Expression and prognosis analyses of CASP1 in acute myeloid leukemia

Transcriptional expression of CASP1 in leukemia

We performed a retrieval of CASP1 mRNA expression levels in a variety of cancers by using the GEPIA database. We discovered that mRNA expression levels of CASP1 in patients with leukemia were upregulated compared with normal samples (Figure 1A–1D).

Figure 1. CASP1 expression in leukemia (GEPIA). (A, B) CASP1 expression in multiple cancers. (C, D) CASP1 expression in AML.

Using the Oncomine dataset, we discovered that CASP1 ranked in the top 15% based on mRNA expression levels in the Coustain-Smith Leukemia and Haslinger Leukemia datasets (Figure 2A–2C). The DNA copy number levels of CASP1 were higher in leukemia than in normal blood samples (Figure 2D), while significance difference was not observed. To further analyze the correlation between CASP1 expression and clinicopathological features, we conducted a subgroup analysis by UALCAN database. As shown in Figure 3, an increased CASP1 was found in AML patients with M3/M4/ M5 subtype, Caucasian ethnicity (Figure 3 and Tables 1, 2, P<0.05). Thus, the expression of CASP1 may be a promising diagnostic indicator for leukemia.

Figure 2. CASP1 transcription in leukemia (Oncomine). CASP1 mRNA expression and DNA copy number were highly in leukemia than in normal blood samples. (A–C) CASP1 mRNA expression levels were shown by box plot in Constant-Smith leukemia, and Haslinger leukemia respectively. (D) CASP1 copy number was shown by box plot in The Cancer Genome Atlas (TCGA) Leukemia 2 datasets.

Figure 3. CASP1 transcription in the leukemia subgroups was stratified according to FAB classification, ethnicity, gender, age, FLT3 mutation status (UALCAN) and PML/RAR-fusion status. (A) CASP1 mRNA expression with different FAB classification was shown in leukemia. (B) CASP1 mRNA expression with different ethnicity was shown in leukemia. (C–F) Boxplot showed relative expression of CASP1 in LAML in gender, age, FLTS mutation status, PML/RAR-fusion, respectively. Using t-test to assess the differences in gene expression levels between groups.

CASP1 is upregulated in AML cell lines and patients

We inquired the CCLE database to analyse the CASP1 expression in multiple cell lines and discovered a higher transcriptional expression of CASP1 in leukemia cell lines compared to other cancer types (Figure 4A). We then performed quantitative real-time PCR assay to examine the CASP1 expression in several AML cell lines and peripheral blood mononuclear cells (PBMCs). As expected, the expression of CASP1 showed higher in most AML cell lines than in PBMCs (Figure 4B). Interestingly, we found that the CASP1 expression was dramatically higher in MLL-rearranged AML cell lines and cytarabine-resistant cell (Figure 4B and Supplementary Figure 1). Further analysis of AML patients showed that CASP1 expression was higher in initial relapse of AML (Sample 1) than in newly diagnosed of AML (Sample 2) (Figure 4C). In sum, the expression of CASP1 was consistent with that of database.

Figure 4. CASP1 expression in leukemia cell lines and AML patients. (A) CASP1 expression in multiple cancer cell lines (CCLE). (B) CASP1 expression in AML cell lines and PBMC. (C) CASP1 expression of 2 patients at initial relapse of AML (sample 1) and at newly diagnosed of AML (sample 2). *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Evaluating the prognostic value of CASP1 in AML

We evaluated the relationship between CASP1 expression and prognosis information in AML through the Kaplan-Meier survival curves (Figure 5). Patients were grouped into low-CASP1 and high-CASP1 expression group based on the median value of CASP1 expression. The survival analysis revealed that patients with elevated CASP1 expression had a shorter OS than those with low levels of CASP1 expression (Figure 5A, P < 0.05). Similarly, in LinkedOmics, increased CASP1 was associated with poor OS in AML (Figure 5B, P < 0.05). Therefore, these results indicated that CASP1 may be a factor for predicting the prognosis of AML patients.

