Identification of death-associated protein-like 1 (DAPL1) as a novel prognostic biomarker of breast cancer

Background: It is meaningful to identify the potential clinical prognosis-associated oncogenes for cases with breast cancer, considering the complicated pathogenesis of breast cancer. Methods: We first utilized the bioinformatics approach to investigate the role of the DAPL1 (death-associated protein-like 1) in breast cancer, based on the available datasets of TCGA and GEO. Results: DAPL1 is lowly expressed in breast cancer tissues compared with the normal tissues. For the breast cancer cases of the TCGA-BRCA cohort, we observed a correlation between lowly expressed DAPL1 gene and poor clinical prognosis of overall survival ( P =0.0028). Based on the survival data of GEO, the low DAPL1 expression was associated with a poor prognosis of distant metastasis free survival ( P =0.0023), and relapse free survival ( P =0.0065). DAPL1 expression was linked to the mutation status or copy number variation o f several genes, such as MAP3K1 , NUP98 , and CCDC59. The infiltration level of immune cells (e.g., M1 macrophage, Follicular B helper T cells, etc.) may be involved in the etiology of breast cancer. Based on the DAPL1 -correlated genes, GSEA, GO, and KEGG analysis data indicated the association between DAPL1 expression and a series of biological issues, such as DNA packaging complex, DNA repair complex, nucleotide excision repair, ubiquitin-like protein binding, and ubiquitin proteasome pathway. We also identified several DAPL1 -associated phosphorylation kinases, such as MAPK, PRKACA, and GSK3B. Conclusions: DAPL1 gene is first identified as a prognosis biomarker of breast cancer, and the underlying molecular mechanism involves protein phosphorylation, immune cell infiltration, and DNA repair or protein ubiquitin-associated cellular pathways. is meaningful to explore the molecular mechanisms of breast cancer-related oncogenes or epigenetic biomarkers. In this study, we focused on the potential mechanism of the death-associated protein-like 1 (DAPL1) gene in the prognosis or pathogenesis of breast cancer.

The death-associated protein-like 1 (DAPL1) protein, encoded by DAPL1 gene, was reported to be associated with the pathogenesis of age-related macular degeneration (AMD) [2] and the regulation of cell proliferation in the retinal pigment epithelium (RPE) [3,4]. A whole genome expression profiling data suggested that DAPL1 gene may be linked to the chewing-tobacco-associated oral cancer [5]. However, there are still no reports investigating the potential effect of the DAPL1 gene in the prognosis of human breast cancer diseases. As a public funded project, TCGA (The Cancer Genome Atlas) archives the multiple-genomics data from more than thirteen types of cancer, including expression level, genetic alteration, genome methylation, and clinical information, etc. [6,7]. In addition, GEO (Gene Expression Omnibus) molecular data sets also provide a large number of clinical cancer-related gene expression data [8,9]. These available datasets help to identify the prognosis-associated oncogenes for breast cancer cases [10].
In the present study, we first identified that the DAPL1 is lowly expressed in breast cancer, mainly based on the data of TCGA and GEO. Also, we explored the possible molecular mechanisms of DAPL1 in breast carcinogenesis from the perspectives of gene expression, genetic alteration, DNA methylation, immune cell infiltration, and enrichment analysis of DAPL1-correlated genes.

Expression analysis
We first obtained the basic information of the DAPL1 gene through Genecards (https://www.genecards.org/) and analyzed the by the Ensembl (http://asia.ensembl.org/index.html).
Then, we analyzed the expression profile of the human DAPL1 gene in the breast cancer tissues and corresponding control tissues in the TCGA-BRCA (Breast invasive carcinoma) cohort by the GEPIA 2 [http://gepia2.cancer-pku.cn/#analysis] [11]. Besides, we investigated the expression difference of the DAPL1 gene between normal tissue and breast invasive lobular carcinoma tissues in the dataset of Radvanyi Breast [12], through Oncomine databases [https://www.oncomine.org/resource/login.html]. In addition, we utilized the LinkedOmics tool [13] to analyze the association between DAPL1 expression and a series of clinical factors, including the intrinsic subtype, pathological stage, ER (estrogen receptor) status, HER2 (human epidermal growth factor receptor-2) status, PR (progesterone receptor) status, race, ethnicity, and radiation therapy [13]. Immunohistochemistry analysis data on the expression of DAPL1 in normal tissue and breast duct or lobular carcinoma tissue were obtained from the online HPA (Human Protein Atlas) database (https://www.proteinatlas.org/pathology).

