Pan-cancer analysis of the PDE4DIP gene with potential prognostic and immunotherapeutic values in multiple cancers including acute myeloid leukemia

1 Introduction

Phosphodiesterase 4D (PDE4D) is an important member of the nucleotide phosphodiesterase superfamily, which can specifically hydrolyze the intracellular second messenger cyclic adenosine monophosphate (cAMP), and plays a vital regulatory role in various activities of cells and the development of malignant tumors [1]. Many studies have pointed out that PDE4D is a tumor-promoting molecule in some cancers, including hematological malignancies, lung cancer, prostate cancer, colorectal cancer, gastric cancer, melanoma, head and neck cancer, breast cancer, ovarian cancer, endometrioma, pancreatic ductal adenocarcinoma (PDAC), etc. [1–5]. For example, Rahrmann et al. revealed that PDE4D is a common insertion site in prostate cancer, and PDE4D knockout can reduce the growth and migration rate of prostate cancer cells in vitro [3]. Liu et al. discovered that the up-regulation of PDE4D expression in patients with PDAC is closely related to poor prognosis and multiple clinicopathological characteristics, indicating that the PDE4D gene may be a potential target for the prognosis and treatment of PDAC [5]. In addition, Dong and Zhang et al. indicated that there are differences in the expression of PDE4D in hematological malignancies, i.e., compared to peripheral blood mononuclear cells from healthy adults, the expression of PDE4D is significantly reduced in cells extracted from patients with chronic lymphocytic leukemia, while is more abundant in CEM and Jurkat T leukemia cell lines [6,7]. All in all, the above studies demonstrated that PDE4D may be a potential target for multiple cancer treatments.

Related studies further found that phosphodiesterase 4D interacting protein (PDE4DIP) is a protein that can interact with PDE4D and anchor PDE4D in the centrosome/Golgi region of cells, and is mainly expressed in the brain and heart [8]. Some studies have shown that there is a close association between PDE4DIP gene mutations and atrial fibrillation, stroke, and heart failure [9]. However, there are relatively few reports on the PDE4DIP gene and tumor development, only including leukemia, breast cancer, and pineoblastoma. Yao et al. found for the first time in their research on gene mutation profiles in Chinese leukemia patients that high mutations in the PDE4DIP gene are significantly associated with the occurrence of leukemia [10]. Onyeisi et al. pointed out that in breast cancer, the most common malignancy in women, the abnormal expression of syndecan-4 is affected by transcriptional and post-transcriptional mechanisms, including PDE4DIP gene mutations [11]. Furthermore, Snuderl et al. found that there is a small duplication of the PDE4DIP gene in patients with pineal cell tumors and this gene mutation may lead to the overexpression of PDE4DIP, which is closely related to the development of pineal cell tumors [12]. However, the full picture of PDE4DIP in pan-cancer has not been reported.

In order to fully elucidate the important role of PDE4DIP in multiple cancer types, we conducted a systematic pan-cancer analysis of the PDE4DIP gene based on The Cancer Genome Atlas (TCGA) [13] and the Genotype-Tissue Expression (GTEx) [14] public databases in this study. We evaluated the differential expression of PDE4DIP in various cancers and its relationship with patients’ prognosis, and further analyzed the relevance between PDE4DIP expression and genetic variation, DNA methylation, and immune characteristics, as well as the potential biological functions of PDE4DIP in pan-cancer. In addition, because previous research has found that PDE4DIP is associated with hematological malignancies, we further explored the specific role of PDE4DIP in acute myeloid leukemia (LAML) in depth, providing a new target for the future diagnosis and treatment of various cancers, including LAML.

