Avalilação PREreview de Divergent ontogeny of Tissue Resident Memory and Tissue Resident Exhausted CD8⁺ T cells underlies distinct functional potential

Manuscript title: Divergent ontogeny of Tissue Resident Memory and Tissue Resident Exhausted CD8+ T cells underlies distinct functional potential

Summary: This manuscript presents an ambitious and comprehensive dissection of CD8+ tissue-resident memory (TRM) and tissue-resident exhausted (TR-TEX) CD8+ T cell states arising after acute vs chronic LCMV infection, with an emphasis on distinguishing bona fide tissue-resident memory (TRM) cells from tissue-resident exhausted (TR-TEX) cells. The authors conclude that TRM and TR-TEX cells are phenotypically similar but transcriptionally and developmentally distinct. TRM cells can convert into exhausted states, whereas committed TEX cells lack the ability to generate TRM. The authors further propose new transcriptional signatures for the cells and illustrate human relevance. The study is exceptionally thorough and provides mechanistic clarity to the long-standing overlap that is known between tissue-resident and exhausted T cells in disease contexts. The paper demonstrates a Tox-dependent requirement for TR-TEX residency, defines asymmetric developmental plasticity between TRM and TEX lineages, and links transcriptional state to functional responsiveness to PD-1 blockade. Overall, this is a highly impactful manuscript, but could benefit from clarification, additional data, or modified interpretation.

Major Comments

1.        The claim of irreversible lineage divergence (TEX cannot become TRM) would benefit from more support. There is strength in the sort- rechallenge experiments, as they support developmental potential for each starting population in these conditions. However, fate mapping or lineage tracing would be required to discount for example that there is rare TEX-PROG → TRM conversion or that TR-TEX and TRM derive from distinct clones. This reviewer suggests the authors soften the language claiming absolute lineage isolation unless additional lineage-tracing approaches are added.

2.        The authors demonstrate promising but limited human data. The manuscript currently lacks full characterization of TR-TEX in human tissues. Consider including this as it would strengthen translational claims or consider commenting on the limitations of the human data in results or discussion.

3.        Throughout the manuscript there are statements that cells proliferated or adoptively transferred cell subsets survived poorly without direct measurements of proliferation or cell death to support the statements. To improve clarity, these data could be provided or statements qualified. Example: (Page 18, line 416-417).

4.        Only some tissues showed functional changes to TEX cells after anti-PD-L1. Thus, it is unclear if functional divergence between TRM and TR-TEX after anti-PD-L1 treatment reflected cell-intrinsic responsiveness or local environment responsiveness. This could be addressed through further interrogation of tissue-specific changes.

Minor Comments

1. Labeling inconsistencies.

a.        There are some labeling inconsistencies throughout the text that the authors could consider clarifying.  “Spl Tcircm”, “Arm Spl Tmem/eff”, “Arm Spl Tmem”, “TCIRCM". Example: Ext. Fig. 3a figure legend says Arm Spl TCIRCM, the text (line 247) says Spl Arm TCIRM, the chart says Arm Spl Tmem/eff, and the key for heat map says Arm Spl Tmem (only one arm of TCIRCM?) for the same data. Consider choosing a naming convention early and applying it consistently across figure panels, legends and text. If there are instances where one arm of TCIRCM is used, please make this clear.

b.        There are many places throughout the text, legends and figures where authors could clarify which tissue is being discussed. Example: Fig. 2e; please insert SI into the figure and text (page 11, line 247).

