Scalable transcription factor mapping uncovers the regulatory dynamics of natural and synthetic transcription factors in human T cell states

Why this preprint matters?

Transcription factors (TFs) regulate gene networks that determine cell fate and state. Although many TFs have been linked to immune cell identities, their precise roles in coordinating gene programs and fate decisions remain unclear. This preprint introduces a method to map paralogous TF binding sites and gene targets in human primary CD8⁺ T cells across distinct activation states. Defining TF dynamics in states such as exhaustion can guide strategies to rewire dysfunctional T cells, providing a powerful foundation for engineering immune cells to fight chronic infections and cancer.

Background and goal

Heterogeneous cytotoxic CD8⁺ T cell states shape immune response effectiveness. Beyond effector and memory phenotypes essential for clearing infections and cancer, chronic antigen exposure can drive dysfunctional exhausted states. These states arise from gene programs coordinated by TFs. For example, TCF7 is linked to stem-like and naïve programs, while TOX promotes exhaustion. Interestingly, both are expressed in human effector memory CD8⁺ T cells. However, the TF binding sites and target genes underlying these states remain undefined, limiting insight into TF binding dynamics and their roles in modulating T cell fate. The lack of reliable antibodies for TFs enriched in T cells has hindered TF mapping using ChIP-seq and CUT&RUN. To address this, the preprint’s authors optimized the transposon-based Calling Cards (CCs) method for primary T cells and developed TFlex, a scalable, multiplexed approach to map paralogous natural and synthetic TFs.

Glossary

Hyperactive piggyBac transposase: An engineered, high activity piggyBac enzyme that inserts transposons at TTAA sites. When fused to a TF, it deposits “calling card” tags near TF binding sites, enabling genomic mapping of TF activity.

DESynR TFs: Synthetic, evolution-guided TFs designed to combine customizable DNA-binding domains with regulatory modules that control gene expression in precise ways.

Experimental Design

 

Key Findings

Application of Calling Cards in primary human CD8 T cells

To validate CCs method in primary T cells, the authors started by quantifying peaks per TF (TF-binding sites) and comparing with ATAC-seq peaks previously identified in peripheral CD8 T cells in vivo, showing considerable overlap between in vitro and in vivo data. Additionally, genes bound within 1 kb of transcription starting sites were enriched for T cell-associated GO terms, indicating CCs accuracy in determining TF-binding sites in T cells.

1. TCF7 and TOX binding converge at enhancers of human memory CD8 T cells. Both TCF7 and TOX co-bound enhancers of genes linked to stemness and central memory subsets, but most TOX-only, TCF7-only, and co-bound enhancers (~65–75%) were associated with genes expressed in effector memory. This data indicates that TCF7 and TOX role in human CD8 T cells is similar to their function in mouse CD8 T cell exhaustion, where TCF7 and TOX regulate an axis of memory- and effector- associated genes.TOX and
2. TCF7 coordinate related gene programs in both memory and exhaustion states. As TCF7-bound stemness programs declined, TOX-bound effector programs increased. Similarly, TCF7-bound stemness program was reduced in central memory and progenitor exhausted states and TOX was enriched in terminally exhausted subsets. Overall, these findings indicate that TOX and TCF7 co-regulate related gene programs, with TCF7 sustaining stemness and TOX directing either exhaustion or terminal effector differentiation based on cell state.
3. TF expression did not correlate tightly with the activity of its target gene module. This suggests that subset-specific cofactors can modulate TF function, leading to distinct TF-induced gene programs rather than a shared one.

Map paralogous transcription factors in primary T cells with TFlex

To validate TFlex, the authors simultaneously mapped eight bZIP family TFs (including JUN, FOSL1, and BATF) and five domain-swapped DESynR TFs that enhance anti-tumour T cell functions ³. Although most binding sites were shared, TF-specific peaks were detected, confirming TFlex’s ability to resolve unique binding sites.

1. Synthetic domain-swapped bZIP TFs acquire unique DNA-binding properties and target gene effects. Even though binding sites were enriched for bZIP motifs similar to the natural JUN, FOSL1 and BATF TFs, the authors identified unique binding sites for the synthetic construct JUN.FOS.BATF (about 15%). Furthermore, genes co-bound by JUN.FOS.BATF and the parent TF show discordant expression, where the gene was decreased by the parent TF but increased by JUN.FOS.BATF and vice versa

Key conclusions

The authors present a scalable, antibody-free method to map TF binding in primary human T cells. This approach reveals gene networks that define cell states and TF functions, and it can distinguish natural or engineered TFs that antibody-based methods cannot. The data suggest that TFs may regulate subset-specific gene programs rather than having a single universal function. TOX and TCF7 appear to interact through direct enhancer binding, extending on previous findings of their co-expression in memory and exhausted human CD8 T cells and cooperation in mice exhausted CD8 T cells. Finally, synthetic TFs acquire unique binding properties that cannot be estimated by the sum of TFs and may underlie their anti-tumour function gain effect.

Questions

1. Do you anticipate a limit to the number of TFs that can be studied simultaneously using TFlex method?
2. You suggest that TOX coordinates distinct programs in exhaustion and memory. Based on your TF-binding data and associated gene sets, do you observe differential pathway or biological-process enrichment among these subsets?

References

1. Moudgil, A. et al. Self-Reporting Transposons Enable Simultaneous Readout of Gene Expression and Transcription Factor Binding in Single Cells. Cell 182, 992-1008.e21 (2020).
2. Wang, H., Johnston, M. & Mitra, R. D. Calling cards for DNA-binding proteins. Genome Res.1284 17, 1202–1209 (2007).
3. Takacsi-Nagy, O. et al. Evolutionarily guided transcription factor design programs novel T cell states. 2024.11.06.622344 Preprint (2024).

Tags: human cd8 t cells, t cell states, t-cell exhaustion, transcription factors mapping

Read preprint (No Ratings Yet)