Differential expression of SNRP members in patients with HCC
Firstly, The genes of SNRP members were determined to be located on definite genomic sites [24, 25] (Table 1). We used Oncomine database [16] to analyze the transcriptional levels of SNRP members in various cancer types and corresponding normal tissues (Fig. 1). The result showed that there were a total of 26, 32, 22, 12, 39, 15 and 23 significant unique analyses for SNRPB, SNRPD1, SNRPD2, SNRPD3, SNRPE, SNRPF and SNRPG, respectively. In tumor tissues, SNRPB was significantly increased in tumor tissues, especially in breast, bladder, cervical, colorectal, gastric, head and neck, kidney, and liver cancers. Additionally, a dataset of SNRPB illustrated downregulation in breast cancer. Compared to normal tissues, SNRPD1, D2, E showed high mRNA levels in most tumor tissues but were downregulated in brain and CNS cancer, breast cancer and esophageal cancer, respectively. On the contrary, there were few cancer types in high levels of SNRPD3. For SNRPF, G datasets also showed increased expression in varying cancers.
Table 1
The chromosomal locations of SNRP members
SNRP familys | SNRPB | SNRPD1 | SNRPD2 | SNRPD3 | SNRPE | SNRPF | SNRPG |
Chromosomal location | 20p13 | 18q11.2 | 19q13.2 | 22q11.23 | 1q32 | 12q32.1 | 2p13.3 |
The results from Oncomine[16] described that the transcription levels of SNRPB, D1, D2 were significantly higher in HCC tissues in two datasets, respectively [24, 26]. Conversely, SNRPD3, F, G was no datasets to analysis the expression. For SNRPE mRNA levels, there was upregulated in three data analysis of TCGA database [24–26]. All p-values with the results for statistically significance were summarized in Fig. 1 and Table 2.
Table 2
Differential expression analyses of SNRP family in transcription level in hepatocellular carcinoma (ONCOMINE).
| Types of cancer vs. normal | Fold change | p-value | t-test | References |
SNRPB | Hepatocellular carcinoma vs. normal | 2.315 | 4.23E-75 | 22.843 | Roessler et al., 2010 |
| Hepatocellular carcinoma vs. normal | 2.338 | 1.12E-6 | 5.744 | Roessler et al., 2010 |
SNRPD1 | Hepatocellular carcinoma vs. normal | 3.270 | 2.55E-97 | 27.765 | Roessler et al., 2010 |
| Hepatocellular carcinoma vs. normal | 2.880 | 7.91E-9 | 7.287 | Roessler et al., 2010 |
SNRPD2 | Hepatocellular carcinoma vs. normal | 2.160 | 4.05E-82 | 24.017 | Roessler et al., 2010 |
| Hepatocellular carcinoma vs. normal | 2.052 | 5.00E-9 | 7.882 | Roessler et al., 2010 |
SNRPE | Hepatocellular carcinoma vs. normal | 2.971 | 1.81E-103 | 28.959 | Roessler et al., 2010 |
| Hepatocellular carcinoma vs. normal | 2.046 | 1.10E-25 | 12.246 | Chen et al., 2002 |
| Hepatocellular carcinoma vs. normal | 2.344 | 2.12E-7 | 6.554 | Roessler et al., 2010 |
We also compared the transcription levels of SNRPs between HCC and normal tissues by using UALCAN [17] (Fig. 2a-2g). We found that SNRPB, D1, D2, D3, E, F, G were all upregulated in tumor tissues. Besides, we analysed the correlative expression of SNRP members in HCC tissues and determined that SNRPD2 was the highest expression (Fig. 2h). Taken together, our results showed that transcriptional expressions of SNRPB, D1, D2, D3, E, F, G were overexpressed in patients with HCC.
Correlation between mRNA expression and tumor stages of SNRP members in HCC patients
We characterized the correlation between the mRNA expression of SNRP members and cancer stages of patients undergoing HCC using GEPIA [18]. SNRPB, D1, D2, D3, F, and G groups were obviously different in stage I, II, III, and IV, while no difference between SNRPE groups and tumor stages (Fig. 3). That is, the results above illustrated that mRNA expressions of SNRPB, D1, D2, D3, F, G were appreciably related to patients’ cancer stages in HCC, and patients in advanced cancer stages tended to express higher mRNA expression of SNRPB, D1, D2, D3, F, G.
Prognostic value of SNRP members in patients undergoing HCC
By online database analysis of the KM plotter [19], prognostic value of mRNA including OS, DFS, and PFS came under observation in HCC patients. It could be found that, patients were grouped into low (black) and high (red) risk based on respective cutoff value (Fig. 4 and Table 3). High mRNA levels of SNRPD1, E could get worse OS; however, there was no significant relations between OS and SNRPB, D2, D3, F, G in HCC patients. And high mRNA levels of SNRPB, D1 led to shorter DFS, and not statistical difference in groups of SNRPD2, D3, E, F, G. Furthermore, only increased SNRPD1 mRNA expression was related to PFS. In short, it was seen that, only mRNA expression of SNRPD1 in high risk was associated with prognosis with OS, DFS and PFS.
