The Expression of LIMK1 is Related to the Poor Prognosis and Immune Function of Hepatocellular Carcinoma

Background: According to reports, LIMK1 may have the effect of promoting tumor progression. However, the effect of the expression of LIMK1 on the healing of patients with hepatocellular carcinoma and its effect on the immune function are still not clear. Therefore, we analyzed the effect of LIMK1 on the healing of patients with hepatocellular carcinoma and its correlation with immunity through bioinformatics analysis. Methods: Download the transcriptional expression prole of LIMK1 in hepatocellular carcinoma tissues and normal tissues in TCGA, and study its expression in hepatocellular carcinoma. Study the expression of LIMK1 in hepatocellular carcinoma through CPTAC and HPA database. The Kaplan-Meier method was used to evaluate the effect of LIMK1 expression on the survival of patients with hepatocellular carcinoma. Use the STRING database to construct a protein-protein interaction (PPI) network. Use the "ClusterProler" package for feature-rich analysis. Use TISIDB database and Xiantao platform to study the relationship between LIMK1 mRNA expression and immune inltration. Results: The expression of LIMK1 in hepatocellular carcinoma tissues was signicantly up-regulated. Increased expression of LIMK1 mRNA is related to high TNM staging. In the ROC curve, when the cut-off level is 1.813, the sensitivity and specicity of LIMK1 to distinguish hepatocellular carcinoma from adjacent controls are 80.7% and 86%, respectively.The Kaplan-Meier curve shows that the higher the expression of LIMK1, the worse the survival of patients with hepatocellular carcinoma (42.2 months vs. 70 months, P = 0.001). Correlation analysis studies have shown that the expression of LIMK1 mRNA in hepatocellular carcinoma is related to immune cell inltration. Conclusion: Up-regulation of LIMK1 may affect the survival rate and immune invasion of hepatocellular carcinoma. Studies have shown that LIMK1 may be related to the poor prognosis of hepatocellular carcinoma, and has a certain relationship with the immune inltration of hepatocellular carcinoma.


Introduction
Hepatocellular carcinoma (HCC) is still one of the malignant tumors with poor prognosis, and its effective treatment methods are very limited [1]. In addition, only a few HCC patients have the opportunity to undergo radical treatments such as liver transplantation, surgical resection, and radiofrequency ablation [2]. If some patients are found early, after surgical resection, their 5-year survival rate can reach more than 70% [3]. However, because early hepatocellular carcinoma has no obvious clinical symptoms, most patients with hepatocellular carcinoma have extensive metastasis when they appear, and the patients have missed the opportunity for radical surgery [4]. Lack of monitoring methods and insu cient accuracy of early diagnosis methods are the reasons for the poor prognosis and high mortality of HCC. Therefore, it is urgent to nd new biomarkers to promote the prognosis of hepatocellular carcinoma.
LIMK1 is a molecule belonging to the serine/threonine kinase family and has the function of regulating actin polymerization [5]. Phosphorylated LIMK1 is related to a variety of cell functions, including cell Page 3/18 cycle, progression of cell metastasis, angiogenesis, and vascular proliferation [6,7]. Previous studies have shown that the expression of LIMK1 may affect tumor progression and is related to the peritoneal metastasis of colorectal cancer, gastric cancer and other tumors [8][9][10]. In addition, the up-regulation of LIMK1 is associated with prostate cancer metastasis and lower survival rate [11]. In addition, in pancreatic tumors, knockdown of LIMK1 can inhibit tumor invasion and metastasis, as well as inhibit angiogenesis [12,13]. In addition, some recent studies have shown that in hepatocellular carcinoma, the down-regulation of LIMK1 can inhibit tumor migration [14]. Therefore, we speculate that LIMK1 may be a biomarker for predicting the poor prognosis of hepatocellular carcinoma.
Through this study, we explored the prognostic value of LIMK1 in hepatocellular carcinoma and its correlation with immune in ltration. Through the Cancer Genome Atlas (TCGA) data, we found that the level of LIMK1 is related to the survival rate of hepatocellular carcinoma. The expression of LIMK1 is increased in hepatocellular carcinoma, and its up-regulation is related to the unfavorable clinical features of hepatocellular carcinoma. We further studied the relationship between LIMK1 in hepatocellular carcinoma, diagnosis and patient recovery. At the same time, we used TISIDB database and Xiantao platform to study the effect of LIMK1 on the immune function of hepatocellular carcinoma. We believe that LIMK1 may be a star target for diagnosing liver cancer and predicting the long-term recovery of patients.

Material And Method
2. 1 We download clinical information about LIMK1 from the o cial website of the Cancer Genome Atlas (TCGA) [15]. We analyzed the expression of LIMK1 in 33 tumor types.

2.2
We downloaded data about the expression of LIMK1 in liver cancer from TCGA, and extracted information about 374 hepatocellular carcinomas and 59 normal tissues adjacent to the cancer. We also extracted relevant clinical information, including gender, age, weight, BMI, AFP, T staging, N staging, M staging, TNM staging, tumor status, histological grade and OS events.

