LCP1 is a prognostic biomarker correlated with immune infiltrates in gastric cancer

Abstract

BACKGROUND:

Previous studies have identified LCP1 as a diagnostic and prognostic marker in several cancers. However, the role of LCP1 in gastric cancer (GC) and its effect on tumor immune infiltration remain unclear.

OBJECTIVE:

The aim was to explore the role of LCP1 in GC and its effect on tumor immune infiltration.

METHODS:

We explored the expression of LCP1 relative to clinicopathology in GC patients by bioinformatics analysis and immunohistochemistry. Using cBioportal database, we analyzed the characteristic genetic variations of LCP1 in GC. In addition, we evaluated the correlation between LCP1 expression and tumor-infiltrating lymphocytes (TILs) using R software, TIMER and TISIDB databases. Finally, we analyzed the biological functions in which LCP1 may participate and the signaling pathways it may regulate.

RESULTS:

Here, we showed that LCP1 expression is significantly correlated with tumor aggressiveness and poor prognosis in GC patients. Additionally, the results indicated that LCP1 was associated with TILs, including both immunosuppressive and immunosupportive cells, and was strongly correlated with various immune marker sets in GC. GSEA analysis demonstrated that LCP1 expression played an important role in lymphocyte formation and immune reaction.

CONCLUSIONS:

LCP1 may be a potential prognostic biomarker for GC patients and a marker for tumor immunotherapy.

Keywords

LCP1 gastric cancer prognosis immune infiltrates

1. Introduction

Gastric cancer (GC) remains the third leading cause of cancer-related deaths, behind lung and liver cancer in males and behind breast and lung cancer in females [1]. The main treatment options for GC include surgery, chemotherapy, radiotherapy, targeted therapy and immunotherapy. However, due to the lack of effective screening methods for early GC, many patients are first diagnosed with advanced GC, leading to a five-year survival rate of approximately 30% [2]. Currently, immunotherapy is considered to be a promising approach for tumor patients, and PD-1/PD-L1 monoclonal antibodies have been proven to increase the overall survival (OS) rate in advanced GC, but only for a portion of patients [3]. Therefore, it is urgent to search for new immune targets and investigate the immune regulatory mechanism of GC tumors [4].

Lymphocyte cytosolic protein 1 (LCP1) is an actin-binding protein which was first isolated from a human fibroblast neoplasm [5]. As subsequent studies showed that this protein was also expressed in activated mouse macrophages, LCP1 is also known as leukocyte-specific plastin or L-plastin [6]. LCP1 protein has the characteristics of highly conserved and calcium-dependent phosphoproteins in the fimbrin family [7, 8]. Researchers also found that activated LCP1 induces cellular adhesion, increased actin binding and actin assembly [9]. LCP1 has been identified as a diagnostic and prognostic marker in oral, colon [10, 11], kidney [12] and lung cancers [13]. In addition, there is a similar effect in breast cancer, and high expression of LCP1 facilitates apoptotic resistance, invasion and metastatic spread in breast cancer cell lines [14]. These studies indicated that LCP1 plays an important role in the progression, invasion and metastasis of tumors. However, until now, the expression of LCP1 and its role in GC remain unclear.

Tumor immunotherapy implies stimulation of the immune system, proliferation of effector cells, infiltration of stimulated effector cells to the tumor microenvironment (TME), and destruction of the tumor cells [15]. Recently, most researchers have focused attention on the importance of tumor-infiltrating lymphocytes (TILs) in the treatment of various tumors. Anti-CTLA-4 (T-lymphocyte-associated antigen 4) mAbs block the CTLA-4 pathway, prolonging T-cell stimulation and restoring T-cell proliferation, thus enhancing their tumor-destructing ability [16]. TILs include various types of immune cells, for example, CD4 $+$ T cells, CD8 $+$ T cells, B cells, natural killer (NK) cells, dendritic cells (DCs) and macrophages. LCP1 is reported to be an actin-binding protein that regulates integrin-mediated cell movement, which is important for the immune system. Additionally, LCP1 is an essential regulator of cell-cell interaction, adhesion, migration and chemokine-induced polarization of immune cells [17, 18, 19]. Whether LCP1 affects immune cells infiltrating into the GC microenvironment remains unclear. Therefore, it is necessary to investigate the role of LCP1 in the prognosis of GC with regard to TILs.

In this study, we explored the expression of LCP1 as it relates to clinicopathology in GC patients by bioinformatic analysis and immunohistochemistry (IHC) By analyzing data from the cBioportal database, we demonstrated that LCP1 genetic variations correlate with patient survival in GC. In addition, we evaluated the correlation between LCP1 expression and TILs in the TME using R software, TIMER (the Tumor Immune Estimation Resource) and TISIDB (a web portal for tumor and immune system interaction) databases. Finally, we analyzed the TCGA database using GSEA (Gene Set Enrichment Analysis) software 4.0.3 and explored the biological functions and signaling pathways in which LCP1 may participate.

2. Materials and methods

2.1 Analysis of LCP1 expression in various types of tumors

The expression of LCP1 in various kinds of cancer was analyzed using TIMER and the Gene Expression Profiling Analysis (GEPIA) database [20, 21]. TIMER is a comprehensive resource for systematical analysis of immune infiltrates across diverse cancer types. We used the ‘Correlation’ module of TIMER, which draws expression scatterplots between a pair of user-defined genes in a given cancer type, together with the Spearman’s correlation and estimated statistical significance. GEPIA is a newly developed interactive web server for analyzing RNA sequencing expression data from the TCGA and the GTEx projects. Dot plots profiling gene expression across multiple cancer types and paired normal samples were generated by the Gene Expression Profile website. The threshold was determined as follows: fold change of 1.0, $P$ -value of 0.01.

2.2 Immunohistochemistry and evaluation of immunostaining intensity

This study consisted of 140 samples from 90 GC patients (90 GC tissues and 50 gastric mucosal tissues) at the First Affiliated Hospital of Nanchang University from November 2019 to April 2020. These tissue paraffin blocks were sliced into 3.5 $\mu$ m-thick sections. The sections were incubated with rabbit anti-LCP1 (13025-1-AP, 1:200, Proteintech) at 4 ${{}^{\circ}}$ C overnight, after deparaffinization, rehydration, and antigen retrieval. All slides were covered with poly-HRR goat anti-rabbit antibody and counterstained with diaminobenzidine (DAB) solution (3–5 minutes) and hematoxylin. The IHC staining intensity was classified according to the following criteria: 0 (none); 1 (weak); 2 (medium); 3(strong). The percentage of positive staining was classified as follows: 0% (0 points), 1–25% (1 point), 26–50% (2 points), 56–75% (3 points), and $>$ 75% (4 points). A score ranging from 0 to 12 was calculated by multiplying the staining extent score with the staining intensity score. Finally, samples with scores of $<$ 6 were considered to have low LCP1 expression, and those with scores $\geqslant$ 6 were considered to have high LCP1 expression. All experiments involved in this study were approved by the Ethic Committee of the First Affiliated Hospital of Nanchang University and patients.

2.3 Analysis of the correlation between LCP1 expression and patient survival in GC

The correlation between LCP1 mRNA expression and patient survival in GC was analyzed using the Kaplan-Meier plotter, which is capable of assessing the effect of 54,000 genes on survival in 21 cancer types, including the largest datasets of breast cancer ( $n=$ 6,234), ovarian cancer ( $n=$ 2,190), lung cancer ( $n=$ 3,452), and GC ( $n=$ 1,440) [22]. We used three probe sets (including 201683_s_at, 201684_s_at and 217448_s_at) to analyze the survival of GC patients, with probe set “201683_a_at” chosen to investigate the relationship between LCP1 expression and clinical characteristics of the GC patients. The hazard ratio (HR) with 95% confidence intervals and log-rank $P$ -value were also computed as the prognostic indicators of high- and low-expression groups.

2.4 Analyses of LCP1 gene mutations and copy number alterations (CNA) in GC

Genetic mutations and CNAs of LCP1 in GC were analyzed in three datasets, including the TCGA Firehose Legacy, Nature 2014 and PanCan Atlas, using the cBioPortal database [23, 24]. The genetic variations of LCP1 were evaluated with the default parameter settings. In addition, we used the ‘Survival’ module to analyze overall survival in the genetically altered and unaltered groups.

2.5 Analysis of the characteristics of immune cells in GC and expression differences compared with normal tissue

There were 374 RNA-seq gene expression profiles of GC patients included in the FPKM and count format search of the TCGA database. R software (R foundation for Statistical Computing) was used to screen data to obtain the mRNA matrices for subsequent analysis [25]. There were 257 RNA-seq gene expression profiles of patients (including 15 normal counts and 242 tumor counts) filtered by R software. Analyses of the abundance ratio, correlation and expression between GC tissues and normal tissues were then conducted among the 22 immune cells in the 257 samples.

2.6 Analysis of the correlation between GC LCP1 expression and immune cell infiltration

TISIDB is a web portal for tumor and immune system interaction that integrates multiple heterogeneous data types [26] and can evaluate correlations between the target genes and lymphocytes. In this study, we employed this website to analyze the correlation between LCP1 expression and lymphocytes in GC. To confirm the significance of the above results, we employed TIMER to infer the abundance of tumor-infiltrating immune cells (TIICs) from gene expression profiles with LCP1 expression. We analyzed the correlation between LCP1 expression and tumor purity and abundance of immune infiltrates, including B cells, CD4 $+$ T cells, CD8 $+$ T cells, neutrophils, macrophages, and dendritic cells. In addition, correlations between LCP1 expression and the gene markers for tumor-infiltrating immune cell subsets were explored via the ‘correlation’ modules in the TIMER and GEPIA databases. We calculated the Spearman’s correlation and the estimated statistical significance for the expression scatter plots in GC. The gene markers for these infiltration immune cells were reported in prior studies [27, 28].

2.7 Analysis of function and pathway enrichment

A total of 374 GC RNA-seq gene expression profiles were downloaded from the TCGA database and further processed into the prepared files using R software. These files were composed of the sample expression files (format gct.) and phenotypic classification files (format cls.). GSEA software is a computational method that determines whether an a priori defined set of genes shows statistically significant, concordant differences between two biological states [29]. We employed GSEA software 4.0.3 to analyze these datasets for biological functions and related biological pathways of genes coexpressed with LCP1. A threshold nominal $P$ -value (NOM $P$ -value) of $<$ 0.05 and false discovery rate (FDR $P$ -value) of $<$ 25% indicated statistically significant differences.

