Immunological function and prognostic value of lymphoid-specific helicase in liver hepatocellular carcinoma

Abstract

BACKGROUND:

Lymphoid-specific helicase (HELLS), a SNF2-like chromatin-remodeling enzyme, plays a key role in tumor progression via its DNA methylation function. However, the effects of HELLS on immune infiltration and prognosis in liver hepatocellular carcinoma (LIHC) remain uncertain.

METHODS:

The Tumor Immune Estimation Resource (TIMER) database was employed to explore the pan-cancer mRNA expression of HELLS and its correlation with immunity. GEPIA2 was used to verify the correlation between HELLS expression and survival. The role of HELLS in cancer was explored via gene set enrichment analysis (Gene Ontology and Kyoto Encyclopedia of Genes and Genomes) and the construction of gene-gene and protein-protein interaction networks (PPI). Additionally, correlations between DNA methylation, HELLS expression, and immune-related genes were explored in LIHC. HELLS expression in LIHC clinical samples was determined using qRT-PCR and western blotting. The effects of downregulated HELLS expression in hepatocellular carcinoma cells was explored via transfection experiments in vitro.

RESULTS:

High HELLS mRNA expression was identified in several cancers and was significantly associated with poorer prognosis in LIHC. Furthermore, HELLS expression was positively correlated with tumor-infiltrating lymphocytes and immune checkpoint genes in LIHC. Bioinformatics analysis suggested that DNA methylation of HELLS may be associated with the immune response. Results from the TCGA-LIHC dataset, clinical samples, and functional analysis indicated that HELLS contributed to tumor progression in LIHC.

CONCLUSION:

The study findings demonstrate that HELLS is an important factor in promoting LIHC malignancy and might serve as a potential biomarker for LIHC.

Keywords

HELLS immune tumor-infiltrating prognosis DNA methylation liver hepatocellular carcinoma

1. Introduction

Cancer incidence and mortality are booming worldwide [1]. The incidence of gastrointestinal and hepatic cancers continues to increase rapidly, which has also contributed to the continuing increase in cancer-related deaths [2]. Liver hepatocellular carcinoma (LIHC), the most common subtype of hepatic cancer, ranked third in tumor-related mortality worldwide. Although some patients with early LIHC can be cured surgically, approximately 70% of patients relapse after five years [3]. Meanwhile, increasing evidence indicates that immunotherapy is a potential prospect for LIHC treatment [4].

Lymphoid-specific helicase (HELLS, also known as LSH) is a 97 kDa transmembrane protein found in many cell types and organs. HELLS is important for normal embryonic development because it interacts with DNA methyltransferases [5]. Strikingly, accumulating evidence has identified HELLS as an oncogenic molecule that promotes tumorigenicity. In addition, scattered evidence has also suggested that HELLS is needed for malignant progression and stem cell maintenance [6, 7]. More significantly, the prognosis of patients is not optimistic for those with upregulated HELLS expression. Nevertheless, the underlying mechanism of HELLS in tumor progression remains unclear.

DNA methylation, chromatin structure regulation, and histone post-translational modifications, which are epigenetic mechanisms, are critical for interactions between tumor and immune cells [8]. Immune-related mechanisms are key points in hepatic cancers, while at the same time, immunotherapeutic strategies are regarded as a promising direction for the treatment of hepatic cancers [9]. Cancer immunotherapy has attracted significant attention, and treating LIHC with immunotherapy is a tantalizing prospect. DNA methylation and gene expression have potential complementarity in the prediction of immune responses [10]. Therefore, the potential correlation between immunity and DNA methylation in LIHC caught our attention.

In this study, we assessed the expression signature and survival value of HELLS gene in pan-cancer by Tumor Immunity Estimation Resource (TIMER) and TCGA database. we evaluated the potential correlation between HELLS and tumor-infiltrating lymphocytes in LIHC. In addition, the correlation of DNA methylation and HELLS in LIHC were also evaluated by bioinformatics. Finally, we verified the HELLS expression in LIHC by clinical LIHC samples. According to these results, it is suggested that HELLS could make a difference in prognosis of LIHC by affecting tumor immunity, and on this basis, HELLS could be a new promising biomarker for LIHC immunotherapy.

In the current study, we hypothesized that HELLS may be an important biomarker in LIHC and contribute to tumor immunity. Thus, the aims of the study were to: 1) assess the pan-cancer expression signature and survival value of HELLS; 2) evaluate the potential correlation between HELLS expression and tumor-infiltrating lymphocytes in LIHC; 3) explore the correlation between DNA methylation and HELLS expression in LIHC; 4) verify HELLS expression in clinical LIHC samples. The study findings may provide a promising new promising biomarker for LIHC immunotherapy.

2. Methods

2.1 TIMER database analysis

TIMER (https://cistrome.shinyapps.io/timer/, Version1.0) is an online database for the comprehensive investigation of tumor-infiltrating immune cells [11]. IMER was used to explore pan-cancer HELLS expression and investigate the correlation between HELLS expression and tumor infiltrating immune cells, including B cells, CD8 $+$ T cells, CD4 $+$ T cells, macrophages, neutrophils, and dendritic cells in cancers. Additionally, the interdependency of HELLS expression and immune checkpoint (ICP) genes was investigated.

2.2 GEPIA2 analysis

Survival analysis was performed using Gene Expression Profiling Interactive Analysis (GEPIA2) (http://gepia2.cancer-pku.cn/#index), a database that contains RNA sequencing data from 9k tumor samples and 8.5k normal tissue samples from The Cancer Genome Atlas (TCGA) database and the Genotype Tissue Expression (GTEx) project [12]. Furthermore, correlations were examined between HELLS and ICP genes, including programmed death-ligand 1 (PD-L1), cytotoxic T-lymphocyte-associated protein 4 (CTLA4), and programmed cell death protein-1 (PD-1). DNA methylation levels of HELLS and related genes were evaluated using GEPIA2 [12].

2.3 LinkedOmics database analysis

Gene expression profiles were investigated using the LinkedOmics platform (http://www.linkedomics.org/login.php), which contains data for 32 cancer types from the TCGA database [13]. The co-expression of HELLS and ICP genes was determined via Pearson’s correlation analysis and exhibited using heat maps. Gene Ontology (GO) [14]. function analysis and KEGG pathway [15] analysis for HELLS and correlated ICP genes were performed using gene set enrichment analysis (GSEA).

