CXC Chemokine Expression in COAD Patients
We used the TCGA database to obtain sixteen CXC chemokines (excluding CXCL15). Using UALCAN, we determined the CXC chemokines expression levels in both normal tissues and the COAD tissues. The CXCL1 expression levels (p <1e−12), CXCL2 (p = 1.62e−12), CXCL3 (p <1e−12), CXCL5 (p =1.05e−07), CXCL6 (p =5.27e−11), CXCL11 (p = 4.20e−11), CXCL16 (p <1e−12), and CXCL17 (p = 3.83e−05) in COAD tissues were considerably elevated, whereas the levels of transcription of CXCL12 (p = 2.02e−10), CXCL13 (p = 1.44e−02), and CXCL14 (p = 1.61e−02) were significantly attenuated. These results are presented in Figure 1.
We also analyzed the relative expression levels regarding CXC chemokines in COAD tissues and discovered that CXCL16 had the highest relative expression level of any of the CXC chemokines we assessed (Figure 2). For the purpose of finding other CXC chemokines that are correlated with the progression, carcinogenesis, as well as clinical outcomes of COAD patients, we analyzed all of the CXC cytokines shown to be differently expressed between COAD tumors and normal tissues (CXCL1, CXCL2, CXCL3, CXCL5, CXCL6, CXCL11, CXCL12, CXCL13, CXCL14, CXCL16, and CXCL17). CXCL4, CXCL7, CXCL8, CXCL9, CXCL10, and CXCL15 were inapplicable in the subsequent investigation because their expression levels were similar both in COAD tumors and normal tissues. We investigated the relationship between the differentially expressed CXC chemokines and the clinical stage of COAD patients, where we observed a considerable correlation between the pathological stage and the CXCL9 (p = 0.003), CXCL10 (p = 0.006), and also CXCL11 (p = 0.004) (Figure 3). The expression levels of CXCL9, CXCL10, and CXCL11 were elevated with the progression of the tumor. According to the findings, these CXCL chemokines perform a critical role in the occurrence and progression of COAD tumors.
Prognostic significance of CXC Chemokines in COAD Patients
We used GEPIA to evaluate CXC chemokine expression in COAD progression, comparing the associated expression of CXC chemokines with clinical outcomes. Figure 4 depicts disease-free survival curves. The presence of high transcriptional levels of CXCL9 (p = 0.026), CXCL10 (p = 0.0011), and CXCL11 (p = 0.0045) in COAD patients was shown to be correlated with a prolonged disease-free survival (DFS) time. The expression of the CXC chemokine was also examined in relation to the survival of COAD patients. We discovered that COAD patients with high transcriptional levels of CXCL2 (p = 0.042), CXCL3 (p = 0.047), and CXCL14 (p = 0.045) had substantially longer overall lifespan (Figure 5).
CXC Chemokine Gene Mutation, Interaction, and Co-expression analysis in Patients with COAD
We carried out a thorough investigation of the molecular features of the CXC chemokines. Genetic changes were investigated using the TCGA database. We found that CXCL1, CXCL2, CXCL3, CXCL8, CXCL9, CXCL10, and CXCL17 were altered in 5, 5, 5, 5, 5, 5, and 5% of the retrieved COAD samples (Figure 6A). There is a potential co-expression among the CXC chemokines that are differently expressed. We used STRING to create a PPI network to examine CXC chemokines with differential expression for the purpose of investigating the possible connections between them. In the PPI network, we discovered multiple nodes with an 8 and several edges with a 43 that were acquired from it (Figure 6B).
The inflammatory response, as well as the chemokine signaling pathway, were shown to be correlated with the role of CXC chemokines. The findings of GeneMANIA also indicated that the roles of CXC chemokines with differential expression were mostly correlated with chemokine activities, cell chemotaxis, as well as chemokine receptor binding, (Figure 6C).
Transcription Factor Target COAD Patients
In view of the considerable difference in the CXC chemokines expression between COAD tissues and normal tissues, we investigated the possibility that differentially expressed CXC chemokines had transcription factor and kinase targets utilizing TRRUST databases., including CXCL1, CXCL2, CXCL5, CXCL8, CXCL10, CXCL12, as well as CXCL14. Three transcription factors (NFKB1, RELA, and SP1) were discovered to be correlated with the modulation of CXC chemokines(Table 1). The hub transcriptional factors for CXCL1, CXCL2, CXCL5, CXCL8, CXCL10, and CXCL12 were RELA and NFKB1. Furthermore, SP1 was the primary transcriptional factor involved in the expression of CXCL1, CXCL5, and CXCL14.