2.8 Statistical analysis

Data analysis was calculated using SPSS 26.0 software (IBM Corporation, Armonk, NY, USA), and diagrams were created using the GraphPad Prism 8.0 software (GraphPad Software, La Jolla, CA, USA). For IHC data, the $t$ -test was used to analyze measurement data, whereas the chi-square test and Fisher exact test were used to analyze count data.

Figure 1.

LCP1 expression levels in various human cancers. Abbreviations of cancer types are presented in Supplementary Table S1. (a) The expression of LCP1 mRNA in different types of cancer and normal tissue from the TCGA database, analyzed by TIMER ( ${}^{*}P<$ 0.05, ${}^{**}P<$ 0.01, ${}^{***}P<$ 0.001). (b) LCP1 expression levels in various types of cancer shown as dot plots based on RNA-seq data through GEPIA. Cancer types in which expression of LCP1 was significantly higher or lower than normal tissues are represented as red or green, respectively. (c) The protein expression of LCP1 in GC tissues (90 samples) and gastric mucosa tissues (50 samples) were compared using IHC. (d) GC and gastric mucosa tissue staining.

Table 1

Correlation of LCP1 expression and clinicopathologic features in gastric cancer

Variable	N	LCP1 expression		$\chi^{2}$	$P$ -value
		Low ( $<$ 6)	High ( $\geqslant$ 6)
Sex				1.44	0.230
Male	6	36	24
Female	3	14	16
Age (years)				0.36	0.549
$<$ 6	3	18	12
$\geqslant$ 60	6	32	28
Tumor size (cm)				3.25	0.071
$<$ 4	5	32	18
$\geqslant$ 4	4	18	22
Differentiation				10.64	0.001
High or moderately	65	43	22
Poor	25	7	18
Stage T				10.24	0.001
T1–T2	32	25	7
T3–T4	58	25	33
Stage N				19.57	$<$ 0.001
N	34	29	5
N1–N3	56	21	35
Stage TNM				8.82	0.003
I $+$ II	45	32	13
III $+$ IV	45	18	27
Lauren classification				2.90	0.089
Intestinal	56	35	21
Diffuse	34	15	19

3. Results

3.1 The mRNA expression levels of LCP1 in various types of cancer

To explore the expression of LCP1 in tumor and normal tissues, the mRNA levels in different types of cancer and their normal tissues were analyzed using TCGA RNA-seq data in TIMER. As the results show in Fig. 1a, the LCP1 expression levels were higher in most cancer types, including BRCA (breast invasive carcinoma), ESCA (esophageal carcinoma), HNSC (head and neck squamous cell carcinoma), KIRC (kidney renal clear cell carcinoma), KIRP (kidney renal papillary cell carcinoma), STAD (stomach adenocarcinoma) and UCEC (uterine corpus endometrial carcinoma) tumor issues. However, LCP1 expression was significantly lower in COAD (colon adenocarcinoma), LIHC (liver hepatocellular carcinoma), LUAD (lung adenocarcinoma), LUSC (lung squamous cell carcinoma) and READ (rectum adenocarcinoma) tumor issues compared to normal issue counterparts.

To further confirm the expression of LCP1 in tumor and normal tissues analyzed by TIMER, the GEPIA database was used. From Fig. 1b, it is evident that the mRNA expression of LCP1 was different in tumor and normal tissues among various cancers. LCP1 expression was significantly higher in BRCA, CESC (cervical squamous cell carcinoma and endocervical adenocarcinoma), ESCA, KIRC, KIRP, LGG (brain lower grade glioma), OV (ovarian serous cystadenocarcinoma), PAAD (pancreatic adenocarcinoma), PRAD (prostate adenocarcinoma), SKCM (skin cutaneous melanoma), STAD, TGCT (testicular germ cell tumors) and THCA (thyroid carcinoma) compared with adjacent normal tissues. However, a lower expression level of LCP1 was detected only in THYM (Thymoma) compared with adjacent normal tissues. In total, BRCA, ESCA, KIRC, KIRP and STAD had significantly higher LCP1 expression compared to normal tissue in these databases. STAD (GC) was chosen for follow-up study.

We collected 90 GC tissues and 50 gastric mucosal tissues to explore LCP1 protein expression using IHC. Additionally, chi-square and Fisher exact tests were used to detect correlations of LCP1 expression with clinicopathologic features in GC. According to the statistical results of $t$ -tests, LCP1 expression was significantly higher in GC tissues than in gastric mucosal tissues ( $P$ -value $<$ 0.0001, Fig. 1c and d). As shown in Table 1, LCP1 expression was significantly correlated with GC tumor differentiation ( $P$ -value $=$ 0.001), tumor depth of invasion ( $P$ -value $=$ 0.001), lymphatic metastasis ( $P$ -value $=$ 0.000) and stage TNM ( $P$ -value $=$ 0.003). However, there was no significant relationship between LCP1 expression and age ( $P$ -value $=$ 0.549), gender ( $P$ -value $=$ 0.230), tumor size ( $P$ -value $=$ 0.71) and Lauren classification ( $P$ -value $=$ 0.089).

Figure 2.

Kaplan-Meier survival curves comparing the high and low expression of LCP1 in three probe sets in the Kaplan-Meier plotter databases. (a)–(c) OS, PFS and PPS survival curves of GC (201683_s_at). (d)–(f) OS, PFS and PPS survival curves of GC (201684_s_at). (g)–(i) OS, PFS and PPS survival curves of GC (217448_s_at). OS, overall survival; PFS, progression-free survival; PPS, postprogression survival.

Table 2

Correlation of LCP1 mRNA expression and clinical prognosis in gastric cancer with different clinicopathological factors by Kaplan-Meier plotter

Clinicopathogical characteristics	Overall survival ( $n=$ 876)			Progression-free survival ( $n=$ 641)			Post progression survival ( $n=$ 499)
	N	Hazard ratio	$P$ -value	N	Hazard ratio	$P$ -value	N	Hazard ratio	$P$ -value
Sex
Female	236	1.76 (1.19–2.61)	0.0039	201	1.53 (1.02–2.31)	0.039	149	2.43 (1.57–3.76)	3.9e-05
Male	545	2.01 (1.53–2.65)	3e-07	438	1.84 (1.36–2.48)	5.1e-05	349	2.74 (2.09–3.57)	1.6e-14
Stage
1	67	7.64 (0.99–58.83)	0.021	60	5.50 (0.70–43.5)	0.069	31	8.59 (1.03–71.65)	0.017
2	140	2.50 (1.23–5.10)	0.0088	131	2.15 (1.17–3.94)	0.012	105	2.96 (1.54–5.71)	0.00068
3	305	1.94 (1.37–2.73)	0.00012	186	2.09 (1.42–3.09)	0.00014	142	3.55 (2.25–5.61)	9.1e-09
4	148	2.45 (1.50–4.02)	0.00024	141	1.80 (1.12–2.91)	0.014	104	2.52 (1.49–4.27)	0.00038
Stage T
2	241	2.46 (1.43–4.23)	0.00081	239	2.43 (1.41–4.17)	0.00089	196	2.27 (1.45–3.55)	0.0024
3	204	1.80 (1.17–2.75)	0.0062	204	1.38 (0.93–2.07)	0.11	150	2.41 (1.62–3.59)	7.6e-06
4	38	4.67 (1.61–13.55)	0.0025	39	3.81 (1.45–10.0)	0.0042	29	2.21 (0.79–6.18)	0.12
Stage N
0	74	3.67 (1.06–12.72)	0.028	72	4.3 (0.99–18.68)	0.034	41	5.05 (1.36–18.77)	0.0074
1	225	2.58 (1.71–3.89)	2.8e-06	222	2.13 (1.42–3.18)	0.00017	169	4.38 (2.75–6.96)	1.3e-11
2	121	2.83 (1.79–4.48)	3.5e-06	125	2.24 (1.45–3.46)	0.00019	105	3.51 (2.11–5.84)	3.2e-07
3	76	2.65 (1.41–4.99)	0.0017	76	2.54 (1.35–4.77)	0.0029	63	3.26 (1.37–7.74)	0.0047
1 $+$ 2 $+$ 3	422	2.07 (1.59–2.70)	4.1e-08	423	1.69 (1.31–2.18)	4.1e-05	337	2.90 (2.16–3.89)	1.3e-13
Stage M
0	444	2.14 (1.56–2.94)	1.2e-06	443	2.13 (1.48–3.06)	2.8e-05	342	3.12 (2.30–4.23)	1.5e-14
1	56	2.15 (1.10–4.20)	0.022	56	1.41 (0.71–2.83)	0.33	36	4.07 (1.74–9.53)	0.00057
Lauren classification
Intestinal	320	2.77 (1.94–3.98)	6.7e-09	263	2.77 (1.69–4.52)	2.2e-05	192	5.66 (3.66–8.76)	1e-16
Diffuse	241	1.56 (1.08–2.24)	0.017	231	1.45 (1.02–2.05)	0.037	176	1.76 (1.20–2.60)	0.0037
Differentiation
Poor	165	1.33 (0.89–1.97)	0.16	121	0.75 (0.46–1.22)	0.24	49	0.54 (0.25–1.15)	0.1
Moderate	67	0.64 (0.33–1.23)	0.18	67	0.52 (0.28–0.97)	0.035	24	2.05 (0.84–5.01)	0.11
Well differentiated	32	0.53 (0.22–1.28)	0.15	–	–	–	–	–	–
HER2 status
Negative	532	1.93 (1.48–2.52)	8.9e-07	408	1.82 (1.30–2.56)	0.00042	334	2.58 (1.93–3.45)	2.9e-11
Positive	344	1.39 (1.04–1.87)	0.026	233	1.31 (0.89–1.93)	0.16	165	2.12 (1.48–3.04)	2.8e-05

Bold values indicate $P<$ 0.05.