2.4 Gene interaction analysis

The GeneMania database (https://genemania.org/) was used to search for interactions, co-expression, and co-localization of related genes with HELLS [16]. The String (https://cn.string-db.org/) database was used to construct protein-protein interaction (PPI) networks and conduct functional enrichment analysis for HELLS and related genes [17].

2.5 DNA methylation-related database analysis

The Molecular Signatures Database (MSigDB) (http: //www.gsea-msigdb.org/gsea/msigdb) contains defined sets of known genes [18]. DNA methylation-related GO function sets for HELLS were obtained using MSigDB. Moreover, promoter methylation levels for HELLS were obtained using the UALCAN portal (http://ualcan.path.uab.edu/index.html) [19]. The cBioPortal database (http://www.cbioportal.org/) was used to search for correlations between DNA methylation and mRNA expression [20].

Table 1
The clinical characteristics of patients in the TCGA dataset

Characteristic	Low expression of HELLS	High expression of HELLS	P
$n$	187	187
T stage, $n$ (%)			0.032
T1	104 (28%)	79 (21.3%)
T2	42 (11.3%)	53 (14.3%)
T3	31 (8.4%)	49 (13.2%)
T4	7 (1.9%)	6 (1.6%)
Pathologic stage, $n$ (%)			0.013
Stage I	97 (27.7%)	76 (21.7%)
Stage II	39 (11.1%)	48 (13.7%)
Stage III	32 (9.1%)	53 (15.1%)
Stage IV	4 (1.1%)	1 (0.3%)
Tumor status, $n$ (%)			$<$ 0.001
Tumor free	119 (33.5%)	83 (23.4%)
With tumor	60 (16.9%)	93 (26.2%)
AFP (ng/ml), $n$ (%)			$<$ 0.001
$<=$ 400	129 (46.1%)	86 (30.7%)
$>$ 400	15 (5.4%)	50 (17.9%)
Age, $n$ (%)			0.011
$<=$ 60	76 (20.4%)	101 (27.1%)
$>$ 60	111 (29.8%)	85 (22.8%)
Age, median (IQR)	64 (55, 69)	59 (51, 68)	0.005
Histologic grade, $n$ (%)			$<$ 0.001
G1	40 (10.8%)	15 (4.1%)
G2	102 (27.6%)	76 (20.6%)
G3	38 (10.3%)	86 (23.3%)
G4	4 (1.1%)	8 (2.2%)

2.6 TCGA database analysis

HELLS expression and clinical data were obtained from the TCGA-LIHC dataset, including 374 tumor samples and 50 normal tissue samples. Cancerous subtype in the TCGA-LIHC dataset was defined using non-negative matrix factorization (NMF) with the “cancertypes” packages in R software (version 3.63). Table 1 lists the relevant information of clinical samples obtained from the TCGA-LIHC dataset.

2.7 Sample collection

To further verify HELLS expression in LIHC, samples of tumors and matched normal tissues were obtained from 8 patients with primary hepatocellular carcinoma (HCC). The samples were collected at the First Affiliated Hospital of Kunming Medical University. Signed informed consent was provided by patients or their family members. This study was approved by the Ethics Committee of the First Affiliated Hospital of Kunming Medical University.

2.8 qRT-PCR

qRT-PCR was performed as previously described[21]. gDNA eraser and cDNA was synthesized using reverse transcription with the PrimeScript RT Reagent Kit (TaKaRa, Kusatsu, Japan). PCR was performed using the following specific primers: HELLS-Forward 5 ${}^{\prime}$ -TAGAGAGTCGACAGAAATTCGG-3 ${}^{\prime}$ , HELLS-Reverse 5 ${}^{\prime}$ -CCTCATAACTGGCTTCTCTTCA-3 ${}^{\prime}$ ; GAPDH-Forward 5 ${}^{\prime}$ -CAGGAGGCATTGCTGATGAT-3 ${}^{\prime}$ , GAPDH-Reverse 5 ${}^{\prime}$ -GAAGGCTGGGGCTCATTT-3 ${}^{\prime}$ . The qRT-PCR was performed by using FastStart Universal SYBR Green Master (Roche Diagnostics, Mannheim, Germany) in a RochLightCycle 480II PCR. Relative mRNA expression levels of genes were analyzed by the 2 ( $-\Delta\Delta$ Ct).

2.9 Western blotting

Western blotting was performed as previously described [22]. HELLS rabbit polyclonal antibody (11955-1-AP) and GAPDH (10494-1-AP) were purchased from Proteintech (Wuhan, China). The membranes were detected by using AI600 (GE Healthcare Life Sciences, America) and densitometric analyses by ImageJ software.

2.10 Cell culture and transfection

The LIHC line SKHep-1, Huh7, SNU-387 were obtained from Pricella (Wuhan, China). Dulbecco’s Modified Eagle’s medium (Gibco, USA) containing 10% fetal bovine serum and 1% penicillin streptomycin (Gibco) was used to culture the cells. The cells were incubated at 37 ${}^{\circ}$ C under an atmosphere containing 5% CO2.

Short hairpin RNA (shRNA) was constructed to regulate HELLS expression. The shRNAs targeting HELLS were designed by Shanghai Genechem (Shanghai, China) as follows: shRNA-1:5 ${}^{\prime}$ -gaAGTGAATATCCCT GTAGAA-3, shRNA-2: 5 ${}^{\prime}$ -gtTGTTGTTTATCGCCTTGTT-3 ${}^{\prime}$ and shRNA-3: 5 ${}^{\prime}$ - tcCAGTGAGAAAGAAACAATT-3 ${}^{\prime}$ . The shRNAs were transfected into SKHep-1 cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s instructions.

2.11 Cell migration

For the cell migration studies, SKHep-1 cells (5 $\times$ 10 ${}^{4}$ cells/insert) were suspended in serum-free DMEM and plated in the top chamber of a non-coated culture insert (Corning Inc., Corning, NY, USA). The lower chamber contained medium with 10% FBS. After incubation for 16 h, the cells on the lower surface of the membrane were stained with crystal violet dye. The migration ability of the cells was quantitated by counting the number of crystal violet-stained cells. Stained cells were observed with BX43 of Olympus (Tokyo, Japan).

