Correlated high expression of FXR and Sp1 in cancer cells confers a poor prognosis for pancreatic cancer: A study based on TCGA and tissue microarray

Oncotarget. 2017 May 16;8(20):33265-33275. doi: 10.18632/oncotarget.16633.

Abstract

Bile acids (BAs) was critical in the initiation and progression of various tumors. However, their prognostic significance in pancreatic cancer was still illusive. In the present study, the expression and biological significance of FXR, a major receptor of BAs, in the lethal disease were evaluated in mRNA and protein levels. We found that FXR protein was elevated in the cancerous tissues, which was significantly higher than the adjacent tissues (p < 0.05). Meanwhile, our data showed that FXR was positively correlated with primary tumor location (p = 0.04) and poor survival (p = 0.002). Finally, COX regression model indicated that FXR protein was an independent prognostic factor (p = 0.01; HR = 2.15; 95% CI 1.27-3.63). Consistently, we also found a significant difference of FXR expression between the high and low groups in mRNA level (p < 0.001), and that high FXR expression confers a poor prognosis (p < 0.001). More importantly, the correlation assay showed that FXR was positively correlated Sp1 in both protein (r = 0.351, p = 0.008) and mRNA levels (r = 0.263, p < 0.01), with the simultaneously high expression indicated the worst prognosis on protein (p < 0.001) and mRNA levels (p < 0.001). Additionally, we also showed that FXR was elevated in the pancreatic cancer cells responsible for proliferation and migration. Overall, the data suggested co-high expression of the two factors was an independent prognostic factor (p < 0.001; HR = 3.27; 95% CI 1.86-5.76). Based on these data, we proposed a model to link FXR to Sp1, which included triggered FXR, p38/MAPK and/or PI3K/AKT signaling and phosphorylated Sp1, to illustrate the potential crosstalk between the two factors.

Keywords: FXR; Sp1; immunohistochemistry; pancreatic cancer; prognosis.

MeSH terms

  • Adult
  • Aged
  • Cell Line, Tumor
  • Cell Movement / genetics
  • Cell Proliferation
  • Computational Biology / methods
  • Databases, Genetic
  • Female
  • Gene Expression Profiling
  • Gene Expression*
  • Humans
  • Kaplan-Meier Estimate
  • Male
  • Middle Aged
  • Models, Biological
  • Neoplasm Grading
  • Neoplasm Staging
  • Pancreatic Neoplasms / genetics*
  • Pancreatic Neoplasms / mortality*
  • Pancreatic Neoplasms / pathology
  • Prognosis
  • RNA, Messenger / genetics
  • Receptors, Cytoplasmic and Nuclear / genetics*
  • Receptors, Cytoplasmic and Nuclear / metabolism
  • Sp1 Transcription Factor / genetics*
  • Sp1 Transcription Factor / metabolism
  • Tissue Array Analysis

Substances

  • RNA, Messenger
  • Receptors, Cytoplasmic and Nuclear
  • Sp1 Transcription Factor
  • farnesoid X-activated receptor