Author ORCID Identifier

https://orcid.org/0009-0007-0400-006

Date Available

1-11-2025

Year of Publication

2024

Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation

College

Medicine

Department/School/Program

Toxicology and Cancer Biology

First Advisor

Dr. Christine Fillmore Brainson

Second Advisor

Dr. Xiaoqi Liu

Abstract

Non-small cell lung cancer (NSCLC) heterogeneity is a major challenge for determining effective treatment strategies. Adenocarcinomas (ADCs) and squamous cell carcinomas (SCCs) are histologically and epigenetically distinct subtypes of NSCLC. Patients with ADC tumors harboring mutations in both KRAS and LKB1 (aka STK11) have lower survival rates than those with KRAS-only tumors. KRAS/LKB1 tumors are not only aggressive, but also respond poorly to immunotherapy. However, these data are limited to ADCs, and it is unclear if SCCs with this genotype are also resistant to immunotherapy. We developed a mouse model of Krasmut/Lkb1mut capable of producing aggressive ADCs, and a proportion of tumors transition to a SCC state. This lineage switch is driven by a reduction in Polycomb Repressive Complex 2 (PRC2) activity, which can be targeted directly with EZH2 inhibitors (EZH2i). Due to this difference, we hypothesize that EZH2i will be more effective at de-repression of genes involved in immune recognition in SCCs when compared to ADCs. To test this hypothesis, we developed tumoroid models that are ADC and SCC in phenotype and have mutant Kras with either WT or mutant Lkb1. When treated with a combination of EZH2i and IFN-gamma, PD-L1, MHCI, and MHCII expression were increased in the SCC but not the ADC tumoroids. These data suggest that combining PRC2 inhibition with immunotherapy may be very efficacious in KRAS/LKB1 tumors with SCC characteristics, but not those with ADC characteristics. However, there may be other ways to improve treatment of Krasmut/Lkb1mut ADCs. For example, PRC2 requires S-adenosyl methionine (SAM) to methylate histones, and SAM production requires exogenous methionine or methionine recycling. Therefore, we predict that methionine restriction (MR) will enhance treatment efficacy in KRAS/LKB1 mutant NSCLCs via lowering PRC2 activity. To test this hypothesis, we placed cohorts of Krasmut/Lkb1mut mice on control methionine and methionine restricted (MR) diets during tumor progression, and during carboplatin treatment. We observed that Krasmut/Lkb1mut mice on MR diets had reduced tumor burden compared to mice on a control methionine diet in both the prevention and chemotherapy studies, and these tumors were predominantly ADC in phenotype. In addition, the tumor immune microenvironment (TIME) likely plays a large role in the response of NSCLCs to various therapies. We determined that MR increased the number of tumor-infiltrating macrophages and decreased tumor-infiltrating neutrophils. This change in the TIME of the MR mice could indicate that they may respond differently to immunotherapy than those on a control methionine diet. Next, human NSCLC cell lines were grown under high, regular, and low methionine conditions to investigate potential mechanisms of these responses and lineage switching. High methionine protected cells from carboplatin, while low methionine conditions sensitized cells to carboplatin. Rescue of LKB1 increased carboplatin survival in regular and low methionine conditions, and reduced carboplatin-induced apoptosis and cell cycle arrest in high methionine conditions. We observed that LKB1 rescue increased the mRNA and protein levels of the methionine pathway protein cystathionine-β-synthase (CBS). CBS condenses homocysteine and serine for downstream glutathione production, and may therefore reduce methionine recycling and modulate PRC2 activity. Knock-down of CBS in LKB1 mutant and rescued cells could sensitize cells to carboplatin treatment. Together these data suggest an important role for PRC2, methionine restriction, and LKB1-mutational status in NSCLC lineage fate and therapy response. This also demonstrates the importance of considering not only the genotype of NSCLCs, but also the lineage phenotype when selecting the best course of treatment for each individual patient.

Digital Object Identifier (DOI)

https://doi.org/10.13023/etd.2024.265

Funding Information

National Cancer Institute - R01 CA237643

American Cancer Society Grant - 133123-RSG-19-081-01-TBG

American Institute for Cancer Research Grant - 710410

Toxicology and Cancer Biology Research Fellowship - 2021

Dr. Matt Devalaraja Graduate Student Career Support Award - 2022

Markey Cancer Center Travel Award - 2022

Available for download on Saturday, January 11, 2025

Share

COinS