2.12 Wound healing assay

SKHep-1 cells were cultured to 90% confluence in 6-well plates. A sterile 200 $\mu$ L pipette tip was used to scratch the cell layer. After washing three times with phosphate-buffered saline, serum-free medium was added. An inverted microscope (BX43, Olympus, Tokyo, Japan) was used to observe the migration of cells across the wound at 0, 24, and 48 h.

2.13 CCK-8 assay

SKHep-1 cells (5 $\times$ 10 ${}^{3}$ cells/well) were cultured in 96-well plates and cell proliferation was measured using the Cell Counting Kit-8 (CCK-8) assay (Mei5, China). The absorbance of microscope (BX53, Olympus, Tokyo, Japan) at 450 nm was measured at 0, 12, 24, 36, 48, 60, and 72 h.

2.14 EDU assay

SKHep-1 cells were cultured in 24-well plates and cell proliferation was measured using the Click-iT ${}^{\text{TM}}$ EdU cell proliferation kit (Beyotime Biotech, Jiangsu, China). The plates examined under a fluorescence microscope (BX53, Olympus, Tokyo, Japan).

2.15 Statistical analysis

GraphPad Prism (Ver. 8.0.1) was performed to statistical analysis and presented as mean $\pm$ SD. For the normally distributed variables, the $P$ -values between groups was assessed using the student’s $t$ test. The results are presented as $P$ -values based on log-rank tests and hazard ratio (HR) values. Spearman’s relation coefficients and $P$ -values were used to evaluate gene correlations. $P$ -values $<$ 0.05 were considered to be statistically significant.

3. Results

3.1 Pan-cancer HELLS expression and prognosis analysis

To investigate differential HELLS expression in multiple types of cancer, HELLS mRNA levels were analysed in tumor and adjacent normal tissues of different cancer types in the TIMER database (Fig. 1A). The results indicated that HELLS mRNA expression differed significantly between tumor and adjacent normal tissues in 19 cancer types. Additionally, HELLS mRNA expression was inconsistently higher or lower in diverse cancer types. For example, HELLS mRNA levels were significantly higher in bladder urothelial carcinoma (BLAC), breast invasive carcinoma (BRCA), cholangial carcinoma (CHOL), colon adenocarcinoma (COAD), esophageal carcinoma (ESCA), head and neck squamous cell carcinoma (HNSC), kidney chromophobe (KICH), LIHC, lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), prostate adenocarcinoma (PRAD), rectum adenocarcinoma (READ), stomach adenocarcinoma (STAD), thyroid carcinoma (THCA), and uterine corpus endometrial carcinoma (UCEC). In contrast, HELLS mRNA levels were significantly lower in kidney renal clear cell carcinoma (KIRC) and skin cutaneous melanoma (SKCM).

Next, GEPIA2 was employed to obtain an in-depth understanding of the correlation between HELLS expression and prognosis in cancers which all Fig. 1A types. (Fig. 1B and C) Higher HELLS expression was correlated with worse overall survival (OS) ( $n=$ 4746, HR $=$ 1.5, $P<$ 0.001) and disease-free survival (DFS) ( $n=$ 4746, HR $=$ 1.2, $P<$ 0.001), even over 200 months. However, higher HELLS expression was correlated with better DFS in some cancers. Generally, upregulated HELLS expression led to a pessimistic prognosis for patients with cancer.

Figure 1.

HELLS mRNA expression and prognostic significance in multiple types of cancers. A. HELLS mRNA levels in different tumor types from the TIMER database. ${}^{***}P<$ 0.001. B. Correlation between HELLS expression and overall survival for multiple types of cancers in the TCGA. C. Correlation between HELLS expression and disease-free survival for multiple types of cancers in the TCGA dataset. D. Survival analysis of gastrointestinal and hepatic cancers with high and low HELLS mRNA levels analysed via GEPIA2.

Notably, HELLS expression was significantly higher in gastrointestinal and hepatic cancers in the TIMER database. Thus, RNA sequencing data from the TCGA dataset were analyzed to determine the prognostic potential of HELLS in gastrointestinal and hepatic cancers via GEPIA2. Surprisingly, higher HELLS expression was correlated with significantly poorer OS (HR $=$ 1.5, $P<$ 0.001) and DFS (HR $=$ 1.7, $P<$ 0.001) in LIHC, which compared with gastrointestinal cancers (Fig. 1D). These results suggested that HELLS expression influences the prognosis of LIHC.

3.2 Correlations between HELLS expression, tumor-infiltrating lymphocytes, and ICP genes in cancer

Tumor-infiltrating lymphocytes are known to be closely related to tumor grade, TNM stage, and prognosis, especially in gastrointestinal and hepatic cancers [23, 24]. Thus, the correlation between HELLS expression and immune cell infiltration in gastrointestinal and hepatic cancers was investigated in the TIMER database (Fig. 2A). The results showed that high HELLS expression was highly positively correlated with immune cell infiltration in gastrointestinal and hepatic cancers. HELLS expression was positively correlated with levels of infiltrating CD8 $+$ T cells in COAD ( $r=$ 0.342, $P<$ 0.001) and in READ ( $r=$ 0.347, $P<$ 0.001). Notably, HELLS expression was significantly and positively correlated with levels of all infiltrating lymphocytes, including B cells ( $r=$ 0.462, $P<$ 0.001), CD8 $+$ T cells ( $r=$ 0.351, $P<$ 0.001), CD4 $+$ T cells ( $r=$ 0.332, $P<$ 0.001), macrophages ( $r=$ 0.447, $P<$ 0.001), neutrophils ( $r=$ 0.375, $P<$ 0.001), and dendritic cells ( $r=$ 0.484, $P<$ 0.001) in LIHC.

Next, correlations between HELLS expression and ICP genes among gastrointestinal and hepatic cancers were explored (Supplementary Fig. 1). Among 72 ICP genes, strong correlations with HELLS expression were identified in LIHC. HELLS expression was positively correlated with 59 of 72 ICP genes (82%) in the non-purity group (Supplemental Table 1). Moreover, in LIHC, 66 of 72 ICP genes (92%) were positively correlated with HELLS expression in the purity group. However, HELLS expression was weakly correlated with ICP genes in COAD, CHOL, READ, and STAD.