3.2 The correlation between LCP1 mRNA expression and patient survival in GC

To evaluate the correlation between LCP1 expression levels and patient survival rate in GC patients, we used the Kaplan-Meier plotter database. The overall survival (OS), progression-free survival (PFS) and postprogression survival (PPS) were calculated with the Start KM Plotter for GC module. There were three candidate probe sets (201683_s_at, 201684_s_at and 217448_s_at) for gene symbol LCP1 (Fig. 2). Interestingly, the LCP1 expression levels showed significant negative correlation with patient survival rate in the three probe sets: 201683_s_at (OS HR $=$ 1.91, 95% CI $=$ 1.54 to 2.37, $P=$ 3e-09, PFS HR $=$ 1.62, 95% CI $=$ 1.29 to 2.03, $P=$ 3.2e-05 and PPS HR $=$ 2.46, 95% CI $=$ 1.96 to 3.07, $P=$ 4.4e-16), 201684_s_at (OS HR $=$ 1.54, 95% CI $=$ 1.29 to 1.83, $P=$ 1.1e-06, PFS HR $=$ 1.45, 95% CI $=$ 1.18 to 1.77 $P=$ 0.00031 and PPS HR $=$ 2.46, 95% CI $=$ 1.96 to 3.07, $P=$ 4.4e-16) and 217448_s_at (OS HR $=$ 1.58, 95% CI $=$ 1.33 to 1.89, $P=$ 2.3e-07, PFS HR $=$ 1.67, 95% CI $=$ 1.34 to 2.07, $P=$ 3.1e-06 and PPS HR $=$ 2.44, 95% CI $=$ 1.95 to 3.06, $P=$ 1e-15).

3.3 Correlation of LCP1 expression with clinicopathological characteristics of GC in kaplan-meier plotter database

To better evaluate the correlation and discover the mechanisms underlying the contribution of LCP1 expression level to GC, we analyzed the impact of LCP1 expression on clinical GC characteristics using the Kaplan-Meier plotter database (Table 2). There were 876, 641 and 499 patients with OS, PFS and PPS, respectively. High LCP1 expression was significantly associated with worse OS, PFS and PPS in almost all subtypes of stage 1–4, stage T2-4, stage M0-1 and HER2 status. However, there were no significant correlations with PFS in stage 1 (HR $=$ 5.50, 95% CI $=$ 0.70 to 43.5, $P=$ 0.069), stage T3 (HR $=$ 1.38, 95% CI $=$ 0.93 to 2.07, $P=$ 0.11) and stage M1 (HR $=$ 1.41, 95% CI $=$ 0.71 to 2.83, $P=$ 0.33) or with PPS in stage T4 (HR $=$ 2.21, 95% CI $=$ 0.79 to 6.18, $P=$ 0.12). Specially, overexpression of LCP1 was inversely correlated with OS, PFS and PPS in sex and Lauren classification as well as lymph node involvement (stage N categories). In contrast, no significant correlation was found between LCP1 expression and types of differentiation.

Figure 3.

LCP1 expression and genetic variations in GC. (a) Correlation analysis of LCP1 expression and clinical stages in gastric cancer through GEPIA. (b) A schematic diagram represents location and frequency of each mutation across the coding sequence of LCP1 in 478 cases from TCGA, Firehose legacy (TCGA); 295 cases from TCGA, nature 2014 (TCGA Pub); and 440 cases from TCGA, PanCancer Atlas (TCGA PanCan) via the cBioPortal database. (c) Frequencies of LCP1 mutations and copy number alterations (CNA) in the three datasets. (d) Kaplan-Meier survival curves comparing patients with or without LCP1 alterations in 393 cases from TCGA, Firehose legacy and 434 cases from TCGA, PanCancer Atlas.

Figure 4.

The correlation and expression level of immune cells in GC. (a) The abundance ratio of immune cells in the 257 RNA-seq samples. Each column represents a sample, and each row represents the abundance ratio of immune cells via different colors and heights. (b) The relationship between the abundance ratio of different immune cells. This value represents the correlation value. Red is positive, and blue is negative. (c) The expression differences of immune cells in tumor tissues compared to normal tissues. Red represents tumor tissue, and blue represents normal tissues. $P$ , $P$ -value.

3.4 LCP1 gene alteration patterns and patient survival in GC

To better understand LCP1 expression levels in the different TNM stages of GC, we employed the GEPIA database. As is shown in Fig. 3a, LCP1 expression level was significantly different from stage I to IV ( $F$ value $=$ 2.84, Pr $=$ 0.0376). Using the cBioPortal database to investigate the genetic variations of LCP1, we found three datasets (478 cases from TCGA, Firehose legacy; 295 cases from TCGA, Nature 2014; and 440 cases from TCGA, PanCancer Atlas). There were 8 mutations in the LCP1 coding region from the three studies (Fig. 3b). Missense and splice mutations, respectively, were located in the EF-hand domain pair (amino acid sequence 14 to 18) and calponin homology (CH) domain (amino acid sequence from 397 to 502). In addition, additional mutations were found in the amino acid sequence from 100 to 627, which contains four calponin homology (CH) domains. Moreover, the alteration of LCP1 comprised deep deletions, amplifications and mutations in the three studies. The TCGA Firehose legacy dataset contains 1.78% mutations, 2.8% amplifications and 0.51% deep deletions (Fig. 3c). We further analyzed whether LCP1 alteration status correlated with patient survival. As shown in Fig. 3d, LCP1 alteration only demonstrated a significant positive correlation with patient survival in the TCGA ( $P$ -value $=$ 0.0496) and TCGA PanCan datasets ( $P$ -value $=$ 0.0449).

3.5 Characteristics of immune cells in GC and expression differences compared with normal tissues

We found that there was the abundance ratio of 22 immune cells in the 257 GC RNA-seq samples (Fig. 4a). To further investigate the underlying mechanisms of this immune response, we analyzed the correlation of the 22 immune cells in GC (Fig. 4b). CD8 T cells significantly correlated with activated memory CD4 T cells, which, in contrast, negatively correlated with resting memory CD4 T cells, M0 macrophages, activated mast cells and neutrophils. Activated memory CD4 T cells were positively correlated with M1 macrophages, gamma delta T cells and resting NK cells, while an inverse correlation was determined with regulatory T cells (Tregs) and resting memory CD4 T cells. In addition, activated mast cells showed a positive relationship with M0 macrophages, activated dendritic cells, resting NK cells and neutrophils, which were negatively correlated with resting memory CD4 T cells. Interestingly, there was an obvious negative correlation between Tregs and most effective immune cells. Then, we analyzed the infiltration level of the 22 immune cells in GC tissues compared to normal gastric tissues (Fig. 4c). The infiltration levels of activated memory CD4 T cells and M0, M1 and M2 macrophages were significantly higher in GC tissues, while plasma immune cells, monocytes and resting mast cells infiltrated to lower levels in tumor tissues than normal tissue counterparts.

Figure 5.

Spearman’s correlation of LCP1 with lymphocytes via TISIDB and TIMER in GC tissues. (a) The correlation between LCP1 expression and the abundance of TILs by the TISIDB database. (b) Detailed information on the correlation between LCP1 expression and the abundance of CD8 $+$ T cells, CD4 $+$ T cells, NK cells and DC cells in the left picture. (c) The correlation between LCP1 expression and abundance of infiltrating immune cells in STAD and PRAD by the TIMER database.

Table 3

Correlation analysis between LCP1and relate genes and markers of immune cells in TIMER

