A protein called HS1BP3 was found at higher levels in memory CD8+ T cells, the cells that can respond again when the same target returns. In antigen-specific mouse T-cell experiments, cells without Hs1bp3 had fewer cells with a memory-like profile, lower levels of memory-related markers and genes, weaker measures of mitochondrial oxidative phosphorylation and reduced recall potential. The pattern suggests that HS1BP3 is tied to both the identity of memory cells and the energy system that supports them.
That finding gives researchers a possible way to think about why some T cells retain a stronger second response than others. It remains an association in a preclinical study, however. The experiments do not show that changing HS1BP3 would improve immunity in people, and the supplied results do not give exact effect sizes or confidence intervals. The direction of the finding is clearer than its size or clinical importance.
The signal emerged from a comparison of T-cell states
The signal first appeared in a comparison of effector and memory CD8+ T cells. Using proteomics, a method that measures many proteins at once, the researchers identified 759 proteins that differed between the two states. Hs1bp3 showed the greatest differential expression in that comparison. Memory cells also showed mitochondrial enrichment, and a separate transcriptomic analysis found enrichment for oxidative phosphorylation, the process cells use to generate energy, in the memory state.
Researchers then tested the pattern in antigen-specific OT-1 CD8+ T cells. In the central infection experiment, naïve CD45.1+ cells were transferred into age-matched CD45.2+ C57BL/6 recipients. The recipients were challenged with Lm-OVA the next day, and the donor cells were later analyzed or sorted. This setup let the researchers examine the transferred cells after an immune challenge and compare their memory-related features.
Removing HS1BP3 weakened several memory measures
Compared with controls, Hs1bp3-deficient OT-1 CD8+ T cells showed fewer CD44+CD62L+/− cells, the memory-related populations measured in the study. They also had lower levels of memory-associated markers and genes. Mitochondrial oxidative-phosphorylation proteins were reduced, as were oxygen-consumption measurements and spare respiratory capacity. The latter is a measure of how much extra energy-producing capacity a cell has available when demand increases.
The metabolic results sat alongside a functional result: the altered cells had impaired recall potential. In other words, they were less able to produce the later response measured by the researchers after the initial challenge. The same overall direction appeared across cell phenotype, memory-related molecular signals, mitochondrial proteins and respiration. Even so, the supplied summary does not quantify those group differences, so it is not possible to judge the size of each change from the information provided.
Human samples provided context, not a treatment test
The human part of the work was limited to samples used for laboratory analysis. It included human peripheral-blood samples and samples from people with lung cancer. They were collected under Ethics Committee approval 2025‐KY‐1023, with informed consent from all participants. These samples add a human-cell reference point, but they do not amount to a clinical test of HS1BP3-engineered T cells.
The study's experimental design sets a clear boundary around the result. Its central infection work used transferred naïve OT-1 cells and C57BL/6 recipients, rather than patients receiving an engineered cell product. The evidence therefore cannot establish whether HS1BP3-engineered CAR-T cells are safe or effective in people, whether the proposed relationship holds in other antigens or CAR designs, or whether any benefit would persist over time.
For now, HS1BP3 is best described as a testable lead in T-cell memory research. The work connects higher HS1BP3 expression in memory cells with mitochondrial and oxidative-phosphorylation enrichment, then shows that deleting Hs1bp3 coincided with a weaker memory and recall profile. Confirming the finding will require testing the relationship in additional human T-cell products and cancer models.
Paper data and sources
Original title: Nicotinamide Metabolism Constrains Memory CD8T Cell Formation Through a Putative HS1BP3-SIRT1-FOXO3-BCL6 Axis.
Authors: Siyang Wang, Xiaoqian Fan, Yifan Huang et al.
Journal/Repository: Advanced science (Weinheim, Baden-Wurttemberg, Germany)
Status: Peer-reviewed
First online: 2026-08-21
DOI: 10.1002/advs.77284
Original paper