIAPaCa-2 cells transfected with siOVOL1#1, siOVOL1#2, or siCtrl. (F) Western blot analysis of the indicated proteins in MIAPaCa-2 cells transfected with siOVOL1 or siCtrl (Left panel), and the band intensities were quantified using ImageJ software (Right panel). (G) Correlation analysis of OVOL1 expression and Ki67 expression in TCGA pancreatic cancer cohort. (H) Ki67 expression in low- and high-OVOL1 pancreatic tumors in TCGA cohort. *p < 0.05, **p < 0.01, and ***p < 0.001.

Full size image

Hypomethylation of OVOL1 is associated with its expression and prognosis in pancreatic cancer

Methylation is an important process that controls OVOL1 expression12,14. Therefore, we investigated whether methylation alterations caused the increase in OVOL1 expression in pancreatic cancer. The results showed that pancreatic cancer tissues had considerably lower OVOL1 methylation levels than normal tissues in the TCGA cohort (Fig. 8A, p < 0.0001). Correlation analysis revealed a negative correlation between OVOL1 methylation and its expression (Fig. 8B, R = − 0.670, p < 0.001). Compared to patients with low OVOL1 methylation, those with high OVOL1 methylation had lower OVOL1 expression (Fig. 8C, p < 0.001). Notably, OVOL1 expression can be increased via a strategy known as “methyl-CpG targeted transcriptional activation” (MeTA)21 (Fig. 8D). Intriguingly, pancreatic cancer patients with low OVOL1 methylation had shorter OS than those with high OVOL1 methylation (Fig. 8E, p = 0.026) and tended to have poorer DSS (Fig. 8F, p = 0.155) and PFI (Fig. 8G, p = 0.073). These findings suggest that epigenetic alterations are involved in the control of OVOL1 expression and that the methylation status of OVOL1 predicts the prognosis of pancreatic cancer patients.

Fig. 8
figure 8

OVOL1 is frequently hypomethylated in pancreatic cancer and is associated with poor prognosis. (A) Methylation levels of OVOL1 in TCGA pancreatic cancer cohort. (B) Correlation analysis of OVOL1 methylation and expression in TCGA pancreatic cancer. (C) The expression of OVOL1 in TCGA pancreatic cancer with high and low OVOL1 methylation. (D) The expression of OVOL1 in pancreatic cancer cells (AsPC-1, MIAPaCa-2, and PANC-1) and normal pancreatic ductal HPDE cells treated with MeTA, a method that can reactivate genes that are transcriptionally silenced by promoter hypermethylation, using a GEO dataset GSE26215. (E–G) Kaplan–Meier survival analysis was conducted to analyze the association of OVOL1 methylation with OS (E), DSS (F), and PFI (G) outcomes in TCGA pancreatic cancer patients. OS, overall survival; DSS, disease-specific survival; and PFI, progression-free interval. ***p < 0.001 and ****p ＜0.0001.

Full size image

Discussion

In this study, we found that OVOL1 expression and methylation were increased and decreased, respectively, in pancreatic cancer, and that these deregulations were associated with a poor patient prognosis. Moreover, OVOL1 expression was found to be related to immune cell infiltration and the TMB. In addition, OVOL1 expression was associated with various biological terms and signaling pathways. Furthermore, in vitro experiments indicated that OVOL1 is essential for pancreatic cell proliferation and is involved in activating the ERK/JNK/p38 MAPK signaling pathway.

The expression of OVOL1 has been reported to be deregulated in several types of cancers10,11,12,13. Similarly, we found that OVOL1 expression was elevated in two independent pancreatic cancer cohorts. Furthermore, we showed that the expression of OVOL1 could accurately distinguish pancreatic cancer tissues from normal tissues. In support of these findings, OVOL1 could be used as a biomarker to differentiate breast cancer tissues from nontumor tissues13. Further investigation is needed to investigate whether it may be detected in biofluids such as peripheral blood as a non-invasive diagnostic biomarker for pancreatic cancer. Notably, increased OVOL1 expression was associated with poor OS, DSS, and PFI outcomes. Several lines of evidence have suggested that OVOL1 is a prognostic marker for breast cancer13,14, cutaneous melanoma12, and clear cell renal cell carcinoma10. In addition, we developed two nomogram survival models for predicting OS and DSS, respectively, in patients with pancreatic cancer. These two models showed high accuracy.