Survival curve analysis
Based on the data of the TCGA-BRCA cohort, we performed the OS (overall survival) analysis by the UALCAN web-portal (http://ualcan.path.uab.edu/) [14]. The factors of cancer type, race, menopause staus were included as well for the survival analyses. We utilized the Kaplan-Meier plotter (http://kmplot.com/analysis/) to perform the OS, DMFS (distant metastasis free survival), RFS (relapse free survival), PPS (post progression survival) by the expression level of the DAPL1 gene in a total of 1809 breast cancer cases within the Affymetrix HGU133A and HGU133+2 microarrays of GEO [15].
The auto-select best cutoff was set. The clinical factors, including ER status, PR status, HER2 status, TP53 (Tumor Protein P53) status, grade, intrinsic subtype, lymph node status, and pietenpol subtype, were also considered in the subgroup analysis.

DNA methylation analysis
We analyzed the DNA methylation status of DAPL1 in the breast cancer cases of the TCGA-BRCA cohort through the MEXPRESS [16,17]. We also utilized the UALCAN web-portal (http://ualcan.path.uab.edu/) [14] to analyze the association between the gene expression and promoter methylation of DAPL1 between the normal and primary breast cancer tissues. The factors of pathological stage and intrinsic subtype were considered. In addition, the waterfall plot of the methylation level of DAPL1 gene was obtained through the MethSurv database (https://biit.cs.ut.ee/methsurv/) [18].

Genetic alteration analysis
Based on the cBioPortal database (https://www.cbioportal.org/) [19,20], we analyzed the genetic alteration features of DAPL1 in the TCGA-BRCA cohort, respectively. The overall survival and disease/progression-free survival analyses were conducted by the genetic alteration of DAPL1 in the total cancer cases of the TCGA project. In addition, we performed the Wilcoxon test or Spearman test to analyze the correlation between the DAPL1 expression level and the mutation or copy number variation (CNV) status of genes for the breast cancer cases of TCGA-BRCA, through the LinkedOmics (https://www.biostars.org/p/287820/) [18].

Immune cell infiltration analysis
We utilized the GEPIA 2 approach to conduct the pair-wise gene correlation analysis between DAPL1

DAPL1-correlated gene enrichment analysis
We performed the cluster analysis of the DAPL1-correlated significant genes, through LinkedOmics.
The heat map and association plot targeting the DAPL1 positively or negatively correlated significant genes were provided, respectively. We also analyzed the expression correlation between DAPL1 and the selected genes, through the cBioPortal database. In addition the data of GO (Gene ontology) enrichment analysis, KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analysis and GSEA (Gene Set Enrichment Analysis) profiles were provided, respectively.

DAPL1-correlated kinase analysis
We first utilized the PhosphoSitePlus (https://www.phosphosite.org/homeAction.action) to predict the potential phosphorylation sites of DAPL1. Then, the features of these phosphorylation sites (e.g. maximum kinase specificity, hydrophobicity, or P-site similarity score, etc.) was analyzed by the PhosphoNET tool (http://www.phosphonet.ca/). The potential kinases were measured by the KinasePhos tool [21]. In addition, we also performed an enrichment analysis of kinase pathway, based on the DAPL1-correlated genes.

Expression analysis data
The human DAPL1 gene is located on chromosome 2q24.1 (Fig. 1). Fig. S1 shows the evolutionary tree of the DAPL1 gene in different species. Our study aims at exploring the functional links of the human DAPL1 gene with breast cancer from different perspectives. We first performed gene expression analysis. As shown in Fig. 1B, based on the dataset of the TCGA-BRCA cohort, we observed a lower expression level of DAPL1 in the tissues of breast invasive carcinoma (P<0.01), compared with the normal tissues. A similar result was detected in the DAPL1 expression analysis of Radvanyi breast dataset (Fig. 1C, P<0.01). The immunohistochemical result further showed that the expression level of DAPL1 protein was higher in the normal tissues than that in the tissues of breast duct carcinoma and lobular carcinoma (Fig. 1D). Besides, the DAPL1 gene showed the different expression levels among different intrinsic subtypes and pathologic stages (Fig. 1E, P<0.01). Nevertheless, there is no significant statistical correlation between DAPL1 expression and some other factors of ER status, HER2 status, PR status, race, ethnicity, and radiation therapy (Fig. 1E). Therefore, the above suggested the potential role of the DAPL1 gene in the etiology of breast cancer.

