Background

In 2020, the World Health Organization (WHO) reported that cancer was one of the leading causes of death, claiming nearly ten million lives [1]. Among all types of cancer, colon and rectal cancer, also known as colorectal cancer (CRC), is the third most commonly diagnosed type worldwide, and it caused 935,000 deaths in the year 2020, making it the second deadliest type of cancer ahead of liver, stomach, and breast cancer [2]. As of now, there are three main treatment options for CRC: surgical resection, chemotherapy, immunotherapy, or any combination of these therapies. Nevertheless, the effectiveness varies from patient to patient, especially in cases of locally invasive or metastatic cancer. We must therefore understand how this disease progresses in order to identify biomarkers that can be used to identify potential therapeutic targets in order to improve the prognosis and treatment of CRC patients.

It is possible to improve treatment for CRC with targeted therapeutic agents that target unique biological features and pathways involved in tumor progression. In contrast to other therapeutics that kill both cancerous and normal cells, these work best in cancer treatment because they target the biological features of cancerous cells only [3]. Several pathways and processes in cancerous cells can be turned off by targeted therapy, including angiogenesis, proliferation, apoptosis inhibition, differentiation, the RAS pathway, the Wnt signaling pathway, the PI3K pathway, and the cell cycle pathway [4].

A number of studies shown that activation of the highly conserved Wnt pathway in a deviant manner is a driving factor in the tumorigenesis of most human cancers, with a strong emphasis on CRC [5]. This pathway controls β-catenin, a key modulator for signal transduction in CRC through phosphorylation and ubiquitin-mediated degradation. This regulation involves key scaffold proteins such as AXIN and disheveled (DVL) which disrupt the β-catenin destruction complex that is contains 3 core proteins; adenomatous polyposis coli (APC), glycogen synthase kinase 3 beta (GSK3β), and casein kinase 1 (CK1) [6]. When the destruction complex is disrupted, β-catenin will no longer be degraded hence leading to its accumulation as free β-catenin in the cytoplasm (Piawah and Venook, 2019b), which is a hallmark of CRC progression (Cheng et al. 2019) when it translocates to the nucleus [5]. This translocated beta-catenin in association with two major transcriptional factors, i.e., T cell factor (TCF) and lymphoid enhancer-binding factor (LEF), displaces their repressor molecule Groucho. The β-catenin/TCF/LEF complex, in association with other co-activators, forms a transcriptional complex that leads to the expression of target genes of Wnt, which include MYC, CCND1, AXIN2, Cyclin D1, among others [7, 8]. These target genes are mostly oncogenes [9]. Moreover, the abnormal up-regulation of the Wnt signaling pathway is facilitated by APC mutations, which is a negative regulator of this pathway [6]. These mutations mainly lead to the loss-of-function of APC hence upregulating the Wnt signaling pathway and facilitating CRC cell proliferation and enhanced anti-apoptosis abilities through overexpression of the target genes of this pathway [9].

Wnt signaling pathway is characterized as either canonical, which is β-catenin dependent or non-canonical, which is β-catenin independent. However, the initiation of signaling events in both pathways involves the binding of Wnt molecules to frizzled receptors and other related receptors like the Low-density lipoprotein Receptor-related Protein 5/6 (LRP5/6)/ROR2/RYK for signal transduction initiation [9]. Frizzled (FZD) is a family of 10 transmembrane proteins which serve as receptors of the Wnt pathway, with every FZD member having a favored Wnt ligand [10]. Various studies have indicated that excessive activation of the Wnt signaling pathway may be a result of a loss-of-function mutation in E3 ubiquitin ligases ring-finger protein 43 (RNF43), through ubiquitin-mediated degradation blockage of frizzled receptors and LRP5/6 co-receptors. Since this is a frequently detected phenomenon in CRC [8], it, therefore, indicates that signal transduction by the Wnt pathway can be influenced as levels of expression for key components of the pathway get altered [5]. Since Wnt is the most implicated pathway in CRC, disrupting the pathway signal transduction through downregulating the expression of crucial pathway components such as FZD receptors can be a therapeutic strategy for CRC.

Human frizzled homolog 3 protein (FZD3) is located on chromosome 8p21 and is expressed in skeletal muscles, pancreas, cerebellum, stomach, kidney, and among other tissues [11]. A number of studies have shown that FZD3 is up-regulated in tissues from lung squamous cell carcinomas, lymphomas, Ewing sarcomas, and myeloma, among other cancers [12, 13]. In their study, Wong and colleagues (2013) reported that FZD3 was 100% expressed in CRC spacemen, 89% in colorectal adenomas, and 75% in colorectal polyp spacemen [12]. Therefore, there is no doubt that FZD3 plays a critical role in the development and progression of CRC, making it a potential candidate for preventative interventions.

Recent studies have demonstrated that microRNAs (miRNAs) may be effective inhibitors of proliferation, growth, and metastasis of CRC cells by targeting FZD receptors and oncogenes [13,14,15]. These single-stranded noncoding RNAs bind to the 3′ untranslated region (3′ UTR) of the target gene mRNA, thereby negatively regulating its expression. This results in the cleavage of the target gene or repression of its translation, thereby inhibiting the production of the target protein [14]. The interaction of MicroRNAs with FZD mRNAs influences the expression of FZD proteins and the Wnt pathway as a result (Smith et al. 2021). Identifying miRNAs that inhibit the expression of FZD genes in different cancers is a reported therapeutic strategy for human cancer [10]. In comparison to other FZD receptors, the FZD3 receptor has received relatively little attention in human cancers, especially colorectal cancer. It is, therefore, the purpose of this study to identify a suitable miRNA target for FZD3 receptor mRNA and to demonstrate its effectiveness as an inhibitor. The finding of this study would allow the clinical evaluation of the potential of miRNA in inhibiting CRC progression and its consideration as a therapeutic strategy in CRC treatment.

