The Potential Dual Role of H2.0-like Homeobox in the Tumorgenesis and Development of Colorectal Cancer and Its Prognostic Value

Background H2.0-like homeobox (HLX) is highly expressed in several hematopoietic malignancies. However, the role of HLX in the carcinogenesis and progression of colorectal cancer (CRC) patients has rarely been reported. Methods In this study, the data were collected from The Cancer Genome Atlas and Gene Expression Omnibus databases. The diagnostic value of HLX was analyzed by the R package “pROC.” The overall survival was estimated using the “survival” and “survminer” packages. A nomogram was established to predict 1-, 3-, and 5-year overall survival of CRC patients. The CIBERSORT software was employed to calculate the relative proportions of 22 immune cells. Results HLX expression was downregulated in CRC patients. Remarkably, HLX expression was increased with stage (stage I–stage III) of CRC, and the CRC patients with high HLX expression exhibited a poor prognosis. The promoter methylation level of HLX was prominently increased in CRC samples compared to paracancerous samples. We also found that the six miRNAs target HLX genes, leading to its downregulation, and HLX expression had a negative correlation with its downstream target gene BRI3BP in both CRC and normal samples. Finally, we found that the 12 immune infiltrating cells were observably different between high and low HLX expression groups. The HLX had a significant positive correlation with 8 immune checkpoints (PD-1 (PDCD1), CTLA4, PDL-1 (CD274), PDL-2 (PDCD1LG2), CD80, CD86, LAG3, and TIGIT) expressions. Conclusion HLX probably played a carcinostasis role in the early stages of CRC but exhibited a cancer-promoting effect in the advanced stages. Meanwhile, HLX could serve as a reliable prognostic indicator for CRC.


Introduction
Colorectal cancer (CRC) is one of the leading causes of tumor-related death in the world, and its mortality accounts for 9.4% of cancer deaths worldwide [1,2].In 2020, approximately 1.9 million new cases of CRC were diagnosed, and about 900,000 deaths occurred in the world [2].Te morbidity of CRC has been continually rising in many medium or high human development index countries, such as South Eastern Europe, South America, and Eastern Europe [3].In addition, about 25% of CRC patients have developed metastatic disease at initial diagnosis, and almost 30% of CRC patients with early-stage disease eventually develop metastatic disease [1,4,5].Te most common metastatic site of CRC was the liver, followed by the lung, distant lymph nodes, and peritoneum [6].It has been documented that aberrant expression of genes is involved in the development and progress of CRC.Wang et al. have revealed that TUG1 knockdown could inhibit the proliferation, invasion, and migration of CRC cells in vitro [7].Te mutations of KRAS, p53, SMAD4, and BRAF increase the risk of distant metastasis of CRC [8].Accordingly, further exploration of the molecular mechanisms of CRC may contribute to diagnosis and treatment of CRC at an early stage.
H2.0-like homeobox (HLX) belongs to the family of NKL homeobox genes.NKL homeobox genes are key regulators of essential processes, such as diferentiation, proliferation, and apoptosis, and their expression is cell type-specifc [9].HLX is highly expressed in hematopoietic progenitors and lowly expressed in activated lymphocytes [10], and it is a downstream mediator of hepatocyte growth factor (HGF)/ c-met induction of cell survival, cell proliferation, and trophoblast migration [11].Murthi et al. have found that the low HLX1 expression is associated with abnormal placental development in idiopathic fetal growth restriction [12].Previous studies have shown that HLX is correlated with the initiation and progression of tumor.In gastric cancer, the low HLX expression is tightly associated with the expression of T-bet and RUNX3 [13].Liu et al. have indicated that the level of HLX1 mRNA is remarkably decreased in hepatocellular carcinoma tissues compared to adjacent nontumorous tissues and downregulation of HLX1 could promote the invasion, migration, and proliferation of HCC cells [14].In addition, HLX is highly expressed in acute myeloid leukemia (AML) [15,16].In zebrafsh and human hematopoietic stem progenitor cells (HSPCs), HLX overexpression could block myeloid diferentiation by regulating metabolic pathways in hematopoietic cells [17].However, to our knowledge, the potential role of HLX in CRC has rarely been reported.
Tus, in this study, we explored the role of HLX expression in carcinogenesis and the progression of CRC through the methods of bioinformatics research.Our study could provide valuable information for researchers regarding diagnosis and treatment of CRC at an early stage.