Table 3
The prognostic values of SNRP family members in liver and gastric cancer patients (Kaplan–Meier plotter) .
SNRP family | OS | | RFS | | PFS |
Cases | HR | 95%Cl | p | | Cases | HR | 95%Cl | p | | Cases | HR | 95%Cl | p |
SNRPB | 364 | 1.38 | 0.97–1.95 | 0.07 | | 316 | 1.52 | 1.09–2.12 | 0.013 | | 370 | 1.32 | 0.98–1.77 | 0.065 |
SNRPD1 | 364 | 1.45 | 1.02–2.05 | 0.036 | | 316 | 1.59 | 1.14–2.21 | 0.0057 | | 370 | 1.47 | 1.1–1.97 | 0.0096 |
SNRPD2 | 364 | 1.16 | 0.82–1.64 | 0.4 | | 316 | 1.38 | 0.99–1.92 | 0.053 | | 370 | 1.3 | 0.97–1.74 | 0.082 |
SNRPD3 | 364 | 1.04 | 0.74–1.47 | 0.81 | | 316 | 1.2 | 0.87–1.67 | 0.27 | | 370 | 1.13 | 0.84–1.51 | 0.41 |
SNRPE | 364 | 1.5 | 1.06–2.12 | 0.022 | | 316 | 1 | 0.72–1.39 | 0.98 | | 370 | 0.97 | 0.73–1.31 | 0.86 |
SNRPF | 364 | 1.14 | 0.81–1.61 | 0.46 | | 316 | 1.14 | 0.82–1.58 | 0.45 | | 370 | 1.08 | 0.8–1.44 | 0.63 |
SNRPG | 364 | 1.2 | 0.85–1.69 | 0.31 | | 316 | 1.35 | 0.97–1.88 | 0.073 | | 370 | 1.33 | 0.99–1.79 | 0.055 |
Bold values mean p < 0.05 |
Genetic alterations and correlations of SNRP members in HCC patients
We analyzed gene mutations and interactions of SNRP members by cBioPortal online tool [20] for HCC patients (INSERM Cancer Cell 2014 dataset, MSK Clin Cancer Res 2018 dataset, INSERM Nat Genet 2015 dataset, MSK PLOS One 2018, AMC Hepatology 2014 dataset, RIKEN Nat Genet 2012 dataset, TCGA Firehose Legacy dataset, TCGA PanCancer Atlas dataset). The results in liver hepatocellular carcinoma databases (TCGA PanCancer Atlas dataset) illustrated that the percentages of genetic alterations were 0.27% gene mutation (1/372), 0.53% deep deletion (2/372), 7.53% amplification (28/372) in SNRP members, respectively (Fig. 5a). The alteration frequency of SNRPs by using the AMC Hepatology 2014 dataset, TCGA Firehose Legacy dataset, TCGA PanCancer Atlas dataset were analyzed (SNRPB, 0.5%; SNRPD1, 0.3%; SNRPD2, 0.8%; SNRPD3, 0.3%; SNRPE, 5%; SNRPF, 0.2%; SNRPG, 0.2%) (Fig. 5b). Then, we presented the correlations between SNRPs genetic alterations and survival outcome for patients. Unfortunately, we found no correlations between patients with SNRPs genetic alterations and OS, DFS or PFS outcomes, respectively (p = 0.792, 0.0977, 0.662, Fig. 5c-5e). The reason why the SNRPs genetic alterations and prognosis seemed to be no significance might be a result of small sample size in varying researchs’ backgrounds and materials.
Interaction of correlated genes and proteins of SNRP members in HCC patients
We stated breifly the correlations among SNRP members at the gene level by GeneMANIA online tool [21] (Fig. 6a). The results demonstrated SNRP members were very close sharing of genetic thresholds. Markedly, relations were noticed in co-localization among SNRPD1, D3, E, F. Additionally, the relations were found among SNRP members in protein-protein interactions. We identified the correlations of SNRP members at the protein expression level using STRING [22] (Fig. 6b).
Functional enrichment analysis of SNRP members in HCC
To see the functions of SNRP members and their neighboring proteins, we used GO and KEGG pathways by Metascape [23]. The result indicated top 8 expand enrichment (Fig. 7a&7b), structutral complexes: Sm core complex; CORUM: SMN1-SIP1-SNRP complex, SMN comlex, U7 snRNA specific; cellular components: plCIn-Sm protein complex, telomerase holoenzyme complex; biological progresses: spliceosomal complex assembly; chemical and genetic perturbations: CHIAS RB1 TARGETS LOW SERUM; immunologic signatures. The top 5 GO enrichment (Fig. 7c&7d) were cellular components: methylosome, plCIn-Sm protein complex, U5 snRNP, U7 snRNP; molecular functions: U1 snRNP binding. The top 2 KEGG enrichment (Fig. 7e&7f) were structural complexes: Sm core complex; pathway: systemic lupus erythematosus. Subsequently, we analysed protein-protein interaction enrichment (Fig. 7g), and found that the protein-complex functions mostly involved in intron transcription splicing process.