Decoding T Cell Receptor Cross-Reactivity

Although specificity is considered a hallmark of adaptive immunity, the vast universe of potential peptides that can be presented by an MHC protein makes T cell and TCR cross-reactivity a necessity. It has been estimated that, for a functioning immune system, any individual TCR must be able to productively recognize at least one million different peptide-MHC complexes. TCR structural adaptability, the flexibility of the peptide and MHC, and molecular mimicry can all contribute to cross-reactivity. This complexity makes it difficult to predict the cross-reactivity of TCRs from structural (much less sequence) information. Access to more detailed data about what kind of ligands any individual TCR prefers and how this relates to structural and sequence data is needed to better understand cross-reactivity and specificity. Along with techniques such as yeast or mammalian cell display, combinatorial and positional scanning libraries (PSLs) can be used for investigating the peptides compatible with a given TCR. Analyzing results from co-culture experiments performed with positional scanning libraries allows for identification of amino acid substitutions in the peptide that are acceptable for a given TCR. Scoring these results allows for further investigation into more unique peptides that can be recognized. The recognition of a new index peptide can then be examined via another PSL. This analysis can be performed for multiple rounds, interrogating the landscape of possible peptides a given TCR can and cannot recognize. The results presented here underscore the effectiveness of this iterative extended PSL (ePSL) approach for mapping the cross-reactivity profiles of TCRs, enabling the identification of distinct peptides with substantial divergence from the cognate sequence. By applying the ePSL with structurally well-characterized systems, we can generate datasets indicating which peptides are and are not recognized that can be directly integrated with structural and biophysical analyses. This thorough approach enhances our understanding of TCR specificity and cross-reactivity and establishes a framework for future studies aimed at predicting TCR cross-reactivity and informing the rational design of TCR-based therapeutics.尽管特异性被视为适应性免疫的标志性特征，但主要组织相容性复合体（Major Histocompatibility Complex, MHC）蛋白可呈递的潜在肽段谱系极为庞大，这使得T细胞与T细胞受体（T cell receptor, TCR）的交叉反应性成为必然。据估算，对于功能正常的免疫系统而言，任意单个TCR必须能够有效识别至少100万种不同的肽-MHC复合物。TCR的结构适应性、肽段与MHC的柔性，以及分子模拟，均可能促成交叉反应性。这种复杂性使得从结构信息（更遑论序列信息）预测TCR的交叉反应性极具难度。为更好地理解交叉反应性与特异性，亟需获取更详尽的数据，明确单个TCR偏好的配体类型，以及该偏好如何与结构和序列数据相关联。除酵母或哺乳动物细胞展示等技术外，组合与位置扫描文库（positional scanning libraries, PSLs）可用于探究与特定TCR兼容的肽段。通过分析基于位置扫描文库的共培养实验结果，能够鉴定出肽段中可为该TCR所接受的氨基酸替换。对这些结果进行评分后，可进一步研究可被该TCR识别的更独特的肽段。随后，可通过另一组PSL对新鉴定的索引肽段的识别情况进行验证。该分析可开展多轮，以全面探查特定TCR能够识别与无法识别的肽段范围。本文呈现的结果证实了这种迭代式扩展位置扫描文库（extended PSL, ePSL）方法在绘制TCR交叉反应性图谱方面的有效性，能够识别出与同源序列存在显著差异的独特肽段。通过结合结构特征明确的系统应用ePSL，我们可生成包含可识别与不可识别肽段的数据集，该数据集可直接整合至结构与生物物理分析中。这种全面的研究方法加深了我们对TCR特异性与交叉反应性的理解，并为未来旨在预测TCR交叉反应性、指导基于TCR的治疗药物理性设计的研究建立了框架。