Data Collection.
Te mRNA expression profling data of 623 CRC patients (colon adenocarcinoma (COAD) and rectum adenocarcinoma (READ)) along with the corresponding clinical information were collected from Te Cancer Genome Atlas (TCGA, https://tcga-data.nci.nih.gov/tcga/) database.Tese 632 patients included a total of 638 CRC samples and 51 paracancerous samples, and 590 patients contained complete survival information.Te methylation 450K chip data of 408 CRC samples (COAD + READ) were downloaded.

Analysis of the Diagnostic Value of HLX.
Te TCGA cohort was employed to make a receiver operating characteristic (ROC) curve through the R language (Version 4.2.1, the same below) function package pROC to analyze the diagnostic value of the HLX gene.

Survival Analysis.
Te R language survival package and survminer package were used to estimate the overall survival of diferent groups based on the Kaplan-Meier method, and the log-rank test was used to test the signifcance of diferences in survival between diferent groups.
Te multivariate Cox regression model was used to analyze whether the target genes could predict the survival of CRC patients.

Gene Set Enrichment Analysis (GSEA).
In the TCGA cohort, the samples were split into HLX high and low expression groups according to the median expression of HLX using the R language limma function package.Ten, the diferentially expressed genes (DEGs) between the two groups were screened based on the |log2FC| > 0.5 and p. adjust <0.05.Te DEGs were subjected to GSEA using the R language function package ReactomePA and ClusterProfler [18,19].

Establishment of the Nomogram Prediction Model.
To predict 1-, 3-, and 5-year overall survival of CRC patients, the R language rms (https://CRAN.R-project.org/package=rms) package was used to establish the nomogram for all independent prognostic factors identifed by multivariate Cox regression analysis.Te calibration curve of the nomogram was plotted, and the relationship between the nomogram predicted probability and actual incidence was observed.For each patient, three lines were drawn upward to determine the points obtained from each factor in the nomogram.Te sums of these points were located on the "Total Points" axis, and a line was then drawn from the total points axis to determine the probability of a 1-, 3-, and 5-year survival rate for CRC patients.

Immune Cell Infltration.
Te software CIBERSORT [20] was employed to calculate the relative proportions of 22 immune cells in the samples.CIBERSORT software characterizes the composition of immune infltrating cells according to gene expression matrices using a deconvolution algorithm based on a preset set of 547 barcode genes.Te immune scores of the samples were calculated using the "estimate" function package.

Statistical Analysis.
Te Wilcoxon rank sum test was used to compare gene expression diferences and infltration diferences of immune cells among diferent groups using the UALCAN online database [21].p < 0.05 was considered statistically signifcant.All the above statistical analyses were performed using the R software.

HLX Expression Was Closely Correlated with Carcinogenesis and Progression of CRC.
First, in the TCGA cohort, we analyzed the expression of 11 genes (DBX1, DBX2, BSX, BARX1, BARX2, BARHL1, BARHL2, LBX1, LBX2, HLX, and HHEX) which belong to the subfamily in which HLX is located within the NKL family.As shown in Figure 1(a), the HLX expression was prominently downregulated in CRC samples.In the GSE41258 dataset, HLX expression was also decreased in CRC samples (Figure 1(b)).In addition, we analyzed the expression of HLX in CRC and normal samples 2 Canadian Journal of Gastroenterology and Hepatology in the CELL cell line database.Te results unequivocally demonstrated a remarkable and substantial downregulation of HLX expression in CRC cell lines, which was visually represented in Figure 1(c).ROC curves showed that HLX might serve as a potential diagnostic marker for CRC in the TCGA cohort (AUC � 0.832, Figure 1(d)).Te CRC patients with high HLX expression exhibited poor prognosis in both the TCGA and GSE17538 cohorts (Figures 1(e) and 1(f )).Moreover, we found that the HLX expression increased as the stage increased from stage I to stage III (Figures 1(g) and 1(h)).Tese results indicated that the HLX expression was closely correlated with carcinogenesis and progression of CRC.