UALCAN
is an open online network data resource used to analyze openly accessible cancer data [16].
In this study, we used UALCAN to analyze LIMK1 protein expression.
Human Protein Atlas (HPA) contains the expression information of human-related proteins [17]. Through the HPA database, we studied the protein expression of LIMK1 in hepatocellular carcinoma [17].
2.4 Using the STRING line database, a protein-protein interactions (PPI) Networks with was constructed [18]. Use the "ClusterPro ler" package and the "ggplot2" package to analyze and visualize the gene ontology (GO) enrichment of co-expressed genes and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway [19].
2. 5 We use TISIDB7 online Web to study the effect of LIMK1 on the immune function of hepatocellular carcinoma [20].
We use the Xiantao platform (www.xiantao.love) to collect relevant gene expression pro les and relevant clinical information in hepatocellular carcinoma, and to study the correlation between LIMK1 and immune cells in hepatocellular carcinoma.
Kaplan-Meier plotter was used to study the effect of LIMK1 expression on the survival of hepatocellular carcinoma patients [21].

Statistical Analyses
Use R (V 3.6.3) to perform statistical analysis on all the data, and use the R package ggplot2 to visualize the difference in expression. The difference between hepatocellular carcinoma tissue and normal tissue was studied using paired t test and Mann-Whitney U test. The detection of the cut-off value of the ROC curve is performed using the pROC package [22]. In order to study the correlation between LIMK1 and the survival rate of hepatocellular carcinoma patients, we used the survminer package to perform Kaplan-Meier.

The expression of LIMK1 in hepatocellular carcinoma and other tumors from the perspective of pan-cancer
We use the data of the TCGA platform to show that compared with adjacent normal tissues, the expression of LIMK1 in 19 types of tumors increased signi cantly, including: BLCA, BRCA, CESC, CHOL, COAD, ESCA, HNSC, KICH, KIRC, KIRP, LIHC, LUAD, LUSC, PCPG, PRAD, READ, STAD, THCA, UCEC. This study shows that in hepatocellular carcinoma, the expression of LIMK1 is up-regulated compared to normal tissues.

LIMK1 mRNA expression and protein expression are upregulated in HCC
Using data from TCGA and HPA, the expression of LIMK1 in hepatocellular carcinoma was studied. Studies have shown that in matched hepatocellular carcinoma samples, the RNA expression level of LIMK1 in tumor samples (n = 50) was signi cantly increased ( Fig. 2A, 1.637 ± 0.839 vs. 0.760 ± 0.322, P < 0.001). In the unpaired samples, the results consistent with the matched samples were obtained, and the expression of LIMK1 in hepatocellular tumor specimens was signi cantly up-regulated (Fig. 2B, 2.687 ± 1.035 vs. 1.413 ± 0.501, P < 0.001). The UALCAN database was used to analyze the expression of LIMK1, and the results showed that the mRNA expression of LIMK1 in hepatocellular carcinoma tissues was signi cantly increased (Fig. 2C). Using HPA to study the protein expression of LIMK1, the results showed that the protein expression of LIMK1 in cancer tissues was signi cantly up-regulated. In summary, the expression of LIMK1 is up-regulated in hepatocellular carcinoma tissues.

3.3
The expression level of LIMK1 mRNA in patients with HCC is correlated with clinical features As shown in Table 1 and Fig. 3A-I, in patients with HCC, the high expression of LIMK1 is associated with higher T staging (P = 0.009), high TNM staging (P = 0.003), high AFP value (P = 0.003) and high histological grade (P < 0.001) are signi cantly correlated. However, our research showed that there was no signi cant correlation between the expression level of LIMK1 and age (P = 0.151) and gender (P = 0.595). In general, the above results show that the expression of LIMK1 is signi cantly correlated with higher T staging and higher TNM staging. This may suggest that LIMK1 may be a potential marker for predicting poor prognosis of HCC.  (Fig. 4B).
Subgroup K-M analysis showed that OS in patients with hepatocellular carcinoma was associated with high LIMK1 and high TNM staging, further indicating that LIMK1 may predict the poor outcome of patients with hepatocellular carcinoma (Fig. 4C, D, E). The results of the Kaplan-Meier curve database also indicate that the high expression of LIMK1 is related to the poor OS of hepatocellular carcinoma (Fig. 4F). The above research results indicate that LIMK1 may indicate that patients with hepatocellular carcinoma have poor recovery.

In HCC, construct PPI network and functional annotations
In HCC, we found that LIMK1 and 10 genes form a network of co-expressed genes through string database (Fig. 5A). At the same time, GO and KEGG analysis showed that the changes in the biological process of LIMK1 are related to the regulation of actin cytoskeleton, axon guidance, regulation of actin cytoskeleton organization Related, regulation of actin lament-based processes, and actin lament organization (Fig. 5B). Functional annotations indicate that these genes are involved in Rho GTPase binding. Correlation analysis showed that there is a correlation between LIMK1 and HCC overexpressed genes ( Fig. 5C-I).