Figure 5. The prognostic value of mRNA level of CASP1 in AML (GEPIA and LinkedOmics). (A) GEPIA showed the prognostic value of CASP1 mRNA expression levels in AML. (B) LinkedOmics showed the prognostic value of CASP1 mRNA expression levels in AML.

Since CASP1 is significantly overexpressed in AML and Cytarabine-resistant cells, and is associated with poor prognosis, we inhibited CASP1 with Caspase-1 inhibitor Belnacasan to investigate its role. We evaluated the effect of CASP1 on proliferation by using CCK-8 assay. We found that CASP1 inhibition reduced the IC50 and proliferation of THP1 and Cyt70-THP1 in a time-dependent manner (Figure 6A, 6B). And Cyt70-THP1 seemed to show a more pronounced decrease in IC50 than THP1 (Figure 6A, 6B). These suggested that CASP1 inhibition could impair the proliferation of AML cells.

Figure 6. Inhibition of CASP1 impairs the proliferation of THP1 and Cyt70-THP1. (A) CCK-8 assay was used to investigate the IC50 of THP1 and Cyt70-THP1 after CASP1 inhibition for 24 and 48h. (B) CCK-8 assay was used to investigate the proliferation of THP1 and Cyt70-THP1 after CASP1 inhibition for 24 and 48h. * p < 0.05, ** p < 0.01, ***p < 0.0005.

CASP1 co-expression networks in leukemia

We used the LinkedOmics function module to analyze co-expression genes of CASP1 for exploring its biological meaning in AML. Figure 7A showed that 2,600 and 4,666 genes were significantly positively and negatively correlated with CASP1, respectively (false discovery rate, FDR < 0.01). Furthermore, Figure 7B, 7C showed that top 50 genes were significantly negatively and positively correlated with CASP1, respectively. There was a strong positively correlation between CASP1 and CTSS (Pearson correlation = 0.83, P = 8.427E-47), TLR5 (Pearson correlation = 0.79, P = 3.134e-38), CARD16 (Pearson correlation = 0.78, P = 3.271E-37), which reflect inflammatory and immune response. The correction scatter diagrams of the top 3 significant genes correlated with CASP1 are shown in Supplementary Figure 2.

Figure 7. Genes differentially expressed in correlation with CASP1 in AML (LinkedOmics). (A) Using Pearson test to analyze genes differentially expressed in correlation with CASP1 in AML. (B, C) Heat maps showed that top 50 genes were significantly negatively and positively correlated with CASP1, respectively.

We found that these co-expression genes associated with CASP1 were involved primarily in phagocytosis, interleukin-1 production, lymphocyte mediated immunity, cellular defense response and leukocyte, while activities base excision repair, RNA 3'-end processing, spinal cord development, protein alkylation, DNA-templated transcription, termination, chromatin assembly or disassembly and ribonucleoprotein complex biogenesis were strongly inhibited through Gene Ontology (GO) term analysis (Figure 8A). In addition, we performed Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, confirmed that co-expression gene of CASP1 enrichment in leishmaniasis, lysosome, legionellosis, endocytosis, cytokine receptor interaction, NOD-like, and B cell receptor signaling pathway (Figure 8B). In summary, CASP1 has widespread impact on immune response in leukemia.

Figure 8. GO annotations and KEGG pathways enriched with CASP1 co-expressed genes in leukemia (LinkedOmics). Using GSEA to analyze the significantly enriched GO_BP and KEGG pathways of CASP1 co-expression genes in LAML. (A) Gene Ontology_Biological processes, analyzed by GSEA. (B) KEGG pathway, analyzed by GSEA. The LeadingEdgeNum and the false discovery rate (FDR) are shown in blue and orange column, respectively.

