Investigating nuclear transport proteins as secreted cancer biomarkers

Okpara, Michael Obinna

Investigating nuclear transport proteins as secreted cancer biomarkers

Master Thesis

2019

Abstract

Previous studies in our laboratory using microarray gene expression analysis identified members of the nuclear transport protein family as significantly upregulated in cervical cancer biopsies compared to normal cervical epithelial tissues. These results were validated at both mRNA and protein levels, and similar upregulation observed in oesophageal cancer. Recent mass spectrometry (MS) analysis of cancer cell secreted proteins identified elevated levels of 13 members of the nuclear transport protein family in the secretomes of transformed, cervical cancer and oesophageal cancer cell lines. The nuclear transport proteins have functions in many cellular processes including proliferation, mitosis, maturation of RNA, activation of the actin cytoskeleton and restructuring of the nuclear envelope. In addition, they are required for the nuclear import and export of numerous cargo proteins such as transcription factors, oncoproteins and kinases, which often display deregulated activity in cancer cells. The aims of this study were to 1) independently validate the MS data showing elevated levels of the nuclear transport proteins in the secretomes of cervical and oesophageal cancer cell lines, 2) investigate the diagnostic potential of members of the nuclear transport protein family using cervical and oesophageal cancer serum samples and 3) identify the potential binding partners of Kpnβ1, a key member of the nuclear transport protein family, in normal and cancer cells. This study investigated the levels of endogenous expression and secretion of 8 members of the nuclear transport protein family; Kpnβ1, IPO5, IPO7, TNPO1, CRM1, CAS, Kpnα2 and Ran in a normal epithelial cell line (hTERT-RPE1) in comparison to transformed (SVWI38 and CT-1), cervical cancer (HeLa and CaSki) and oesophageal cancer (WHCO5 and KYSE 30) cell lines using Western blot analysis. Our data revealed differential endogenous expression in the cell lines. An analysis of the secretomes of the cell lines showed that all 8 proteins assayed were secreted at elevated levels by the transformed, cervical cancer and oesophageal cancer cell lines compared to the normal cell line. These results validate previous MS data generated in our laboratory. To investigate whether members of this protein family can be detected in the serum of cancer patients, ELISA for Kpnβ1, CRM1, Kpnα2 and CAS proteins were performed using commercially available ELISA kits. The results showed significantly elevated levels of Kpnβ1, CRM1 and CAS in the serum of cervical cancer patients compared to the non-cancer controls. Serum levels of Kpnβ1, CRM1, Kpnα2 and CAS were elevated in the oesophageal cancer patients compared to the non-cancer controls. To investigate the diagnostic potential of these proteins, logistics regression analysis was performed. Our results showed that CAS was the best performing individual candidate biomarker in discriminating between cervical cancer cases and non-cancer controls. It had the highest AUC (0.85±0.03) and highest sensitivity (55%) at 95% specificity compared to those of Kpnβ1 (AUC=0.77±0.04 with 35% sensitivity at 95% specificity), CRM1 (AUC=0.64±0.05 with 20% sensitivity at 95% specificity) and Kpnα2 (AUC=0.51±0.05 with <10% sensitivity at 95% specificity). The combination of Kpnβ1, CRM1, Kpnα2 and CAS as a panel of biomarkers had an improved AUC of 0.89 with a sensitivity of 100% at 60% specificity. In discriminating oesophageal cancer cases from the non-cancer controls, CAS (AUC=0.86±0.03 with 56% sensitivity at 95% specificity) similarly performed better compared to Kpnβ1 (AUC=0.62±0.05 with 15% sensitivity at 95% specificity), CRM1 (AUC=0.75±0.04 with 32% sensitivity at 95% specificity) and Kpnα2 (AUC=0.73±0.04 with 21% sensitivity at 95% specificity). The combination of Kpnβ1, CRM1, Kpnα2 and CAS as a panel of biomarkers had the highest diagnostic capacity with an AUC of 0.90 and 84% sensitivity at 86% specificity. These results suggest that individual members of the nuclear transport protein family have potential as diagnostic biomarkers for both cervical and oesophageal cancers, with a combination of Kpnβ1, CRM1, Kpnα2 and CAS being the best predictor. Our investigation aimed at identifying the binding partners of Kpnβ1 in normal, cervical cancer and oesophageal cancer cell lines using immunoprecipitation coupled to mass spectrometry (IP-MS) identified 100 potential Kpnβ1 binding partners in hTERT-RPE1 normal cell extracts, 179 in HeLa cervical cancer cell extracts, 147 in WHCO5 cell extracts and 176 in KYSE30 oesophageal cancer cell extracts. Venn Dis JavaFX-based Venn and Euler diagram software was used to identify common and unique Kpnβ1 binding partners. 38 proteins were identified as common binding partners of Kpnβ1 in normal and cancer cells and 56 common binding partners of Kpnβ1 in the three cancer cell lines. Of these, 18 proteins were found to be unique to the three cancer cell lines and of these, 10 could be linked via protein-protein interaction mapping using STRING bioinformatic analysis. These include nucleoporin 214 (Nup214), Prem RNA 3'-end-processing factor FIP1 (FIP1L1), cell division cycle and apoptosis regulator 1 (CCAR1), cleavage and polyadenylation specific factor 7 (CPSF7), ribosomal protein L7 (RPL7), ribosomal protein L10 (RPL10), ribosomal protein L13A (RPL13A), ribosomal protein S6 (RPS6), ribosomal protein S4, X isoform (RPS4X) and Ras-related nuclear protein (Ran). Among these, FIP1L1, CCAR1 and CPSF7 have not been previously described as binding partners of Kpnβ1. In conclusion, elevated levels of nuclear transport proteins in the extracellular environment of cancer cells and in cancer patient serum samples suggest that they have potential as diagnostic biomarkers for cervical and oesophageal cancers, with a combination of Kpnβ1, CRM1, Kpnα2 and CAS being the best predictor. In addition, this study shows that Kpnβ1 interacts with several different proteins in normal and cancer cells, with some of the interactions unique to cancer cells presenting as novel binding partners for further investigation.

Keywords

Medical Biochemistry

Reference:

Collections

Masters

Full item page