Subsequently, correlations between HELLS expression and PD-L1, CTLA4, and PD-1 were explored in gastrointestinal and hepatic cancers using ENCORI (Fig. 2B). HELLS expression was positively correlated with PD-L1 ( $r=$ 0.176, $P<$ 0.001), CTLA4 ( $r=$ 0.315, $P<$ 0.001), and PD-1 ( $r=$ 0.312, $P<$ 0.01) in LIHC.

Figure 2.

Correlations between HELLS expression, tumor-infiltrating lymphocytes, and ICP genes in gastrointestinal and hepatic cancers by TIMER. A. Correlations between HELLS expression and different tumor-infiltrating lymphocytes. B. Correlations between HELLS expression and ICP genes.

Finally, correlations between HELLS expression, tumor-infiltrating lymphocytes, and ICP genes in gastrointestinal and hepatic cancers were analysed. HELLS upregulation was correlated with increased immune-cell infiltration, suggesting that HELLS expression could have a relationship with the tumor immune microenvironment (TME), especially in LIHC. Thus, these results further confirmed that HELLS expression was correlated with infiltrating immune cells in LIHC.

Figure 3.

Correlations between HELLS expression and genes in LIHC analysed using the LinkedOmics database. A. Top 50 genes positively correlated genes with HELLS in LIHC. B. Top 50 genes negatively correlated with HELLS in LIHC. C. Top 20 GO molecular function terms associated with HELLS-related genes in LIHC. D. KEGG pathway enrichment analysis of HELLS-related genes in LIHC. E. Gene-gene interaction network for HELLS in LIHC. F. Protein-protein interaction network for HELLS in LIHC. G. GO enrichment analysis of HELLS-interacting genes in LIHC.

3.3 HELLS function and signaling pathways in LIHC

GSEA was performed to help elucidate the molecular mechanisms affected by HELLS in LIHC using data from the TCGA. Heat maps were used to illustrate the top 50 co-expressed genes that were positively and negatively correlated with HELLS in LIHC (Fig. 3A, B). Genes highly correlated with HELLS were determined by Pearson’s correlation analysis in the LIHC cohort (Supplementary Fig. 2A). Next, GO and KEGG enrichment analyses were used to explore the functions and signaling pathways related to HELLS in LIHC.

Hepcidin-related pathways and biological functions were analysed for 300 genes that were positively correlated with HELLS via KEGG and GO enrichment analyses. The top 20 significant GO biological process (BP) terms (Supplementary Fig. 2B), cell component (CC) terms (Supplementary Fig. 2C), and molecular function (MF) terms (Fig. 3C), as well as KEGG enriched pathways (Fig. 3D), were presented. Notably, in terms of GO MFs, co-expressed genes were enriched in DNA-methylation related functions such as histone binding, methyl-CpG binding, etc. In terms of KEGG pathways, co-expressed genes were enriched in pathways associated with the cell cycle, DNA replication, and cancer-related pathways. These results indicated that HELLS expression might be involved in regulating the immune response in LIHC via DNA-related functions.

The GeneMania database was searched to construct the gene-gene interaction network for HELLS. The network indicated that the 20 most frequently alteredgenes, including DNA methyltransferase 3A (DNMT3A) and DNA methyltransferase 3B (DNMT3B), were closely correlated with HELLS expression (Fig. 3E). These genes were chosen for GO enrichment analysis (Fig. 3G), which identified many DNA methylation-related GO terms, such as DNA methylation and DNA methylation activity. Subsequently, a PPI network was constructed for HELLS using the STRING database (Fig. 3F). The PPI network included 11 nodes and 33 edges, which contained DNA-methyltransferase 1 (DNMT1) and DNMT3B. The results suggested that DNA methylation may play a key role of HELLS in LIHC.

3.4 Correlation between HELLS expression and DNA methylation

Tumor-infiltrating lymphocytes and DNA methylation are regarded as key influencing factors of tumor progression [25]. MSigDB was used to investigate the correlation between HELLS expression and DNA methylation. GO BP terms, including maintenance of DNA methylation (GO-0010216) and DNA methylation-dependent heterochromatin assembly (GO-0006346) were identified, and gene correlation analysis was performed via GEPIA2 (Fig. 4A, B). CTCF, DNMT1, UHRF1, UHRF2, USP7, and ZNF445 were positively correlated with HELLS expression in GO-0010216. Similarly, AICDA, APOBEC1, ATF7IP, DNMT3A, HDAC1, MBD2, MBD3, MORC2, MPHOSPH8, PPHLN1, SETDB1, SETDB2, SIRT1, TET1, TRIM28, and ZNF304 were positively associated with HELLS expression in GO-0006346. HELLS might be closely related to DNA methylation in HCC and might affect the occurrence and development of HCC through DNA methylation. At present, DNA methylation markers are widely associated with differential cancer survival. Thus, promoter methylation expression was analysed using UALCAN ( $P<$ 0.05). In the normal group, HELLS promoter methylation was higher than that in the primary tumor group (Fig. 4C). Further, the correlation between HELLS methylation and mRNA expression was analyzed using cBioPortal. Receiver operating curve (ROC) analysis of HELLS expression in LIHC was conducted based on the TCGA dataset (AUC $=$ 0.942, CI: 0.913–0.971, $P<$ 0.001) (Fig. 4D).