Description	Gene markers	STAD				PRAD
		None		Purity		None		Purity
		Cor.	P	Cor.	P	Cor.	P	Cor.	P
CD8 ${}^{+}$ T cell	CD8A	0.702	${}^{***}$	0.673	${}^{***}$	$-$ 0.024	0.597	$-$ 0.059	0.233
	CD8B	0.565	${}^{***}$	0.546	${}^{***}$	0.017	0.713	0.002	0.961
T cell (general)	CD3D	0.778	${}^{***}$	0.746	${}^{***}$	$-$ 0.042	0.355	$-$ 0.048	0.331
	CD3E	0.748	${}^{***}$	0.712	${}^{***}$	$-$ 0.014	0.751	$-$ 0.019	0.694
	CD2	0.813	${}^{***}$	0.790	${}^{***}$	$-$ 0.006	0.887	$-$ 0.058	0.234
B cell	CD19	0.552	${}^{***}$	0.519	${}^{***}$	$-$ 0.062	0.167	$-$ 0.058	0.234
	CD79A	0.623	${}^{***}$	0.586	${}^{***}$	$-$ 0.038	0.401	$-$ 0.027	0.580
Monocyte	CD86	0.839	${}^{***}$	0.826	${}^{***}$	$-$ 0.056	0.210	$-$ 0.080	0.105
	CD115 (CSF1R)	0.751	${}^{***}$	0.753	${}^{***}$	$-$ 0.008	0.851	$-$ 0.037	0.450
TAM	CCL2	0.458	${}^{***}$	0.408	${}^{***}$	0.001	0.990	$-$ 0.013	0.788
	CD68	0.514	${}^{***}$	0.491	${}^{***}$	$-$ 0.040	0.374	$-$ 0.044	0.376
	IL10	0.644	${}^{***}$	0.614	${}^{***}$	0.020	0.649	0.011	0.619
M1 macrophage	INOS (NOS2)	0.141	${}^{*}$	0.128	0.012	0.089	0.047	0.074	0.133
	IRF5	0.358	${}^{***}$	0.322	${}^{***}$	$-$ 0.216	${}^{***}$	$-$ 0.216	${}^{***}$
	COX2 (PTGS2)	0.052	0.292	0.010	0.842	0.065	0.146	0.026	0.500
M2 macrophage	CD163	0.697	${}^{***}$	0.688	${}^{***}$	$-$ 0.019	0.679	$-$ 0.030	0.539
	VSIG4	0.604	${}^{***}$	0.597	${}^{***}$	$-$ 0.077	0.087	$-$ 0.098	0.046
	MS4A4A	0.727	${}^{***}$	0.714	${}^{***}$	$-$ 0.060	0.184	$-$ 0.071	0.147
Neutrophils	CD66b (CEACAM8)	0.035	0.477	0.030	0.564	0.060	0.179	0.068	0.164
	CD11b (ITGAM)	0.709	${}^{***}$	0.696	${}^{***}$	$-$ 0.018	0.682	$-$ 0.044	0.370
	CCR7	0.719	${}^{***}$	0.682	${}^{***}$	0.028	0.533	0.020	0.689
Natural killer cell	KIR2DL1	0.390	${}^{***}$	0.387	${}^{***}$	0.028	0.532	0	0.992
	KIR2DL3	0.355	${}^{***}$	0.335	${}^{***}$	0.017	0.704	0.032	0.520
	KIR2DL4	0.403	${}^{***}$	0.363	${}^{***}$	0.041	0.357	0.031	0.534
	KIR3DL1	0.363	${}^{***}$	0.372	${}^{***}$	0.102	0.023	0.084	0.862
	KIR3DL2	0.432	${}^{***}$	0.406	${}^{***}$	0.092	0.039	0.110	0.025
	KIR3DL3	0.127	${}^{*}$	0.144	${}^{*}$	$-$ 0.030	0.498	$-$ 0.052	0.290
	KIR2DS4	0.347	${}^{***}$	0.347	${}^{***}$	0.066	0.139	0.048	0.326
	KLRK1	0.732	${}^{***}$	0.698	${}^{***}$	$-$ 0.067	0.138	$-$ 0.099	0.044
	NCR1 (NKp46)	0.572	${}^{***}$	0.556	${}^{***}$	0.098	0.029	0.100	0.041
	NCR2 (NKp44)	0.180	${}^{**}$	0.152	${}^{*}$	0.002	0.971	0	0.998
	NCR3 (NKp30)	0.646	${}^{***}$	0.607	${}^{***}$	$-$ 0.006	0.894	$-$ 0.004	0.937
Dendritic cell	HLA-DPB1	0.685	${}^{***}$	0.642	${}^{***}$	$-$ 0.100	0.025	$-$ 0.004	0.937
	HLA-DQB1	0.554	${}^{***}$	0.488	${}^{***}$	$-$ 0.048	0.285	$-$ 0.112	0.022
	HLA-DRA	0.708	${}^{***}$	0.673	${}^{***}$	0.006	0.890	$-$ 0.006	0.188
	HLA-DPA1	0.690	${}^{***}$	0.650	${}^{***}$	$-$ 0.042	0.351	0.012	0.813
	BDCA-1 (CD1C)	0.587	${}^{***}$	0.542	${}^{***}$	0.032	0.476	0.012	0.813
	BDCA-4 (NRP1)	0.474	${}^{***}$	0.444	${}^{***}$	$-$ 0.037	0.414	$-$ 0.056	0.251
	CD11c (ITGAX)	0.768	${}^{***}$	0.751	${}^{***}$	$-$ 0.071	0.116	0.014	0.769
Th1	T-bet (TBX21)	0.740	${}^{***}$	0.721	${}^{***}$	$-$ 0.056	0.213	$-$ 0.071	0.150
	STAT4	0.787	${}^{***}$	0.770	${}^{***}$	$-$ 0.075	0.096	$-$ 0.097	0.048
	STAT1	0.404	${}^{***}$	0.390	${}^{***}$	0.171	${}^{**}$	0.174	${}^{**}$
	IFN- $\gamma$ (IFNG)	0.548	${}^{***}$	0.529	${}^{***}$	$-$ 0.026	0.559	$-$ 0.022	0.657
	IFN- $\alpha$ (TNF)	0.398	${}^{***}$	0.329	${}^{***}$	$-$ 0.027	0.552	$-$ 0.047	0.338
Th2	GATA3	0.521	${}^{***}$	0.489	${}^{***}$	0.042	0.351	$-$ 0.018	0.714
	STAT6	0.062	0.207	0.053	0.304	0.124	${}^{*}$	0.109	0.026
	STAT5A	0.511	${}^{***}$	0.483	${}^{***}$	$-$ 0.018	0.695	$-$ 0.022	0.662
	IL13	0.127	${}^{*}$	0.104	0.043	$-$ 0.045	0.321	$-$ 0.06	0.224
Tfh	BCL6	0.228	${}^{***}$	0.175	${}^{**}$	$-$ 0.049	0.272	$-$ 0.087	0.077
	IL21	0.533	${}^{***}$	0.517	${}^{***}$	0.018	0.681	0.045	0.359
Th17	STAT3	0.321	${}^{***}$	0.298	${}^{***}$	0.245	${}^{***}$	0.242	${}^{***}$
	IL17A	0.104	0.033	0.093	0.071	0.094	0.035	0.088	0.072
Treg	FOXP3	0.705	${}^{***}$	0.666	${}^{***}$	0.076	0.092	0.086	0.079
	CCR8	0.706	${}^{***}$	0.699	${}^{***}$	0.080	0.076	0.079	0.107
	STAT5B	0.365	${}^{***}$	0.373	${}^{***}$	0.021	0.645	0.004	0.932
	TGF- $\beta$ (TGFB1)	0.415	${}^{***}$	0.38	${}^{***}$	$-$ 0.166	${}^{**}$	$-$ 0.190	${}^{***}$

Table 3, continued
Description	Gene markers	STAD				PRAD
		None		Purity		None		Purity
		Cor.	P	Cor.	P	Cor.	P	Cor.	P
T cell exhaustion	PD-1 (PDCD1)	0.657	${}^{***}$	0.622	${}^{***}$	$-$ 0.099	0.028	0.118	0.016
	CTLA4	0.659	${}^{***}$	0.616	${}^{***}$	$-$ 0.074	0.097	$-$ 0.067	0.179
	LAG3	0.613	${}^{***}$	0.582	${}^{***}$	$-$ 0.138	${}^{*}$	$-$ 0.127	${}^{*}$
	TIM-3 (HAVCR2)	0.790	${}^{***}$	0.775	${}^{***}$	$-$ 0.070	0.119	$-$ 0.074	0.134
	GZMB	0.538	${}^{***}$	0.490	${}^{***}$	0.042	0.349	0.058	0.242

STAD, stomach adenocarcinoma; PRAD, prostate adenocarcinoma; TAM, tumor-associated macrophage; Th, T helper cell; Tfh, follicular helper T cell; Treg, regulatory T cell; Cor. R value of Spearman’s correlation; None, correlation without adjustment. Purity, correlation adjusted by purity. ${}^{*}P<$ 0.01; ${}^{**}P<$ 0.001; ${}^{***}P<$ 0.0001.

Table 4

Correlation between LCP1 and markers of CD8 ${}^{+}$ T, T cell (general), B cell, monocyte, TAM, M2 macrophage, dendritic cell and T cell exhaustion in GEPIA

Description	Gene markers	STAD				PRAD
		Tumor		Normal		Tumor		Normal
		R	P	R	P	R	P	R	P
CD8 ${}^{+}$ T cell	CD8A	0.700	${}^{***}$	0.850	${}^{***}$	0.170	${}^{**}$	0.490	${}^{**}$
	CD8B	0.570	${}^{***}$	0.770	${}^{***}$	0.160	${}^{**}$	0.500	${}^{**}$
T cell (general)	CD3D	0.750	${}^{***}$	0.850	${}^{***}$	0.023	0.610	0.510	${}^{**}$
	CD3E	0.740	${}^{***}$	0.920	${}^{***}$	0.130	${}^{*}$	0.420	${}^{*}$
	CD2	0.820	${}^{***}$	0.910	${}^{***}$	0.150	${}^{**}$	0.500	${}^{**}$
B cell	CD19	0.500	${}^{***}$	0.870	${}^{***}$	0.000	0.880	0.260	0.060
	CD79A	0.580	${}^{***}$	0.870	${}^{***}$	0.050	0.210	0.900	${}^{**}$
Monocyte	CD86	0.850	${}^{***}$	0.840	${}^{***}$	0.210	${}^{***}$	0.600	${}^{***}$
	CD115 (CSF1R)	0.770	${}^{***}$	0.730	${}^{***}$	0.240	${}^{***}$	0.500	${}^{**}$
TAM	CCL2	0.470	${}^{***}$	$-$ 0.180	0.310	0.096	0.034	0.220	0.110
	CD68	0.560	${}^{***}$	0.510	${}^{*}$	0.260	${}^{***}$	0.580	${}^{**}$
	IL10	0.670	${}^{***}$	0.580	${}^{**}$	0.230	${}^{***}$	0.240	0.013
M2 macrophage	CD163	0.650	${}^{***}$	$-$ 0.091	0.600	0.047	0.300	0.330	0.018
	VSIG4	0.620	${}^{***}$	0.230	0.180	0.150	${}^{**}$	0.470	${}^{**}$
	MS4A4A	0.740	${}^{***}$	0.150	0.390	0.140	${}^{**}$	0.450	${}^{**}$
Dendritic cell	HLA-DPB1	0.685	${}^{***}$	0.642	${}^{***}$	$-$ 0.100	0.025	$-$ 0.004	0.937
	HLA-DQB1	0.554	${}^{***}$	0.488	${}^{***}$	$-$ 0.048	0.285	$-$ 0.112	0.022
	HLA-DRA	0.708	${}^{***}$	0.673	${}^{***}$	0.006	0.890	$-$ 0.006	0.188
	HLA-DPA1	0.690	${}^{***}$	0.650	${}^{***}$	$-$ 0.042	0.351	0.012	0.813
	BDCA-1 (CD1C)	0.587	${}^{***}$	0.542	${}^{***}$	0.032	0.476	0.012	0.813
	BDCA-4 (NRP1)	0.474	${}^{***}$	0.444	${}^{***}$	$-$ 0.037	0.414	$-$ 0.056	0.251
	CD11c (ITGAX)	0.768	${}^{***}$	0.751	${}^{***}$	$-$ 0.071	0.116	0.014	0.769
T cell exhaustion	PD-1 (PDCD1)	0.657	${}^{***}$	0.622	${}^{***}$	$-$ 0.099	0.028	0.118	0.016
	CTLA4	0.659	${}^{***}$	0.616	${}^{***}$	$-$ 0.074	0.097	$-$ 0.067	0.179
	LAG3	0.613	${}^{***}$	0.582	${}^{***}$	$-$ 0.138	${}^{*}$	$-$ 0.127	${}^{*}$
	TIM3 (HAVCR2)	0.790	${}^{***}$	0.775	${}^{***}$	$-$ 0.070	0.119	$-$ 0.074	0.134
	GZMB	0.538	${}^{***}$	0.490	${}^{***}$	0.042	0.349	0.058	0.242

STAD, stomach adenocarcinoma; PRAD, prostate adenocarcinoma; TAM, tumor-associated macrophage; R, R value of Spearman’s correlation; ${}^{*}P<$ 0.01; ${}^{**}P<$ 0.001; ${}^{***}P<$ 0.0001.