OVOL1 has been implicated in the regulation of immunity17,18. Consistently, our findings showed that OVOL1 is related to the infiltration of various immune cells. The infiltration levels of multiple types of favorable immune cells, including CD8+ T cells, B cells, and NK cells, were reduced in tumors with high OVOL1 expression. Furthermore, OVOL1 was shown to be associated negatively with immune and ESTIMATE scores in pancreatic cancer. These findings indicate that tumors with high OVOL1 expression have an immunosuppressive immune microenvironment. An immunosuppressive tumor microenvironment has been found to strongly impact the outcome of therapy for pancreatic cancer patients19. Notably, OVOL1 was shown to be predominantly expressed in pancreatic cancer cells but not immune cells, suggesting that OVOL1 may indirectly influence immune cell infiltration. Further investigations into how OVOL1 modulates immune cell infiltration are needed. Immunotherapy has revolutionized the treatment of most solid cancers3; however, only a small proportion of pancreatic cancer patients benefit from immunotherapy4,5. Hence, it is crucial to discover specific markers to predict the response to immune checkpoint inhibitors in pancreatic cancer patients. A high TMB is the most important predictor of immunotherapy efficacy for pancreatic cancer, as it is associated with a high amount of cancer neoantigens22,23. We found that the TMB and neoantigen load were much higher in pancreatic tumors with high OVOL1 expression than in those with low OVOL1 expression. These findings suggest that OVOL1 may be a useful marker for identifying pancreatic cancer patients who would benefit from immune checkpoint inhibitors.

OVOL1 has been shown to inhibit cancer growth and metastasis by accelerating TGF-β type I receptor degradation and repressing ZEB114,15,24. In this study, we found that knocking down OVOL1 inhibited pancreatic cell proliferation while having a minor effect on cell motility. Further investigation revealed that knocking down OVOL1 suppressed the ERK/JNK/p38 MAPK signaling pathway, which controls cell proliferation25, suggesting that OVOL1 plays an oncogenic role in pancreatic cancer. These findings show that OVOL1 may function as a tumor suppressor or promoter in different types of cancer; however, this hypothesis requires further investigation. Notably, OVOL1 knockdown downregulated total JNK, and there was a positive association between OVOL1 and JNK mRNA expression in pancreatic cancer (Fig. S10). Because OVOLs directly repress the transcription of target genes by recruiting corepressor histone deacetylases26,27,28, it appears that OVOL1 does not diretly regulate JNK or ERK and p38. More research is needed to unravel the mechanisms that support OVOL1’s effect on the activation of the MAPK signaling pathway. Intriguingly, in addition to the MAPK signaling pathway, OVOL1 was found to be related to multiple pathways, such as “Retinol metabolism”, “Estrogen signaling pathway”, and “Staphylococcus aureus infection.” More research is needed to determine whether OVOL1 regulates these pathways and what biological implications they have.

In cancer, DNA methylation changes include a wave of global hypomethylation and locus-specific hypermethylation, which play important roles in regulating specific gene transcription29. In this study, we found that OVOL1 was hypomethylated in pancreatic cancer. Importantly, OVOL1 methylation was negatively correlated with its expression, suggesting that hypomethylation of OVOL1 could be one of the mechanisms driving its upregulation. In support of this hypothesis, MeTA, a method that can reactivate genes transcriptionally silenced by promoter hypermethylation21 could increase the expression of OVOL1 in pancreatic cancer cells, suggesting that methylation regulates OVOL1 expression. Indeed, the OVOL1 promoter is methylated in cutaneous melanoma12, and it may be methylated in breast cancer because its expression is elevated by 5-aza-2’-deoxycytidine14, a DNA demethylation agent30. Both OVOL1 and OVOL2 are downstream targets of Wnt signaling27,31; however, methylated OVOL2 does not respond to Wnt signaling27. Given that OVOL1 is methylated in normal tissue but hypomethylated in pancreatic cancer tissue, and that the Wnt pathway is persistently activated in pancreatic cancer32, it is reasonable to speculate that Wnt signaling may also contribute to the upregulation of OVOL1 in pancreatic cancer, which requires further investigation.

Nonetheless, the findings of this study are mainly based on public databases, and the clinical significance of OVOL1 needs to be validated in other pancreatic cancer patient cohorts. Furthermore, the effect of OVOL1 on pancreatic cancer cell behavior needs to be verified in a mouse model. In addition, the mechanism by which OVOL1 modulates the ERK/JNK/p38 MAPK signaling pathway needs to be further defined.