Table 2
The information of clinical samples ( $n=$ 8)

Characteristic	HELLS
	Low expression ( $n=$ 4)	High expression ( $n=$ 4)
Age, median	45	63.5
Gender, male ( $n$ )	2 (50%)	4 (100%)
HBV (positive)	1 (25%)	3 (75%)
AFP ( $>$ 400)	2 (50%)	4 (100%)
Liver cirrhosis	0	4 (100%)
Outcome
Cure	1 (25%)	0
Improve	3 (75%)	2 (50%)
Death		2 (50%)
T
T1, T2	3 (75%)	0
T3, T4	1 (25%)	4 (100%)
M
M0	4 (100%)	3 (75%)
M1		1 (25%)
Stage
I/II	2 (50%)	3 (75%)
III/IV	2 (50%)	1 (25%)

TNM is the most widely used tumor clinical staging system, with T representing the size of the tumor. Tumor grade assesses the pathologic degree of the tumor tissue, including the number of nuclear divisions, arrangement, degree of differentiation, and local infiltration of cancer cells. Tumor status is used to describe whether the tumor has spread. The expression of HELLS in LIHC at different TNM stages, pathologic stages, tumor status, alpha-fetoprotein (AFP) levels, and age was analysed in the TCGA dataset using R software (Table 1). Compared with normal tissues, HELLS expression was significantly higher in LIHC tissues (Fig. 4E). HELLS expression was significantly higher in stages T1 $+$ T2 than in stages T3 $+$ T4 (Fig. 4F). Likewise, pathologic degree and tumor status were associated with higher HELLS expression in stages III $+$ IV and with tumor-free status (Fig. 4G, H). Interestingly, AFP, a biomarker for evaluating the prognosis of patients with LIHC, was potentially correlated with HELLS (Fig. 4I). Furthermore, HELLS expression was higher in patients with LIHC under age 60 (Fig. 4J). Finally, HELLS had a highly predictive power for predicting cancer and normal outcomes (AUC $=$ 0.942, CI $=$ 0.913–0.971). For normal outcomes, HELLS expression was significantly lower at Normal (Fig. 4K).

Figure 4.

Analysis of HELLS DNA methylation and expression in LIHC. A. Co-expression analysis of the GO term maintenance of DNA methylation (GO-0010216) in LIHC by MSigDB. B. Co-expression analysis of the GO term DNA methylation-dependent heterochromatin assembly (GO-0006346) in LIHC by MSigDB. C. Promoter methylation levels of HELLS in LIHC by UALCAN. D. ROC analysis of HELLS expression in LIHC based on the TCGA dataset. E. Correlation between HELLS expression in normal and LIHC samples based on the TCGA dataset. F. Correlation between HELLS expression and tumor stage (T) in LIHC based on the TCGA dataset. G. Correlation between HELLS expression and pathologic stage of patients with LIHC based on the TCGA dataset. H. Correlation between HELLS expression and tumor status in LIHC based on the TCGA dataset. I. Correlation between HELLS expression and alpha-fetoprotein (AFP) level in LIHC based on the TCGA dataset. J. Correlation between HELLS expression and age in LIHC based on the TCGA dataset. K. Correlation between HELLS expression and number of affected lymph nodes (N) in LIHC based on the TCGA dataset.

Figure 5.

Downregulated of HELLS expression suppressed LIHC growth and migration ( ${}^{*}P<$ 0.05, ${}^{**}P<$ 0.01, ${}^{***}P<$ 0.001, ${}^{****}P<$ 0.0001). A, B. Relative protein expression levels of HELLS in clinical samples. The histogram shows the densitometric analysis of the bands. C. Relative mRNA expression levels of HELLS in clinical samples. D. Relative mRNA expression levels of HELLS in LO-2, Huh7, SKHep-1, and SNU387 cells. E, F. Relative protein expression levels of HELLS in LO-2, Huh7, SKHep-1, and SNU387 cells. The histogram shows the densitometric analysis of the bands. G, H. Relative protein expression levels of HELLS in SKHep-1 cells transfected with NC, shRNA-1, shRNA-2, and shRNA-3. The histogram shows the densitometric analysis of the bands. I. CCK8 assay results for SKHep-1 cells transfected with NC, shRNA-1, and shRNA-2. J. EdU assay results for SKHep-1 cells transfected with NC, shRNA-1, and shRNA-2 (magnification $\times$ 200). K. Wound healing assay results for SKHep-1 cells transfected with NC, shRNA-1, and shRNA-2 (magnification $\times$ 100). L. Transwell migration assay results for SKHep-1 cells transfected with NC, shRNA-1, and shRNA-2 (magnification $\times$ 200).

3.5 Validation of HELLS expression in clinical samples

To validate HELLS expression in clinical samples, HELLS protein and mRNA levels were assessed in matched tumor and normal tissues of an HCC cohort ( $n=$ 8) via western blotting and qRT-PCR (Fig. 5A–C). In clinical samples, HELLS expression was higher in tumor tissues both protein and mRNA levels ( $P<$ 0.05).

3.6 Effect of HELLS knockdown on LIHC malignant behaviors

According to the results, HELLS expression was higher in tumor tissues than in matched normal tissues. Additionally, HELLS mRNA and protein levels were increased in LO2, Huh7, SKHep-1, and SNU-387 tumor cells. Compared with LO-2, the level of mRNA and protein differences of Huh7, SKHep-1, and SNU-387 were statistically significant ( $P<$ 0.05). Since the protein and mRNA levels of SKHep-1 are neither the highest nor the lowest, this can avoid bias in the experiment, SKHep-1 cells were selected for shRNA transfection (Fig. 5D–F). Transfection of shRNA-1, -2, and -3 into SKHep-1 cells revealed that shRNA-1 and -2 significantly knocked down HELLS protein expression ( $P<$ 0.001) (Fig. 5G, H). The results of the CCK-8 and EDU assays demonstrated that downregulation of HELLS expression significantly suppressed the proliferation of SKHep-1 cells ( $P<$ 0.05) (Fig. 5I, J). Meanwhile, the results of the wound-healing and transwell migration assays showed that downregulation of HELLS expression significantly decreased the migration abilities of SKHep-1 cells ( $P<$ 0.05) (Fig. 5K, L). Transfection with shRNA-1 and -2 significantly suppressed the wound-healing (approximately 93% and 50%, respectively) and transwell migration abilities (approximately 43% and 37%, respectively) of SKHep-1 cells ( $P<$ 0.05).

4. Discussion

HELLS, a DNA helicase related to the SWI/SNF2 family, affects DNA methylation and chromatin packaging. HELLS has been shown to promote tumor development and progression in various cancers, as well as the proliferation and invasion of cancer cells [26]. In spite of the key roles played by HELLS in cancer, HELLS has not been sufficiently studied in the context of immunity and oncology. Therefore, this study explored the pan-cancer expression of HELLS and revealed its potential immunological role in gastrointestinal and hepatic cancers via bioinformatics analysis.

