T cells are the immune system’s circulating assassins. Each carries on its surface a T-cell receptor, or TCR, a complex protein assembly capable of recognizing short fragments of foreign or abnormal proteins displayed by other cells. When a TCR binds its target with sufficient strength, the T cell becomes activated, proliferates, and destroys the presenting cell. This mechanism underlies both natural immunity to viruses and the emerging clinical field of engineered T-cell therapy, in which a patient’s own T cells are genetically modified to recognize tumors. Yet natural and therapeutic TCRs vary enormously in their stability and signaling capacity, and weak receptors translate directly into weak therapeutic cells, a persistent bottleneck in the field.

The Bar-Ilan and Weizmann team attacked this bottleneck with the tools of computational protein design. Rather than screening random mutations in the laboratory, the researchers used two protein-design algorithms to predict specific amino acid changes that would improve the stability and function of the T-cell receptor scaffold. Crucially, they focused on conserved regions of the receptor, the structural framework shared by virtually all TCRs, rather than the variable loops that determine what each receptor recognizes. Mutations identified by the two algorithms were combined to create a single enhanced receptor scaffold, the SET construct, which could in principle be grafted onto many different target-recognition modules.

The logic of targeting the conserved framework is what gives the approach its unusual generality. Most efforts to improve a therapeutic receptor must start from scratch for each new candidate, a process that can take years of trial and error. Because the SET mutations act on the shared structural core of the receptor rather than on its target-binding surface, the researchers expected, and then experimentally confirmed, that a single set of strategically designed changes could favorably influence very different TCR types. In the words of Prof. Fleishman, who led the computational design work together with Dr. Jake Parker, a visiting student from Australia, protein design methods are making inroads into the most challenging areas of biotherapeutic engineering, and the designed mutations are expected to improve any TCR rather than one candidate at a time.

The experimental results were unambiguous. In laboratory cultures, T cells carrying the SET receptor produced substantially stronger immune responses upon encountering their targets and demonstrated a greater ability to kill cancer cells than otherwise identical T cells carrying the original receptor. The enhancement was not a subtle shift in a single readout; the engineered cells responded more vigorously across the functional measures the team examined, consistent with the idea that a more stable receptor scaffold improves the fundamental mechanics of T-cell activation.

The most dramatic findings came from mouse models bearing human tumors. After 83 days, tumors treated with T cells containing the enhanced SET receptor were approximately 35 percent smaller than tumors in untreated control animals. By day 127 of the experiment, every mouse that had received the enhanced cells was still alive, while fewer than half of the control mice had survived. In cancer research, survival differences of this magnitude in preclinical models are rare, and they suggest that the computational improvements translated into a genuine therapeutic advantage rather than a laboratory artifact.

Equally important was the breadth of the effect. The researchers applied the same design strategy to T cells engineered to recognize several distinct cancer-associated targets, including targets linked to melanoma and other difficult-to-treat cancers. They then extended the approach to receptors recognizing viral targets, including components of Epstein-Barr virus and SARS-CoV-2, the virus responsible for COVID-19. In every case, the SET-enhanced T cells showed stronger immune activity than their unmodified counterparts. The consistency across such different targets, from tumor antigens to proteins of two unrelated viruses, is the strongest evidence that the method acts on a universal property of the TCR scaffold rather than on the quirks of any one receptor.

Maria Radman of Bar-Ilan University’s Goodman Faculty of Life Sciences, the study’s first author, described the breadth of the enhancement as one of the most exciting findings of the work. According to the published release, she noted that the team observed enhanced activity not only against cancer-associated targets but also against viral targets, suggesting that the approach could be broadly useful for engineering T cells with stronger and more effective immune responses. That generality matters because the pipeline of candidate therapeutic receptors is far larger than the capacity of any laboratory to optimize them one by one.

Prof. Cohen framed the results as the opening of a new direction for therapy. Computational protein design, he said, is giving researchers a much more effective way to engineer the immune system, and in this study it made T cells substantially more powerful at attacking their targets across both cancer and viral applications. His laboratory has long worked at the interface of T-cell biology and cancer immunotherapy, and the collaboration with the Fleishman group at Weizmann, one of the world’s leading centers for computational protein design, brought together the biological and algorithmic expertise needed to move from calculated mutations to functioning therapeutic cells in animal models.

The implications extend well beyond a single study. If a shared set of designed mutations can reliably strengthen any T-cell receptor, the timeline for translating a promising receptor candidate into a clinically testable therapy could shrink dramatically, replacing years of empirical optimization with a rational design step. The strategy could also complement existing engineered-cell platforms, potentially improving receptors used against solid tumors, where weak signaling and poor persistence remain major obstacles, as well as antiviral T-cell therapies. The findings, however, remain preclinical, demonstrated in laboratory experiments and animal models, and further research will be required to establish the safety and potential clinical applications of the approach before patients can benefit. Even so, the study offers a striking demonstration that algorithms designed to understand protein structure can now hand immunologists molecules that work better than the ones evolution or conventional engineering provided, and that a computation performed on a conserved protein scaffold can ripple outward into stronger immune responses against cancer and viruses alike.

Subject of Research: Computational protein design of T-cell receptors to enhance engineered immune cell therapies for cancer and viral diseases

Article Title: Bar-Ilan University and Weizmann Institute researchers use computational design to engineer immune cells to fight cancer and viruses

Article References: Bar-Ilan University and Weizmann Institute researchers use computational design to engineer immune cells to fight cancer and viruses. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: T cells, T-cell receptor, computational protein design, cancer immunotherapy, SET receptor, Bar-Ilan University, Weizmann Institute, Science Advances, cellular immunotherapy, Epstein-Barr virus, SARS-CoV-2, preclinical study