2 Methods and materials

3 Results

3.1 Aberrant expression of PDE4DIP in pan-cancer

To determine the basic landscape of PDE4DIP expression, multi-omics data on PDE4DIP levels in various cancers were analyzed. Differential analysis of PDE4DIP mRNA expression from the TCGA and GTEx databases revealed that PDE4DIP expression was significantly up-regulated in 14 cancers compared to that in healthy tissues, including adrenocortical carcinoma (ACC), bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), brain lower grade glioma (LGG), liver hepatocellular carcinoma (LIHC), ovarian serous cystadenocarcinoma (OV), pancreatic adenocarcinoma (PAAD), pheochromocytoma and paraganglioma (PCPG), prostate adenocarcinoma (PRAD), stomach adenocarcinoma (STAD), testicular germ cell tumors (TGCT), thyroid carcinoma (THCA), uterine corpus endometrial carcinoma (UCEC), and uterine carcinosarcoma (UCS) (all p < 0.05). In contrast, it was significantly down-regulated in eight tumors: kidney chromophobe (KICH), kidney renal clear cell carcinoma (KIRC), kidney renal papillary cell carcinoma (KIRP), LAML, lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), rectum adenocarcinoma (READ), and skin cutaneous melanoma (SKCM) (all p < 0.05) (Figure 1a). The average expression levels of PDE4DIP in 33 tumors and healthy tissues are shown in Figure 1b and c. Furthermore, we analyzed the expression of PDE4DIP at the protein level in different cancers based on the CPTAC data set. The results showed that the expression of the PDE4DIP protein in PAAD, KIRC, and colon adenocarcinoma (COAD) (all p < 0.05) was significantly down-regulated compared with that in normal tissues, while it was significantly up-regulated in BRCA, glioblastoma multiforme (GBM), head and neck squamous cell carcinoma (HNSC), LIHC, LUAD, and UCEC (all p < 0.05) (Figure 2). The detailed results of the differential expression analysis of PDE4DIP at mRNA and protein levels are shown in Tables S2 and S3.

Figure 1

PDE4DIP expression pattern at mRNA level: (a) differential expression of PDE4DIP in tumor tissues and normal tissues, (b) mean expression of PDE4DIP in 33 cancer tissues, and (c) mean expression of PDE4DIP in normal tissues. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 2

PDE4DIP expression pattern in tumor and normal tissues at protein level: (a)–(l) PAAD, KIRC, COAD, BRCA, GBM, HNSC, LIHC, LUAD, UCEC, STAD, PRAD, and OV. p < 0.05 indicates statistically significant.

3.2 Prognostic value of PDE4DIP in pan-cancer

Prognostic analysis aims to elucidate the association between the expression of a specific gene and the prognosis of cancer patients, thus determining the prognostic value of that gene [15]. Through Cox regression analysis, we studied the relevance between PDE4DIP expression and four survival data of cancer patients, including OS, DSS, DFI, and PFI, to further reveal the potential prognostic value of PDE4DIP in different cancers. The results of the Cox proportional hazards model showed that the expression of PDE4DIP was significantly correlated with lower OS in patients with LAML (p = 0.001), PCPG (p = 0.020), thymoma (THYM) (p = 0.002), and UVM (p = 0.003) (Figure 3a). Univariate Cox regression analysis showed that high PDE4DIP expression was associated with a significant decrease in DSS in PCPG (p = 0.028), THYM (p = 0.043), and UVM (p = 0.006) patients (Figure 3b), as well as poor DFI in HNSC (p = 0.023) patients, while it was significantly associated with better DFI in PAAD (p = 0.033) and PCPG (p = 0.034) patients (Figure 3c). For PFI, high PDE4DIP expression was associated with the significant reduction in PFI in KICH (p = 0.030), THYM (p = 0.006), and UVM (p = 0.016) patients, while it was significantly associated with better PFI in PAAD (p = 0.003) patients (Figure 3d). In addition, KM curves further showed that the high expression of PDE4DIP was significantly related to the shortened survival time of LAML (p = 0.003) and UVM (p = 0.002) (Figure 3e and f) patients, indicating that PDE4DIP had a poor prognosis in LAML and UVM patients.

Figure 3

Prognostic analysis of PDE4DIP in pan-cancer: (a)–(d) Forest plots of Cox regression analysis for OS, DSS, DFI, and PFI, (e) and (f) KM curves of OS in LAML and UVM. Only significant results are shown and p < 0.05 was considered statistically significant.

