Systematic characterization of the expression, prognosis and immune characteristics of PLOD family genes in breast cancer

Oncomine

Oncomine (http://www.oncomine.org/) is an integrated data-mining platform that collects cancer microarray data [5]. Oncomine was used to analyze the expressions of PLOD family genes (PLOD1, PLOD2, PLOD3) in different cancer tissues compared with normal tissues. The thresholds were as follows: P-value < 1E-4, fold change >2, and gene rank top 10%.

cBio cancer genomics portal (cBioPortal)

cBioPortal (https://www.cbioportal.org/) for cancer genomics is a comprehensive and open accessible web resource that analyzes cancer genomic datasets, such as copy number variation, nonsynonymous mutations, mRNA and microRNA expressions, expressions and phosphorylation levels of proteins, DNA methylation, and clinical data [6].

Gene expression profiling interactive analysis (GEPIA)

GEPIA (http://gepia.cancer-pku.cn/) is a newly developed interactive web server that estimates mRNA expression data from TCGA and GTEx [7]. Here, GEPIA was used to validate the differential expressions of PLOD family genes in breast cancer and healthy donor samples.

Breast cancer gene-expression miner (bc-GenExMiner 4.6)

bc-GenGxMiner 4.6 (http://bcgenex.ico.unicancer.fr/) is a statistical mining tool based on published annotated breast cancer transcriptomic data including DNA microarrays and RAN-seq, and offers the possibility to explore gene expressions of genes of interest in breast cancer [8]. The expressions of PLOD family genes in different subtypes were analyzed using bc-GenGxMiner.

UALCAN

UALCAN (http://ualcan.path.uab.edu) is a comprehensive and interactive web resource based on TCGA database [9]. We analyzed the relative transcriptional expressions of PLOD family genes in breast cancer and normal samples, and the expressions of PLOD family genes with clinical stages P < 0.01 was considered significant.

PrognoScan

PrognoScan (http://www.prognoscan.org) is a new database for meta-analyzing the prognostic values of PLOD family genes in breast cancer. The recent availability of published cancer microarray datasets with clinical annotations facilitates the analysis of gene expressions for prognosis. P < 0.05 was regarded as meaningful.

GeneMANIA

GeneMANIA (http://www.genemania.org) is a common website for constructing protein-protein interactions [10]. The online tool analyzes genes or gene lists through gene co-expression, gene co-location, gene enrichment analysis, physical interaction, and web prediction. We predicted the functions of PLOD family genes and visualized gene networks using GeneMANIA.

Metascape

Metascape (http://metascape.org) is a well-maintained gene-list analysis tool for gene annotation and analysis using Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) tools [11]. We used Metascape to conduct progress and pathway enrichment analyses of the PLOD family genes and neighboring genes.

Tumor immune estimation resource (TIMER)

TIMER (https://cistrome.shinyapps.io/timer/) is a comprehensive analysis of tumor-infiltrating immune cells in various cancer and infers the abundance of tumor-infiltrating immune cells (TIICs) from TCGA [12]. We explored the expressions of the PLOD family genes of TIICs in breast cancer, including B cells, CD8+ T cells, CD4+ T cells, macrophages, neutrophils, and dendritic cells. In addition, we used immuneeconv to assess the reliable results of immune score evaluation. It is an R software package including MCP-counter.

Tissue microarrays in IHC

IHC is the gold standard for detecting the expressions of in situ protein biomarkers in formalin-fixed, paraffin-embedded tumor tissues. First, tissue microarrays were baked at 60°C for 12 h and dewaxed in xylene, followed by gradient descent alcohol hydration. Sodium citrate was used for 6 minutes of antigen retrieval at 121°C. Endogenous enzyme blocking reagents and nonspecific blocking reagents (Kit-9710, Maixin, China) were used for incubation with primary antibodies at 4°C overnight in a wet box: PLOD1 (29480-1-AP, 1:100, Proteintech, China), PLOD2 (66342-1-Ig, 1:500, Proteintech, China) and PLOD3 (11027-1-AP, 1:500) (all Proteintech, China). On the second day, the assays were incubated with donkey anti-mouse/rabbit secondary antibodies and Streptomyces anti-biotin proteinperoxidase (Kit-9710, Maixin, China) for 10 min each. Finally, chromogenic detection was performed using a DBA detection kit (Kit-2031, Maixin, China) and nuclei were stained with hematoxylin. Additionally, the assays were dehydrated in a gradient concentration of ethanol and xylene and sealed with a neutral resin. IHC staining was scored by two independent pathologists, and the results were presented as levels (0, none; 1+, weak; 2+, moderate; 3+, strong) and the percentage of positively stained cells (0: <5%; 1:6–25%; 2:26–50%; 3:51–75%; and 4:75%). Finally, the score was expressed as (staining intensity × percentage of positively stained cells).