CXC Chemokines Immune Cell Infiltration in COAD Patients
COAD patients have been demonstrated to have altered clinical outcomes after being exposed to CXC chemokines, which are correlated with immune cell infiltration and inflammatory reactions. In view of this, we utilized the TIMER algorithm to examine the relationship between immune cell infiltration and differentially expressed CXC chemokines. The findings illustrated a negative association between macrophages infiltration (Cor = −0.15, p = 2.46e−03) and CXCL1 expression, and a positive association between CXCL1 expression and neutrophils infiltration (Cor = 0.364, p = 4.93e−14), CD8+ T cells infiltration (Cor = 0.122, p = 1.36e−02), and dendritic cells infiltration (Cor = 0.177, p = 3.72e−04) (Figure 7A).
CXCL2 expression was correlated with the macrophages infiltration (Cor = −0.197, p = 6.80e−05) in a negative manner, and correlated with the neutrophils infiltration (Cor = 0.265, p = 7.14e−08) (Figure 7B) in a positive manner.
CXCL3 expression was shown to be negatively correlated with the infiltration of macrophages (Cor = −0.229, p = 3.48e−06), and positively correlated with the infiltration of neutrophils (Cor = 0.257, p = 1.86e−07) (Figure 7C).
CXCL5 expression was shown to exhibit a positive correlation with the macrophages infiltraton (Cor = 0.148, p = 2.88e−03), CD8+ T cells infiltration (Cor = 0.151, p = 2.27e−03), neutrophils infiltration (Cor = 0.435, p = 5.46e−20), as well as dendritic cells infiltration (Cor = 0.22, p = 8.14e−06) (Figure 7D).
Furthermore, CXCL6 expression was found to be positively correlated with the macrophages infiltration (Cor = 0.218, p = 9.80e−06), CD8+ T cells infiltration (Cor = 0.226, p = 4.07e−06), neutrophils infiltration (Cor = 0.456, p = 6.06e−22), and also dendritic cells infiltration(Cor = 0.277, p = 1.67e−08) (Figure 7E).
CXCL9 expression was illustrated to positively correlated with the CD8+ T cells infiltration (Cor = 0.488, p = 1.25e−25), B cells infiltration (Cor = 0.267, p = 4.99e−08), CD4+ T cells infiltration (Cor = 0.292, p = 2.45e−09), neutrophils infiltration (Cor = 0.684, p = 1.27e−56), dendritic cells infiltration (Cor = 0.632, p = 3.63e−46) and macrophages infiltration (Cor = 0.428, p = 1.81e−19) (Figure 7F).
The same outcomes were acquired for CXCL10, CXCL11, CXCL12, CXCL13, CXCL14, CXCL16, and CXCL17. CXCL10, CXCL11, CXCL12, CXCL13, and CXCL16 expression was shown to be positively correlated with the infiltration of CD8+ T cells, B cells, CD4+ T cells, neutrophils, dendritic cells, as well as macrophages (all p < 0.05) (Figures 7G–J, L).
Expression of CXCL14 was negatively correlated with neutrophils infiltration (Cor = -0.163, p = 1.09e−03), CD8+ T cells infiltration (Cor = -0.141, p = 4.33e−03), and dendritic cells infiltration (Cor = -0.109, p = 2.91e−02), and showed positive correlation withthe infiltration of CD4+ T cells (Cor = 0.186, p = 1.80e−04) (Figure 7K).
Expression of CXCL17 was positively correlated with B cell infiltration (Cor = 0.201, p = 4.59e−05) (Figure 7M).
The relationship between CXC chemokines with differential expression and the infiltration status of immune cells was investigated. Specifically, the Cox proportional hazards model was employed in the present research, which was modified to take into account the confounders below: dendritic cells, CD8+ T cells, macrophages, B cells, CD4+ T cells, neutrophils, CXCL1, CXCL2, CXCL3, CXCL5, CXCL6, CXCL9, CXCL10, CXCL11, CXCL12, CXCL13, CXCL14, CXCL16, and CXCL17. The expression of CXCL13 (p = 0.020) and CXCL14 (p = 0.012) was shown to have a substantial correlation with the clinical outcomes of COAD patients (Table 2).