Kinase, miRNA and transcription factor of CASP1 networks in AML

Here we used LinkedOmics to further explore the regulators of CASP1 in AML, in which we analyzed the enrichment in kinases, miRNAs and transcription factors (TF) of CASP1 positively related genes. Lck/yes-related novel protein tyrosine kinase (LYN) and lymphocyte-specific protein tyrosine kinase (LCK) are significant kinases primarily associated with positive genes (Table 3). In addition, mutated and overexpressed LCK is driving leukemia cell proliferation [19]. LYN kinase was significantly associated with the OS of AML [20]. There are no significantly enriched miRNA for CASP1 coexpressed genes (Table 3 and Supplementary Table 1). The E2F transcription factor family, including V$E2F_01, V$E2F1_Q6_01, V$E2F_Q6_01, V$E2F1_Q4_01, V$E2F_Q3_01, V$E2F_03, V$E2F_Q4_01 and V$E2F_Q4_01, is significant TF correlated with positive genes (Table 3 and Supplementary Table 2). The transcription factor E2F family is identified as an oncogene or antioncogene. The properties of these regulators in AML may give us insight into the pathogenesis of leukemia [21]. It has become clear that E2F1 expression can regulate the prognosis of leukemia [22]. Pellicano et al. [23] shown that E2F1 plays a crucial regulatory role in the proliferation state of chronic myeloid leukemia (CML) stem/progenitor cells (SPC).

Cell culture

The AML cell lines THP1, MOLM13, HL60, U937 and NB4 were grown in RPMI-1640 medium (Gibco) containing 10% fetal bovine serum (FBS), 2 mM glutamine, 100μg/ml penicillin, and streptomycin (Sigma) at 37° C with 5% CO2. Cells were passaged every 2–3 days.

Cytarabine selection

Multiple step selections with gradually increasing Cytarabine concentrations (cat. 147-94-4, Huateng) was performed on THP1 cells for the establishment of drug-resistant sublines, using a starting dose of approximately its original IC30 (1uM, 72hours) values; the latter were obtained by growth inhibition assays as detailed below. THP1 cells were continuously grown in 1uM Cytarabine for 28 days until cells resumed their original doubling time, yielding a drug-resistant subline termed Cyt30-THP1 (THP1 Ara-C IC30 resistant); at this passage (day 28 from initiation of drug selection), Cyt30-THP1 cells were frozen down in aliquots and thawed for any experiment that required the original cells. Cyt30-THP1 cells were also transferred into 1.6μM (IC50) Cytarabine, resulting in the sunlines Cyt50-THP1. Cyt50-THP1 cells were also transferred to grow in either 2.7μM (IC70) Cytarabine, resulting in the sunlines Cyt70-THP1. Following their establishment, Cyt70-THP1 cells were also grown in drug-free medium to evaluate the stability of their drug resistance phenotype.

Patient samples

Two AML patients who presented to the Seventh Affiliated Hospital of Sun Yat-Sen University were included in this study. Diagnosis of AML was based on morphology, immunophenotype, cytogenetics and molecular analysis (MICM). In addition, one healthy volunteer in whom cancers were ruled out by a comprehensive medical examination were enrolled as normal controls. Mononuclear cells were isolated from peripheral whole blood of AML patients and volunteers using density gradient centrifugation and then dispersed in Trizol reagent (NucleoZol, Germany) and cells were stored at −80° C for further analysis. Informed consent was obtained from all the participants and the study was approved by the Ethics Committee of the Seventh Affiliated Hospital of Sun Yat-Sen University.

Growth inhibition assays

Belnacasan (VX-765, cat. T6090), was purchased from TOPSCIENCE. For the analysis of cell lines, THP1 cells were grown in RPMI-1640 medium. Cyt70-THP1 cells continuously growing in Cytarabine-containing medium were grown in drug-free medium for 3 days prior to experiments. THP1 and Cyt70-THP1 cells were seeded into a 96-well plate (1×10⁴ cells/well) and cultured in RPMI 1640 medium at 37° C for 24 h. Then, the cells were treated with various concentrations of Belnacasan for 48 h. Cells treated with DMSO instead of Belnacasan served as the control group. Finally, cell viability was detected using a CCK-8 kit according to the manufacturer’s instructions. The experiment was performed at least 3 times.