Discussion
Research shows that the expression of LIMK1 is increased in hepatocellular carcinoma. At the same time, when the TNM stage of hepatocellular carcinoma patients is higher, the mRNA expression value of LIMK1 is also higher. According to the ROC curve results, LIMK1 may be a potential marker for differentiating tumor tissues from normal tissues in hepatocellular carcinoma. According to Kaplan-Meier curve and subgroup analysis, when LIMK1 is highly expressed, the OS of hepatocellular carcinoma patients is shortened. In addition, the expression of LIMK1 in HCC may be related to tumor immune in ltration.
In recent years, there have been many studies on the carcinogenic effects of LIMK1 in human tumors such as pancreatic tumors, stomach tumors and HCC [23][24][25]. In addition, the increased expression of LIMK1 in various cancers has a certain correlation with the poor prognosis of the tumor [11,26]. However, the expression of LIMK1 in patients with HCC and its correlation with the prognosis have not yet been fully studied. In this study, the pan-cancer analysis of the TCGA database was rst used, and the results obtained were consistent with the reported results of over-expression of LIMK1 mRNA in various tumors such as lung tumor and gastric tumor. At the same time, based on paired data analysis, we also con rmed that the expression of LIMK1 is signi cantly increased in hepatocellular carcinoma. Our research results are consistent with those previously obtained by Guo et al. [14]. Correlation analysis shows that there is a certain correlation between the expression of LIMK1 and high TNM staging, and the results show a positive correlation [27]. These ndings show that LIMK1 may be related to the prognosis of hepatocellular carcinoma and can be used to predict the adverse prognosis of HCC.
At present, the role of LIMK1 in HCC has not been fully reported. According to previous research results, miRNA-27-3p and miRNA-520-3p are the targets of LIMK1, and they can inhibit tumor development [14,27,28]. At the same time, studies of potential mechanisms have shown that LIMK1 has a certain in uence in tumor progression [23,29]. Through previous research, we speculate that LIMK1 may be a potential substance or new target that affects HCC [30]. Since the expression of LIMK1 mRNA is signi cantly increased in HCC, we speculate that LIMK1 may have a role in predicting the prognosis of HCC. We use the ROC curve to study the role of LIMK1 in the diagnosis of HCC. ROC studies have shown that LIMK1 has a high AUC value in the diagnosis of hepatocellular carcinoma, and the ROC curve shows that its sensitivity is 80.7% and speci city is 86%. Based on our ndings, we believe that LIMK1 can be used to distinguish HCC tissues from normal tissues. Given that the up-regulation of LIMK1 is positively correlated with high TNM, we speculate that LIMK1 may promote the progression of hepatocellular carcinoma. In addition, when the TNM stage is higher, the patient's survival rate is lower. We speculate that the up-regulation of LIMK1 is a potential marker of a poorer prognosis for patients. In addition, the results of K-M curve and log-rank test showed that compared with patients with low LIMK1 expression, patients with hepatocellular carcinoma with high LIMK1 mRNA expression had a lower survival rate. Based on our results, we speculate that LIMK1 can be used to predict the prognosis of patients with HCC.
LIMK1 is an important part of the Rac1/PAK1/LIMK1 signaling pathway and is related to a variety of cancers [31]. For example, in esophageal tumors, miR-384 can regulate this pathway to promote cell apoptosis in esophageal tumors and increase the sensitivity of patients to chemotherapy [32]. Studies have shown that inhibiting Rho GDP dissociation inhibitor 2 in gastric cancer can inhibit tumor progression by regulating the LIMK1 signaling pathway [30,33]. Research by Lu et al. showed that the expression of LIMK1 has an effect on the expression of Rac1 and PAK1 [34]. Based on the research results, we believe that the overexpression of LIMK1 will have in uence of the expression of the entire pathway of Rac1/PAK1/LIMK1. However, this result should be veri ed in other experiments.
Through the TISIDB database, we found that the expression of LIMK1 in HCC is correlated with some tumor in ltrating immune cells (Tcm-CD8 + T cells, Tcm-CD4 + T cells, Treg cells, Act-CD4 cells, CD56bright cells). At the same time, through the Xiantao platform, we also observed that the expression of LIMK1 has an effect on T, B, NK, Th1, Th2 and CD56bright cells. These ndings indicate that LIMK1 may affect the immune in ltration function of patients with hepatocellular carcinoma to play a role. In the future, we will design further studies to verify this relationship.
However, this study has several aws. First of all, this study only used online public databases such as TCGA and TISIDB to study the effect of LIMK1 expression on the prognosis of hepatocellular carcinoma. However, these results need to be veri ed by further clinical studies. Second, there is a lack of relevant experiments to verify the mechanism of LIMK1's effect on hepatocellular carcinoma immune in ltration.
In conclusion, through this study, we found that the expression of LIMK1 mRNA in HCC was up-regulated, and it was positively correlated with high TNM staging. At the same time, our research shows that LIMK1 may be a potential target for predicting the poor outcome of hepatocellular carcinoma patients, and may play a certain role in immune in ltration.

Declarations
Chen, Xuewen Wang andHaoxian Gou contributed to photo editing and data analysis; all authors reviewed the manuscript.