Nomogram Model Could Efectively Predict the Prognosis of CRC Patients.
Multivariate Cox regression analysis including gender, age, stage, race, BRAF mutation, and KRAS mutation showed that HLX genes could be used as prognostic factors in CRC (Figure 2(a)).In addition, three independent prognostic factors, HLX, age, and TNMstage, were used to construct the nomogram model (Figure 2(b)).In the TCGA cohort, the AUC of 1-, 3-, and 5-year overall survival was 0.58, 0.62, and 0.54, respectively (Figure 2(c)).Te results suggested that the nomogram model could effectively predict the prognosis of CRC patients.Meanwhile, the corrected curve in the calibration plot was relatively close to the ideal curve (a 45-degree line passing through the origin of the coordinate axis and with a slope of 1), indicating that the prediction agreed well with the real outcome (Figures 2(d)-2(f )).

Increased Promoter Methylation Repressed the Expression of HLX in CRC Patients
. Furthermore, we analyzed the promoter methylation level of HLX in TCGA-COAD and TCGA-READ using the online analysis tool UALCAN (http://ualcan.path.uab.edu/)[21].As shown in Figures 3(a) and 3(b), the promoter methylation level of HLX was prominently increased in cancer tissues compared to paracancerous tissues in COAD and READ patients.Remarkably, compared to paracancerous tissues, the promoter methylation level of HLX was also prominently increased in diferent cancer development stages and diferent lymph node metastases of COAD and READ (Figures 3(c Te TCGA CRC methylation data were used to explore which sites in the promoter region of HLX that increased methylation level would afect the silencing of the HLX gene.Te 18 cg sites with high methylation levels were screened by methylation level mean >0.5 and median >0.5.Further screening was performed according to the diferent sites methylation levels and HLX expression correlation absolute values >0.3 and p < 0.05.Te results showed that eight methylation sites were associated with the downregulation of HLX expression (Figures 3(g)-3(n), Table 1).
Subsequently, the GRNdb (https://www.grndb.com/)database was used to inquire HLX's target genes, and results were visualized by Cytoscape (Figure 4(b)).We found a possible binding site for the transcription factor HLX around 20 bp upstream of the BRI3BP promoter (Table S1) via FIMO (https://meme-suite.org/meme/tools/fmo).We also found that the HLX expression had a negative correlation with BRI3BP in both CRC samples and normal samples (Figures 4(c) and 4(d)).Te results suggested that the HLX negatively regulated the BRI3BP in CRC.

Te Result of GSEA.
Te GSEA showed that 138 pathways were prominently activated, such as the PI3K-Akt signaling pathway, Rap1 signaling pathway, pathways in cancer, JAK-STAT signaling pathway, and toll-like receptor signaling pathway (Figures 5(a)-5(e)), and 8 pathways were inhibited in the HLX high expression group compared to the HLX low expression group (Table S2).Te top 10 signifcantly enriched pathways are displayed in Figure 5(f ).

HLX Involved in the Immune Cell Infltration in CRC
Patients.Te relative proportions of 22 immune infltrating cells in CRC samples in the TCGA cohort were calculated by the CIBERSORT algorithm (Figure 6(a), Table S3).Te 12 immune infltrating cells (B.cells.naive,Plasma.cells,T.cells.CD8, T.cells.CD4.naive,T.cells.CD4.memory.activated,T.cells.follicular.helper,Macrophages.M0, Macrophages.M1, Macrophages.M2, Dendritic.cells.activated,Mast.cells.resting,and Mast.cells.activated)were signifcantly diferential between HLX high and low expression groups (Figure 6(b)).To avoid possible errors, we additionally recalculated the content of immune infltrating cells in CRC samples from the TCGA cohort using the Xcell method.Te results showed that the contents of B.cells.naive (Figure 6(c)) and T.cells.CD8 (Figure 6(d)) were signifcantly diferent between the high and low HLX expression groups, and the trend was consistent with that found using the CIBERSORT algorithm.
Next, we used the results of CIBERSORT to further analyze the Spearman correlation between the HLX and the 12 signifcantly diferent immune infltrating cells.Te results showed that HLX expression exhibited a negative correlation with Plasma.cells,T.cells.CD8, T.cells.