3.4 Association of PDE4DIP expression with DNA methylation

DNA methylation is an epigenetic modification, which is closely related to gene expression and regulation [20]. During cancer progression, abnormal DNA methylation patterns are often accompanied, which will result in abnormal expression of some genes and disorder of normal cell functions, thus promoting the development of tumors [21,22]. DNA methylation analysis showed that in GBM, COAD, LUSC, PAAD, HNSC, LGG, UCEC, and BLCA patients, PDE4DIP expression was significantly positively related to DNA methylation, while was significantly negatively related to DNA methylation in STAD, KIRC, SARC, MESO, THCA, THYM, PCPG, and UVM patients (Figure S2b). Furthermore, we evaluated the prognostic value of PDE4DIP methylation based on OS, DSS, DFI, and PFI. The results showed that PDE4DIP methylation was a prognostic factor for OS of KIRC, PAAD, LGG, SKCM, UCEC, and UVM patients. To be specific, the high level of PDE4DIP methylation was significantly related to the increase of OS in patients with PAAD, LGG, SKCM, UCEC, and UVM (Figure 4b–f), but only to the decrease of OS in patients with KIRC (Figure 4a). In addition, PDE4DIP methylation was a prognostic factor for DSS of DLBC, LGG, SKCM, UCEC, UCS, and UVM patients (Figure S3), and for PFI in patients with LGG, MESO, UVM, UCEC, UCS, and ACC (Figure S4). However, there was no significant correlation between PDE4DIP methylation and DFI in any cancers.

Figure 4

Association of PDE4DIP methylation with OS in DLBC, LGG, SKCM, UCEC, UCS, and UVM: (a)–(f) only significant results are shown.

3.5 Association of PDE4DIP expression with immune-related characteristics

Tumor-infiltrating immune cells are typically dysfunctional, fail to control tumor growth and may even promote tumor development and immune escape [23]. Moreover, abnormal expression of immune-related genes can lead to reduced antigen presentation, and increased production of immunosuppressive substances, and tumor immune escape, thereby impacting tumor immune therapy [24]. Based on the TIMER 2.0 online website, we studied the relevance between TAM infiltration and PDE4DIP expression in the tumor microenvironment (TME). The results showed that there was a significant positive correlation between TAM infiltration levels and PDE4DIP expression in most cancers, mainly including M2-like TAMs (Figure 5a). Furthermore, based on the TISIDB online website, we further evaluated the potential relevance between immune-related genes and PDE4DIP expression in the TME. Notably, gene co-expression analysis revealed that PDE4DIP expression was significantly positively related to the expression of most immunoinhibitors in cancers (Figure 5b; Figure S5a–d).

Figure 5

Correlation between PDE4DIP expression and immune cell infiltration: (a) correlation between PDE4DIP expression and macrophages in pan-cancer and (b) correlation between PDE4DIP expression and immunosuppressive factors in pan-cancer. *p < 0.05, **p < 0.01, and ***p < 0.001.

3.7 Role of PDE4DIP in LAML

Based on the TCGA-LAML data, we analyzed the correlation between PDE4DIP expression and immune cell infiltration in LAML. The results revealed that PDE4DIP expression was significantly positively related to the infiltration of B cell naive and plasma cells, but negatively related to the infiltration of dendritic cell resting, T cell follicular helper, and mast cell resting (Figure 6).

Figure 6

Correlation between PDE4DIP expression and immune cell infiltration in LAML.

In order to better clarify the role of PDE4DIP in LAML and its relationship with immune response, TME, and tumor purity, we conducted correlation analyses between PDE4DIP expression and estimate score, immune score, normal score, and TumorPurity score. The results revealed that there was a significant positive association between PDE4DIP expression and immune score, indicating that the PDE4DIP gene might have a certain impact on the immune response process of LAML, but there was no significant correlation between the PDE4DIP gene and estimate score, normal score, and TumorPurity score (Figure 7a–d).

Figure 7

Correlation matrixes between PDE4DIP gene expression and (a) estimate score, (b) immune score, (c) stromal score, and (d) TumorPurity score in LAML.

We further explored the potential function of PDE4DIP in LAML. GSEA showed that PDE4DIP was most markedly enriched in immune-related pathways in LAML, including primary immunodeficiency, intestinal immune network for IgA production, viral protein interaction with cytokine and cytokine receptor, cell adhesion molecules, chemokine signaling pathway, B cell receptor signaling pathway, Th17 cell differentiation, T cell receptor signaling pathway, and JAK-STAT signaling pathway (Figure 8a). Gene ontology (GO) functional enrichment analysis revealed that in LAML, PDE4DIP was mainly enriched in immunoglobulin complex, T cell receptor complex, chromosomal region, ATP-dependent activity acting on DNA, methyltransferase activity, double-strand break repair, immune response-regulating signaling pathway, etc. (Figure 8b–d).

Figure 8

Enrichment analysis of PDE4DIP in LAML: (a) GSEA, (b) GO enrichment analysis about cellular component, (c) biological process, and (d) molecular function.

