Graduation Year

2019

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Biology (Cell Biology, Microbiology, Molecular Biology)

Major Professor

Douglas W. Cress, Ph.D.

Committee Member

John Cleveland, Ph.D.

Committee Member

Eric Haura, M.D.

Committee Member

Jamie Teer, Ph.D.

Committee Member

David Carbone, Ph.D.

Keywords

GABA, type II pneumocyte, NKX2-1, PD-L1, polyamines, STK11, putrescine, neuroendocrine

Abstract

This dissertation covers a variety of the genetic and molecular abnormalities of lung adenocarcinoma with an emphasis on STK11 loss and its implications on immunotherapy response. Given that lung cancer is the leading cancer killer, novel therapies are in great demand. In particular, immunotherapy has shown some of the most promise in the last decade but remains limited due to nearly 80% of patients not significantly responding. This dissertation aims to molecularly characterize lung adenocarcinoma while attempting to explain the reason why patients with STK11 loss do not respond to immunotherapy.

In the first chapter we discuss the relationship between ancestry and mutational frequency in lung adenocarcinoma. We performed next generation sequencing on over 100 tumors from Hispanic patients. From whole blood, we calculated ancestry using a panel of single nucleotide polymorphisms. Our analysis revealed that Indigenous American ancestry is highly associated with an increased rate of EGFR mutations despite smoking status. Further studies will be needed to determine whether this change in frequency can be attributed to the environment or if polymorphisms associated with this ancestry can predispose patients to the acquisition of lung cancers driven by EGFR.

Chapter 2 is focused on the classification and characterization of patients with mutations in STK11. First, we developed a novel signature for the prediction of STK11 loss of function in lung adenocarcinoma patients. We further describe the inverse relationship of the polyamine pathway with gene expression markers of immune response suggesting that polyamines may be the result of immune suppression in patients with STK11 loss of function. Finally, using LC-MS we show that patients with loss of function in STK11 upregulate putrescine and GABA amongst other metabolites. This reveals that not only do patients upregulate genes regulating polyamine metabolism (ODC1) but that the metabolites downstream of its control are upregulated. These data also suggest a non-canonical utilization of putrescine for the synthesis of GABA.

Chapter 3 highlights the limitations of studying STK11 loss in vitro through its impact on cell lineage identity. We reveal that cell line, mouse models, and patient derived xenografts of STK11 loss unanimously lose their marker of cell lineage (NKX2-1). In addition to this loss in cell lineage, a small subset of patient tumors that lose NKX2-1 also lose hallmarks of immune modulation through inflammatory signaling and polyamine metabolism. These patient tumors that lack NKX2-1 have similar genetic changes as ex vivo models of STK11 loss suggesting that we are not accurately studying STK11 loss as it occurs in a majority of patients.

Chapter 4 is a description of in vitro experiments performed using cell lines with a small report of in vivo biomarkers of STK11 loss. We restore STK11 expression in 3 cell lines lacking STK11 function: A549, NCIH1437, and NCIH1944. We confirm that the addition of STK11 to these cell lines did not alter ODC1 expression or other patient specific biomarkers of STK11 loss. We further attempted to recapture the patient tumor phenotype of STK11 loss through the modification of metabolic stress. Other experiments focused on sensitivities or resistance as a result of STK11 loss. We conclude by displaying that KIT can be used as a biomarker for STK11 loss in vivo.

Chapter 5 is a molecular characterization of lung adenocarcinoma. We performed untargeted LC-MS on 123 patients from our MLOS cohort for proteins and metabolites. We analyzed the relationship between networks of protein, metabolite, and gene expression. We then molecularly characterized this cohort by mutations in STK11, EGFR, KRAS, TP53, and KEAP in addition to by two immune signatures for inflammatory and interferon gamma response. Finally, using the overall survival data in this cohort we identified novel metabolite, protein, and gene expression networks that can be used in patient prognosis.

Share

COinS