Figure 6.

Correlation between LCP1 expression and various subsets of immune cell infiltrates in STAD and PRAD via the TIMER database. Markers include CD8A and CB8B for CD8 ${}^{+}$ T cells, CD3D, CD3E and CD2 for T cells (general), CD19 and CD79A for B cells, CD86 and CD115 for monocytes, CCL2, CD68 and IL10 for TAMs, CD163, VSIG4 and MS4A4A for M2 macrophages, HLA-DPB1, HLA-DQB1, HLA-DRA, HLA-DPA1, BDCA-1, BDCA-4 and CD11c for dendritic cells (DCs) and PD-1, CTLA4, LAG3, TIM-3 and GZMB for T cell exhaustion. Scatterplots of correlations between LCP1 expression and gene markers of CD8 ${}^{+}$ T cells, T cells (general), B cells, monocytes, TAMs, macrophages, dendritic cells and T cell exhaustion.

Figure 7.

Functional and pathway enrichment analyses of genes coexpressed with LCP1 expression via GSEA software 4.0.3. (a) Graph representing enriched biological functions (GO) of genes coexpressed with LCP1 expression in GC. (b) Graph representing enriched biological pathways (KEGG) of genes coexpressed with LCP1 expression in GC.

3.6 LCP1 expression is correlated with immune infiltration into the GC microenvironment

TILs imply that immune cells act as prognostic factors of sentinel lymph node status and survival in various cancers [30, 31]. To investigate whether LCP1 expression was an independent predictor of GC immune cell infiltration, we employed the TISIDB database to analyze the correlation between LCP1 expression and immune cell infiltrates in various types of cancer. Figure 5a shows that LCP1 expression had significant correlations with immune cell infiltrates in most types of cancer, including STAD; in contrast, no significant relationship was found in PRAD. We confirmed that LCP1 expression was strongly correlated with the amounts of infiltrating CD8 ${}^{+}$ T cells (rho $=$ 0.652, $P<$ 2.2e-16), CD4 ${}^{+}$ T cells (rho $=$ 0.547, $P<$ 2.2e-16), NK cells (rho $=$ 0.693, $P<$ 2.2e-16) and dendritic cells (rho $=$ 0.618, $P<$ 2.2e-16) in GC tissues (Fig. 5b).

Then, to confirm the correlation between LCP1 expression and immune cell infiltration in STAD and PRAD, LCP1 expression was analyzed using the TIMER database. Figure 5c shows that LCP1 expression is significantly inversely correlated with tumor purity (cor. $=$ $-$ 0.276, $P=$ 4.68e-08) and B cells (cor. $=$ $-$ 0.167, $P=$ 1.31e-03) in GC. Higher LCP1 expression in STAD tissues markedly increased the number of infiltrating immune cells, such as CD8 ${}^{+}$ T cells (cor. $=$ 0.683, $P=$ 4.63e-52), CD4 ${}^{+}$ T cells (cor. $=$ 0.342, $P=$ 1.68e-11), macrophages (cor. $=$ 0.455, $P=$ 2.48e-20), neutrophils (cor. $=$ 0.771, $P=$ 3.63e-74) and dendritic cells (cor. $=$ 0.822, $P=$ 2.37e-92) in tumor tissues. However, LCP1 expression did not correlate with tumor purity or infiltrating levels of B cells, CD8 $+$ T cells, macrophages, neutrophils, and dendritic cells in PRAD (Fig. 5c). According to our results that LCP1 expression levels are inversely correlated with tumor purity and that overexpression of LCP1 correlates with poor prognosis in GC patients, we suggest that the infiltration of immune cells may be an important source of LCP1 expression in the GC microenvironment.

3.7 Correlation between LCP1 expression and immune marker sets

To further investigate the correlation between LCP1 and diverse subsets of immune cell infiltrations in STAD, we employed the TIMER databases to analyze the relationships between LCP1 expression and immune cell markers including T cells, B cells, monocytes, M1 and M2 macrophages, neutrophils, NK cells, Tregs and dendritic cells, using PRAD as the control. Table 3 shows that LCP1 expression level is significantly correlated with most immune cell markers in STAD, whereas LCP1 expression significantly correlated with only 6 gene markers in PRAD.

Obviously, the LCP1 expression level was strongly correlated with the expression levels of most subsets of CD8 ${}^{+}$ T cells, T cells (general), B cells, monocytes, TAMs, M2 macrophages, dendritic cells and exhausted T cells in STAD (Table 3). At the same time, as shown in Fig. 6, the marker sets CD8A and CB8B of CD8 ${}^{+}$ T cells; CD3D, CD3E and CD2 of T cells (general); CD19 and CD79A of B cells; CD86 and CD115 of monocytes; CCL2, CD68 and IL10 of TAMs; CD163, VSIG4 and MS4A4A of M2 macrophages; HLA-DPB1, HLA-DQB1, HLA-DRA, HLA-DPA1, BDCA-1, BDCA-4 and CD11c of dendritic cells (DCs); and PD-1, CTLA4, LAG3, TIM-3, and GZMB of T cell exhaustion were strongly correlated with LCP1 expression in STAD ( $P<$ 0.0001). Similarly, we investigated the relationship between the above subsets of immune cell infiltrates and LCP1 expression in the GEPIA database. LCP1 expression level showed strong correlations with the expression levels of most subsets of CD8 ${}^{+}$ T cells, T cells (general), B cells, monocytes, TAMs, M2 macrophages, dendritic cells and exhausted T cells, in a manner similar to those in the TIMER analysis (Table 4). Therefore, these results provide further evidence that LCP1 expression is specifically correlated with immune cells infiltrating into the GC microenvironment.

3.8 Correlation between LCP1 expression and function and pathway enrichment in GC

To analyze biological functions and pathways of genes coexpressed with LCP1 expression, we employed GSEA software 4.0.3 to investigate these datasets. After adjusting the thresholds to a nominal $P$ -value (NOM $P$ -value) of $<$ 0.05 and a false discovery rate (FDR $P$ -value) of $<$ 25%, the top 20 enriched biological functions and pathways of the gene set were identified (Tables S2 and S3). Interestingly, we found that the majority of biological functions and pathways identified were related to immune reaction pathways. Among the biological functions of gene coexpressed with LCP1 expression, there was significant correlation with immune substances of leukocytes and cytokines (Fig. 7a). LCP1 expression had a special role in the activation, proliferation and differentiation of lymphocytes, especially for T cells and B cells (GO:0042110, GO:002286 and GO:0042113). Furthermore, the results indicated its involvement in cytokine production, such as interleukin-1 production and secretion (GO:0032612, GO:0050701 and GO:0050663). Similarly, various immunoreactions were included in the signaling pathway results enriched for the genes coexpressed with LCP1 expression, including hsa04060 (chemokine signaling pathway), hsa04620 (toll-like receptor signaling pathway), hsa04650 (natural killer cell mediated cytotoxicity), hsa04514 (cell adhesion molecules (CAMs)), hsa04612 (antigen processing and presentation), hsa04670 (leukocyte transendothelial migration), hsa04660 (T cell receptor signaling pathway), hsa04672 (intestinal immune network for IgA production), hsa04664 (Fc epsilon RI signaling pathway), hsa04662 (B cell receptor signaling pathway), hsa04666 (Fc gamma R-mediated phagocytosis) and hsa04622 (RIG-I-like receptor signaling pathway) (Fig. 7b).

4. Discussion

LCP1 has two EF-hand calcium-binding motifs at the N-terminus and two actin-binding domains (ABD1), each of which contains two calponin-homology regions (CH), at the C-terminus [32]. LCP1 was first isolated from neoplastic human fibroblasts and soon identified as significantly over-expressed in many cell lines and solid tumors [33, 34]. Researchers have found that LCP1 is a critical regulator of cell shape and behavior in both the adaptive and innate immune systems, including macrophages, B cells, T cells and NK cells [35, 36, 37, 38, 39]. In this study, we first investigate the expression level of LCP1 in GC, which was more highly expressed in GC than in normal gastric tissues. Overexpression of LCP1 showed negative correlation with GC patient survival. Furthermore, there were significant correlations between the immune cell infiltration levels/immune marker sets and LCP1 expression level. According to enriched biological function and pathway analysis, LCP1 was mainly involved in lymphocyte formation and immune reaction. Therefore, this study explored the role of LCP1 in GC and the underlying mechanism of LCP1 in TILs.

Many studies have reported that LCP1 is a regulator of tumor growth and metastasis, indicating its use as a biomarker indicating poor prognosis in different types of cancer [40, 41, 42]. These studies suggest that LCP1 participates in the progression, invasion and metastasis of tumor cells. Even though oral and colon cancers of the digestive system have been studied, the expression and role of LCP1 in GCs have not been reported [10, 11]. Here, we analyzed the expression levels of LCP1 in various types of cancer using the TIMER and GEPIA databases. According to the TIMER database analysis, we found that LCP1 was significantly overexpressed in breast, esophageal, head and neck, kidney, stomach and uterine cancers, but showed low expression in colon, liver, lung and rectal cancers. Additionally, the GEPIA database showed that LCP1 expression was higher in most cancers, such as BRCA, ESCA, KIRC, KIRP, and STAD; interestingly, only lower expression was in THYM compared to normal tissues. Collectively, the expression of LCP1 was higher in BRCA, ESCA, KIRP, KIRC and STAD compared to the normal tissues. Moreover, the expression of LCP1 was significantly correlated with cancer stage in GC, which was similar to the results in colorectal cancer [40]. We found that LCP1 expression was significantly higher in GC tissues than in gastric mucosal tissues using IHC. In addition, LCP1 expression was significantly correlated with GC tumor differentiation, tumor depth of invasion, lymphatic metastasis and stage TNM.

