Identification of genes differentially expressed among microarray datasets

To acquire the gene expression CRC datasets, the microarray data were downloaded from the GEO database. Based on the inclusion/exclusion criteria, CRC and normal or adjacent mucosa tissue gene expression profile of GSE41328, GSE41657, GSE9348 and GSE146587 were selected. By using q-value < 0.05 and [logFC] > 1 as cut-off criterion, 861, 1932, 1568, and 1107 differentially expressed genes (DEGs) were extracted from the expression profile datasets of GSE41328, GSE41657, GSE9348 and GSE146587, respectively (Supplementary Table 1).

GUCA2A was founded as a DEG which is significantly differentiated only in colon adenocarcinoma

To determine a DEG which is significantly differentiated only in colon adenocarcinoma, we first extracted TCGA-COAD associated DEGs from GEPIA2 database. With the q-value < 0.01 and [logFC] > 1 as cut-off criterion, 5356 DEGs were detected (Supplementary Table 1). After integrated bioinformatics analysis, a total of 88 common DEGs were identified among the five profile datasets (Supplementary Table 1), including 59 up-regulated genes and 29 down-regulated genes in the colorectal cancer tissues compared to normal colon tissues (Fig. 2A and B, respectively).

Fig. 2

A total of 88 common DEGs were identified among the five profile datasets. A Common up-regulated DEGs between datasets. B Common down-regulated DEGs between datasets

Second, all 88 common were analyzed in a pan cancer model using the GEPIA2 database. Finally, GUCA2A was founded as a DEG which is significantly differentiated only in colon adenocarcinoma (Fig. 3A). GUCA2A was not only expressed at low levels in tumor tissues compared to adjacent normal tissues, but also than the normal tissues from normal samples (Fig. 3B). The box plots for GUCA2A expression in COAD patients compared to adjacent normal colon tissue and normal colon tissue in normal patients are shown in Fig. 3C and D.

Fig. 3

Pan cancer model for GUCA2A expression. A GUCA2A was founded as a DEG which is significantly differentiated only in colon adenocarcinoma among 33 cancers in tumor tissues compared to adjacent normal tissues and normal tissues from normal samples. B GUCA2A was founded as a DEG which is significantly differentiated only in colon adenocarcinoma among 33 cancers in tumor tissues compared to normal tissues from normal samples. Box plots show that GUCA2A significantly downregulated in COAD samples compared to adjacent normal tissues C and normal tissues in normal patients D

Oncovirus analyzing of GUCA2A was performed using OncoDB database. Cytomegalovirus and Herpes virus do not significantly change the expression of GUCA2A in colon adenocarcinoma (Supplementary File 2 Fig. 1). In the clinical parameter analysis of GUCA2A by expression level, GUCA2A expression in was significantly different between races (p value = 0.003) (Supplementary File 2 Fig. 2). There were no significant differences in GUCA2A expression between gender, BMI, and pathological stages (Supplementary File 2 Fig. 2).

Investigating the diagnostic and prognostic value of GUCA2A

To investigate the diagnostic values of GUCA2A in colon adenocarcinoma, expression level of GUCA2A in COAD and normal patients in OncoDB database was used. The Area under the ROC Curve (AUC) was used to calculate sensitivity and specificity. With 98% sensitivity, 95% Specificity and 99.6% AUC, GUCA2A can be used as a diagnostic factor for COAD (Fig. 4A). To investigate the prognostic values of GUCA2A in colon adenocarcinoma, data from colon adenocarcinoma TCGA datasets were used to perform the survival analyses utilizing UALCAN database. Low expression of GUCA2A was significantly affected on patient’s overall survival (Fig. 4B).

Fig. 4

Diagnostic and prognostic value of GUCA2A. A ROC curve of GUCA2A for colon adenocarcinoma. With 0.95 sensitivity, 0.98 specificity, and 0.996 AUC, GUCA2A can be identified as a diagnostic biomarker for colon adenocarcinoma. B Survival analysis of GUCA2A. Low expression of GUCA2A was significantly affected on patient’s overall survival

Methylation of GUCA2A

Further, we examined the Methylation levels within the GUCA2A gene using OncoDB database. The results showed significant differential methylated levels of the gene between normal and COAD tissues in ‘’cg13452215’’ probe which was hypo methylated (Fig. 5). In addition, in the clinical parameter analysis of GUCA2A by methylation level, significant differences were seen between BMI (p value = 0.029) and pathological stages (p value = 0.0093). There were no significant differences in GUCA2A methylation regarding Race and gender (Supplementary File 2 Fig. 3).