Survival curve analysis data
Subsequently, we performed a correlation analysis between DAPL1 gene expression and the clinical overall survival outcome of breast cancer patients in TCGA-BRCA cohort. As shown in Fig (Table   1, P=0.04, HR=0.72) and DMFS (Table S1, P=0.011, HR=0.48), but a better prognosis of OS (Table S2, P= 0.0042, HR=2.84) and PPS (Table S3, P= 0.0016, HR=2.87) for the breast cancer patients with pathological grade 3. These results provide evidence regarding the association between the DAPL1 expression and clinical outcomes of breast cancer.

DNA methylation analysis data
We attempted to exploit the potential molecular mechanism from the point of DAPL1 DNA methylation based on the methylation data of TCGA-BRCA. Fig. 3A presents the waterfall plot of DAPL1 DNA methylation levels for each breast cancer case. We did not observe the statistically significant difference of the promoter methylation levels of DAPL1 between the normal and tumor tissues, or among the different pathological analyses and tumor types (Fig. 3B, all P>0.05). We also observed a similar negative result using the MEXPRESS method (Fig. S2). Thus, DNA methylation is not involved in the molecular mechanism of DAPL1 in the pathogenesis of breast cancer.

Genetic alteration analysis data
Based on the TCGA-BRCA data, we explore the possible molecular mechanisms from the perspective of genetic alteration. As shown in Fig. 4A, we only observed a genetic alteration frequency of ~ 0.5% for the breast ductal carcinoma cases of TCGA. We also did not observe a correlation between DAPL1 genetic alteration and the prognosis of overall survival (Fig. 4B, P=0.992), disease/pression free survival (Fig. 4C, P=0.317) for the total cancer patients. We then analyzed the correlation between DAPL1 gene expression and different mutation status of specific genes. As shown in Fig. 4D, compared with the wildtype status, the DAPL1 gene was lowly expressed in the mutated genes of MAP3K1, NUP98, CEP152M, and ZNF300 (P<0.001), whereas it was highly expressed in the mutated KIF2IB gene (P=0.003). Furthermore, we observed that the expression of DAPL1 gene was negatively correlated with the copy number variation (CNV) of the CCDC59 gene (Fig. 4E, R=-0.18, P<0.0001).

Immune cell infiltration analysis data
Also, we aimed to investigate whether the DAPL1 gene is involved in the etiology of breast cancer through immune cell infiltration. As shown in Table 2
The GO analysis data of Fig. 6 identified the DAPL1-associated cellular component (e.g. DNA packaging complex, methyltransferase complex, etc.), molecular function (e.g. ubiquitin-like protein binding, ribonocleoprotein complex binding, etc), and biological process (e.g. ribonucleoprotein complex localization, translational elongation, etc.). KEGG pathway analysis data further showed a series of pathways, such as ubiquitin mediated proteolysis, RNA transports, cell adhesion molecules (CAMs), chemical carcinogenesis, and so on (Fig. 7A). In addition, GSEA data also suggested the correlation between DAPL1 expression and the biological issuses of DNA repair complex, nucleotide excision repair, translational elongation, ubiquitin-like protein binding, ubiquitin proteasome pathway, and proteasome (Fig. 7B).

DAPL1-correlated kinase analysis
After the prediction of potential phosphorylation sites and kinases, five sites of human DAPL1 protein [NP_001017920.2], namely S10, T35, T77, T86, and Y99, were identified (Fig. 8A). And the CDC2, MAPK were predicted as the potential catalytic kinases of DPAL1 protein at S10 site, while MAPK and PKA were for the T86 site (Fig. 8A). In addition, we performed an enrichment analysis of kinase pathway, and also identified several DAPL1-correlated kinases, such as PRKACA, PRKCA, GSK3B, ROCK, and CHEK1 (Fig. 8B).