Discussion
In this study, we explore the association between HLX and carcinogenesis and progression of CRC by systematically collecting and analyzing the clinical and genomic data of CRC patients.We found that HLX was downregulated in CRC samples compared to paracancerous samples, and HLX expression increased as stage increased from stage I totage III.Te CRC patients with high HLX expression exhibited a worse prognosis.We also found that the downregulation of HLX was regulated by six miRNAs, and the HLX negatively regulated its downstream target gene BRI3BP in CRC.
Previous studies have demonstrated that HLX is abnormally expressed in multiple cancers.For example, in anaplastic large cell lymphoma and difuse large B-cell lymphoma, the HLX was overexpressed in cancer tissues 1 year survival 0.5 0.7 0.9 0.9 3 year survival 0.3 0.5 0.7 0.9 0.9 5 year survival 0.05 0.1 0.3 0.5 0.7 0.9 0.9  6 Canadian Journal of Gastroenterology and Hepatology [22,23].In this study, we found that HLX expression was prominently downregulated in CRC samples.Kawahara et al. have reported that the HLX was highly expressed in 87% of acute myeloid leukemia patients, and patients with high HLX expression had an inferior prognosis [24].In addition, the inhibition of HLX expression could reduce the proliferation and clonogenicity of leukemia cells and prolong the survival rate [24].Zhu et al. have indicated that low HLX expression could reduce the proliferation of acute myelogenous leukemia cells by regulating the JAK/STAT signaling pathway [25].Tis evidence suggested that HLX downregulation might be involved in the progression of some tumors.In this study, we also found that the HLX expression was increased with tumor stage (stage I-stage III) in CRC, and the high expression of HLX was correlated with poor prognosis of the CRC patients.Tese results showed    Promoter DNA methylation generally represses the transcription expression by regulating the binding of transcription factors [26], and DNA methylation is a critical epigenetic process that contributes to the progression of CRC [27].Since HLX was downregulated in CRC samples, we analyzed whether the promoter methylation level of HLX was responsible for HLX gene silencing.We found that the promoter methylation level of HLX was prominently increased in cancer tissues compared to paracancerous tissues in CRC patients.Tese results suggested that the downregulation of HLX in CRC was associated with its promoter methylation level.Moreover, miRNAs could lead to degradation or translational repression of the mRNA by targeting the form of complementary mRNAs [28].In difuse large B-cell lymphoma, the expression of HLX was regulated by EBV-mediated STAT3 activation [29], and STAT3 was also confrmed to regulate the expression of HLX in Hodgkin lymphoma [30].In this study, we found that the downregulation of HLX was regulated by six miRNAs, and the HLX negatively regulated its downstream target gene BRI3BP in CRC.BRI3BP was mapped to chromosome 12q24.2-qter in humans, and it was highly expressed in the brain, liver, and kidney, and BRI3BP might be involved in apoptosis [31].BRI3BP overexpression promoted apoptosis in 293T cells (human embryonic kidney) challenged with etoposide (an anticancer agent).In addition, it has been reported that BRI3BP expression is decreased in cancer samples [32].Tus, we infer that HLX might regulate the expression of BRI3BP to be involved in the carcinogenesis and progression of CRC.