The initial investigation revealed that HELLS expression was higher in different malignancies than in normal tissues and was positively correlated with lower survival rates in patients. The high HELLS expression found in various types of tumors was consistent with the results reported in other recent studies [27, 28]. Notably, HELLS expression was correlated with the poorest prognosis in LIHC, which was comparable with gastrointestinal tumors. These results indicated that HELLS promotes tumor progression and development in malignancies, and may provide a potential prognostic biomarker in LIHC.

The immune system has great significance in the monitoring and clearing of tumor cells [29]. Therefore, pan-cancer HELLS expression was investigated in various immune-related molecular subtypes to determine the potential underlying mechanism. HELLS expression was correlated with a variety of immune cells in gastrointestinal and hepatic cancers, especially in LIHC. HELLS expression was positively correlated with levels of CD8 $+$ T cells, CD4 $+$ T cells, macrophages, neutrophils, and dendritic cells in the TME of LIHC. Immunotherapy and the modulation of ICP-mediated signaling pathways have attracted increasing attention in oncology, emerging as a potential therapeutic approach for the treatment of several cancers, including LIHC [30]. Thus, correlations between HELLS and ICP genes were explored in the current study. B cells and related tertiary lymphoid structures are known to be associated with ICPs that block immunotherapy responses and affect OS [31]. In the current study, B cell-related genes were positively correlated with HELLS expression in LIHC. To date, only a few studies have focused on the influence of B cells in the TME during anti-tumor immunotherapy; however, B cells may provide a potential anti-tumor strategy [32]. In addition, T (general), CD8 ${+}$ T, T follicular helper, T helper types 1, 2, 9, 17, and 22, Treg, and exhausted T cells were positively correlated with HELLS expression. A growing body of evidence has demonstrated that T cells have an important function within immunotherapy, and CTLA-4, PD-1, and PD-L1 are the most well-known immunotherapy targets [33, 34]. Meanwhile, HELLS expression was positively correlated with PD-L1 (CD274), CTLA4, and PD-1 in LIHC. Moreover, some studies have provided insight into potential treatment strategies for LIHC that are sensitive or not to PD-1 and PD-L1 inhibitors and into mechanisms by which DNA methylation may affect LIHC [35, 36]. Taken together, these results suggest that HELLS has a potential function in the regulation of tumor immunology in LIHC.

Further exploring the functions and signaling pathways affected by HELLS in LIHC through GSEA revealed that HELLS was mainly associated with DNA-related functions such as DNA methylation. In the constructed gene-gene and protein-protein interaction networks, some notable genes and proteins interacted with HELLS, including DNMT1, DNMT3A, and DNMT3B. These results suggest that HELLS and its interacting genes and proteins may take part in regulating the immune response via DNA methylation in LIHC. Previous studies have reported that HELLS is a key epigenetic driver of cancers[26, 37]. Therefore, the correlation between HELLS expression and DNA methylation was explored in the current study. Unsurprisingly, the MSigDB data mining results revealed that HELLS was associated with GO terms such as maintenance of DNA methylation and DNA methylation-dependent heterochromatin assembly. Meanwhile, Han et al. provided evidence verifying that HELLS drives DNA methylation via DNMT1, and downregulation of HELLS contributed more to DNA methylation than downregulation of DNMT3A and DNMT3B [38]. Moreover, Termanis et al. reported that the ATP-binding function of HELLS was important for cell-autonomous de novo DNA methylation [39]. HELLS expression was negatively correlated with DNA methylation in the current study. Thus, we hypothesize that HELLS, as a methylation gene, may regulate the methylation of other DNA methylation-related genes in LIHC.

Recent seminal studies have discovered that HELLS expression is upregulated in several types of tumors, such as prostate cancer, LUAD, and colorectal cancer [7, 40, 41, 42]. and HELLS is probably involved in tumor progression. In the current study, drastic upregulation of HELLS expression was verified in LIHC via bioinformatics analysis and in clinical samples, and HELLS expression was remarkably correlated with TNM stage, pathologic stage, tumor status, AFP, and age in patients with LIHC. Additionally, high HELLS expression was significantly correlated with worse OS and DFS in LIHC. These findings highlight the potential significance of HELLS upregulation in LIHC progression and prognosis. Further, the transfection experiments confirmed that downregulation of HELLS could inhibit the proliferation and migration of HCC cells.

In summary, HELLS was upregulated in gastrointestinal tumors and LIHC. HELLS expression was closely correlated with poor prognosis and levels of infiltrating CD8 ${+}$ T cells, CD4 ${+}$ T cells, macrophages, neutrophils, and dendritic cells in the TME of LIHC. Meanwhile, the DNA methylation function of HELLS warrants further investigation. DNA methylation of HELLS was positively correlated with prognosis in LIHC. Hence, HELLS likely plays a crucial role in immune cell infiltration and may provide a prognostic biomarker in LIHC. The underlying mechanism is worthy of further exploration.

5. Conclusion

This study found abnormal HELLS expression in a variety of tumors, including LIHC, which was significantly associated with poor prognosis in LIHC. The correlation between HELLS expression and immune cell infiltration was demonstrated via bioinformatics analysis and further validated in clinical samples using western blotting and qRT-PCR. Targeting HELLS may provide a new direction for the clinical diagnosis and treatment of LIHC. Nevertheless, given the unclear mechanism underlying the role of HELLS in the occurrence and development of LIHC, further studies are warranted to elucidate the impact of HELLS and its epigenetics on the TME of LIHC.

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the First Affiliated Hospital of Kunming Medical University.

Consent for publication

All authors have read and agreed to the published version of the manuscript.

Availability of data and material

The data used to support the findings of this study are available from the corresponding author upon request.

Funding

National Natural Science Foundation of China (No. 81960124). Funded by the Scientific Research Fund of the Education Department of Yunnan Province (2022Y199).

Author contributions

Conceptualization and financial support: HanFei Huang, YingLei Miao and Zhong Zeng.

Analyzed the data, carried out the main experiments, and drafted this manuscript: Yuan Fang, WeiQiang Tang and Dan Zhao. These authors contributed equally to this work.