4 Discussion

As we all know, PDE4DIP is a protein that can interact with cAMP-specific PDE4D and is widely expressed in various cell types, including neurons, cardiomyocytes, immune cells, and tumor cells, and participates in cell proliferation, migration, and cell cycle [8,25,26]. Previous studies have found that PDE4DIP exhibits abnormal expression in various diseases, including malignant tumors [9–12], but its effect on the diagnosis, prognosis, and TME of tumors remains to be further explored. Therefore, our study is the first to comprehensively and systematically analyze the expression, prognosis, genetic changes, DNA methylation, immune cell infiltration, immune-related genes, and biological functions of PDE4DIP in many cancers. Our research revealed that the PDE4DIP gene had potential prognostic and immunotherapeutic values in various cancers, including LAML, providing a certain theoretical basis for using PDE4DIP as a biomarker in future cancer treatment.

Previous studies have pointed out that different tumors exhibit tumor heterogeneity due to genetic and epigenetic changes in the evolution and development of tumors [27]. Consistent with previous reports, our study found that PDE4DIP expression was heterogeneous in different cancers, where it was up-regulated in some cancer types and down-regulated in others. Although the exact mechanisms remain unclear, multiple factors could result in the differential expression of PDE4DIP. On the one hand, differences in gene expression regulatory networks between different cancer types might result in different regulatory mechanisms of PDE4DIP, leading to different expression patterns in cancers [10–12]. On the other hand, PDE4DIP expression can be influenced by various factors, including transcription factor regulation, epigenetic modifications, and activation of signaling pathways. And relevant studies have revealed that the PDE4DIP expression may be related to some signaling pathways, such as PI3K/AKT, RAS/ERK, and NF1/RAS pathways [28,29]. However, the specific regulatory mechanism of PDE4DIP expression in different cancers needs further exploration in subsequent studies. It is worth noting that PDE4DIP was significantly up-regulated in many cancer types, including ACC, BLCA, BRCA, LGG, LIHC, OV, PAAD, PCPG, PRAD, STAD, TGCT, THCA, UCEC, and UCS. It is speculated that PDE4DIP may serve as an oncogene in these cancers, possibly promoting tumor development. In addition, we compared the expression patterns of PDE4DIP at the mRNA and protein levels (Table S4), and the results showed that in some cancer types, the expression patterns of PDE4DIP at the mRNA and protein levels were inconsistent, such as in PAAD and LUAD. We speculated that this may be related to the changes in post-transcriptional regulation [30], the differences in post-translational modification [31], and the differences in sample sources. In future research, we will collect more clinical data or design relevant experiments to further clarify the expression pattern of PDE4DIP in different cancers.

Prognostic analysis based on OS, DSS, DFI, and PFI indicated that high expression of PDE4DIP had a potential prognostic effect on some cancers, with a poor prognosis in LAML, PCPG, THYM, UVM, and HNSC patients, and a good prognosis in PAAD and PCPG patients. Previous studies have reported that the treatment of LAML remained unchanged for most of the twentieth and early twenty-first centuries, and the survival curve remained stagnant for decades [32]. UVM is one of the most invasive and deadly skin cancers, and its mortality increases with the increase of incidence rate [33]. Immunotherapy and targeted therapy can markedly improve the survival rate of patients, but due to the complexity of cancer treatment, seeking the best treatment still requires continuous efforts. In this study, we found that PDE4DIP had potential prognostic value in pan-cancer, including LAML and UVM, giving a new guidance for the diagnosis and treatment of cancers in the future.

PDE4DIP is an anchor protein, and its genetic mutations can result in impaired cell function, further leading to changes in intracellular targeting and cell survival, and the occurrence of diseases such as atrial fibrillation, stroke, and heart failure [9,34]. In this study, we further found that PDE4DIP had a higher frequency of mutations in some cancers, including endometrial cancer, bladder cancer, melanoma, NSCLC, hematological malignancies, and BRCA. Chang et al. conducted whole-exome or targeted gene sequencing of tumor regions in patients with NSCLC and found that many gene mutations occur in patients’ bodies, and certain gene-specific mutation patterns may affect targeted treatment of cancers [35]. Onyeisi et al. found that mutations in the PDE4DIP gene may be a key factor leading to the imbalance of expression of cell surface heparan sulfate proteoglycan syndecan-4, thereby promoting the development of BRCA [11]. In addition, Yao et al. found that the PDE4DIP gene is one of the highly mutated genes during the occurrence and development of leukemia [10]. All in all, given the high mutation frequency of PDE4DIP in various cancers, we speculated that PDE4DIP mutations might be closely related to the development of cancers, which needs to be verified through subsequent experiments in different cancers.