Higher expression of LCP1 also correlated with poorer prognosis in GC patients, as analyzed by the Kaplan-Meier plotter database. Furthermore, high expression of LCP1 was correlated with poor prognosis of GC patients in the following categories: stage 1–4, stage T2-4, stage M0-1, HER2 status, sex and Lauren classification, and lymph node involvement (stage N categories) had the highest HR. These results differ somewhat with the IHC data, perhaps due to sample size contributing to different conclusions. Mechanisms that may affect gene expression, such as gene mutation, deletion and epigenetic changes [43, 44], we investigated the LCP1 coding region and identified 8 mutations including missense and splice mutations. LCP1 alteration was significantly correlated with GC patient survival. This result suggests that the genetic variation of LCP1 might be beneficial for patients with gastric cancer, warranting further study into its mechanism. Therefore, LCP1 has the potential to be a novel and valuable biomarker for human GC.

LCP1 is highly expressed in normal leukocytes and plays an important role in the organization of the actin cytoskeleton by localizing to actin-rich membrane structures involved in immune defense, locomotion and adhesion [6, 40]. As TILs, NK cells, T cells and dendritic cells induce cell death as anti-tumor effecting cells [15]. Previous studies showed that LCP1 affected NKG2D-mediated regulation of NK cell chemotaxis by participating in NKG2D receptor clustering upon activation [39]. It is also well known that LCP1 is considered an important factor for chemokine-mediated T cell polarization and migration [17]. Considering that LCP1 is related to anti-tumor effector cells, we speculate that LCP1 will also affect GC tumor immune infiltration. First, our results show that immune cells had significant correlation of cell-cell interactions, and infiltration of some immune cells, such as activated memory CD4 T cells and M0 to M2 macrophages, was significantly higher compared to normal gastric tissue. High levels of activated memory CD4 T cells and M1 macrophages in the TME represented a favorable prognosis in colorectal cancer patients [45]. We can hypothesize that these immune cells could also be positive factors for GC patients. Second, this study indicates that LCP1 expression is significantly correlated with immune cell infiltration including CD8 ${}^{+}$ and CD4 ${}^{+}$ T cells, dendritic cells and NK cells in GC, in contrast to a lack of significant association in PRAD, via the TISIDB database. Furthermore, the LCP1 expression level demonstrated a strong positive correlation with the infiltration level of CD8 ${}^{+}$ and CD4 ${}^{+}$ T cells, macrophages, neutrophils and dendritic cells, while a significant negative correlation was observed with tumor purity and B cells in GC, based on analysis of the TIMER database. In these immune cells, CD8 ${}^{+}$ T cells and NK cells directly recognize and destroy cells in a tumor as anti-tumor effecting cells [46, 47]. In addition, CD4 ${}^{+}$ T cells have an important role in facilitating the endurance of memory CD8 ${}^{+}$ T cells that induce tumor cell death in the tumor environment [48]. During immunization with cancer vaccines, dendritic cells deliver the proper signals that determine the specific type and quantity of T-cell response [15]. According to this study, it is possible to treat GC patients by targeting LCP1 to increase the amount of these effector cells infiltrating into the TME. Thus, LCP1 may be a potential therapeutic target for immunotherapy in GC patients.

As such, our study represents the first clinical evidence that LCP1 expression is negatively correlated with patient survival in GC. However, this result is not consistent with the notion that high level of TILs in the TME is a favorable prognosis. This study showed that LCP1 expression was significantly correlated with most immune markers in GC. Although there were positive correlations between LCP1 expression and miscellaneous effector cells stimulating markers of CD8 ${}^{+}$ T cells, TAMs, monocytes, NK cells and dendritic cells, such as CD8A and CB8B of CD8 ${}^{+}$ T cells, CCL2, CD68 and IL10 of TAMs and others, the increase in LCP1 expression positively correlated with T cell exhaustion and Treg markers such as PD-1, CTLA-4, LAG3, TIM-3, FOXP3, CCR8, STAT5B and TGF- $\beta$ , which will suppress the function of anti-tumor effector cells [49]. Previous studies showed that the immune checkpoints PD-1 and CTLA-4 downregulated T-cell activation, which led to infiltrating immune anergy and tumor immune escape [49, 50]. It is also well known that TIM-3 (mucin-domain-containing molecule 3) is upregulated by highly dysfunctional TA-specific CD ${}^{+}$ 8 T cells as a coinhibitory molecule which suppresses the anti-tumor functions [51]. FOXP3 also may play a special role in inhibiting effective immunity [15]. Because CD8 ${}^{+}$ T cells constitute the majority of TILs, these immune inhibitors could significantly downregulate the functions of effector immune cells [47]. It has been reported that anti-PD-1 and anti-CTLA-4 can be used to expand infiltrating T cells and reduces Tregs, producing a survival benefit in clinical trials [52]. This study demonstrated that LCP1 was positively correlated with PD-1, CTLA-4, TIM-3, FOXP3 and other immune suppressive molecules in GC, and there was an obvious negative correlation between Tregs and most effector immune cells. Therefore, LCP1 could be an immune inhibitor with an important role in effector functions and recruitment of TILs in GC.

Additionally, to investigate the role of LCP1 in the functions of TILs in GC, we have presented the top 20 enriched biological functions and pathways of genes coexpressed with LCP1 expression, as calculated by GSEA software. Interestingly, the results mainly indicated involvement in cell proliferation, differentiation, activation and immune response. Previous studies showed that LCP1 served as a regulator of the actin cytoskeleton, with cellular functions such as leukocyte function, cell migration, DNA repair and endocytosis by stabilizing the clasped conformation of $\alpha$ M $\beta$ 2 integrin [17]. Toll-like receptor signaling was correlated with LCP1 expression in GC. In this signaling pathway, TLRL (toll-like receptor ligand) has been investigated as a protein vaccine, with pleiotropic functions such as stimulating APCs as well as NK cells and mediating tumor cell death [53]. The TME comprises tumor cells, inflammatory cells and extracellular matrices [54]. In the TME, there are both immunosuppressive and immunosupportive mechanisms, but the immunosuppressive mechanism may take over [55]. Although this study showed that LCP1 expression facilitates infiltrating effector cells into the TME of GC, the immune reaction is a complex and dynamic process. Therefore, we suggest that further study is necessary to explore the role of LCP1 involvement in the immune response in GC.

In conclusion, this study has identified that overexpression of LCP1 is correlated with poor OS in GC patients and increased infiltrating levels of CD8 ${}^{+}$ T cells, CD4 ${}^{+}$ T cells, macrophages, neutrophils and dendritic cells. However, the increase in LCP1 expression was also positively correlated with markers of T cell exhaustion and Tregs, which further suppress anti-tumor functions. Therefore, our studies reveal that LCP1 may function as a potential biomarker for GC patients and play a potential role in lymphocyte formation and immune cell infiltration.

Footnotes

Conflict of interest

The authors declare that they have no competing interests.

Funding

This work was supported by the National Natural Science Foundation of China (No. 81860428, 81960503), the Science and Technology Plan of Health Commission of Jiangxi Province (No. 20191026), the spark promotion plan of grassroots health appropriate technology of Health Commission of Jiangxi Province (No. 20198012) and the Graduate Innovation Fund of Jiangxi Province (YC2019-S057).

Supplementary data

Supplementary Table S1

Tumor Abbreviations

Abbreviation	Type of cancer
ACC	Adrenocortical carcinoma
BLCA	Bladder Urothelial Carcinoma
BRCA	Breast invasive carcinoma
CESC	Cervical squamous cell carcinoma and endocervical adenocarcinoma
CHOL	Cholangio carcinoma
COAD	Colon adenocarcinoma
DLBC	Lymphoid Neoplasm Diffuse Large B-cell Lymphoma
ESCA	Esophageal carcinoma
GBM	Glioblastoma multiforme
HNSC	Head and Neck squamous cell carcinoma
KICH	Kidney Chromophobe
KIRC	Kidney renal clear cell carcinoma
KIRP	Kidney renal papillary cell carcinoma
LAML	Acute Myeloid Leukemia
LGG	Brain Lower Grade Glioma
LIHC	Liver hepatocellular carcinoma
LUAD	Lung adenocarcinoma
LUSC	Lung squamous cell carcinoma
MESO	Mesothelioma
OV	Ovarian serous cystadenocarcinoma
PAAD	Pancreatic adenocarcinoma
PCPG	Pheochromocytoma and Paraganglioma
PRAD	Prostate adenocarcinoma
READ	Rectum adenocarcinoma
SARC	Sarcoma
SKCM	Skin Cutaneous Melanoma
STAD	Stomach adenocarcinoma
TGCT	Testicular Germ Cell Tumors
THCA	Thyroid carcinoma
THYM	Thymoma
UCEC	Uterine Corpus Endometrial Carcinoma
UCS	Uterine Carcinosarcoma
UVM	Uveal Melanoma

Supplementary Table S2

Top 20 enriched biological functions in analysis of genes co-expressed with LCP1 expression in GC via GSEA