Conclusion

In summary, OVOL1 expression and methylation were found to be deregulated in pancreatic cancer, which is clinically significant. Upregulation of OVOL1 was associated with the immunosuppressive microenvironment in pancreatic cancer. OVOL1 promotes pancreatic cell proliferation, which may be attributed to its stimulation of the ERK/JNK/p38 MAPK signaling pathway. Therefore, our findings support OVOL1 as a prognostic and immunological marker, as well as a therapeutic target for pancreatic cancer.

Materials and methods

Tissue microarray and immunohistochemistry analysis

A tissue microarray (TMA) consisting of 29 pancreatic cancer tissues and paired adjacent normal tissues was obtained from Shanghai Outdo Biotech Co., Ltd. (Shanghai, China). The TMA was processed and incubated with an anti-OVOL1 antibody (Abcam, # ab235949), followed by a secondary antibody, and then stained with 3,3’-diaminobenzidine (DAB). The staining immunohistochemical staining was evaluated as previously described33. The study was reviewed and approved by The Ethics Committee of Shanghai Outdo Biotech Co., Ltd.

Cell culture

The human pancreatic cancer cell lines MIAPaCa-2 and AsPC-1 were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured in DMEM and RPMI-1640, respectively, each supplemented with 10% fetal bovine serum (FBS; Gibco, NY, USA) and antibiotics (100 units/ml penicillin and 100 mg/ml streptomycin), at 37 °C in an incubator with 5% CO2.

Small interfering RNA (siRNA) transfection

Cells were seeded into 6-well plates. After reaching 70% confluence, the cells were transfected with siRNAs targeting OVOL1 (siOVOL1) or a scrambled negative control siRNA (siCtrl) using Lipofectamine 3000 reagent (Invitrogen, USA). After 72 h, additional experiments were performed on the transfected cells. siOVOL1 and siCtrl were purchased from GenePharma Co., Ltd. (Shanghai, China). The siOVOL1 sequences used were as follows: siOVOL1#1 sense: 5’-GGAACCCUCUGUGGCCGAATT-3’; siOVOL1 #2 sense: 5’-GCAUGCUGAACCGCCACAUTT-3’; and siOVOL1 #3 sense: 5’-GCCACGUCCUGCACCUGAATT-3’. The siCtrl sequence was as follows: sense: 5’-UUCUCCGAACGUGUCACGUTT-3’.

Cell counting Kit-8 assay

Cells transfected with siOVOL1 or siCtrl were seeded into a 96-well plate, and their proliferative capacity was assessed at the indicated time points by Cell Counting Kit-8 (CCK-8) assay (APExBIO, Houston, USA, #K1018). The absorbance at 450 nm was measured using a multipurpose microplate reader (Tecan Group Ltd., Switzerland).

Wound healing assay

Cells were seeded in a 6-well plate and then transfected with siOVOL1 or siCtrl. When the cells reached 90% confluence, they were scraped with a pipette tip and cultured in serum-free media. After 24 h, photographs of the wound healing were taken, and the wound area was quantified with ImageJ software. The wound healing rate = (wound area at 0 h - wound area at 24 h)/wound area at 0 h × 100%.

Western blot analysis

Total proteins were extracted from the cells transfected with siOVOL1 or siCtrl using RIPA buffer (Beyotime Biotechnology, #P0013E), and the protein concentration was determined with a BCA kit (Beyotime Biotechnology, #P0012S). The proteins were separated by 10% SDS‒PAGE and transferred to PVDF membranes. The membranes were blocked with 5% silk milk powder and then cut following the molecular weight of the specific proteins. Subsequently, they were incubated with primary antibodies at 4 °C overnight, followed by incubation with secondary antibodies at room temperature for 1 h. Signals were detected using an enhanced chemiluminescence (ECL) system and ImageLab software (Bio-Rad Laboratories, Hercules, CA, USA). The primary antibodies used were as follows: anti-OVOL1 (Proteintech, #14082-1-AP), anti-p-ERK (Beyotime Biotechnology, #AF5818), anti-ERK (Beyotime Biotechnology, #AF1051), anti-p-JNK (Beyotime Biotechnology, #AF1762), anti-JNK (Beyotime Biotechnology, #AF1048), anti-p-p38 (Beyotime Biotechnology, #AM063), anti-p38 (Beyotime Biotechnology, #AF7668), and anti-GAPDH (ZS-GB Biotech, #TA-08).