QRT-PCR

TRIzol reagent (NucleoZol, Germany) was used to extract total RNA from the AML cells. β-actin was used as an internal control. cDNAs were generated using the HiScript II Q RT reagent kit (Vazyme) in line with the manufacturer’s instructions. ChamQ Universal SYBR qPCR Master Mix (Vazyme) was used to analyze the synthesized cDNAs according to the manufacturer’s instructions. Primer sequences used for real-time PCR were:

CASP1: Forward: 5′GCCTGCCGTGGTGATAATGT3′;

Reverse: 3′TCACTCTTTCAGTGGTGGGC5′.

β-actin: Forward: 5′CCTGTACGCCAACACAGTGC3′;

Reverse: 3′ATACTCCTGCTTGCTGATCC5′.

Statistical analyses

All data are displayed as the mean ± SEM. GraphPad Prism 8 statistical software was used for statistical analyses. Comparisons between groups were performed using the t test or ANOVA. p<0.05 was considered as statistically significant.

Database descriptions

GEPIA database analysis

GEPIA (Gene Expression Profiling Interactive Analysis) is an online database including 8,587 normal samples and 9,736 tumors from TCGA and Genotype-Tissue Expression (GTEx) projects [31]. We performed differential gene expression analysis and survival analysis between tumors with normal samples. It can also be used to perform similar gene detection, correlation analysis, and Principal Component Analysis [31]. The p value was used to assess the significance of the difference, and p < 0.05 was considered to be significant.

Oncomine database analysis

Oncomine (https://www.oncomine.org/) is an online cancer sample database that was used to analyze CASP1 mRNA expression and DNA copy number. This analysis drew on a series of leukemia studies, including Coustan-Smith Leukemia, Haslinger Leukemia, and TCGA Leukemia 2 [32, 33]. We performed differential gene expression and DNA copy number analysis between tumors with normal samples. The p value and fold change were used to assess the significance of the difference, and p < 0.01 and fold change >1.5 was significant.

UALCAN database analysis

UALCAN (http://ualcan.path.uab.edu) is an interactive web-portal including clinical data from 31 cancer types and TCGA level 3 RNA-seq [34]. It can be used to analyze differential gene expression and multiple clinicopathological features between tumors with normal samples. We conducted a subgroup analysis of various clinic-pathological characteristics of AML patients samples from the UALCAN.

CCLE database analysis

The CCLE is a project including of gene expression, chromosomal copy number, and DNA mutations from 947 human cancer cell lines covering more than 30 tissues [35]. This database allows us to explore the expression of CASP1 in different cancer types. CCLE is publicly available at https://www.broadinstitute.org/ccle.

LinkedOmics database analysis

The LinkedOmics (http://www.linkedomics.org/login.php) is an interactive web-portal including 32 TCGA cancer-associated data [36]. Using Pearson test to analyze genes differentially expressed in correlation with CASP1 in AML, shown in heat maps. We used the LinkedOmics function module to analyze co-expression genes of CASP1 for exploring its biological meaning in leukemia. Using GSEA, we performed analysis of GO_BP, KEGG pathways and target enrichment of kinase, miRNA as well as transcription factor. The threshold was determined based on the following values: FDR of 0.05, simulations of 500. In addition, we evaluated its prognosis value in leukemia by clinical analysis. The p value was used to assess the significance of the difference, and p < 0.05 was considered to be significant.

Database statistical analysis

We performed differential gene expression and DNA copy number analysis between tumors with normal samples by using the t-test. The p value and fold change were used to assess the statistical significance of the difference, and p < 0.01 and fold change >1.5 was considered to be significant. We also used t-test to conduct a subgroup analysis of various clinic-pathological characteristics of AML patients’ samples from the UALCAN, and p < 0.05 indicates the significance of difference. We used Kaplan-Meier curves to compare the prognosis of AML patients with different median expression levels of CASP1. The log-rank test p < 0.05 indicates that the prognosis analysis is statistically significant. We calculated the correlation between CASP1 and CTSS, TLR5, and CARD16 gene expression levels using the Pearson correlation coefficient. The threshold of correlation was determined based on the following values: p of 0.05.