All roads lead to “ome”: defining the DRiPome

Class I molecules are encoded by 3 highly polymorphic genes, HLA-A, -B, and -C, in humans, so individuals typically express 6 distinct class I molecules on their cell surface that function to regulate the effector functions of CD8⁺ T cells and natural killer (NK) cells. CD8⁺ T cells are activated by the binding of their clonally restricted T-cell receptor to MHC class I molecules bearing oligopeptides, typically of 8 to 11 residues. MHC I–associated peptides (MIPs) are typically generated from source proteins synthesized by host cell ribosomes. MIPs are liberated as longer peptides by proteasomes and trimmed by aminopeptidases in the cytosol and endoplasmic reticulum. MIPs are used to flag cells needing eradication such as those harboring viruses and other intracellular pathogens, but also to identify tumor cells and tissue transplants, because cellular proteins are constitutively processed into peptides and presented on the surface of virtually all cells in the body.

The near immediate presentation of viral peptides from long-lived viral proteins prompted the DRiP hypothesis (for defective ribosomal products),³  with the central premise that some nascent proteins fail to attain a stable conformation and, as such, will be targeted for proteasomal degradation. While the intimate association between nascent protein synthesis and MIP production has stood the test of time,⁴  true to their name DRiPs have proven slippery to define.⁵  Our understanding of DRiPs has largely been limited to models of genetically manipulated antigens expressed by viral vectors and transfection. The relationship between these artificial antigens and naturally processed cellular antigens is uncertain. New and important information, however, is coming from global “omic” studies that relate the immunopeptidome to the transcriptome, translatome, and degradome^6,7 

Granados et al use high-throughput proteomics to explore the relationship between transcriptome and immunopeptidome of EBV-transformed B-cell lines from disparate pairs of HLA-identical siblings.¹  The MIP repertoire is closely associated with HLA haplotype, with high overlap between siblings and very little overlap between the disparate sibling pairs. Differences in peptide binding to different class I allomorphs is long known⁸  and expected, although the magnitude of the effect is surprising because there is often considerable overlap in peptide binding to disparate class I allomorphs in model experimental systems.

Granados et al found that the approximately 2300 MIPs defined derive from an almost equal number of proteins (approximately 1800). This is surprising because on average each protein should have multiple peptides capable of binding the 6 different class I molecules expressed by each individual. One explanation for “1 protein, 1 peptide” representation would be competition among peptides derived from a single translation product, consistent with the proposal that antigen processing is compartmentalized at the level of translating “immunoribosomes.”⁹  Also surprising is the small overlap in MIP source proteins observed between sibling pairings.

Although most MIPs are derived from high-abundance mRNAs, nearly half derive from low-abundance mRNAs. Speculation that such mRNAs provide a more efficient source of DRiPs due to microRNA (miRNA) targeting led Granados et al to discover that MIP-source mRNAs are greatly enriched in miRNA response elements (MREs). Remarkably, the disparate nature of the MIP-source mRNAs in disparate sibling pairs can be explained by a common set of miRNAs regulating a disparate set of MRE-expressing mRNAs. Importantly, extending this insight to data mining from past studies revealed the preferential generation of both mouse and human MIPs from MRE-containing transcripts.

How miRNAs increase MIP generation is uncertain. MicroRNAs control gene expression by repressing translation and/or targeting transcripts for degradation. Granados et al propose that miRNAs increase MIPs by increasing DRiPs. Several recent reports demonstrate that mRNA regulation can enhance generation of DRiP-derived MIPs.^10,11 

In demonstrating a clear link between miRNAs and the immunopeptidome, Granados et al have established a fertile area for future research. No doubt there are great discoveries to be made in understanding how miRNAs influence DRiP generation and immune recognition of self-antigens for tolerance induction, cancer immunosurveillance, and allo-recognition. Surely, the DRiPome is not too far down the road.