Genome maintenance gene regulation in normal bronchial epithelial cells by CEBP and TP53 transcription factor families is different in COPD subjects
Recommended Citation
Willey J, Yeo J, Chen T, Crawford EL, Zhang X, Khuder S, and Levin A. Genome maintenance gene regulation in normal bronchial epithelial cells by CEBP and TP53 transcription factor families is different in COPD subjects. Am J Respir Crit Care Med 2017; 195.
Document Type
Conference Proceeding
Publication Date
5-23-2017
Publication Title
Am J Respir Crit Care Med
Abstract
Rationale: Transcription factors in the CEBP and TP53 families regulate key antioxidant (AO), DNA repair (DNAR), and cell cycle control (CCC) genes in normal bronchial epithelial cells (NBEC). In turn, these genome maintenance genes are important for protecting NBEC from damage, and sub-optimal function of these gene pathways is associated with risk for COPD. In this study, we investigated expression patterns of CEBP and TP53 transcription factor family gene pathways in NBEC of COPD and non-COPD subjects in an effort to identify a biomarker for COPD risk and to better understand mechanisms of COPD risk.
Methods: NBEC samples were obtained by bronchoscopy from 30 COPD subjects defined by spirometry (FEV1/FVC <0.7), and 30 non-COPD controls. A competitive multiplex PCR-amplicon method was used to generate targeted libraries for next generation sequencing (NGS) analysis of the transcription factors CEBPA, CEBPG, CEBPD, TP53, TP63, and TP73 transcription factors, and 24 AO, DNAR, and CCC genes that are known or putative targets for these transcription factors. Shrinkage Linear Discriminant Analysis (SLDA) and 10-fold cross-validation was used to identify a model COPD classifier.
Results: The model with the best AUC ROC comprised eight gene targets, including KEAP1, XPA, GPX1, TP73, CAT, CEBPG, TP53, CEBPD, with pooled 10-fold cross-validation AUC of 76.3% (95% CI: 72%-93%). To further explore the regulatory relationships underlying this model, we assessed inter-gene correlation difference in COPD by Fisher Z-test, network of correlation analysis, ANCOVA and multivariate linear regression. CEBPD was expressed higher in NBEC of COPD compared to controls. Gene-pairs negatively or non-correlated with CEBPD in controls were positively correlated in COPD (p-value of Z-score <0.05). Further, CEBPA expression was more disperse in COPD compared to non-COPD control (p-value =0.02). TP53 was positively correlated with CDKN1A, ERCC4 and TFE3 in controls and correlation of each gene-pair was significantly lower among COPD subjects (p-value of Z-score <0.05). Further, KEAP1, which is regulated by TP53, was expressed lower in NBEC of COPD compared to controls. In contrast, TP63 and CDKN1A were negatively correlated in controls and positively correlated in COPD (p-value = 0.002).
Conclusion: CEBP and TP53 transcription factor family regulation of many AO, DNAR, and CCC pathway genes in NBEC was different in COPD subjects compared with non-COPD controls. A classifier based on these differences may enable more accurate identification of individuals at highest risk for COPD.
Volume
195
