Role of Nucleolin in Endometrial Precancerous Hyperplasia and Carcinogenesis: Ex Vivo and In Silico Study

Abstract

Endometrial cancer (EC) is the most common gynaecological malignancy. Nucleolin (NCL) is involved in rDNA transcription, cell proliferation, and apoptosis, with high expression associated with worse overall survival (OS) in other adenocarcinomas. Our aims were to assess NCL gene and protein expression and explore the differential expression of NCL-associated genes (NAGs) in endometrial carcinogenesis. Endometrial samples were obtained from 157 women to include healthy, hyperplastic (EH), EC, and metastatic groups. RT-qPCR and immunohistochemistry were employed to assess NCL gene and protein levels. In silico analysis of NAGs in TCGA and GEO datasets was performed, with the prognostic value determined via Human Protein Atlas. NCL mRNA level of EC was lower than in healthy post-menopausal endometrium (p < 0.01). EH samples had lower NCL immuno-expression scores than healthy pre-menopausal (p < 0.001), benign post-menopausal (p < 0.01), and EC (p < 0.0001) samples. Metastatic lesions demonstrated higher NCL quick scores than primary tissue (p = 0.04). Higher NCL Immuno quick scores carried a worse OS in high-grade EC (p = 0.01). Interrogating Uterine Corpus Endometrial Carcinoma (TCGA-UCEC) and Uterine Carcinosarcoma (TCGA-UCS) cohorts revealed NCL to be the most highly upregulated gene in carcinosarcoma, with S100A11, LMNB2, RERG, E2F1 and CCNA2 representing key dysregulated NAGs in EC. Since NCL is implicated in transforming hyperplastic glands into cancer, with further involvement in metastasis, it is suggested to be a promising target for better-informed diagnosis, risk stratification, and management of EC.

Keywords: endometrial cancer; metastasis; nucleolin.

Figure 1

Scatter plot of NCL mRNA expression using RT-qPCR in post-menopausal (n = 6) versus endometrial cancer tissue (n = 20; G1 endometrioid n = 3, G2 endometrioid n = 3, G3 endometrioid n = 6, serous n = 2, carcinosarcoma n = 3, clear cell n = 3); line indicates the median.

Figure 2

(A) Scatter plot showing NCL quickscore comparison across pre-menopausal (n = 15), post-menopausal (n = 18), EH (n = 21) and EC (n = 90) samples. Line indicates the median. (B) Scatter plot showing NCL quickscore comparison across EH (n = 21), G1 (n = 21), G2 (n = 15), G3 (n = 13), serous (n = 11), carcinosarcoma (n = 18), clear cell (n = 11), and mixed (n = 1) EC. Line indicates the median. (C) Kaplan–Meier survival curve showing effect of NCL expression on OS in EC. p = 0.05. (D) Kaplan–Meier survival curve showing effect of NCL expression on OS in HGEC. p = 0.01.

Figure 3

Representative microphotographs of immunolocalised nucleolar NCL expression in different tissues. Positive staining indicated by brown nucleoli. All images seen at ×1000 magnification. Scale bar = 20 μm, applicable to all panels. (A) Pre-menopausal (B) post-menopausal (C) hyperplasia (D) G1 endometrioid EC (E) G2 endometrioid EC (F) G3 endometrioid EC (G) serous EC (H) carcinosarcoma EC (I) clear cell EC (J) mixed. The median scores for each group were 6.8, 5.1, 2.8, 6.5, 5.8, 4.6, 6.6, 5.1, 6.4, and 8.3, respectively.

Figure 4

(A) Graph showing Wilcoxon matched pairs test in women diagnosed with both EH and EC (n = 26). Representative micrographs showing nucleolar NCL staining in (B) EC (median quick score = 4.6) and (C) matched EH (median quick score = 3.6). (D) Graph showing Wilcoxon matched pairs test in women diagnosed with metastatic EC, showing nucleolar NCL quick score in primary endometrial tissue (n = 27) and matched metastatic lesions (n = 35). Representative micrographs showing nucleolar NCL staining in (E) EC (median quick score = 4) and (F) matched metastatic lesion (median quick score = 6). Positive staining indicated by brown nucleoli. All images seen at ×1000 magnification. Scale bar = 20 μm, applicable to all panels.

Figure 5

(A) Table of DEGs commonly upregulated and downregulated between EC subtypes. Venn diagrams displaying common (B) upregulated and (C) downregulated genes between each subtype.

Figure 6

(A) Table of DEGs commonly upregulated and downregulated between those exposed and not exposed to hormonal, radiation, or neoadjuvant therapy. (B) Protein–protein interaction network displaying upregulated and downregulated DEGs in those exposed and not exposed to hormonal, radiation, or neoadjuvant therapy. Green and red nodes represent upregulated and downregulated DEGs, respectively. (C) Protein–protein interaction network of favourable (green) and unfavourable (red) genes in endometrial cancer.

Figure 7

Bar charts of (A) biological processes and (B) KEGG pathways analyses in prognostic DEGs.