Data acquisition

The mRNA-seq expression profiles of 179 pancreatic cancer samples, four paracancerous tissues, and the corresponding clinical data were obtained from The Cancer Genome Atlas (TCGA) via the National Cancer Institute (NCI) Genomic Data Commons (https://gdc.cancer.gov). 178 of 179 patients had completed overall survival (OS) and progression-free interval (PFI) follow-up data, while 172 had completed disease-specific survival (DSS). Except for age and gender, the patient’s clinical characteristics were incomplete. Moreover, OVOL1 mRNA expression data in 167 normal pancreatic tissue samples were collected from the Genotype-Tissue Expression (GTEx) database (https://gtexportal.org/)34. The methylation data of OVOL1 in TCGA pancreatic cancer patients (n = 185) were obtained from UCSC Xena (https://xenabrowser.net/). The immune score and ESTIMATE score for each TCGA pancreatic cancer sample were obtained from the ESTIMATE database (https://bioinformatics.mdanderson.org/estimate/). The expression of OVOL1 in pancreatic cancer was investigated using the Gene Expression Omnibus (GEO; www.ncbi.nlm.nih.gov/geo) dataset GSE7172935. The effect of methylation on OVOL1 expression was investigated using the GSE26215 dataset21. The distribution of the OVOL1 in different single-cell clusters in pancreatic cancer was investigated in the PAAD_CRA001160 dataset using the Tumor Immune Single-cell Hub (TISCH) database (http://tisch.comp-genomics.org/). The study did not require local ethics committee approval because the data was obtained from public databases.

Survival analysis

TCGA pancreatic cancer patients were separated into two groups according to the median OVOL1 expression or methylation value. Kaplan–Meier curves were used to examine the effect of OVOL1 expression or methylation on the OS, DSS, and PFI of pancreatic cancer patients. The correlation between OVOL1 expression and the OS in pancreatic cancer patients from the EMTAB6134 and ICGC_array cohorts was investigated using the LOGpc database (http://bioinfo.henu.edu.cn/PAAD/PAADList.jsp.)36.

Construction and assessment of the nomogram survival models

A nomogram survival model was built using OVOL1 expression and N stage data with the R package “rms” to predict the probabilities of 1-, 2-, and 3-year OS and DSS. The R package “rms” was used to create calibration curves to compare the nomogram survival model-predicted survival probability and the actual survival probability.

Immune cell infiltration analysis

The single-sample gene set enrichment analysis (ssGSEA) method in the R package “GSVA” was used to calculate the infiltration levels of 24 different types of immune cells in the TCGA pancreatic cancer cohort. The correlation between OVOL1 expression and the infiltration levels of various immune cells was investigated using the R package “ggplot2.” Furthermore, the difference in immune cell infiltration between the high- and low-OVOL1 expression groups was investigated.

CAMOIP analysis

The relationships between OVOL1 expression and the mutational landscape, TMB, neoantigen load, lymphocyte infiltration signature score and TCR richness in pancreatic cancer were investigated via the Comprehensive Analysis on Multi-Omics of Immunotherapy in Pan-cancer (CAMOIP) database (https://www.camoip.net/)37.

GO and KEGG pathway enrichment analyses

The differentially expressed genes (DEGs) between the OVOL1High and OVOL1Low TCGA pancreatic cancer tissues were identified using the R package “DESeq2.” Genes with Log2 FC > 1 and adjusted p-value < 0.05 were subjected to the Database for Annotation, Visualization, and Integrated Discovery (DAVID) database (https://david.ncifcrf.gov/) for gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses. The enriched KEGG pathways were visualized using the R package “ggplot2.”

Gene set enrichment analysis

The mRNA profiles of TCGA pancreatic cancer tissues were divided into OVOL1High and OVOL1Low groups according to the median expression value of OVOL1 and subjected to gene set enrichment analysis (GSEA) as previously described38.

Statistical analyses

The statistical analyses were performed using SPSS 17.0 software (SPSS, Inc., USA). Statistical significance was determined using the Student’s t-test or the ANOVA test, as appropriate. The diagnostic value of various variables was determined by receiver operating characteristic (ROC) curve analysis. Univariate and multivariate Cox regression analyses were used to analyze the associations between variables and survival. Pearson’s correlation coefficients were used to assess the relationships between OVOL1 expression and other variables. A p-value ＜0.05 was considered to indicate significance.