No abstract

Highlights d Affinity-tagging protocol enables proteomic profiling of individual HLA-II alleles d Even in ''hot'' tumors, professional APCs-not cancer cellsdrive HLA-II expression d Cellular localization influences which phagocytosed cancer proteins get presented d Machine-learning models for binding and processing improve HLA-II prediction

Background: The ongoing COVID-19 pandemic has created an urgency to identify novel vaccine targets for protective immunity against SARS-CoV-2. Early reports identify protective roles for both humoral and cell-mediated immunity for SARS-CoV-2. Methods: We leveraged our bioinformatics binding prediction tools for human leukocyte antigen (HLA)-I and HLA-II alleles that were developed using mass spectrometry-based profiling of individual HLA-I and HLA-II alleles to predict peptide binding to diverse allele sets. We applied these binding predictors to viral genomes from the Coronaviridae family and specifically focused on T cell epitopes from SARS-CoV-2 proteins. We assayed a subset of these epitopes in a T cell induction assay for their ability to elicit CD8 + T cell responses. Results: We first validated HLA-I and HLA-II predictions on Coronaviridae family epitopes deposited in the Virus Pathogen Database and Analysis Resource (ViPR) database. We then utilized our HLA-I and HLA-II predictors to identify 11,897 HLA-I and 8046 HLA-II candidate peptides which were highly ranked for binding across 13 open reading frames (ORFs) of SARS-CoV-2. These peptides are predicted to provide over 99% allele coverage for the US, European, and Asian populations. From our SARS-CoV-2-predicted peptide-HLA-I allele pairs, 374 pairs identically matched what was previously reported in the ViPR database, originating from other coronaviruses with identical sequences. Of these pairs, 333 (89%) had a positive HLA binding assay result, reinforcing the validity of our predictions. We then demonstrated that a subset of these highly predicted epitopes were immunogenic based on their recognition by specific CD8 + T cells in healthy human donor peripheral blood mononuclear cells (PBMCs). Finally, we characterized the expression of SARS-CoV-2 proteins in virally infected cells to prioritize those which could be potential targets for T cell immunity.

Extracellular matrix (ECM) remodeling is a key component of cell migration and tumor metastasis, and has been associated with cancer progression. Despite the importance of matrix remodeling, systematic and quantitative studies on the process have largely been lacking. Furthermore, it remains unclear if the disrupted tensional homeostasis characteristic of malignancy is due to initially altered ECM and tissue properties, or to the alteration of the tissue by tumor cells. To explore these questions, we studied matrix remodeling by two different prostate cancer cell lines in a three-dimensional collagen system. Over one week, we monitored structural changes in gels of varying collagen content using confocal reflection microscopy and quantitative image analysis, tracking metrics of fibril fraction, pore size, and fiber length and diameter. Gels that were seeded with no cells (control), LNCaP cells, and DU-145 cells were quantitatively compared. Gels with higher collagen content initially had smaller pore sizes and higher fibril fractions, as expected. However, over time, LNCaP- and DU-145-populated matrices showed different structural properties compared both to each other and to the control gels, with LNCaP cells appearing to favor microenvironments with lower collagen fiber fractions and larger pores than DU-145 cells. We posit that the DU-145 cells' preference for denser matrices is due to their higher invasiveness and proteolytic capabilities. Inhibition of matrix proteases resulted in reduced fibril fractions for high concentration gels seeded with either cell type, supporting our hypothesis. Our novel quantitative results probe the dynamics of gel remodeling in three dimensions and suggest that prostate cancer cells remodel their ECM in a synergistic manner that is dependent on both initial matrix properties as well as their invasiveness.

Cell migration is a fundamental process that is crucial to a variety of physiological events. While traditional approaches have focused on two-dimensional (2D) systems, recent efforts have shifted to studying migration in three-dimensional (3D) matrices. A major distinction that has emerged is the increased importance of cell-matrix interactions in 3D environments. In particular, cell motility in 3D matrices is more dependent on matrix metalloproteinases (MMPs) to degrade steric obstacles than in 2D systems. In this study, we implement the effects of MMP-mediated proteolysis in a force-based computational model of 3D migration, testing two matrix ligand-MMP relationships that have been observed experimentally: linear and log-linear. The model for both scenarios predicts maximal motility at intermediate matrix ligand and MMP levels, with the linear case providing more physiologically compelling results. Recent experimental results suggesting MMP influence on integrin expression are also integrated into the model. While the biphasic behavior is retained, with MMP-integrin feedback peak cell speed is observed in a low ligand, high MMP regime instead of at intermediate ligand and MMP levels for both ligand-MMP relationships. The simulation provides insight into the expanding role of cell-matrix interactions in cell migration in 3D environments and has implications for cancer research.