Thanks! This is a great question, we don’t actually see any difference on a transcriptional level between the Tpex biased clones that form memory and the other Tpex. Definitely need to look by other assays though to see if there’s something scRNA-seq is missing

Our results emphasize that clonal heterogeneity underlies the amazing functional flexibility of T cells. While the bulk of a response to chronic infection forms cells adapted to persist in chronic antigen, sacrificing memory capacity, select clones avoid this fate and form protective memory.

fin.

We’re excited to share our results because they suggest that T cells can form protective immunity even in suboptimal setting like tumors and could serve as a source of long-term protection against relapse.

20/

We found one trajectory that was specifically enriched in the aPDL1 group, suggesting that these clones were responding to aPDL1. When we looked at the phenotypes of these clones over time, we saw that these clones preferentially gave rise to Tcm. aPDL1 mobilized clones can make Tcm!

19/

Next, we wanted to understand if aPDL1 mobilized clones formed Tcm. So we again turned to clone tracing to assess clones over the course of aPDL1. Assessing the size and phenotypes of clones at three timepoints, we identified clones that followed several distinct differentiation trajectories.

18/

We treated chronically infected mice with aPDL1 checkpoint blockade and found no differences in T cell subset frequencies or phenotype after viral clearance demonstrating that the treatment did not disrupt Tcm formation

17/

Finally, we wanted to assess clinical relevance of chronic Tcm. Checkpoint blockade immunotherapy can unleash T cells to clear tumors, but can these cells still go on to form Tcm?

16/

To assess if this response is protective, we transferred chronic Tcm and other populations into TCR-/- mice that are unable to clear virus themselves, and found that chronic Tcm significantly reduce pathogen burden to a similar degree as acute Tcm.

15/

Using TCR-seq of the cells from rechallenge, we could determine the pre-rechallenge identity of the clones that expanded best to rechallenge. Chronic Tcm biased clones expanded better than all other chronic clones and as well as acute memory clones.

14/

We looked at the expansion of cells with rechallenge, and as expected, the expansion of polyclonal T cells from the chronic pool was reduced compared to acute memory cells. However, we saw that some clones from chronic donors actually expanded as well as or better than the acute memory clones!

13/

Based on their epigenetic signature, we hypothesized chronic Tcm would have a functional recall response. To test this, we transferred polyclonal T cells from memory mice into new hosts, rechallenged with LCMV, and traced clones over time by sequencing.

12/

We used ATAC-seq to determine if chronic Tcm also shared this epigenetic signature of chronic stimulation. We found that chronic Tcm were almost identical to their acute counterparts and showed minimal signs of exhaustion, while chronic Tem maintained accessibility of exhaustion related loci.

11/

As alluded to earlier, one of the major factors that impairs recall responses chronically stimulated T cells is “epigenetic scarring” or increased accessibility if Tex-associated loci compared to acute memory cells doi.org/10.1038/s41590-021-00975-5

10/

We next wanted to determine where in the Tex pool that Tcm biased clones come from, so we performed longitudinal sampling of mice with LCMV-Cl13 and compared clones between timepoints. We found that Tex-Progenitor biased clones exist during infection and preferentially form chronic Tcm.

9/

Encouraged by this, we wanted to understand the origin of chronic Tcm. We dove into the clonal information in our data and found clones with distinct patterns of differentiation including a clonal trajectory biased to Tcm. In fact, most chronic Tcm were found in clones skewed towards Tcm.

8/

To our surprise, T cells from a cleared chronic infection generally clustered with acute central memory (Tcm) and effector memory cells (Tem). We were especially struck by the presence of “chronic Tcm”, which were transcriptionally almost identical to their acute counterparts.

7/

To address the clonal makeup of the T cell pool after clearance of chronic antigen, we performed scRNA/TCR-seq on antigen-specific T cells after the clearance of chronic LCMV-Cl13. For comparison we also looked at memory cells from acute LCMV-Arm and exhausted (Tex) T cells during LCMV-Cl13.

6/

For example, transfer of one memory CD8 leads to a range of differentiation potentials including clones biased to cytotoxic (SLEC) or stem (MPEC) states upon rechallenge. We hypothesized that there’s similar heterogeneity in exhausted T cells after clearance doi.org/10.1016/j.immuni.2013.07.014

5/

Searching for a model that would explain these findings, we thought about the role that clonal heterogeneity plays in T cell responses. T cell responses are not homogenous, rather made up of individual clones with distinct fate distributions.

4/

The field has been perplexed by contradictory findings: T cells persist long-term after tumor clearance and can protect long-term against relapse. But exhausted T cells also retain a molecular signature of exhaustion, which impairs their functional recall response.

3/

T cells become exhausted with prolonged antigen exposure, such as in tumors and chronic infections, rendering them hypofunctional. Anti-PD1 therapy helps T cells overcome exhaustion and clear tumors, but what happens to the T cells after antigen is cleared?

2/

I’m excited to share our lab’s new preprint! We find that some exhausted T cell clones maintain the capacity form functional memory cells:
www.biorxiv.org/content/10.1...

1/ 🧵

Functional memory T cells are derived from exhausted clones and expanded by checkpoint blockade https://www.biorxiv.org/content/10.1101/2025.02.10.637523v1