Statistical analysis

Continuous variables were described as mean ± standard deviation (SD). Two and three groups were compared using independent sample t-test and one-way analysis of variance (ANOVA), respectively. Statistical and graphical analyses were done with SPSS 22 and GraphPad 8.0. P significance was set at p < 0.05.

Data availability statement (DAS)

The authors confirm that the data supporting the findings of this study are available within its supplementary materials.

Expressions of PLOD1, PLOD2 and PLOD3 in different cancers

We analyzed the mRNA expressions of PLOD1, PLOD2, and PLOD3 in various cancer and normal tissues using Oncomine database. The number of significant datasets for PLOD family genes was analyzed. Red and blue indicate overexpression and downregulation respectively. PLOD1, PLOD2, and PLOD3 levels were upregulated in most cancers (Figure 1A). Furthermore, the mRNA expressions of PLOD1, PLOD2, and PLOD3 were validated using GEPIA, which showed consistent results (Figure 1B).

Figure 1. The mRNA expression patterns of PLOD1, PLOD2 and PLOD3 in various cancers. (A) The expression patterns were analyzed by Oncomine database; (B) The expression patterns were showed by GEPIA database.

Expressions of PLOD1, PLOD2 and PLOD3 in breast cancer

Using UALCAN in TCGA, we discovered that PLOD1 and PLOD3 were significantly upregulated in breast cancer tissues compared to normal tissues (Figure 2A). In clinical stages, PLOD1 and PLOD3 were overexpressed in all stage subgroups compared with normal tissues (Figure 2B). However, PLOD2 expression was not significant in either of the above results (Figure 2A, 2B). In terms of age, the expressions of PLOD1, PLOD2, and PLOD3 were significantly higher in the ≤51 y group than in the >51 y group (Figure 2C). Regarding hormone receptor status, the expressions of PLOD1, PLOD2, and PLOD3 were upregulated in the estrogen receptor-negative (ER-) group versus the ER-positive (ER+) group and in the progesterone receptor-negative (PR-) group versus the PR-positive (PR+) group (Figure 2D, 2E). Additionally, no PLOD family gene was identified in the human epidermal growth factor receptor-2 (HER-2) status (Figure 2F). Besides, the levels of PLOD1, PLOD2 and PLOD3 were higher in triple negative breast cancer (TNBC) patients than non-TNBC patients (Figure 2G).

Figure 2. The expression of PLOD family genes in breast cancer using UALCAN database. (A) The expression of PLOD1, PLOD2 and PLOD3 in breast cancer tissues and normal tissues; (B) The expression of PLOD1, PLOD2 and PLOD3 in different clinical stages of breast cancer; bc-GeneExMiner 4.6 analyze the PLOD family genes in different clinical status. (C) Age; (D) ER; (E) PR; (F) Her-2; (G) TNBC.

Correlation of PLOD1, PLOD2 and PLOD3 expressions with prognosis in breast cancer

We analyzed the prognostic value of PLOD family genes in bc-GeneExMiner 4.6. The Kaplan-Meier (KM) plotter showed that higher levels of PLOD1 and PLOD3 demonstrated worse overall survival (OS), whereas PLOD2 showed no difference in OS (Figure 3A). In addition, high expressions of PLOD1, PLOD2, and PLOD3 were associated with worse disease-free survival (DFS) and distant metastasis free survival (DMFS) (Figure 3B, 3C). Moreover, overexpression of PLOD1, PLOD2, and PLOD3 was associated with inferior DMFS, RFS, and OS (Table 1).

Figure 3. The value of prognosis in PLOD family genes in breast cancer using bc-GeneExMiner 4.6 software. (A) OS; (B) DFS; (C) DMSF.

