Pereira lab publishes “A combinatorial transcription factor screening platform for immune cell reprogramming” in Cell Systems

We are delighted to announce the publication of our article “A combinatorial transcription factor screening platform for immune cell reprogramming” in Cell Systems.

Previous strategies to identify transcription factors (TFs) that drive cellular reprogramming – including iterative “N–1” approaches, computational inference, and multiplexed pooled screens – face key limitations, such as low throughput, incomplete TF coverage, or an inability to resolve combinatorial TF logic. To accelerate immune cell reprogramming and unlock its full therapeutic potential, we developed REPROcode, a high-throughput, single-cell screening platform that systematically interrogates complex TF combinations driving immune cell identity.

We first validated REPROcode by directly reprogramming fibroblasts into type 1 conventional dendritic cells (cDC1s). This enabled the identification and optimization of minimal TF combinations, optimal TF stoichiometries, and key co-factors that enhance reprogramming efficiency and fidelity. Building on the cDC1-inducing PU.1, IRF8, and BATF3 (PIB) combination, we found that GATA2 enhances cDC1 fidelity within a narrow dose window, whereas GFI1B increases reprogramming efficiency at higher expression levels.

Beyond cell fate, REPROcode also identified regulators that instruct distinct cDC1 states, including immature (IκB-α/β), immunostimulatory (MXD1, ID2), and CCR7⁺ migratory (BCL6, IRF7) programs.

To assess REPROcode’s ability to resolve TF logic from large pools and across immune lineages, we applied the platform to a curated set of TFs enriched in several DC lineages. This enabled the programming of immune cell diversity and the construction of a hierarchical TF map guiding immune cell reprogramming, positioning PU.1 at the root of reprogrammed immune cells and KLF4 as a major early branch point.

We further demonstrate the versatility of REPROcode by programming natural killer (NK) cells, a cytotoxic lymphoid population highly relevant for immunotherapy. The combination of TBX21, ETS1, NFIL3, and EOMES generated NK-like cells expressing CD56, NKG7, NCR1, and GZMB, with functional CD107a degranulation and cytokine release.

Led by Ilia Kurochkin, together with Pereira Lab members Abigail Altman, Diogo Cabral, Evelyn Halitzki, Luís Oliveira, Malavika Sreekumar Nair, and Daniel Oliveira, this work established REPROcode as a powerful platform to decode immune cell identity regulators and to enable next-generation immunotherapies across oncology, autoimmunity, and regenerative medicine.

We would like to thank our collaborators at Asgard Therapeutics, as well as the groups of Branko Cirovic, Ewa Sitnicka, and Fabian Theis. We would also like to acknowledge the National Academic Infrastructure for Supercomputing in Sweden (NAISS), and Markus Ringnér and Marcel Martin for the long-term support of the National Bioinformatics Infrastructure Sweden (NBIS). This work was supported by the European Research Council (ERC) and the European Innovation Council (EIC), Horizon Europe, NextGenerationEU, Cancerfonden, Swedish Research Council, Novo Nordisk Fonden, Cancer Research Institute, Knut and Alice Wallenberg Foundation, and FCT.

Click here for the full article!

---

About the illustration

In a strategic game of Mancala, each move involves a calculated distribution of pieces, mirroring how transcription factor combinations direct immune cell fate decisions. We present the REPROcode platform to uncover these combinatorial rules, offering a deliberate and strategic roadmap for reprogramming cells toward defined immune identities.

Illustration by Lilith Lawrence.