Numerous studies have pointed out that TME has a notable effect on the progression of tumors, including immune cells, blood vessels, fibroblasts, lymphocytes, endothelial cells, and extracellular components (cytokines, hormones, etc.) [36,37]. The content of various immune cell infiltrating tumor tissues and the expression level of immune-related genes are closely related to the immune regulation and prognosis of patients [38]. Our study revealed that PDE4DIP expression was markedly correlated with immune cell infiltration and the expression of immune-related genes in various cancers. In LAML, PDE4DIP expression was positively related to B cell naive and plasma cell infiltration, and negatively related to dendritic cell resting, T cell follicular helper, and mast cell resting infiltration. Moreover, TME score showed that there was a significant positive association between PDE4DIP expression and immune score in AML. These results indicate that PDE4DIP plays a vital role in the possible immune regulation of tumors, which may become a new marker for cancer immunotherapy.

Finally, through functional enrichment analysis, we found that PDE4DIP had a critical impact on the immunotherapy of various cancers, i.e., PDE4DIP expression was mainly associated with immune-related pathways such as JAK-STAT signaling pathway, chemokine signaling pathway, and NOD-like receptor signaling pathway, and in LAML, PDE4DIP expression was also related to immune modulators such as immunoglobulin complexes and T-cell receptor complexes. Currently, immunotherapy is regarded as a promising option for cancer treatment [39]. Our findings further provide more theoretical basis for using PDE4DIP as a biomarker for future immunotherapy of various cancers.

Inevitably, there are some shortcomings in our research that are worth considering. Our research mainly studied the potential role of PDE4DIP expression in multiple cancer types based on online public databases, which may have systematic errors and lack of large-scale clinical cohort data for further analysis and verification. At the same time, the specific role of PDE4DIP in cancers, especially in LAML, still needs to be further clarified through molecular experiments.

5 Conclusion

In summary, our study first conducted a comprehensive analysis of the PDE4DIP gene in pan-cancer, revealing its potential prognostic and immunotherapeutic values in various cancers, including LAML, and providing a theoretical basis for the future use of the PDE4DIP gene as a potential target for the treatment of cancers.

Abbreviations

ACC

adrenocortical carcinoma

LAML

acute myeloid leukemia

LGG

brain lower grade glioma

BLBC

basal-like breast cancer

BLCA

bladder urothelial carcinoma

BRCA

breast invasive carcinoma

CHOL

cholangiocarcinoma

COAD

colon adenocarcinoma

CNV

copy number variation

CLL

chronic lymphocytic leukemia

DSS

disease-specific survival

DFI

disease-free interval

ESCA

esophageal carcinoma

GTEx

genotype-tissue expression

GSEA

gene set enrichment analysis

GO

gene ontology

GBM

glioblastoma multiforme

HNSC

head and neck squamous cell carcinoma

KM

Kaplan–Meier

KICH

kidney chromophobe

KIRC

kidney renal clear cell carcinoma

KIRP

kidney renal papillary cell carcinoma

LIHC

liver hepatocellular carcinoma

LUAD

lung adenocarcinoma

LUSC

lung squamous cell carcinoma

MHC

major histocompatibility complex

MESO

mesothelioma

NES

Normalized Enrichment Score

OV

ovarian serous cystadenocarcinoma

OS

overall survival

PDE4DIP

phosphodiesterase 4D interacting protein

PDE4D

phosphodiesterase 4D

PFI

progression-free interval

PDAC

pancreatic ductal adenocarcinoma

PAAD

pancreatic adenocarcinoma

PCPG0

prostate adenocarcinoma

READ

rectum adenocarcinoma

SARC

sarcoma

SKCM

skin cutaneous melanoma

STAD

stomach adenocarcinoma

TCGA

The Cancer Genome Atlas

TIME

tumor immune microenvironment

TAMs

tumor-associated macrophages

TGCT

testicular germ cell tumors

THCA

thyroid carcinoma

THYM

thymoma

UCEC

uterine corpus endometrial carcinoma

UCS

uterine carcinosarcoma

UVM

uveal melanoma