The assumption that antigen presentation is primarily a function of protein abundance has always been an attractive simplification. In the case of p53, this assumption becomes even more tempting: a stress-activated protein, stabilized in tumors, transcriptionally active, and deeply embedded in the machinery of cellular surveillance. If any protein should be visible to the immune system, it should be p53.
Yet the emerging evidence suggests something more nuanced. What determines whether p53 is seen is not its abundance, but whether its fragments survive a highly selective intracellular processing environment long enough to be displayed.
This distinction, hinted at in earlier work examining the natural ligandome of HLA-C alleles, becomes far more consequential when viewed alongside recent population-scale genetic findings. In the Codondex analysis of 2022, a TP53-derived peptide—TAKSVTCTY—was identified among ligands presented by HLA-C*02:02, alongside tumor-associated antigens such as MAGEA3. That observation traces back to immuno-peptidomic work characterising HLA-C motifs and naturally presented ligands, where TP53-derived peptides were directly eluted from HLA complexes.
The implication was straightforward: p53 can, at least under some conditions, enter the HLA-C presentation pathway. But, this observation left unresolved a deeper question. If the peptide can bind, why is there so little evidence of dominant immune responses driven through this route?
A partial answer emerges when we contrast this with the much more extensively validated role of HLA-A*02:01. Across multiple studies, p53-derived epitopes presented by HLA-A*02:01 consistently generate measurable CD8⁺ T-cell responses, including classic epitopes such as p53₍264–272₎ (LLGRNSFEV) and related variants (see for example: https://pubmed.ncbi.nlm.nih.gov/10811890/ and https://pubmed.ncbi.nlm.nih.gov/24504111/). This allele appears repeatedly in vaccine design, adoptive T-cell strategies, and immuno-peptidomic datasets. It behaves, in effect, as a reliable conduit between intracellular mutation and immune recognition.
The difference between these two alleles is not simply one of binding affinity. It is a difference in how the intracellular system prepares peptides before they ever encounter the HLA binding groove.
The recent study published in Nature does not mention p53 directly, but it introduces a critical shift in how we should think about antigen presentation. By examining viral DNA loads across large human cohorts, the study demonstrates that variation in antigen presentation is strongly influenced not only by HLA alleles themselves, but by upstream processing enzymes—most notably ERAP1 and ERAP2. These enzymes trim peptide precursors within the endoplasmic reticulum, effectively determining which peptides reach the optimal length and composition required for stable HLA binding. Their functional importance has been demonstrated mechanistically in earlier work showing how ERAP variants reshape the immuno-peptidome (for example: https://www.nature.com/articles/ng.2558 and https://pubmed.ncbi.nlm.nih.gov/21804560/).
This introduces a gating mechanism that sits upstream of presentation. A peptide may be theoretically compatible with an HLA allele, yet never appear on the cell surface if it is over-trimmed, under-trimmed, or otherwise destabilized during processing.
Seen through this lens, the difference between HLA-A02:01 and HLA-C02:02 becomes more intelligible. HLA-A*02:01 appears to sit in a favorable position within this processing landscape. Its binding motif aligns well with peptides commonly generated by proteasomal cleavage and refined by ERAP trimming, as demonstrated in motif and structural studies of peptide–MHC stability. The result is a relatively high probability that suitable p53-derived peptides will survive the journey from cytosol to cell surface.
HLA-C*02:02, by contrast, operates under tighter constraints. Its peptide-binding preferences are narrower, and its expression levels on the cell surface are lower. More importantly, it appears more sensitive to the exact output of ERAP-mediated trimming. A peptide such as TAKSVTCTY may exist as a valid ligand, but whether it is produced consistently and in sufficient quantities may depend on subtle variations in ERAP1 and ERAP2 activity.
This introduces a form of biological contingency. In one individual, a TP53-derived peptide may be efficiently trimmed and presented by HLA-C*02:02. In another, the same peptide may be degraded before it ever reaches the binding groove. The difference is not in the tumor, nor in the TP53 sequence itself, but in the inherited configuration of the antigen-processing machinery.
What makes HLA-C particularly interesting is that its role extends beyond classical T-cell presentation. Unlike HLA-A, HLA-C molecules are primary ligands for killer immunoglobulin-like receptors (KIRs) on natural killer cells. This means that the peptide presented by HLA-C can influence not only whether a T cell recognizes a target, but whether an NK cell is inhibited or activated. Importantly, peptide sequence itself can modulate KIR binding affinity, introducing a second layer of regulation beyond simple HLA presence.
In this context, a TP53-derived peptide presented by HLA-C*02:02 may have a dual function. It may serve as a weak or conditional T-cell antigen, while simultaneously modulating NK cell behavior through peptide-dependent effects on KIR binding. The immune consequence of p53 presentation through HLA-C may therefore be less about direct cytotoxic targeting and more about altering the threshold at which NK cells engage.
This possibility reframes the earlier observation. The presence of a TP53 peptide in the HLA-C*02:02 ligandome is not simply evidence of presentation; it may represent a regulatory signal embedded within the innate immune interface.
The 2026 findings reinforce this interpretation by showing that small differences in HLA and ERAP combinations can produce measurable differences in viral control at the population level. If such differences can influence the handling of persistent viral DNA, as shown in the Nature study, it is reasonable to infer that they could also shape the presentation of endogenous stress signals such as p53.
What emerges is a layered model. HLA-A02:01 represents a pathway of efficiency and dominance: peptides are reliably generated, stably presented, and frequently recognized. HLA-C02:02 represents a pathway of conditionality: peptides may be presented, but only under specific processing conditions, and with consequences that extend into NK cell regulation.
Within this framework, p53 is not a single antigenic entity but a source of potential signals whose visibility depends on the alignment of multiple systems—proteasomal cleavage, ERAP trimming, HLA binding, and receptor engagement on immune cells.
This aligns closely with the broader Codondex perspective, in which sequence context—particularly intronic and repetitive elements—may influence not only transcriptional behavior but also downstream processing dynamics. If short RNA or peptide fragments derived from TP53 can alter condensate formation or proteostasis, as discussed in related Codondex work, they may also indirectly influence which peptides are generated and presented.
The question is no longer whether p53 can be presented by a given HLA allele. That has already been answered in the affirmative for both HLA-A02:01 and HLA-C02:02. The more precise question is under what intracellular conditions that presentation becomes stable, visible, and immunologically consequential.
It is within this narrower and more mechanistically grounded space that new opportunities emerge—particularly in understanding how antigen processing variability might be leveraged, or corrected, to expose signals that are otherwise hidden in plain sight.