Fig. 5

GUCA2A methylation levels. Significant hypo methylation in ‘’cg13452215’’ probe was observed in COAD tissues compared to normal

Association between GUCA2A with immune-related signatures

It is well established that the immune system plays an important role in tumor development, progression, and therapeutic response, and immune modulators in the tumor immune microenvironment (TIME) may have a significant influence on the infiltrating lymphocytes’ activity [42, 43]. Also, immune cells may serve as an independent predictor of survival and response to chemotherapy [44, 45]. Therefore, it is necessary to investigate the relationship between GUCA2A with TIME. The correlation between GUCA2A expression and immune signatures in colon adenocarcinoma performed via TISIDB (Fig. 6A). There was significant correlation between GUCA2A expression and CD4 + T cell (Act-CD4), central memory CD4 + T cell (Tcm_CD4), Effector memory CD4 + T cell (Tem-CD4), Type 17 T helper cell (Th17), Type 2 T helper cell (Th2), Activated B cell (Act-B cell), Natural killer cell (NK), Natural killer T cell (NKT), Eosinophil, Mast cell, Monocyte, and Neutrophil (Fig. 6B).

Fig. 6

The relationship between GUCA2A expression and immune cells infiltration in colon adenocarcinoma. (A) Analysis between GUCA2A expression and immune infiltration levels. (B) Significant Correlations between GUCA2A expression and immune infiltration levels. * p value < 0.05, **p value < 0.01, ***p value < 0.001

Additionally, the correlation analysis was performed between GUCA2A expressions and immune stimulators in colon adenocarcinoma (Fig. 7A). A significant correlation was observed between GUCA2A expression and CD27, CD40LG, CD48, CXCR4, ENTPD1, HHLA2, ICOSLG, IL6, KLKR1, MICB, NT5E, PVR, TMEM173, TMIGD2, TNFRSF17, TNFRSF25, TNFRSF4, TNFSF13B, TNFSF4, TNFSF9, and ULBP1 (Fig. 7B).

Fig. 7

The relationship between GUCA2A expression and Immune stimulators in colon adenocarcinoma. A Analysis between GUCA2A expression and immune stimulators levels. B Significant Correlations between GUCA2A expression and immune stimulators levels. * p value < 0.05, **p value < 0.01, ***p value < 0.001

Also, correlation GUCA2A expression with immuno-inhibitors in colon adenocarcinoma performed (Fig. 8A). There was significant correlation between GUCA2A expression and ADORA2A, CD160, CD244, CD274, IDO1, IL10RB, TGFBR1, and VTCN1 (Fig. 8B).

Fig. 8

The relationship between GUCA2A expression and immune-related signatures in COAD. (A) Analysis between GUCA2A expression and immune infiltration levels. B Analysis between GUCA2A expression and immuno-inhibitors levels. *P-value < 0.05, **p value < 0.01, ***p value < 0.001

Furthermore, correlation between GUCA2A expression with MHC molecules in colon adenocarcinoma showed significant correlation between GUCA2A expression and HLA-A, HLA-DMA, HLA-DQA2, HLA-G, and TAP2 (Fig. 9).

Fig. 9

The relationship between GUCA2A expression and MHC molecules in COAD. (A) Analysis between GUCA2A expression and MHC molecule levels. B Significant Correlations between GUCA2A expression and MHC molecules levels. * P-value < 0.05, **p value < 0.01, ***p value < 0.001

In addition, the role of GUCA2A expression on immune and molecular subtypes among human cancers was determined using TISIDB website. Immune subtypes were classified into six types, including C1 (wound healing), C2 (IFN-gamma dominant), C3 (inflammatory), C4 (lymphocyte depleted), C5 (immunologically quiet) and C6 (TGF-b dominant). The results showed that GUCA2A expression was related to different immune subtypes in COAD (Fig. 10A). Furthermore, GUCA2A expression differed in different immune subtypes of COAD (Fig. 10B). Based on the above results, we concluded that GUCA2A expression significantly differs in various immune subtypes as well as molecular subtypes of COAD.

