Prostate cancer (PCa) is the second most common cancer among men worldwide. Localized PCa is usually treated with surgery and radiation therapy, which are ineffective at the distant metastasis stage (Sartor and de Bono, 2018). PCa is characterized by distant metastasis, which most commonly occurs in the bones, liver, lungs, and brain (Kfoury et al., 2021). Metastatic and advanced PCa induces drug resistance to current therapies, which contributes to the poor prognosis. Nearly 80% of patients treated with androgen deprivation therapy finally become unresponsive, resulting in a median survival of only 14 months (Shafi et al., 2013). Therefore, the identification of pathophysiological mechanism and diagnostic biomarkers needs further investigation.

Recently, developments in microRNAs (miRNAs), which are endogenous single-stranded noncoding RNAs regulating gene expression post-transcriptionally, provided new insights into the pathogenesis of cancer (Lu and Rothenberg, 2018; Kanavarioti et al., 2024; Wu et al., 2024; Hor et al., 2023). Different types of tumors can be regulated by miRNAs, which function as either tumor suppressors or oncogenes, such as miR-21, which has both oncogenic and onco-suppressor functions (Hashemi et al., 2023). MiRNAs involved in PCa tumorigenesis are usually found to be deregulated, influencing many processes at the molecular and cellular levels (Padmyastuti et al., 2023; Slabáková et al., 2021; Wang et al., 2024; Lu et al., 2023; Armstrong et al., 2024). For example, miR-24, functioning as a cancer suppressor, is frequently downregulated in PCa cells (Cheng et al., 2021). Another oncogene, miR-888, promotes PCa growth by suppressing retinoblastoma-like protein 1, which can directly bind to the transcriptional factor E2F and regulate cell cycle progression from the G1 to S phase (Hasegawa et al., 2018). Furthermore, a study reports an exovesicle-derived miR-20a-5p, which can regulate PCa cell proliferation and inflammation through the RORA gene (Sánchez et al., 2024). The potential utility of miRNA as biomarkers has been widely reported in the past decade (Fabris et al., 2016). Despite that, there are very few studies analyzing the miRNA–mRNA regulatory network in PCa. Research on the role of miRNA in PCa is crucial for early diagnosis and effective treatment.

In this research, we screened out differentially expressed miRNAs (DE-miRNAs) in serum and tissue samples of PCa using bioinformatics methods (Barrett et al., 2013; Blum et al., 2018). As predicted by miRTarBase, miR-146a-3p and miR-136-3p are two of the most downregulated miRNAs. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were used for detecting potential biological functions of the 948 target genes through Database for Annotation, Visualization, and Integrated Discovery (DAVID). We also developed the protein–protein interaction (PPI) network using Cytoscape to reveal regulatory mechanisms of miRNA–mRNA networks. The expression of the top 10 target genes was further validated using the UALCAN database. We further validated our findings using the UALCAN database by obtaining serum samples from PCa and benign prostatic hypertrophy patients and measuring expression levels of miR-146a-3p and miR-136-3p in these samples. Only miR-136-3p maintained consistency in the serum and tissue. In this study, we aim to identify PCa-associated miRNAs through various bioinformatic analyses and validate the consistency of miR-136-3p expression between serum and tissue samples. For this reason, our findings may provide a simpler diagnosis of PCa using blood without biopsy.

PCa claims thousands of lives every year, mainly due to its drug resistance and invasiveness, despite multiple new drug approvals in recent years. The early diagnoses of PCa have become a very important issue. Liquid biopsy is a scientific source of biomarkers and is currently a major focus of clinical research. This approach can provide direct assistance for disease diagnosis using blood, urine, and other body fluids and allows for sustainable monitoring of the disease’s response to treatment (Raza et al., 2022; Detassis et al., 2024; Sequeira et al., 2023; Mao et al., 2023; Vaidyanathan et al., 2016). However, the heterogeneity of biomarkers in serum and tissue remains to be considered. MiRNAs play a critical role in the regulation of cancer progression and serve as biomarkers at various stages of PCa. Scientists have reported that inhibiting miR-4719 and miR-6756-5p to upregulate IL-24 may represent a therapeutic strategy for aggressive PCa (Das et al., 2019).

In this study, we aimed to identify potential biomarkers for PCa by detecting miRNAs from serum and tissue datasets. The differential expression analysis was performed using a public database. Two miRNAs, miR-146a-3p and miR-136-3p, both downregulated in serum and tissue, were identified. Recent research suggested that miR-146a-3p is related to the occurrence and progression of diseases, including asthma, allergic rhinitis, and Paget’s disease (Duan et al., 2023; Xia et al., 2023; Stephens et al., 2020). Similarly, miR-136-3p is also reported to inhibit tumorigenesis (Xu, 2020). However, it has not been reported that miR-146a-3p and miR-136-3p may participate in the progression of PCa. It is significant to explore the functions of miR-146a-3p and miR-136-3p in PCa and elucidate their mechanisms. Next, the researchers predicted 948 target genes that might be regulated using the two common downregulated miRNAs and performed functional enrichment analysis. The results demonstrated that these target genes were enriched in protein binding, porphyrin and chlorophyll metabolisms, and the mTOR signaling pathway.