Expressions of PLOD1, PLOD2 and PLOD3 correlated with TIICs in breast cancer

TIICs are an important complex in the tumor microenvironment of breast cancer and promote or suppress the development and growth of tumors. Therefore, we investigated PLOD family genes in TIICs of breast cancer. PLOD1 expression was positively correlated with the infiltrating levels of CD4+ T cells, macrophages, neutrophils, and dendritic cells (Figure 4A). PLOD2 expression was positively correlated with the infiltrating levels of B cells, CD8+ T cells, CD4+ T cells, macrophages, neutrophils, and dendritic cells (Figure 4B). PLOD3 expression was positively correlated with the infiltrating levels of CD8+ T cells, CD4+ T cells, macrophages, and dendritic cells (Figure 4C).

Figure 4. Correlation of PLOD family genes with TIICs in breast cancer. (A) The expression of PLOD1 in TIICs; (B) The expression of PLOD2 in TIICs; (C) The expression of PLOD3 in TIICs; (D) PLOD1 of immune score evaluation; (E) PLOD1 of immune score evaluation; (F) PLOD1 of immune score evaluation. Negative values indicate negative correlations and positive values indicate positive correlations, the deeper the color, the stronger the correlation. ^*p < 0.05, ^**p < 0.01 and ^***p < 0.001, (^*) stand for significance levels.

We further assess the reliable results of immune score evaluation through MCP-counter. PLOD1 was negatively correlated with monocyte (P < 0.001) and macrophage (P < 0.001), and positively correlated with T cell (P < 0.01), neutrophil (P < 0.001) and B cell (P < 0.001) (Figure 4D). PLOD2 was negatively correlated with monocyte (P < 0.001) and macrophage (P < 0.001) and positively correlated with CD8+ T cell (P < 0.05), T cell (P < 0.001), NK cell (P < 0.05) and B cell (P < 0.001) (Figure 4E). PLOD3 was negatively correlated with monocyte (P < 0.001), macrophage (P < 0.001), and was negatively correlated with B cell (P < 0.01) (Figure 4F).

Co-expression, interaction analyses and functional enrichment analysis of PLOD family genes

We analyzed the relationship in the co-expression of PLOD1, PLOD2, and PLOD3 at the gene level using GeneMANIA (Figure 5A). PLOD family genes and their neighboring top 20 genes were chosen and analyzed using GO and KEGG in Metascape. The co-expression was associated with protein hydroxylation, collagen biosynthesis and modifying enzymes, collagen metabolic processes, RNA splicing, extracellular matrix organization, the VEGFA-VEGFR2 signaling pathway, and skeletal system development (Figure 5B). It was also related to biological processes, such as metabolic processes, responses to stimuli, cellular component organization, and biogenesis and development processes (Figure 5C). To better understand the relationship among PLOD family members, we constructed a protein-protein interaction network and MCODE components using Metascape (Figure 5D–5F). Results showed the biological processes were mainly related to lysine degradation, collagen biosynthesis, and enzyme modification.

Figure 5. Protein-protein interaction network in POLDs in GeneMANIA. (A) Co-expression of PLOD family genes; (B) Heatmap of GO enriched terms and biological processes colored by the p-value; (C) Networks of GO-enriched terms and biological processes using the p-value; (D, E) Protein-protein interaction network and significant MCODE components from protein-protein network. (F) Independent functional enrichment analysis of MCODE components.

Expressions of PLOD family genes in breast cancer tissues by IHC

To further validate the role of PLOD family genes in breast cancer progression, we examined their levels in breast cancer and non-tumor breast tissues using IHC. Tissue microarrays including 85 tumor and 85 para-tumor tissue samples were used here. PLOD1 and PLOD2 and PLOD3 were mainly expressed in the cytomembrane and plasma (Figure 6A–6C). IHC staining to show that the levels of all PLOD family proteins were significantly higher in breast cancer tissues than para-tumor tissues (Table 2, P < 0.001). In breast cancer tissues, the total immunostaining score of PLOD1 expression was significantly associated with the ER status (P = 0.04384), the TNBC status (P = 0.01592) and tumor grade (P = 0.001192). PLOD2 expression was positively correlated with the Ki-67 status (P = 0.01167). PLOD3 expression was positively correlated with age (P = 0.006977) and tumor grade (P = 0.0004829) (Table 3).

Figure 6. Representative IHC staining of PLOD family genes in breast cancer. Brown staining in cytomembrane and plasm indicates positive staining. (A) PLOD1; (B) PLOD2; (C) PLOD3.