Fig. 10

GUCA2A expression and immune and molecular subtypes in colon adenocarcinoma. (A) The relationship between GUCA2A expression and COAD immune subtypes. B The relationship between GUCA2A expression and COAD molecular subtypes

GUCA2A co-expression analysis and PPI network

GUCA2A co-expressed genes were retrieved using LinkedOmics [30], which provided the data from TCGA-COAD cohort. The heatmap of the top 50 positively and negatively GUCA2A correlated genes are demonstrated in Fig. 11A and B, respectively. Besides, the volcano plot of the genes positively and negatively correlated with GUCA2A is shown in Fig. 11C. While GUCA2B, CA4, and TMIGD1 were found to be the highest positively correlated genes, CLUAP1, C5orf46, and KIR2DL3 were found to be the top 3 highest negatively correlated genes with GUCA2A. GUCA2A interacting genes were screened for protein–protein interactions (PPI) network construction using STRING database. A PPI network containing 11 genes with 5.82 average node degree and average local clustering coefficient of 0.824 was constructed (Fig. 11D). We also performed an intersection analysis between these 11 genes and the top 100 genes that have similar expression pattern with GUCA2A obtained from GEPIA2, which identified TMIGD1, SLC26A3, PDZD3 (NHERF4), and GUCA2B as common genes.

Fig. 11

Co-expressed genes with GUCA2A and interaction network of proteins. Top 50 positively A and negatively B similar genes with GUCA2A in COAD. C Volcano plot demonstrating the positively and negatively correlated genes with GUCA2A. (D) Protein–protein interaction network of GUCA2A E Intersection analysis of GUCA2A similar and interacted genes

GO and KEGG pathway enrichment analysis

GO and KEGG pathway enrichment analysis were performed with GUCA2A and its interacting genes. For GO enrichment analysis, GUCA2A and its interacting genes were significantly enriched in 31 Biological processes (BPs) and 17 Molecular functions (MFs). Although sixteen cellular components were enriched, none of them were significant. The most significant biological process and molecular function were receptor guanylyl cyclase signaling pathway (GO:0007168; q-value = 8.52E-4, RF = 0.2) (Fig. 12A) and guanylate cyclase activator activity (GO:0030250; q-value = 7.69E-5, RF = 0.4) (Fig. 12B). Also, the most significant KEGG pathway was pancreatic secretion (q-value = 0.01, RF = 0.02).

Fig. 12

GO enrichment analysis. A Biological process and B Molecular function (MF) of interacting genes

Integrated analysis of miRNA, mRNA, CircRNA ceRNA network

Upstream regulation of GUCA2A was analyzed by screening miRNAs targeted GUCA2A. Of 1003 miRNAs predicted targeting GUCA2A, 16 was found from miRDB, 966 from miRWalk, and 103 from Target Scan databases (Supplementary Table 1). No miRNA was detected by miRTarBase. As a result, hsa-miR-4747-5p, hsa-miR-4722-5p, hsa-miR-5196-5p, hsa-miR-4731-5p, hsa-miR-1207-5p, hsa-miR-4763-3p, hsa-miR-6851-3p, and hsa-miR-558 were screened as the most vital miRNA regulators by overlapping predictions of miRDB, miRWalk, and Target Scan databases (Fig. 13A). Additionally, 61 circRNAs were identified regarding 8 potential miRNAs regulating GUCA2A (Supplementary Table 1). The ceRNA constructed included GUCA2A, 8 shared miRNAs, and 61 circRNAs (Fig. 13B).

Fig. 13

Integrated analysis of miRNA, mRNA, CircRNA, and ceRNA network. A miRNAs that target GUCA2A were screned using the data of mirDIP, miRWalk and Target Scan. Eight miRNAs were detected as the most vital miRNA regulators. B The ceRNA network includes GUCA2A, 8 miRNAs targeting GUCA2A, and 61 circRNAs sponging identified miRNAs

SNPs/Mutations of GUCA2A

Five mutation sites between amino acids 0 and 115, including 5 missense mutations were identified utilizing CbioPortal database (Table 2). Proteins changed by mutation of GUCA2A expression in COAD were P75S, E96K, A98T, A108V, and G114R. By analyzing whether an amino acid substitution affects protein function, P75S, A98T, A108V, and G114R were found as a deleterious mutation [46,47,48]. Also, A98T, A108V, and G114R were found as a probably damaging mutation.

Table 2 Detailed analysis of mutations of GUCA2A gene in TCGA-COADaLack of GUCA2A protein expression level in colorectal cancer tissues

We focused on the levels of GUCA2A protein and its effects on COAD utilizing HPA database. Based on the IHC data, GUCA2A Protein is expressed in normal colon cells with moderate cytoplasmic and membranous staining intensity in endocrine cells, low moderate staining intensity in enterocytes (less than 25%), and low moderate staining intensity in goblet cells (also less than 25%). In contrast, GUCA2A was not detected in tumor cells in any of the aforementioned cell types, indicating a lack of expression in the tumor microenvironment (Fig. 14).