Name	Exact source	Size	ES	NES	RANK AT MAX	Leading edge
GO Lymphocyte activation involved in immune response	0002285	175	0.78709440	2.8459980	4942	tags $=$ 50%, list $=$ 9%, signal $=$ 55%
GO Leukocyte proliferation	0070661	290	0.77899830	2.8044589	4097	tags $=$ 48%, list $=$ 7%, signal $=$ 51%
GO Cytokine secretion	0050663	217	0.76810414	2.7786455	4097	tags $=$ 44%, list $=$ 7%, signal $=$ 48%
GO T cell activation	0042110	457	0.77435970	2.7599480	4097	tags $=$ 47%, list $=$ 7%, signal $=$ 50%
GO T cell activation involved in immune response	0002286	99	0.80587260	2.7523665	4846	tags $=$ 55%, list $=$ 9%, signal $=$ 60%
GO External side of plasma membrane	0009897	372	0.81166350	2.7361288	4057	tags $=$ 55%, list $=$ 7%, signal $=$ 59%
GO Positive regulation of cytokine secretion	0050715	130	0.77925920	2.7356020	4311	tags $=$ 45%, list $=$ 8%, signal $=$ 49%
GO Regulation of inflammatory response	0050727	342	0.70572730	2.7258835	6074	tags $=$ 43%, list $=$ 11%, signal $=$ 48%
GO Positive regulation of immune effector process	0002699	214	0.77734023	2.7128057	5196	tags $=$ 51%, list $=$ 9%, signal $=$ 56%
GO Regulation of immune effector process	0002697	453	0.79083484	2.7102656	4846	tags $=$ 54%, list $=$ 9%, signal $=$ 58%
GO Positive regulation of leukocyte proliferation	0070665	137	0.81084520	2.7081175	4097	tags $=$ 53%, list $=$ 7%, signal $=$ 57%
GO Regulation of leukocyte proliferation	0070663	219	0.79012245	2.7050890	4097	tags $=$ 51%, list $=$ 7%, signal $=$ 55%
GO Positive regulation of response to external stimulus	0032103	305	0.73972710	2.7049177	6114	tags $=$ 51%, list $=$ 11%, signal $=$ 57%
GO Positive regulation of cell activation	0050867	378	0.81305593	2.7021196	4097	tags $=$ 56%, list $=$ 7%, signal $=$ 60%
GO Lymphocyte differentiation	0030098	345	0.76775960	2.7004917	5503	tags $=$ 50%, list $=$ 10%, signal $=$ 55%
GO Positive regulation of cytokine production	0001819	444	0.73692500	2.6926480	5345	tags $=$ 46%, list $=$ 10%, signal $=$ 50%
GO Interleukin-1 production	0032612	94	0.77275900	2.6899457	4311	tags $=$ 48%, list $=$ 8%, signal $=$ 52%
GO Interleukin-1 secretion	0050701	57	0.78996390	2.6871681	4311	tags $=$ 49%, list $=$ 8%, signal $=$ 53%
GO Regulation of leukocyte differentiation	1902105	268	0.75817823	2.6850412	4650	tags $=$ 47%, list $=$ 8%, signal $=$ 51%
GO B cell activation	0042113	296	0.79974720	2.6849854	4942	tags $=$ 57%, list $=$ 9%, signal $=$ 62%

SIZE, number of genes in the gene set; ES, enrichment score; NES, norminal enrichment score; NOM p-val, nominal $p$ -vlaue; FDR q-val, false discovery rate; FWER $p$ -value, Familywise-error rate.

Supplementary Table S3

Top 20 enriched biological pathways in analysis of genes co-expressed with LCP1 expression in GC via GSEA

Name	Exact source	Size	ES	NES	FDR q-val	FWER p-val	RANK AT MAX	Leading edge
KEGG Cytokine receptor interaction	hsa04060	264	0.77655363	2.8461928	0	0	4704	tags $=$ 49%, list $=$ 9%, signal $=$ 54%
KEGG Chemokine signaling pathway	hsa04062	188	0.76826980	2.6687932	0	0	4739	tags $=$ 49%, list $=$ 9%, signal $=$ 55%
KEGG JAK-STAT signaling pathway	hsa04630	155	0.71712550	2.6559598	0	0	5196	tags $=$ 49%, list $=$ 9%, signal $=$ 56%
KEGG Hematopoietic cell lineage	hsa04640	85	0.85720444	2.6554537	0	0	4097	tags $=$ 49%, list $=$ 9%, signal $=$ 57%
KEGG Toll-like receptor signaling pathway	hsa04620	102	0.78043395	2.5938632	0	0	6137	tags $=$ 49%, list $=$ 9%, signal $=$ 58%
KEGG Natural killer cell mediated cytotoxicity	hsa04650	132	0.76855070	2.5919600	0	0	4605	tags $=$ 49%, list $=$ 9%, signal $=$ 59%
KEGG Cell adhesion molecules (CAMs)	hsa04514	131	0.78309447	2.5152433	0	0	4708	tags $=$ 49%, list $=$ 9%, signal $=$ 60%
KEGG Autoimmune thyroid disease	hsa05320	50	0.87039804	2.4829192	0	0	4209	tags $=$ 49%, list $=$ 9%, signal $=$ 61%
KEGG Antigen processing and presentation	hsa04612	81	0.83043873	2.4644510	0	0	4605	tags $=$ 49%, list $=$ 9%, signal $=$ 62%
KEGG Leukocyte transendothelial migration	hsa04670	116	0.69506377	2.4365847	0	0	5620	tags $=$ 49%, list $=$ 9%, signal $=$ 63%
KEGG T cell receptor signaling pathway	hsa04660	108	0.75259000	2.4267507	0	0	5929	tags $=$ 49%, list $=$ 9%, signal $=$ 64%
KEGG Viral myocarditis	hsa05416	68	0.77337164	2.4135090	0	0	4209	tags $=$ 49%, list $=$ 9%, signal $=$ 65%
KEGG Leishmania infection	hsa05140	70	0.85721195	2.4046340	0	0	2302	tags $=$ 49%, list $=$ 9%, signal $=$ 66%
KEGG Intestinal immune network for IgA production	hsa04672	46	0.90114164	2.3957763	0	0	2884	tags $=$ 49%, list $=$ 9%, signal $=$ 67%
KEGG Fc epsilon RI signaling pathway	hsa04664	79	0.67757046	2.3811480	0	0	5945	tags $=$ 49%, list $=$ 9%, signal $=$ 68%
KEGG B cell receptor signaling pathway	hsa04662	75	0.76037544	2.3644030	0	0	4401	tags $=$ 49%, list $=$ 9%, signal $=$ 69%
KEGG Asthma	hsa05310	28	0.90408826	2.3285637	0	0	2003	tags $=$ 49%, list $=$ 9%, signal $=$ 70%
KEGG Fc gamma R-mediated phagocytosis	hsa04666	96	0.68339510	2.2934237	6.25E-05	0.001	6152	tags $=$ 49%, list $=$ 9%, signal $=$ 71%
KEGG Type I diabetes mellitus	hsa04940	41	0.86884370	2.2838593	0.000118	0.002	4209	tags $=$ 49%, list $=$ 9%, signal $=$ 72%
KEGG RIG-I-like receptor signaling pathway	hsa04622	71	0.64933800	2.2598982	0.000113	0.002	8335	tags $=$ 49%, list $=$ 9%, signal $=$ 73%

SIZE, number of genes in the gene set; ES, enrichment score; NES, norminal enrichment score; NOM p-val, nominal $p$ -vlaue; FDR q-val, false discovery rate; FWER $p$ -value, Familywise-error rate.

References

Niccolai

Taddei

Prisco

and Amedei

, Gastric cancer and the epoch of immunotherapy approaches, World J Gastroenterol 21 (2015), 5778–5793.

Inokuchi

Nakagawa

Tanioka

Okuno

and Kojima

, Long- and short-term outcomes of laparoscopic gastrectomy versus open gastrectomy in patients with clinically and pathological locally advanced gastric cancer: A propensity-score matching analysis, Surgical Endoscopy 32 (2017), 735–742.

Chen

Zhang

Liu

Zhao

Lin

and Li

, A gastric cancer LncRNAs model for MSI and survival prediction based on support vector machine, BMC Genomics 20 (2019), 846.

Pagni

Guerini-Rocco

Schultheis

A.M.

Grazia

Rijavec

Ghidini

Lopez

Venetis

Croci

G.A.

Malapelle

and Fusco

, Targeting immune-related biological processes in solid tumors: We do need biomarkers, International Journal of Molecular Sciences 20 (2019), 5452.

Goldstein

Djeu

Latter

Burbeck

and Leavitt

, Abundant synthesis of the transformation-induced protein of neoplastic human fibroblasts, plastin, in normal lymphocytes, Cancer Research 45 (1985), 5643–5647.

Shinomiya

Hirata

Saito

Yagisawa

and Nakano

, Identification of the 65-kDa phosphoprotein in murine macrophages as a novel protein: Homology with human l-plastin, Biochem Biophys Res Commun 202 (1994), 1631–1638.

Galkin

V.E.

Orlova

Cherepanova

Lebart

M.-C.

and Egelman

E.H.

, High-resolution cryo-EM structure of the F-actin-fimbrin/plastin ABD2 complex, Proceedings of the National Academy of Sciences of the United States of America 105 (2008), 1494–1498.

Wan

Otsuna

Chien

C.-B.

and Hansen

, FluoRender: An application of 2D image space methods for 3D and 4D confocal microscopy data visualization in neurobiology research, IEEE Pacific Visualization Symposium: [Proceedings]. IEEE Pacific Visualisation Symposium (2012), 201–208.

Morley

S.C.

, The actin-bundling protein L-plastin: A critical regulator of immune cell function, International Journal of Cell Biology 2012 (2012), 935173.

10.

Koide

Kasamatsu

Endo-Sakamoto

Ishida

Shimizu

Kimura

Miyamoto

Yoshimura

Shiiba

and Tanzawa

, Evidence for critical role of lymphocyte cytosolic protein 1 in oral cancer, Sci Rep 7 (2017), 43379.

11.

Ning

Gerger

Zhang

Hanna

D.L.

Yang

Winder

Wakatsuki

Labonte

M.J.

Stintzing

and Volz

, Plastin polymorphisms predict gender- and stage-specific colon cancer recurrence after adjuvant chemotherapy, Molecular Cancer Therapeutics 13 (2014), 528–539.

12.

Kim

D.S.

Choi

Y.D.

Moon

Kang

Lim

J.-B.

Kim

K.M.

Park

K.M.

and Cho

N.H.

, Composite three-marker assay for early detection of kidney cancer, Cancer Epidemiology Biomarkers & Prevention 22 (2013), 390–398.

13.

Luo

Schork

N.J.

Marschke

K.B.

S.C.

and Meglasson

M.D.

, Identification of polymorphisms associated with hypertriglyceridemia and prolonged survival induced by bexarotene in treating non-small cell lung cancer, Anticancer Research 31 (2011), 2303–2311.

14.

Janji

Vallar

Tanoury

Z.A.

Bernardin

F.o.

Vetter

Schaffner-Reckinger

Berchem

Friederich

and Chouaib

, The actin filament cross-linker L-plastin confers resistance to TNF-α in MCF-7 breast cancer cells in a phosphorylation-dependent manner, Journal of Cellular and Molecular Medicine 14 (2010), 1264–1275.

15.

Kirkwood

J.M.

Butterfield

L.H.

Tarhini

A.A.

Hassane Zarour

M.D.

Pawel Kalinski

M.D.

and Soldano Ferrone

M.D.

, Immunotherapy of cancer in 2012, Ca A Cancer Journal for Clinicians 62 (2012), 305–339.

16.

Engelhardt

J.J.

Sullivan

T.J.

and Allison

J.P.