To our knowledge, numerous RNA-binding proteins are involved in regulating the post-transcriptional processes and have a profound impact on RNA metabolism (Corley et al., 2020). It has been well documented that protein binding is closely associated with tumor migration and invasion (Wang et al., 2022). The activation of the mTOR pathway is the major promoter of various cellular activities, including protein synthesis, tumor proliferation and invasion, autophagy, and cellular metabolism (Hua et al., 2019; Mossmann et al., 2018). Recent studies have indicated that the complex interactions within the PI3K–AKT–mTOR pathway may promote PCa progression and influence the resistance of tumor cells to mTOR-targeted therapy (Dai et al., 2021).

According to the PPI network, the top 10 hub genes of two downregulated miRNAs were screened out. With further evaluation using the UALCAN database, we found that NSF, HIST2H2BE, IGF2R, and CADM1 were identified as hub genes with higher degrees. A few studies have reported these genes in other diseases and cancers. N-ethylmaleimide-sensitive factor (NSF) is an ATPase involved in intracellular vesicle trafficking, mostly found in eukaryotic cells, and is, therefore, considered a potential therapeutic target (Calvert et al., 2007). HIST2H2BE was demonstrated to regulate cancer progression and development. The upregulation of HIST2H2BE has been found in gastric cancer and invasive ductal carcinoma (Guo et al., 2010; He et al., 2021). However, the association with PCa still needs to be investigated. IGF2R, the hub gene predicted in this study, functions as a receptor for insulin-like growth factor 2. Apart from the intracellular trafficking of lysosomal enzymes, mutation or loss of this gene has been confirmed to be associated with various cancers, including gastrointestinal cancer, renal tumors, and osteosarcoma (Oates et al., 1998; Xu et al., 1997; Broqueza et al., 2021). However, the underlying mechanism by which IGF2R promotes the progression of PCa, especially its crosstalk with microRNA, remains unsolved. The final predicted gene, CADM1, is also involved in several processes, including cell recognition, positive regulation of cytokine secretion, and natural killer cell-mediated susceptibility to cytotoxicity (Li et al., 2021). The functions of these hub genes encompass intracellular transport (NSF), epigenetic regulation (HIST2H2BE), growth factor signaling regulation (IGF2R), and cell adhesion (CADM1), all of which are core biological processes closely associated with cancer development and progression, including tumor cell proliferation, survival, invasion, and metastasis. It is generally assumed that miRNAs negatively regulate their target genes, but the specific regulatory pathway remains to be investigated.

Then, we validated the expression and prognostic roles of miR-146a-3p and miR-136-3p in the public database and clinical PCa patients. Only miR-136-3p was downregulated both in serum and tissue samples according to GEO and TCGA. A consistent trend was confirmed through qRT-PCR analysis of clinical samples. The observed consistency raises the question of whether blood-based assessment of miR-136-3p could replace biopsies for early diagnosis. Compared to the traditional biomarker such as PSA, the clinical applicability of miR-136-3p needs further rigorous experiments and clinical trials. However, miR-146a-3p heterogeneity in serum versus tissue compels us to reflect on the underlying reasons. Previous studies have also found inconsistency; Skog et al. (2008) detected tumor-specific miRNAs in the serum of patients with glioblastoma, but tissue expression levels were not fully correlated with those in serum. The miRNAs in the multi-source serum of circulating miRNAs may originate from organs other than tumor tissue (such as liver and immune cells) or extracellular vesicles (exosomes and microparticles), and miRNAs in tissue samples more directly reflect the local microenvironment (Turchinovich et al., 2011). Valadi et al. (2007) confirmed that extracellular vesicles can transfer miRNAs from donor cells to recipient cells, resulting in incomplete consistency between circulating miRNAs and tissue sources. Non-tumor cells, such as tumor-associated fibroblasts (CAFs) and immune cells, may secrete specific miRNAs into the bloodstream, while miRNAs in tissue samples mainly come from tumor cells themselves. Researchers found that breast cancer cells secrete miRNA through exosomes, but CAFs also contribute to the circulating miRNA (Melo et al., 2014). Disease stages and dynamic changes may also contribute to the secretion mode of miRNA. Early-stage tumors may preferentially secrete specific miRNAs into the bloodstream (such as miR-21 as an early diagnostic marker), while late-stage tumor tissues may experience changes in miRNA release patterns due to necrosis (Chen et al., 2008).

In conclusion, we confirmed that miR-136-3p is poorly expressed in PCa serum and tissue samples and might serve as a biomarker in PCa. However, our study has some limitations: (1) only two target genes with overlapping downregulated miRNAs were identified for further enrichment analysis; (2) the hub genes of miR-136-3p showed low degree, and detailed molecular mechanisms of miR-136-3p downregulation in PCa patients are lacking; (3) more clinical survival data need to be included for detecting prognosis efficiency; and (4) why miR-146a-3p shows a different tendency between serum and tissue remains to be investigated.

In our study, we confirmed that miR-136-3p plays an important role in the progression of PCa through bioinformatics analysis and qRT-PCR validation. These findings provide new approaches for targeting miR-136-3p as a biomarker of PCa.