Fig. 14

The immunohistochemical staining results from HPA. There was medium GUCA2A straining in endocrine cells and low straining in enterocytes and goblet cells in normal tissues. However, no GUCA2A straining was detected in colon cancer tissues

Gene drug interaction

According to the results obtained from DGIdb, Lactose anhydrous[50], Atropin[51], and Volanesorsen sodium [52] were identified as a drug that has interactions with GUCA2A (Table 3).

Table 3 Candidate drugs target GUCA2A GUCA2A expression levels in colorectal cancer tissues have shown significant down-regulation compared to NATs

The clinicopathological characteristics of colorectal cancer patients and those with colorectal adenomatous polyps are presented in Table 4. We investigated the expression levels of the GUCA2A gene in cancerous tissues and colorectal adenomatous polyps in comparison to normal adjacent tissues (NATs). Our results indicate that GUCA2A expression is significantly down-regulated in colorectal cancer tissues compared to NATs (p value = 0.0045, Fig. 15a). In contrast, the expression level of GUCA2A in colorectal adenomatous polyps did not show a significant difference compared to NATs (p value = 0.8468, Fig. 15b).

Table 4 Clinicopathological characteristics of colorectal adenomatous polyp and colorectal cancer patientsFig. 15

The scatter diagram shows the expression levels of GUCA2A gene in colorectal cancer tissues and colorectal adenomatous polyps compared to their normal adjacent tissues (NATs). (A) The average level of GUCA2A expression in colorectal cancer tissues showed a significant decrease compared to NATs. B GUCA2A expression levels in colorectal adenomatous polyps were not significantly different from NATs. **: p value < 0.01, ns: non- significant

The expression of GUCA2A in adenomatous polyps equal to or larger than 5 mm has shown a significant decrease compared to polyps smaller than 5 mm

The expression of the GUCA2A gene was analyzed in relation to various demographic and tumor characteristics of colorectal cancer patients, specifically age (younger than 60 years and 60 years or older), sex (male and female), and tumor location (left side and right side). None of these characteristics showed statistically significant differences (Fig. 16a–c). Similarly, the GUCA2A gene expression in colorectal adenomatous polyps was evaluated based on age ((< 65 and ≥ 65), sex (male and female), polyp location (left side and right side), and type of polyps (tubular and tubulovillus/villous). No statistically significant differences were found in these analyses (Fig. 17a–d). Importantly, the expression of the GUCA2A gene in adenomatous polyps was also assessed based on polyp size ((< 5 mm and ≥ 5 mm). Our findings indicate that GUCA2A expression was significantly decreased in polyps measuring 5 mm or larger compared to those smaller than 5 mm (Fig. 17e).

Fig. 16

Comparison of GUCA2A expression levels in colorectal cancer tissues based on clinicopathological characteristics. Expression levels GUCA2A were compared based on (A) age, (B) sex, and (C) tumor location colorectal cancer patients. It was shown that there is no significant difference in the expression of GUCA2A based on the parameters of age, sex and tumor location. ns: non- significant

Fig. 17

Comparison of GUCA2A expression levels in colorectal adenomatous polyps based on clinicopathological characteristics. Expression levels GUCA2A were compared based on (A) age, (B) sex (C) polyp location, (D) type of polyps and (E) polyp size in patients with colorectal adenomatous polyps. It was found that there is no significant difference in the expression of the GUCA2A gene based on the parameters of age, sex, location of the polyp and the type of polyp but the level of expression of GUCA2A in polyps equal to 5 mm and larger than 5 mm was significantly lower compared to smaller than 5 mm. *: p value < 0.05, ns: non- significant

GUCA2A expression can be used as a biomarker for CRC diagnosis

To investigate and analyze the diagnostic value of GUCA2A expression in colorectal cancer tissues and colorectal adenomatous polyps, we performed a Receiver Operating Characteristic (ROC) curve analysis for each group (Fig. 18a and b). Our results revealed that AUC for colorectal cancer tissues was 77%, which is statistically significant (p value: 0.0083). The analysis indicated a sensitivity of 68.75% and a specificity of 81.25%. In contrast, the ROC curve analysis for colorectal adenomatous polyps showed an AUC of 59%, with a sensitivity of 73.33% and a specificity of 60%. However, this finding was not statistically significant (p value = 0.3837).

Fig. 18

Investigating the biomarker property and diagnostic value of GUCA2A expression in colorectal cancer tissues, and colorectal adenomatous polyps using ROC curve analysis and based on the area under the curve (AUC). A AUC for GUCA2A expression levels in colon cancer tissues was 77% which is statistically significant. (B) GUCA2A expression levels in colon adenomatous polyps with AUC of 59% were not statistically significant