High expression of eight members of the HSP70 family in HCC patient.
The UALCAN web portal was applied to detect the expression levels of the eight members of the HSP70 family in normal and in HCC tissues, respectively. Through the UALCAN, we observed the high expression of HSPA1A, HSPA1B, HSPA1L, HSPA2, HSPA5, HSPA6, HSPA8 and HSPA9 mRNA in 371 patients with liver cancer, as compared to their lower expression in 50 normal liver tissues, which was statistically significant (P<0.05, Fig. 1A-H).
Prognostic value of members of HSP70 in HCC from KM plotter database.
Firstly, the prognostic effect of HSPA1A was assessed using the information collected from the dataset, with the valid gene ID 3303(HSPA1A). Curves that described the survival performance were drawn as demonstrated in Fig. 2A (n=364). It turned out that highly expressed HSPA1A would significantly lead to a worse OS performance in patients with HCC ([hazard ratio (HR) =1.49; 95% confidence interval (CI): 1.03-2.15; P=0.031]).
Then, the prognostic effect of HSPA1B was examined using the information collected in this database, with the valid gene ID 3304(HSPA1B). It turned out that highly expressed HSPA1B would significantly lead to a worse OS performance in patients with HCC (HR=1.49; 95%CI: 1.05-2.12; P=0.026; Fig. 2B).
Subsequently, the prognostic effect of HSPA6 was also examined using the information collected from the database, with the valid gene ID 3310 (HSPA6). It turned out that highly expressed HSPA6 would significantly lead to a worse OS performance in patients with HCC (HR=1.53; 95%CI: 1.06-2.2; P=0.021; Fig. 2F).
Furthermore, the prognostic effect of HSPA8’s was examined using the information collected from the database, with the valid gene ID 3312(HSPA8). It turned out that highly expressed HSPA8 would significantly lead to a worse OS performance in patients with HCC (HR=1.81; 95%CI: 1.21-2.71; P=0.0036; Fig. 2G).
In contrast, no significant correlation has been detected between the high expression of HSPA1L, HSPA2, HSPA5 and HSPA9 and OS performance in patients with HCC. Their ID, HR, CI and P- values are as follows: The HSPA1L gene ID was 3305, HR =1.2 (95% CI: 0.84–1.72), P=0.32 (Figure. 2C). The HSPA2 gene ID was 3306, HR =1.2 (95% CI: 0.81–1.76), P=0.36 (Figure. 2D). The HSPA5 gene ID was 3309, HR =1.25 (95% CI: 0.86–1.81), P=0.24(Figure. 2E). The HSPA9 gene ID was 3313, HR =0.72 (95% CI: 0.49-1.04), P=0.078(Figure. 2H).
In summary, a significant correlation was detected between the members of the HSP70 family, showing high expressions of HSPA1A, HSPA1B, HSPA6 and HSPA8 were associated with OS of HCC patients, while this correlation was not significant between the OS performance of HCC patients and the high expression of HSPA1L, HSPA2, HSPA5 and HSPA9.
Prognostic value of members of HSP70 in HCC from clinicopathological characteristics .
Besides, we further analyzed the related factors influencing the prognosis of HCC from clinicopathological characteristics. Including clinical stage (Table 1), cancer grade (Table 2), AJCC-T (Table 3), vascular invasion (Table 4), gender (Table 5), race (Table 6), alcohol consumption (Table 7), hepatitis virus (Table 8). From table 1, we found that high expression of HSPA1A,HSPA2,HSPA6,HSPA8 and HSPA9 was connected with poor OS in stage I HCC patients; High expression of HSPA1A and HSPA2 with Stage II had a similar outcome. In table 2, high HSPA1A, HSPA1L and HSPA8 mRNA expression was correlated with a poor prognosis for HCC patients with Grade 1; High expression of HSPA6 was associated with a worse OS for HCC patients with Grade 2; but the high expressions of HSPA5 and HSPA6 showed correlations with better OS for HCC patients with Grade 1. The results in table 3 illuminated that high expression of HSPA6 and HSPA8 was related to unfavorable OS of HCC patients with AJCC-T 1; High expression of HSPA1A with AJCC-T 1 showed a similar outcome. As shown in Table 4, High expression of HSPA8 was relevant to poor OS of HCC patients with no vascular invasion; On the contrary, High expression of HSPA8 showed a good OS from HCC patients with micro vascular invasion. As presented in Table 5, High expression of HSPA1A, HSPA1B, HSPA1L and HSPA6 had poor prognosis in female HCC patients; High expression of HSPA8 was associated with a poor prognosis for HCC male patients; In contrast, high expression of HSPA8 was associated with a better prognosis for HCC female patients. In table 6 we further investigated that high expression of HSPA1A, HSPA1B, HSPA1L, HSPA5 and HSPA8 was correlated with poor OS of white HCC patients; High expression of HSPA6 was relevant to worse OS in both white and Asian patients of HCC. In table 7, High expression of HSPA1L was relevant to worse OS for patients of HCC with alcohol consumption; Whether drinking or not, high expression of HSPA6 was associated with poor OS in HCC patients; High expression of HSPA9 showed a worse OS in HCC patients without alcohol consumption. Lastly table 8 demonstrated that high expression of HSPA1B was connected with worse OS in HCC patients, Whether or not infected with hepatitis virus; High expression of HSPA6 and HSPA8 was related to poor OS in HCC patients infected with hepatitis virus; High expression of HSPA1A, HSPA1L and HSPA9 had a connection with poor OS in HCC patients not infected with hepatitis virus.
GO functional annotation analysis and KEGG pathway analysis for members of HSP70.
Functional enrichment analysis together with the corresponding genes was investigated for the members of HSP70, with the usage of KEGG and GO analyses, conducted with DAVID, in which the biological processes (BPs), cellular components (CCs) and molecular functions (MFs) were predicted via GO analysis. According to the analysis, BPs included such as GO:1903265 positive regulation of tumor necrosis factor-mediated signaling pathway, GO:1902236 negative regulation of endoplasmic reticulum stress-induced intrinsic apoptotic signaling pathway, GO:2001240 negative regulation of extrinsic apoptotic signaling pathway in absence of ligand, GO:0030512 negative regulation of transforming growth factor beta receptor signaling pathway, and so on (Figure. 3A). CPs, such as GO:0005913 cell-cell adherens junction, GO:0000151 ubiquitin ligase complex, GO:0005739 mitochondrion,GO:0005829 cytosol,GO:0031012 extracellular matrix,GO:0016234 inclusion body,GO:0070062 extracellular exosome (Figure. 3B) were observed. MFs, such as GO:0031625 ubiquitin protein ligase binding,GO:0005524 ATP binding, GO:0019899 enzyme binding, GO:0016887 ATPase activity, GO:0098641 cadherin binding involved in cell-cell adhesion,GO:0001664 G-protein coupled receptor binding, GO:0044183 protein binding involved in protein folding (Figure. 3C) were detected. KEGG analysis such as, hsa04141:Protein processing in endoplasmic reticulum,hsa04612:Antigen processing and presentation, hsa04010:MAPK signaling pathway (Figure. 3D) were detected.
The PPI network was established with 13 nodes and 59 edges. Moreover, there was a strong co-expression relationship between the genes.
STRING (Figure. 4A) and GeneMANIA (Figure. 4B) were used in the construction of PPI and gene-gene interaction networks, respectively. The findings suggested that members of the HSP70 family interacted with each other in a complex manner. Besides, the co-expression of the HSP70 sub-members was also implied.