Technical support and manuscript revision: XiaoLi Zhang, Na Li, Yang Yang, Li Jin, ZhiTao Li and BenKai Wei.

Supplementary data

The supplementary files are available to download from https://dx-doi-org.web.bisu.edu.cn/10.3233/CBM-230073.

sj-docx-1-cbm-10.3233_CBM-230073.docx - Supplemental material

Supplemental material, sj-docx-1-cbm-10.3233_CBM-230073.docx

Footnotes

Acknowledgments

Not Applicable.

Conflict of interest

The authors declare no conflict of interest.

References

Bray

Ferlay

Soerjomataram

Siegel

R.L.

Torre

L.A.

and Jemal

, Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 68(6) (2018), 394–424.

Siegel

R.L.

Miller

K.D.

and Jemal

, Cancer Statistics. CA Cancer J Clin 67(1) (2017), 7–30.

Brown

Z.J.

Greten

T.F.

and Heinrich

, Adjuvant Treatment of Hepatocellular Carcinoma: Prospect of Immunotherapy. Hepatology 70(4) (2019), 1437–1442.

Zongyi

and Xiaowu

, Immunotherapy for hepatocellular carcinoma, Cancer Lett 470 (2020), 8–17.

Myant

and Stancheva

, LSH cooperates with DNA methyltransferases to repress transcription. Mol Cell Biol 28(1) (2008), 215–26.

Liu

Yang

Shi

Chen

Liu

Wang

Liu

Ouyang

Wang

Lai

Mao

Wang

Cheng

Liu

Wang

Zhou

Cao

Xiao

and Tao

, The cross-talk between methylation and phosphorylation in lymphoid-specific helicase drives cancer stem-like properties, Signal Transduct Target Ther 5(1) (2020), 197.

von Eyss

Maaskola

Memczak

Möllmann

Schuetz

Loddenkemper

Tanh

M.D.

Otto

Muegge

Heinemann

Rajewsky

and Ziebold

, The SNF2-like helicase HELLS mediates E2F3-dependent transcription and cellular transformation. Embo J 31(4) (2012), 972–85.

Cao

and Yan

, Cancer Epigenetics, Tumor Immunity, and Immunotherapy. Trends Cancer 6(7) (2020), 580–592.

Kurebayashi

Ojima

Tsujikawa

Kubota

Maehara

Abe

Kitago

Shinoda

Kitagawa

and Sakamoto

, Landscape of immune microenvironment in hepatocellular carcinoma and its additional impact on histological and molecular classification. Hepatology 68(3) (2018), 1025–1041.

10.

Yan

Ren

Wang

Shu

Hou

and Guo

, A Pan-Cancer Analysis of Predictive Methylation Signatures of Response to Cancer Immunotherapy. Front Immunol 12 (2021), 796647.

11.

Fan

Wang

Traugh

Chen

Liu

J.S.

and Liu

X.S.

, TIMER: A Web Server for Comprehensive Analysis of Tumor-Infiltrating Immune Cells. Cancer Res 77(21) (2017), e108–e110.

12.

Tang

Kang

Chen

and Zhang

, GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res 47(W1) (2019), W556–w560.

13.

Vasaikar

S.V.

Straub

Wang

and Zhang

, LinkedOmics: analyzing multi-omics data within and across 32 cancer types. Nucleic Acids Res 46(D1) (2018), D956–d963.

14.

Gene Ontology Consortium: going forward. Nucleic Acids Res 43 (Database issue) (2015), D1049–56.

15.

Kanehisa

Furumichi

Tanabe

Sato

and Morishima

, KEGG: new perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res 45(D1) (2017), D353–d361.

16.

Warde-Farley

Donaldson

S.L.

Comes

Zuberi

Badrawi

Chao

Franz

Grouios

Kazi

Lopes

C.T.

Maitland

Mostafavi

Montojo

Shao

Wright

Bader

G.D.

and Morris

, The GeneMANIA prediction server: biological network integration for gene prioritization and predicting gene function. Nucleic Acids Res 38(Web Server issue) (2010), W214–20.

17.

Szklarczyk

Gable

A.L.

Nastou

K.C.

Lyon

Kirsch

Pyysalo

Doncheva

N.T.

Legeay

Fang

Bork

Jensen

L.J.

and von Mering

, The STRING database in 2021: customizable protein-protein networks, and functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Res 49(D1) (2021), D605–d612.

18.

Liberzon

Birger

Thorvaldsdóttir

Ghandi

Mesirov

J.P.

and Tamayo

, The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst 1(6) (2015), 417–425.

19.

Chandrashekar

D.S.

Bashel

Balasubramanya

S.A.H.

Creighton

C.J.

Ponce-Rodriguez

Chakravarthi

and Varambally

, UALCAN: A Portal for Facilitating Tumor Subgroup Gene Expression and Survival Analyses. Neoplasia 19(8) (2017), 649–658.

20.

Cerami

Gao

Dogrusoz

Gross

B.E.

Sumer

S.O.

Aksoy

B.A.

Jacobsen

Byrne

C.J.

Heuer

M.L.

Larsson

Antipin

Reva

Goldberg

A.P.

Sander

and Schultz

, The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2(5) (2012), 401–4.

21.

Fang

Yang

Zhang

Yuan

Jin

Bao

Zhao

Zeng

and Huang

, A Co-Expression Network Reveals the Potential Regulatory Mechanism of lncRNAs in Relapsed Hepatocellular Carcinoma. Front Oncol 11 (2021), 745166.

22.

Yuan

Huang

Zhang

Lin

Jin

Yang

and Zeng

, Gastrodin Pretreatment Protects Liver Against Ischemia-Reperfusion Injury via Activation of the Nrf2/HO-1 Pathway. Am J Chin Med 48(5) (2020), 1159–1178.

23.

Federico

Petrillo

Giordano

Bosso

Fabbrocini

Ottaviano

Rosanova

Silvestri

Tufo

Cozzolino

and Daniele

, Immune Checkpoint Inhibitors in Hepatocellular Carcinoma: Current Status and Novel Perspectives. Cancers (Basel) 12(10) (2020).

24.

Malka

Lièvre

André

Taïeb

Ducreux

and Bibeau

, Immune scores in colorectal cancer: Where are we? Eur J Cancer 140 (2020), 105–118.

