Multi-functional gene ZNF281 identified as a molecular biomarker in soft tissue regeneration and pan-cancer progression

Abstract

Regeneration and tumorigenesis are indicated as related processes, while regeneration leads to life and the outcome of tumorigenesis is death. Here, we show the upregulation of zfp281 (zinc finger 281) in our adipose de novo regeneration model through RNA-seq analysis. Then, we validated the upregulation of zfp281 in adipose regeneration via immunofluorescence. Following that, we found that ZNF281 (the human homolog of Zfp281) was upregulated in most types of cancer and related to worse prognosis in 10 tumors. We further investigated the role of ZNF281 in cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), pancreatic adenocarcinoma (PAAD), and stomach adenocarcinoma (STAD) and confirmed the high accuracy in the clinical diagnostic feature. Beyond that, based on these three types of cancers, we analyzed the ZNF281-related tumor immune infiltration and DNA methylation sites and finally built risk prediction models for future disease diagnosis. Taken together, our findings provide new insights into the dual role of ZNF281, and we found that it was a potential biomarker for regeneration and tumor prognosis.

Keywords: ZNF281; biomarker; pan-cancer; regeneration; tumorigenesis.

FIGURE 1

Zfp281 is upregulated at an early stage in adipose de novo regeneration tissues. (A–F) Immunofluorescence staining of perilipin-1 in bFGF-loaded DAM (A–C) and DAM-injected mouse tissuees (D–F) at 12 weeks; (G–L) immunofluorescence staining of zfp281 in bFGF-loaded DAM (G–I) and DAM-injected mouse tissues (J–L) at 1 week. Perilippin-1 is labeled in green. Zfp281 is Labeled in red. Nuclei are labeled with DAPI in blue.

FIGURE 2

Expression levels of ZNF281 were higher in tumors and normal tissues. (A) ZNF281 expression in normal tissues, (B) ZNF281 expression in single-cell types, (C) ZNF281 expression was upregulated in most types of cancers than normal tissues, combining the data of TCGA and the GTEx database, and (D) ZNF281 expression in TCGA tumors and adjacent normal tissues (^* p < 0.05, ^** p < 0.01, and ^*** p < 0.001).

FIGURE 3

Correlations between the ZNF281 expression and prognosis (OS) of patients in different cancers. (A) ACC, (B) CESC, (C) KIRC, (D) KIRP, (E) LUSC, (F) MESO, (G) PAAD, (H) SKCM, (I) STAD, (J) THCA, and (K) UCEC.

FIGURE 4

ROC curve and the relationship with immune cell infiltration for ZNF281 in CESC, PAAD, and STAD. (A) ROC curve for ZNF281 in CESC, (B) ROC curve for ZNF281 in PAAD, and (C) ROC curve for ZNF281 in STAD. (D) Immune infiltration analysis for ZNF281 in CESC, (E) immune infiltration analysis for ZNF281 in PAAD, and (F) immune infiltration analysis for ZNF281 in STAD.

FIGURE 5

ZNF281-associated DNA methylation sites in CESC, PAAD, and STAD. (A–C) Methylation sites in the TSS region of ZNF281 in CESC, (D–I) methylation sites in the TSS region of ZNF281 in CESC, and (J–N) methylation sites in the TSS region of ZNF281 in CESC.

FIGURE 6

Three lncRNA prognostic models associated with ZNF281 showed good accuracy in predicting the prognosis of patients in CESC, PAAD, and STAD. (A–C) Cross validation based on C-index to determine the best choice of genes in the prediction models. (D–F) Genes in the different choices of models and their corresponding coefficients based on different lambda values. (G–I) According to prediction models, the relationship between the survival outcome and risk levels of patients. (J–L) The tdROC curve of 1, 3, and 5 years verified the efficacy of prediction models. (A,D,G,J) CESC, (B,E,H,K) PAAD, and (C,F,I,L) STAD.