, CTLA-4 overexpression inhibits t cell responses through a CD28-B7-dependent mechanism, Journal of Immunology 177 (2006), 1052–1061.

17.

Freeley

O’Dowd

Paul

Kashanin

Davies

Kelleher

and Long

, L-plastin regulates polarization and migration in chemokine-stimulated human T lymphocytes, Journal of Immunology (Baltimore, Md.: 1950) 188 (2012), 6357–6370.

18.

Morley

S.C.

, The actin-bundling protein L-plastin supports T-cell motility and activation, Immunological Reviews 256 (2013), 48–62.

19.

Wabnitz

G.H.

Köcher

Lohneis

Stober

Konstandin

M.H.

Funk

Sester

Wilm

Klemke

and Samstag

, Costimulation induced phosphorylation of L-plastin facilitates surface transport of the T cell activation molecules CD69 and CD25, European Journal of Immunology 37 (2007), 649–662.

20.

Fan

Wang

Traugh

Chen

Liu

J.S.

and Liu

X.S.

, TIMER: A web server for comprehensive analysis of tumor-infiltrating immune cells, Cancer Research 77 (2017), e108–e110.

21.

Tang

Kang

Gao

and Zhang

, GEPIA: A web server for cancer and normal gene expression profiling and interactive analyses, Nucleic Acids Research 45 (2017), W98–W102.

22.

Marcell

S.A.

András

ádám

Susann

F.r.

Kim

G.J.

Alex

Rita

András

and Balázs

G.r.

, Cross-validation of survival associated biomarkers in gastric cancer using transcriptomic data of 1,065 patients, Oncotarget 7 (2016), 49322–49333.

23.

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 Discovery 2 (2012), 401–404.

24.

Gao

Aksoy

B.A.

Dogrusoz

Dresdner

Gross

Sumer

S.O.

Sun

Jacobsen

Sinha

and Larsson

, Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal, Science Signaling 6 (2019), pl1.

25.

Cao

Yang

Yao

Dong

and Tian

, Screening and identifying immune-related cells and genes in the tumor microenvironment of bladder urothelial carcinoma: Based on TCGA database and bioinformatics, Frontiers in Oncology 9 (2020), 1533.

26.

Wong

C.N.

Tong

Zhong

J.Y.

Zhong

S.S.W.

W.C.

Chu

K.C.

Wong

C.Y.

Lau

C.Y.

Chen

Chan

N.W.

and Zhang

, TISIDB: An integrated repository portal for tumor-immune system interactions, Bioinformatics 35 (2019), 4200–4202.

27.

Danaher

Warren

Dennis

D’Amico

White

Disis

M.L.

Geller

M.A.

Odunsi

Beechem

and Fling

S.P.

, Gene expression markers of Tumor Infiltrating Leukocytes, Journal for ImmunoTherapy of Cancer 5 (2017), 18.

28.

Siemers

N.O.

Holloway

J.L.

Chang

Chasalow

S.D.

and Szustakowski

J.D.

, Genome-wide association analysis identifies genetic correlates of immune infiltrates in solid tumors, Plos One 12 (2017), e0179726.

29.

Subramanian

Tamayo

Mootha

V.K.

Mukherjee

Ebert

B.L.

Gillette

M.A.

Paulovich

Pomeroy

S.L.

Golub

T.R.

Lander

E.S.

and Mesirov

J.P.

, Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles, Proceedings of the National Academy of Sciences 102 (2005), 15545–15550.

30.

Letca

A.F.

Ungureanu

Şenilă

S.C.

Grigore

L.E.

Pop

Ş.

Fechete

Vesa

Ş.C.

and Cosgarea

, Regression and sentinel lymph node status in melanoma progression, Medical Science Monitor International Medical Journal of Experimental & Clinical Research 24 (2018), 1359–1365.

31.

Ohtani

, Focus on TILs: Prognostic significance of tumor infiltrating lymphocytes in human colorectal cancer, Cancer Immunity A Journal of the Academy of Cancer Immunology 7 (2007), 4.

32.

Kell

M.J.

Riccio

R.E.

Baumgartner

E.A.

Compton

Z.J.

and Leclair

E.E.

, Targeted deletion of the zebrafish actin-bundling protein L-plastin (lcp1), Plos One 13 (2018), e0190353.

33.

Lin

C.S.

Aebersold

R.H.

Kent

S.B.H.

Varma

and Leavitt

, Molecular cloning and characterization of plastin, a human leukocyte protein expressed in transformed human fibroblasts, Molecular & Cellular Biology 8 (1988), 4659–4668.

34.

Park

Zong

P.C.

and Leavitt

, Activation of the leukocyte plastin gene occurs in most human cancer cells, Cancer Res 54 (1994), 1775–1781.

35.

Deady

L.E.

Todd

E.M.

Davis

C.G.

Zhou

J.Y.

Topcagic

Edelson

B.T.

Ferkol

T.W.

Cooper

M.A.

Muenzer

J.T.

and Morley

S.C.

, L-plastin is essential for alveolar macrophage production and control of pulmonary pneumococcal infection, Infection & Immunity 82 (2014), 1982–1993.

36.

Dubey

Singh

A.K.

Awasthi

Nagarkoti

Kumar

Ali

Chandra

Kumar

Barthwal

M.K.

Jagavelu

Sánchez-Gómez

F.J.

Lamas

and Dikshit

, L-Plastin S-glutathionylation promotes reduced binding to β-actin and affects neutrophil functions, Free Radical Biology & Medicine 86 (2015), 1–15.

37.

Morley and Celeste

, The actin-bundling protein L-plastin supports T-cell motility and activation, Immunological Reviews 256 (2013), 48–62.

38.

Todd

E.M.

Deady

L.E.

and Morley

S.C.

, Intrinsic T- and B-cell defects impair T-cell-dependent antibody responses in mice lacking the actin-bundling protein L-plastin, European Journal of Immunology 43 (2013), 1735–1744.

39.

Serrano-Pertierra

Cernuda-Morollon

Brdika

Hooeji

and Lopez-Larrea

, L-plastin is involved in NKG2D recruitment into lipid rafts and NKG2D-mediated NK cell migration, Journal of Leukocyte Biology 96 (2014), 437–445.

40.

Chaijan

Roytrakul

Mutirangura

and Leelawat

, Matrigel induces L-plastin expression and promotes L-plastin-dependent invasion in human cholangiocarcinoma cells, Oncology Letters 8 (2014), 993–1000.

41.

Foran

McWilliam

Kelleher

Croke

D.T.

and Long

, The leukocyte protein L-plastin induces proliferation, invasion and loss of E-cadherin expression in colon cancer cells, International Journal of Cancer 118 (2006), 2098–2104.

42.

Klemke

Rafael

M.T.

Wabnitz

G.H.

Weschenfelder

Konstandin

M.H.

Garbi

Autschbach

Hartschuh

and Samstag

, Phosphorylation of ectopically expressed L-plastin enhances invasiveness of human melanoma cells, International Journal of Cancer 120 (2007), 2590–2599.

43.

Launonen

, Mutations in the human LKB1/STK11 gene, Human Mutation 26 (2005), 291–297.

44.

Zhao

Wilkerson

M.D.

Shah

Yin

Wang

Hayward

M.C.

Roberts

Lee

C.B.

Parsons

A.M.

and Thorne

L.B.

, Alterations of LKB1 and KRAS and risk of brain metastasis: Comprehensive characterization by mutation analysis, copy number, and gene expression in non-small-cell lung carcinoma, Lung Cancer 86 (2014), 255–261.

45.

Narayanan

Kawaguchi

Yan

Peng

and Takabe

, Cytolytic activity score to assess anticancer immunity in colorectal cancer, Annals of Surgical Oncology 25 (2018), 2323–2331.

46.

Konjević

G.M.

Vuletić

A.M.

Mirjačić Martinović

K.M.

Larsen

A.K.

and Jurišić

V.B.

, The role of cytokines in the regulation of NK cells in the tumor environment, Cytokine 117 (2019), 30–40.

47.

Seo

A.N.

Lee

H.J.

Kim

E.J.

Kim

H.J.

Jang

M.H.

Lee

H.E.

Kim

Y.J.

Kim

J.H.

and Park

S.Y.

, Tumour-infiltrating CD8+ lymphocytes as an independent predictive factor for pathological complete response to primary systemic therapy in breast cancer, British Journal of Cancer 109 (2013), 2705–2713.

48.

Sun and C.

, Defective CD8 T cell memory following acute infection without CD4 T cell help, Science 300 (2003), 339–342.

49.

Wang

Pino-Lagos

Vries

V.C.d.

Guleria

Sayegh

M.H.

and Noelle

R.J.

, Programmed death 1 ligand signaling regulates the generation of adaptive Foxp3∼+CD4∼+ regulatory T cells, Proc Natl Acad Sci U S A 105 (2008), 9331–9336.

50.

Iwai

Ishida

Tanaka

Okazaki

Honjo

and Minato

, Involvement of PD-L1 on tumor cells in the escape from host immune system and tumor immunotherapy by PD-L1 blockade, Proceedings of the National Academy of Sciences of the United States of America 99 (2002), 12293–12297.

51.

Fourcade

Sun

Pagliano

Chauvin

J.-M.

Sander

Janjic

Tarhini

A.A.

Tawbi

H.A.

Kirkwood

J.M.

and Moschos

, PD-1 and Tim-3 regulate the expansion of tumor antigen-specific CD8+ T cells induced by melanoma vaccines, Cancer Research 74 (2014), 1045–1055.

52.

Tsai

K.K.

and Daud

A.I.

, Nivolumab plus ipilimumab in the treatment of advanced melanoma, Journal of Hematology & Oncology 8 (2015), 123.

53.

Fourcade

Sun

Kudela

Janjic

Kirkwood

J.M.

El-Hafnawy

and Zarour

H.M.

, Human tumor antigen-specific helper and regulatory T cells share common epitope specificity but exhibit distinct T cell repertoire, Journal of Immunology 184 (2010), 6709–6718.

54.

Klemm

and Joyce

J.A.

, Microenvironmental regulation of therapeutic response in cancer, Trends in Cell Biology 25 (2015), 198–213.

55.

Tang

Qiao

and Fu

Y.X.

, Immunotherapy and tumor microenvironment, Cancer Letters 370 (2015), 85–90.