Chronic and acute leukemias are responsible for ~25,000 deaths annually in the US - a mortality rate that represents ~5% of all cancer related deaths. While leukemias are susceptible to antigen-specific T-cell mediated cytotoxicity, leukemia antigen (LA) discovery methods are complex and time consuming. Most LAs are derived from proteins that contain a somatic mutation or are aberrantly expressed or localized in the leukemia cell; however, another potential source of many LAs are antigenic peptides derived from leukemia specific protein splice isoforms. Our laboratory performed deep RNA-Sequencing of 8 acute myeloid leukemia (AML) samples, which did not contain spliceosome mutations, and identified >300,000 previously unannotated (i.e. novel) splice junctions. The central hypothesis of this project is that leukemia-specific splice junctions yield peptide epitopes that can be presented by class I HLA and targeted by CD8+ cytotoxic T-cells. Roughly 15% of AML and chronic lymphocytic leukemia (CLL) samples contain spliceosome mutations: primarily U2AF1 mutations in AML and SF3B1 mutations in CLL.
In Aims 1 and 2, we will investigate the effect of these mutations on mRNA splicing, protein isoform generation and neo-antigen presentation in model cell lines modified to contain common U2AF1 and SF3B1 mutations (Aim 1) and then in human AML and CLL samples with or without a U2AF1 or SF3B1 mutation (Aim 2). In collaboration with Dr. Jan Prins, of the UNC Department of Computer Science, we will perform hybrid RNA-Sequencing (RNA-Seq) that combines short-read RNA-Seq data for splice junction discovery with long-read RNA-Seq to determine full mRNA isoform sequences. Following isoform sequencing we will test for protein translation of the isoform by probing for splice junction specific peptides using a prototype differential ion mobility spectrometer - mass spectrometer (DIMS-MS) that we have used with Dr. Gary Glish (UNC Department of Chemistry) on peptide discovery projects before.
Aim 3 of this research proposal will apply these methods, as well as whole exome sequencing and comprehensive peptide epitope prediction performed by the lab of Dr. Catherine Wu, in a clinical protocol where we will genetically and computational predict LAs from AML or CLL patient samples, confirm their expression by DIMS-MS, and generate LA-specific T-cells from autologous patient-derived lymphocytes. At the completion of this project we will have characterized the effects of spliceosome mutations on mRNA splicing and its resultant effect on protein isoform production and epitope presentation. More significantly, we will have developed a methodology for personalized LA discovery, where we will identify and confirm patient-derived LA that are presented by the patient's HLA, and produce patient-derived LA-specific T-cells. The combined sequencing, bioinformatics and proteomics methods could be applied for the development of patient-specific T-cell therapeutics.
Roughly 25,000 Americans die from leukemia annually - a mortality rate that represents ~5% of all cancer related deaths. Most leukemias are susceptible to killing by the immune system; however, identifying leukemia specific targets (called antigens) is time-consuming and inefficient. The purpose of this grant is to identify a new class of leukemia antigens derived from alterations in RNA splicing and then generate immune cells that are capable of killing the leukemia cells that contain these antigens.
| Lansford, Jefferson L; Dharmasiri, Udara; Chai, Shengjie et al. (2018) Computational modeling and confirmation of leukemia-associated minor histocompatibility antigens. Blood Adv 2:2052-2062 |
| Attayek, Peter J; Waugh, Jennifer P; Hunsucker, Sally A et al. (2017) Automated microraft platform to identify and collect non-adherent cells successfully gene-edited with CRISPR-Cas9. Biosens Bioelectron 91:175-182 |
