Identifying tumor-specific antigens is crucial for the development and implementation of personalized cancer immunotherapies. These therapies may leverage novel antigens or “neoantigens” for developing vaccines to trigger tumor-specific T cell responses and tumor cell killing. Alternatively, neoantigen peptides may be used in adoptive cell therapy for ex-vivo enrichment of specific T cell populations from tumor infiltrates or peripheral blood and subsequent reinfusion into patients.
Like other cell-derived and -processed peptides, neoantigen peptides are presented on the cell surface by major histocompatibility complex (MHC, also referred to as HLA) proteins for recognition by CD8 or CD4 T cells. Neoantigen peptides are tumor-specific antigens that arise due to the genomic instability associated with the carcinogenesis process and may show sequence changes consistent with point mutations, short indels, or gene fusion events (Bianchi et al. 2020).
Adapted from “Cancer Vaccine Principle” by BioRender.com (2020). Retrieved from https://app.biorender.com/biorender-templates
One limitation in the neoantigen discovery workflow is that neoantigen peptide identification relies on whole-exome sequencing. Therefore, by following this strategy, investigators primarily focus on mutations found within coding regions. More recently, peptides from non-coding sequences have been identified within the MHC immunopeptidome (Laumont et al. 2018). These peptides result from non-canonical translation and transcription events and provide new opportunities for neoantigen discovery.
The proteasome is a multi-protein complex, consisting of a 20S core catalytic unit and associated regulatory subunits, such as 19S, PA28, and PA200. Together, this complex mediates ubiquitinated-protein degradation. The proteasome’s activity is not only determined by associated regulatory subunits but is also influenced by its 20S core components. Several 20S core subsets, which differ in β subunit composition, give rise to constitutive, immune, and intermediate proteasomes. Among these, the immunoproteasome predominates in lymphoid tissues. However, in non-immune cells, proinflammatory cytokines such as IFN (α, β and γ) and TNF induce the immunoproteasome (Murata et al. 2018, Kors et al. 2020). Because tumors possess an inflammatory microenvironment, a standing question in the field pertains to the role of inflammation in modulating the proteasome’s composition and activity, and consequently, how it shapes peptide presentation and the immunopeptidome.
In recent work by a group at the Broad Institute of MIT and Harvard, ribosomal profiling or Ribo-seq enabled identifying non-canonical HLA-bound peptides. In Ribo-seq, mRNA sequences protected by the ribosome are captured and deep sequenced; therefore, this approach allows identification of sequences undergoing translation.
Ouspenskaia et al. 2020, used Ribo-seq to uncover “novel unannotated open reading frames” (nuORFs) derived peptides expressed in various tumors, including melanoma, chronic lymphocytic leukemia, and glioblastoma.
At the most recent Neoantigen Based Therapies Summit, Tamara Ouspenskaia, Ph. D., shared her findings on identifying thousands of nuORFs derived peptides. By combining Ribo-seq and MHC I immunopeptidome mass spectrometry analysis, Dr. Ouspenskaia found that peptides from translated nuORFs form part of the cancer MHC I immunopeptidome. Some of the identified nuORFs originated from the 5’ and 3’ sequences of annotated ORFs.
For this approach, Dr. Ouspenskaia first performed Ribo-seq analysis of normal and cancerous primary samples and cell lines, allowing her to build a database of predicted translated nuORFs. Ultimately, this database, combined with MHC I immunopeptidome mass spectrometry analysis, allowed her to identify translated nuORFs presented by normal and cancerous cells.
By following this approach, Dr. Ouspenskaia found that nuORFs derived peptides represented ~1.5-2.2% of the MHC I immunopeptidome across various cancer samples. Interestingly, about 50% of the translated nuORFs were detected in multiple patient-derived cancer cell lines, which strengthened the idea that they represent specific peptide products. Additionally, nuORFs were identified that carried melanoma-specific mutations or were preferentially expressed or overexpressed in melanoma. Overall, these findings provide a new avenue for investigators to broaden the repertoire of neoantigens with the potential for cancer immunotherapy.