25.

Chen

Pan

Zhang

Guo

Cheng

Yang

and Ding

, Epigenetic strategies synergize with PD-L1/PD-1 targeted cancer immunotherapies to enhance antitumor responses. Acta Pharm Sin B 10(5) (2020), 723–733.

26.

Tameni

Sauta

Mularoni

Torricelli

Manzotti

Inghirami

Bellazzi

Fragliasso

and Ciarrocchi

, The DNA-helicase HELLS drives ALK(-) ALCL proliferation by the transcriptional control of a cytokinesis-related program. Cell Death Dis 12(1) (2021), 130.

27.

Hou

Yang

Wang

Zhou

Kong

Liu

and He

, HELLS, a chromatin remodeler is highly expressed in pancreatic cancer and downregulation of it impairs tumor growth and sensitizes to cisplatin by reexpressing the tumor suppressor TGFBR3. Cancer Med 10(1) (2021), 350–364.

28.

Wang

F.J.

Jing

Y.H.

Cheng

C.S.

Cao

Z.Q.

Jiao

J.Y.

and Chen

, HELLS serves as a poor prognostic biomarker and its downregulation reserves the malignant phenotype in pancreatic cancer. BMC Med Genomics, 14(1) (2021), 189.

29.

Xie

Zhang

Zhou

Huang

Wang

Jin

Zhang

Zeng

and Zhou

, Targeted Anti-Tumor Immunotherapy Using Tumor Infiltrating Cells. Adv Sci (Weinh), 8(22) (2021), e2101672.

30.

Gutiérrez-Larrañaga

González-López

Roa-Bautista

Rodrigues

P.M.

Díaz-González

Á.

Banales

J.M.

López-Hoyos

Santos-Laso

and Crespo

, Immune Checkpoint Inhibitors: The Emerging Cornerstone in Cholangiocarcinoma Therapy? Liver Cancer 10(6) (2021), 545–560.

31.

Liu

Zhou

and Wang

, Targeting the tumor microenvironment in B-cell lymphoma: challenges and opportunities. J Hematol Oncol 14(1) (2021), 125.

32.

Burger

J.A.

and Wiestner

, Targeting B cell receptor signalling in cancer: preclinical and clinical advances. Nat Rev Cancer 18(3) (2018), 148–167.

33.

Richard

A.C.

Frazer

G.L.

C.Y.

and Griffiths

G.M.

, Staggered starts in the race to T cell activation. Trends Immunol 42(11) (2021), 994–1008.

34.

Iwai

Hamanishi

Chamoto

and Honjo

, Cancer immunotherapies targeting the PD-1 signaling pathway. J Biomed Sci 24(1) (2017), 26.

35.

Tao

Wang

Kang

Jiang

Yin

Zong

and Wang

, PD-1 blockade improves the anti-tumor potency of exhausted CD3(+)CD56(+) NKT-like cells in patients with primary hepatocellular carcinoma. Oncoimmunology 10(1) (2021), 2002068.

36.

Zhu

and Guo

, Combined DNA Methylation and Transcriptomic Assessments to Determine a Prognostic Model for PD-1-Negative Hepatocellular Carcinoma. Front Cell Dev Biol 9 (2021), 708819.

37.

Zocchi

Mehta

S.C.

Wang

Suh

Spitale

R.C.

and Benavente

C.A.

, Chromatin remodeling protein HELLS is critical for retinoblastoma tumor initiation and progression. Oncogenesis 9(2) (2020), 25.

38.

Han

Cao

Huang

Zhu

Zhao

Han

J.J.

Feng

and Wong

, A role for LSH in facilitating DNA methylation by DNMT1 through enhancing UHRF1 chromatin association. Nucleic Acids Res 48(21) (2020), 12116–12134.

39.

Termanis

Torrea

Culley

Kerr

Ramsahoye

and Stancheva

, The SNF2 family ATPase LSH promotes cell-autonomous de novo DNA methylation in somatic cells. Nucleic Acids Res 44(16) (2016), 7592–604.

40.

Peng

Xue

Fan

Wang

Lian

Zhai

Lian

Qin

and Zhao

, Integrated analysis of dysregulated long non-coding RNAs/microRNAs/mRNAs in metastasis of lung adenocarcinoma. J Transl Med 16(1) (2018), 372.

41.

Liu

Hou

Zhou

Kong

Yan

Lei

Xiong

and He

, Downregulation of the Helicase Lymphoid-Specific (HELLS) Gene Impairs Cell Proliferation and Induces Cell Cycle Arrest in Colorectal Cancer Cells. Onco Targets Ther 12 (2019), 10153–10163.

42.

Yano

Ouchida

Shigematsu

Tanaka

Ichimura

Kobayashi

Inaki

Toyooka

Tsukuda

Shimizu

and Shimizu

, Tumor-specific exon creation of the HELLS/SMARCA6 gene in non-small cell lung cancer. Int J Cancer 112(1) (2004), 8–13.

Immunological function and prognostic value of lymphoid-specific helicase in liver hepatocellular carcinoma

Abstract

BACKGROUND:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Methods

2.1 TIMER database analysis

2.2 GEPIA2 analysis

2.3 LinkedOmics database analysis

2.4 Gene interaction analysis

2.5 DNA methylation-related database analysis

Table 1 The clinical characteristics of patients in the TCGA dataset

2.7 Sample collection

2.8 qRT-PCR

2.9 Western blotting

2.10 Cell culture and transfection

2.11 Cell migration

2.12 Wound healing assay

2.13 CCK-8 assay

2.14 EDU assay

2.15 Statistical analysis

3. Results

3.1 Pan-cancer HELLS expression and prognosis analysis

3.4 Correlation between HELLS expression and DNA methylation

Table 2 The information of clinical samples ( n = 8)

3.6 Effect of HELLS knockdown on LIHC malignant behaviors

4. Discussion

5. Conclusion

Ethics approval and consent to participate

Consent for publication

Availability of data and material

Funding

Author contributions

Supplementary data

sj-docx-1-cbm-10.3233_CBM-230073.docx - Supplemental material

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
The clinical characteristics of patients in the TCGA dataset

Table 2
The information of clinical samples ( $n=$ 8)