Canadian Journal of Gastroenterology and Hepatology
GSEA showed immune-related signaling pathways were signifcantly activated in the HLX high expression group compared to the HLX low expression group, such as the PI3K-Akt signaling pathway and the Rap1 signaling pathway.Tese fndings might inform the development of targeted therapies that could improve CRC patient outcomes.Subsequently, we investigated the efect of HLX on immune cell infltration in CRC.We found that the HLX expression exhibited a negative correlation with Plasma.cells,T.cells.CD8, T.cells.CD4.naive,T.cells.CD4.memory.activated, and T.cells.follicular.helperand had a positive association with B.cells.naive.It has been reported that the HLX expression is involved in the activation and growth of T cells [33].Te HLX overexpression could disturb the development of B cells and Tcells in murine lymphocyte [34] and inhibit CD4+T cells development and disrupt thymic involution in transgenic mice [35].Te above commendations indicated that the HLX probably regulates the infltration of B cells and T cells, thereby infuencing the prognosis of CRC patients.
Finally, we discovered that the HLX expression had a signifcant positive correlation with the expression of PD-1, CTLA4, PDL-1, PDL-2, and LAG3.Te interaction between PD-1 and its ligand PD-L1 suppresses T cell proliferation and cytokine release.As a result, PD-1 modulates immunological responses in reverse, allowing tumor cells to evade immune surveillance [36].Previous studies showed that PD-L1 expression was higher in metastatic CRC than in primary tumors [37].Wang et al. have demonstrated that the combination of fruquintinib and antiPD-1 could synergistically inhibit the progression of CRC and alter the tumor microenvironment in favor of anti-tumor immune responses [38].Regorafenib combined with antiPD-1 could enhance M1 macrophage diferentiation and activation and continuously inhibit Treg cell infltration to improve antitumor activity [39].CTLA-4 was an inhibitory immune checkpoint that was overexpressed in the CRC tissues and the CRC cell line SW480, and capecitabine treatment resulted in a signifcant downregulation of CTLA-4 expression in SW480 cells [40].LAG3 was also found to be overexpressed in microsatellite instability (MSI) tumors compared to microsatellite stability (MSS), making it an excellent target for immunotherapy in MSI CRC [41,42].It has been reported that the favezelimab (LAG-3 antibody) in combination with pembrolizumab has promising antitumor activity in CRC patients [43].Considering the positive correlation between HLX expression and PD-1, CTLA4, PDL-1, and LAG3 expressions in CRC patients, HLX might play a role in regulating immune checkpoints and infuencing the response to immunotherapy.Terefore, combining chemotherapy with immunotherapy approaches, such as PD-1 and CTLA-4 inhibitors, might be a promising strategy for improving the treatment of CRC.

Conclusion
Our study demonstrated that HLX probably exhibits dual roles at diferent stages of CRC; HLX might play a carcinostasis role in the early stage of CRC, but exhibit cancerpromoting efects in the advanced-stage.Moreover, the downregulation of HLX was regulated by six miRNAs and the HLX negatively regulated its downstream target gene   BRI3BP in CRC.Our fndings provide valuable insights into the molecular mechanisms underlying CRC development and progression and could potentially inform the development of more efective treatment strategies for CRC patients.

Figure 1 :
Figure 1: HLX expression was closely correlated with the prognosis of CRC patients.(a) Te 11 genes (DBX1, DBX2, BSX, BARX1, BARX2, BARHL1, BARHL2, LBX1, LBX2, HLX, and HHEX) expression in CRC samples.(b) HLX expression in CRC samples in the GSE41258 dataset.(c) Te expression of HLX in CRC and normal samples in the CELL cell line database.(d) Te curve for the diagnostic value of HLX.(e, f ) Te KM survival curves of HLX high and low expression groups in the TCGA and GSE17538 cohorts.(g, h) HLX expression in a diferent stage of CRC patients in TCGA and GSE41258 cohorts.

Figure 2 :
Figure 2: Nomogram model could efectively predict the prognosis of CRC patients.(a) Forest plot of multivariate Cox regression analysis.(b) Nomogram to predict the probability of 1-, 3-, and 5-year overall survival in CRC patients.(c) Curve of the ROC.(d-f ) Calibration curves of the nomogram to predict the probability of 1-, 3-, and 5-year overall survival in CRC patients.

Figure 3 :
Figure 3: Te CRC patients with a high promoter methylation level of HLX.(a, b) Te promoter methylation level of HLX in COAD and READ.(c-f ) Te promoter methylation level of HLX in diferent cancer development stages and diferent lymph node metastases of COAD and READ.(g-n) Te correlation of eight methylation sites with HLX mRNA.

Figure 4 :
Figure 4: Regulation of HLX translation by miRNAs.(a, b) Regulatory relationships visualized by the Cytoscape.(c, d) Te correlation of HLX with BRI3BP.

Figure 5 :
Figure 5: Te result of GSEA.(a-e) Te PI3K-Akt signaling pathway, Rap1 signaling pathway, pathways in cancer, JAK-STAT signaling pathway, and toll-like receptor signaling pathway were signifcantly activated in the HLX high expression group.(f ) Te top 10 signifcantly enriched pathways.

Table 1 :
Correlation of HLX and methylation in the TCGA cohort.