Methods

Gene expression datasets retrieval

Microarray data of five gene expression projects for normal colon and rectum samples were retrieved from the Gene Expression Omnibus database (GEO), searching against query words such as colorectal cancer and CRC, on 27th September 2021 [16]. The study’s criteria for selecting samples were Homo sapiens-derived samples, excluding cell line experiments (Table 1). In the GSE41657 dataset, there were 12 normal cells, 25 carcinoma cells, and 51 dysplastic cells. Since dysplastic cells are not true carcinoma cells, we only included the normal and carcinoma cell samples in our study.

Table 1 Characteristics of the datasets
Full size table

Identification of differentially expressed genes (DEGS)

We used the limma R package (Ritchie et al. 2015) and the GEO2R tool to normalize and identify differentially expressed genes in each dataset [22]. DEGs were selected in all 5 datasets at log fold change (log FC) = 1 or − 1 and P values > 0.05 as the cut-offs. Bioinfokit v2.0.1 tool in Python was used to plot volcano plots for DEGs from each dataset [23].

Functional enrichment analysis

DEGs were arranged in descending order based on their magnitude of Log FC value after obtaining DEGs. We analyzed the top 20 DEGs from each dataset for their enrichment to understand their in-depth significance. The Supplementary Data file mentions details of the parameters submission of the top 20 DEGs. Enrichr was used to annotate the top 20 DEGs from each dataset [37]. Pathways that offer potential sites for targeted therapy in CRC include among others; Wnt/β-catenin, HGF/c-MET pathway, notch, hedgehog, and EGFR-related pathways [37]. In spite of being referred to as non-coding RNA molecules, miRNAs have been shown by a number of studies to play an important role in regulating 60% of human genes post-transcriptionally and in being associated with cancer [38]. Hence, we sought to inhibit FZD3, one of the frequently up-regulated frizzled receptors of the Wnt signaling pathway in CRC, by using miR-98-5p, a rarely reported miRNA in this disease.

FZD3 receptor expression was determined using differential gene expression analysis on five datasets and found that the change in levels of expression between tumor and normal samples was significant enough for it to be a DEG alongside other genes. Following an analysis of gene ontology enrichment, it was found that FZD3 was significantly enriched in the Wnt signaling pathway along with many other genes within the pathway, which correlates with recent studies which implicate Wnt signaling in the development of CRC [9, 39, 40]. Further analysis of the GEDS web server datasets revealed that FZD3 expression was higher in CRCs, which validated our findings. Furthermore, we have constructed a PPI and found FZD3 receptor interacting with key WNT signaling pathway genes, including DVL, WNT1, WNT5, LRP6, and VANGL2 that were also found to be upregulated in our study data. It was also found that sFPR1 (secreted frizzled receptor protein 1), a Wnt antagonist, was downregulated in all the datasets, which was previously reported to be down-regulated by the up-regulation of FZD3 [41]. Finally, it has been reported in several studies that FZD3 expression is correlated with Wnt target genes, Cyclin D1 and c-My, which we also found to be true in our analysis [14]. Therefore, on the basis of these results and literature references, it seems reasonable to suggest that FZD3 plays a crucial role in the Wnt pathway and that its inhibition would inhibit the pathway (Fig. 8).

Fig. 8
figure 8

Inhibition of CRC progression. The illustration depicts the putative progression of Wnt signaling pathway with/out miR-98-5p in CRC cells

Full size image

Recent studies have demonstrated that microRNA miR-98-5p inhibits tumor proliferation, migration and invasion by targeting the Wnt signaling pathway-related genes in various cancers including ovarian cancer [42], glioblastoma [43], gastric cancer [44] in non-small cell lung cancer [45], and pancreatic ductal adenocarcinoma [46]. The findings represent miR-98-5p as a potential target of FZD3, one of the major receptors of the Wnt pathway. Computational analysis showed that the FZD3 mRNA contained the binding sites for miR-98-5p in its 3′-UTR, which is a key feature in the miRNA post-translational gene regulation mechanism. However, miRNA prediction algorithms can barely confirm that miR-98-5p targets FZD3 directly in CRC samples. To validate the study results, luciferase reporter assays should be done to compare the behavior of a wild-type (WT) as well as the mutated (MUT) FZD3 in the 3’UTR binding site. The difference in fluorescence between FZD3-WT and FZD3-MUT will confirm that miR-98-5p is directly targeting the FZD3 gene. FZD3 and miR-98-5p could be forming an axis that inhibits Wnt signaling and CRC in general; however, the involvement of other target genes in the process cannot be ruled out. It is important that all the predicted target genes by at least two miRNA prediction algorithms are enriched in Wnt pathways by gene ontology and KEGG in order to validate the mechanism by which miR-98-5p inhibits Wnt signaling pathways. The mRNA expression levels of such genes can then be measured with miR-98-5p mimic and inhibitor, respectively.

Conclusion

In conclusion, this study demonstrated that FZD3 is upregulated in CRC along with other crucial genes of the Wnt signaling pathway. Moreover, this provides evidence that miR-98-5p may inhibit the expression of FZD3, which may lead to reduced proliferation and metastasis of colorectal cancer cells, and these findings can be used in the development of target-based therapies for CRC patients. It is essential, however, that these findings be validated by basic research in the future to determine the mechanism by which miR-98-5p regulates CRC cells, both in vivo and in vitro.