Discussion
In this study, we focused on the correlation between the DAPL1 gene and breast cancer for the first time. We observed the low expression of DAPL1 in invasive breast cancer tissue, from the levels of both gene and protein, suggesting that DAPL1 may be a kind of tumor suppressor gene. It was found that DAPL1 gene was downregulated in an in vitro cultured ovarian granulosa cell of pig [22], but upregulated in the endometrial stromal and epithelial cells under the treatment of lipopolysaccharide [23]. DAPL1 gene was found to be upregulated in nine hepatoblastomas patients [24]. Based on the expression dataset of TCGA-LIHC (Liver hepatocellular carcinoma) cohort, we also observed an enhanced DAPL1 gene expression in hepatocellular carcinoma tissues, compared to normal tissues (data not shown). Therefore, the low expression characteristics of the DAPL1 gene in breast cancer may be tumor-specific.
Next, we explored the correlation between DAPL1 expression status and clinical prognosis of breast cancer cases, based on the survival datasets of TCGA and GEO databases, respectively. As stated above, the low expression of DAPL1 is correlated with the poor OS prognosis of the breast cancer patients in the TCGA-BRCA cohort. Besides, the poor prognoses of distance metastasis free survival and relapse free survival, but a good post progression survival prognosis, were observed in the breast cancer cases of GEO with the lowly expressed DAPL1 gene. Considering the impact of clinical information on survival analysis, we performed a series of subgroup analyses, and observed the distinct association conclusions. For the OS prognosis, we observed that low expression of DAPL1 was correlated with the poor OS prognosis of breast cancer patients with pathological grade 1, but a good prognosis of breast cancer patients with pathological grade 3. We also noticed that the sample size in grade 1 group was relatively limited, containing only 26 patients. More breast cancer cases and more detailed clinical information are needed to further investigate the complexity of the correlation between DAPL1 expression and clinical prognosis.
Considering the association between DNA methylation and breast cancerogenesis [25][26][27], we want to study whether DNA hypermethylation of DAPL1 is associated with the low expression level of DAPL1 in breast cancer tissues. Our finding showed no obvious methylation modification of DAPL1 and the lack of the correlation between promoter methylation status and gene expression level. In addition, for the breast cancer cases of the TCGA-BRCA cohort, DAPL1 did not exhibit a high-frequency of genetic alteration status. Due to the low mutation rate, we performed the survival analysis of TCGA patients with overall tumors and observed the negative conclusion. These suggested that DNA methylation and genetic alteration are not involved in the role of DAPL1 in breast cancer pathogenesis. Even though this, for the first time, we found that the expression of DAPL1 gene is related to the mutation status of several genes (e.g. MAP3K1, NUP98, and KIF2IB, etc), hinting that the genetic alteration of these genes may be a potential upstream regulatory mechanism for DAPL1 expression in breast cancer.
The infiltration of immune cells is also reported to be associated with the occurrence of breast cancer [28][29][30]. Our preliminary analysis showed a positive correlation of the DAPL1 gene expression with the infiltration level of M1 macrophage, NK cell, B cell, Tfh, naive T cell, effector T cell, effector memory T cell, central memory T cell, resident memory T, but a negative correlation with resting or effect Treg T cell. Although we observe a statistically significant correlation, the relatively small R-value means a weak correlation. The factor of immune cell infiltration may only be involved, but not the most critical issue affecting the onset of breast cancer.
In the present study, we screened out a number of DAPL1-correlated genes in breast cancer tissues, and then performed a series of enrichment analyses. Our data indicated that the protein ubiquitination and DNA repair may be the essential mechanisms of DAPL1 gene in the pathogenesis of breast cancer, providing an analytical perspective for further investigation. Additionally, we tried to conduct an in-depth analysis from the perspective of protein phosphorylation. Unfortunately, we did

Conclusion
In summary, based on the breast cancer cases within TCGA and GEO datasets, we first identified the lowly expressed oncogenic gene DAPL1 as a prognostic gene for breast cancer. Our finding also suggested the involvement of protein phosphorylation, immune cell infiltration, and DNA repair or protein ubiquitin-associated cellular pathways in the role of DAPL1 in the pathogenesis of breast invasive carcinoma.

Funding
Not applicable.

Availability of data and materials
The data of this manuscript can be download from The Cancer Genome Atlas database (https://portal.gdc.cancer.gov/).

Ethics approval and consent to participate
Not applicable.

Consent for publication
Not applicable.     Genetic alteration analysis of DAPL1. We analyzed the genetic alteration features of DAPL1 (A), and perform the OS (B) and disease/progression-free survival (C) analyses, based on the total cancer data of the TCGA project, respectively. We also analyzed the correlation between the DAPL1 expression level and the mutation (D) or CNV (E) status of genes for the breast cancer cases.

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download. Figure S2.tif Table S1-S3.docx Figure S1.tif