Identifying mutational landscape and inflammatory pathways associated with lung squamous premalignant lesion progression
Camassola Breda, Julia
MetadataShow full item record
Lung cancer is the leading cause of death among all cancer types, and a subtype of lung cancer known as lung squamous cell carcinoma (LUSC) is the second most predominant subtype accounting for ~30% of all lung cancer cases. Lung cancer is typically diagnosed at advanced stages, resulting in poor prognosis with a 5-year survival rate of 19% between 2008 and 2014 (Siegel et al., 2019). The lung cancer genome has extensively been profiled in part by The Cancer Genome Atlas (TCGA), identifying mutational signatures and revealing the contributions of tumor microenvironment. TCGA work has enabled the development of biomarkers and therapeutics for both identifying and treating advanced lung cancer. However, due to late diagnosis, the prognosis of lung cancer still remains poor. Therefore, understanding the genomics and immune microenvironment of premalignant lesions (PMLs) that precede the development of LUSC is crucial in order to improve the ability to diagnose disease early and identify intervention strategies to enhance the prognosis of lung cancer. As part of this thesis project, my objective is to validate, in both cancer cell lines and a carcinogenic mouse model, previously identified genes and their correlations with important immuno-regulatory pathways, as well as their impact in the progression or regression of PMLs. The Lung Pre-Cancer Genome Atlas (PCGA) has begun to achieve the aforementioned goals of understanding the early molecular events in the process of lung squamous carcinogenesis (Beane et al., 2019. In the prior study from our group, bronchial PMLs were obtained through bronchoscopy procedures from a high-risk screening cohort at Roswell Park and subject to a molecular and genomic profiling (Beane et al., 2019). Beane et al. identified four molecular subtypes of PMLs via transcriptional profiling, where one subtype, the proliferative subtype, of PMLs were found to be enriched for bronchial dysplastic lesions. Additionally, within samples in the proliferative subtype, it was demonstrated that immunoregulatory pathways, such as interferon signaling and antigen presentation, were suppressed in the progressive/persistent proliferative PMLs. Further computational analyses of the lung PCGA transcriptional data has identified that Glutathione S-transferase P (GSTP1) is up-regulated in PMLs that remain stable as dysplastic lesions or progress into advanced dysplastic lesions, suggesting that there is an anti-correlation between the expression of GSTP1 and immunological responses. Understanding the relationship between GSTP1 and its downstream effectors is important, as targeting GSTP1 may be a strategy to stimulate an immunological response in PMLs to drive a regression rather than progression towards invasive carcinoma. In order to establish the relationship between GSTP1 and immune-regulatory pathways such as the STAT3 and NF-KB pathways, we identified baseline expression of the target genes in cell lines all genes were consistent with the computational findings. Subsequently, the knockdown of GSTP1 demonstrated a significant decrease in the expression of GSTP1 but no change in STAT3 and NF-KB. Based on these results, interleukins such as IL-8 and IL-13 were used to stimulate immunological responses and expected to cause an increase in expression of STAT3 and NF-KB pathways. However, upon 48, 24 and 4 hours of IL-8 and IL-13 stimulation, STAT3 and NF-KB pathways did not present changes in expression. Furthermore, in effort to establish a viable research model for studying premalignant lesions of LUSC, we utilized the NTCU mouse model, a carcinogenic model that has been previously identified to mimic squamous lung carcinogenesis due to smoking in humans. The histology of the PMLs triggered by the NTCU treatment in mice resembles the carcinogenesis observed in human lungs, in which lesions progress from squamous metaplasia to increasing grades of dysplasia. Previous studies have suggested that carcinogenic models share mutational burden and specific gene mutations with humans opposed to genetically engineered mice (Westcott et al., 2015). As part of my work, I aimed to establish the viability of the NTCU mouse model as a representative model of study for PMLs in humans by analyzing the histology of lung lesions in mice at 8, 12 and 16 weeks of NTCU treatment. Furthermore, I established methods for isolating DNA and RNA from the tail and lungs of both control and NTCU treated mice for whole exome sequencing (WES) and RNA sequencing in addition to use for validation of my GSTP1 finding in an in vivo setting. Our goal was to show similarities between genomic changes of PMLs from the NTCU mice and human PMLs in order to establish the NTCU model as a means to further study changes identified in the progression of human disease.