2.        Figure 1.

a.        Page 7 (Lines 127–129) states that CD69+CD103+ P14 cells are abundant and references Figure 1b. Please clarify, is this sentence meant to say kidney CD69+CXCR6+ P14 cells and refer to Ext. Fig1c?

b.        For improved interpretation, please consider listing the organs in the same order for plots Fig 1b, Fig 1c-d.

c.        Page 7 (Lines 130-134): States “similar populations are found after either acute or chronic infection in some tissues”. Due to the quantitative differences seen, consider rephrasing to clarify similar in phenotype but different in number to help with contextualization. 

d.        Page 7-8 (Lines 144–147): Extended Fig. 1e is referenced to support broad similarity in residency-associated molecules across tissues, but the panel only shows SI comparisons and only CD69⁺CD103⁺ TRM/TR-TEX—not all subsets mentioned. Consider clarifying what is directly shown versus inferred and introduce the use of splenic TCIRCM controls in the text.

e.        Figure 1h has inconsistencies in wording of schematic. Consider correcting the wording to be the same for each organ and marker.

f.           Ext. Data Fig 1i-k: Providing mention of CD244, TIM3 and CD39 use in the text could assist in context for the reader. Specifically, the significant difference between acute and chronic TIM3 expression, since TIM3 has been put forth as a marker for terminal exhaustion.

g.         Fig. 1g: P14 is written twice in figure key.

3.        Figure 2.

a.        Figure legend for Fig2c states “see Supplementary Figure 2”, correct to Extended Data 2.

b.        On page 10 (Lines 220-222), TEX subsets are introduced. Consider citing the studies defining TEX cells to help introduce these cells to non-experts.

c.        Numerous transcriptional clusters (e.g., SG Rps/Rplhi, ISG, Prolif, SG Tex-Prog) appear prominently in UMAPs and heatmaps but are not mentioned in Results. Consider introducing concise definitions when they first appear in Fig. 2c or adding explicit descriptions to the figure legends.

d.        Page 11 (Lines 247–250): The text references Extended Data Fig. 3a for SI Arm TRM or SI Cl13 TR-TEX to Spl Arm TCIRCM, but that comparison is shown in Fig. 2e. Extended Fig. 3a instead shows Arm TRM vs Arm Spl TCIRCM overlap with CL13 TR-TEX vs Spl TEX-TERM. It would be helpful for the authors to update figure references here.

e.        It would be informative to reference Table 1 when Fig2e is introduced as it provides the full dataset for that figure.

f.           Page 12 (Line 263) should say TRM and TR-TEX cells.

g.         Fig. 2f-g and Extended Data Fig. 3d show upregulated DEGs only. Consider clarifying in the figure legend and text that only upregulated DEGs are shown in the figure.

h.        Page 12&13 (Lines 276-279) state higher gene expression and chromatin accessibility and refers to Fig 2g-h. Please correct to include Fig 2f (gene expression). Please do the same for page 13 (line 282).

i.            Please correct panel letter k in Figure 2 legend.

4.        Figure 2h and Extended Data Figure 4a. Placing a key defining the peak shading into the figure itself may be helpful for interpretation (Tmem vs Trm-green, Trm vs TR-ex-blue, TR-ex vs Trm-pink).

5.        Extended Data Fig. 2 contains presentation inconsistencies.

a.        2a. Naïve UMAP cluster is mis-colored and does not match key. It would be helpful to update color schemes for clarity.

b.        Naïve cells are listed in some keys and not in others, but naïve cell populations are not apparent in any UMAPs. Can the authors comment on this?

c.        Key for Ext. Fig2i, SG vs Spl. Please clarify what cell subset is being referenced in the population labeled CL13 SG Progenitor.

d.        Inconsistent populations are shown. SG vs Spl included SG progenitor and SG Rps/Rplhi, which are not shown in others figures. Liv vs Spleen heatmaps include Liv CCL3/CCL4hi population which is not described in the manuscript. Please comment on the usefulness of these comparisons in the text.

6.        Extended Data Fig. 3 Can authors add panel letter f to the figure legend? Ext. Fig3 also states that proportion of genes are shown, however, proportion data is not evident in the graphs. Updating the figure or text would provide clarity.

7.        For clarity, please add the group names to Extended Data Fig 4a.

Competing interests

The authors declare that they have no competing interests.

Use of Artificial Intelligence (AI)

The authors declare that they did not use generative AI to come up with new ideas for their review.