Regulatory T cells safeguard liver health during metabolic-associated steatohepatitis

In human single-cell RNA-sequencing datasets, Treg frequency is significantly higher in MASH livers than in control livers.

“Notably, the Treg frequency was significantly elevated in MASH livers relative to that of controls (Fig. 1E), echoing the Treg enrichment observed in the CDAHFD and AMLN models.”

• cites [22] sourceCould not verify

  Extended Data Figure 3 compares T-cell and innate lymphoid-cell fractions in healthy versus cirrhotic livers, but the available text does not report a significant increase in Tregs in MASH livers. Extended Data Figure 7 examines NASH liver samples, but its frequency analysis concerns scar-associated macrophages, not Tregs. Searches for Treg and regulatory T cells found no passage establishing the claimed comparison. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [23] sourceCould not verify

  The ‘Augmented CD8+PD1+ T cells in human-NASH’ section reports human liver scRNA-seq and enrichment of CD8+PD1+ T cells, not Treg cells. Its Fig. 3 legend identifies Treg cells but gives no comparison of their frequency between NASH and control livers. Searches for elevated Treg frequency and regulatory T-cell enrichment found no passage establishing the claimed significant difference. (LLM, conf=0.76)

  Experimental — accuracy not yet measured

In published single-cell RNA-sequencing datasets from MASH patient livers, FOXP3+ cells expressing canonical Treg transcripts such as IL2RA, CTLA4, and TNFRSF18 are present.

“Focusing our analysis on the CD4+ T cell compartment, we identified FOXP3+ cells expressing canonical Treg cell transcripts, including IL2RA, CTLA4 and TNFRSF18 (Fig. 1D, right panel).”

• cites [22] sourcePartly supported

  Extended Data Figure 3 describes single-cell analysis of liver T cells from five healthy and five cirrhotic donors, so the paper provides a relevant T-cell dataset. It does not report FOXP3+ cells coexpressing IL2RA, CTLA4 and TNFRSF18. Its NASH analysis in Extended Data Figure 7 uses whole-liver microarray data, not a NASH-specific single-cell T-cell dataset. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites [23] sourcePartly supported

  The ‘Augmented CD8+PD1+ T cells in human-NASH’ section and Fig. 3 caption describe single-cell RNA-sequencing of liver T cells from patients with NAFLD or NASH; the caption also mentions Treg cells. Neither those passages nor the human scRNA-seq Methods report FOXP3+ cells co-expressing IL2RA, CTLA4 and TNFRSF18, so the specific cell-marker finding is not established by our copy. (LLM, conf=0.78)

  Experimental — accuracy not yet measured

Acute Treg depletion in CDAHFD-fed mice triggers a marked increase in hepatic immune cell numbers, most prominently CD4+ Tconv and CD8+ T cells.

“This acute loss of Tregs triggered a marked increase in hepatic immunocyte numbers, most prominently numbers of CD4+ T conventional (Tconv) cells and CD8+ T cells (Fig. 5B), accompanied by elevated proliferation according to Ki-67 staining (Fig. 5C).”

In Treg-deficient livers, metabolic pathways, particularly bile-acid and fatty-acid metabolism, are suppressed.

“In contrast, metabolic pathways, particularly those related to bile-acid and fatty-acid metabolism, were suppressed.”

The Itgb1+ Treg cluster is marked by S100a10−, S100a4−, Slamf6−, and S1P-receptor-encoding transcripts S1pr1 and S1pr4.

“The Itgb1+ cluster was marked by S100a10−, S100a4−, Slamf6-, and S1P-receptor-encoding (S1pr1, S1pr4) transcripts.”

Expression of the activation markers CD44 and CXCR6 on liver Tregs remains unchanged under CDAHFD and LFD conditions in PpargΔTreg mice.

“Expression of other activation markers, including CD44 and CXCR6, remained unchanged under the two feeding conditions (Fig. 4D, E).”

Acute Treg depletion in CDAHFD-fed mice is accompanied by elevated proliferation according to Ki-67 staining.

“This acute loss of Tregs triggered a marked increase in hepatic immunocyte numbers, most prominently numbers of CD4+ T conventional (Tconv) cells and CD8+ T cells (Fig. 5B), accompanied by elevated proliferation according to Ki-67 staining (Fig. 5C).”

Single-cell RNA-sequencing of liver Tregs from CDAHFD-fed and control Foxp3Gfp mice resolves four discrete subtypes: naïve, Itgb1+, Pparg− reparative, and Pparg+ reparative.

“Integrated cell-clustering analysis of Tregs from the two conditions resolved four discrete subtypes: “naïve,” “Itgb1+,” “Pparg− reparative” and “Pparg+ reparative” (Fig. 3A).”

Genes characteristic of naïve Tregs, including Ccr7, S1pr1, and Tcf7, are significantly downregulated in liver Tregs in both MASH models.

“In parallel, genes characteristic of naïve Tregs, including Ccr7, S1pr1 and Tcf7, were significantly downregulated in both models.”

In AMLN-fed mice, Treg depletion increases plasma transaminases and hepatic triglyceride content.

“Similarly, in AMLN-fed mice, Treg depletion increased plasma transaminases and hepatic triglyceride content (Fig. 6E).”

In a transcriptomic time course of wild-type CDAHFD-fed mice, module C1, enriched for metabolic functions, is highest at baseline and progressively declines as disease unfolds.

“One module (C1), enriched for metabolic functions, was highest at baseline and progressively declined over disease unfolding.”

The distribution of other innate and adaptive T cell subsets is unaffected by Pparg deletion in Tregs.

“The distribution of other T cell subsets, both innate and adaptive, was likewise unaffected (SI Appendix, Fig. S3).”

Liver Tregs from both dietary MASH models acquire transcriptional features characteristic of nonlymphoid-tissue Tregs.

“As expected, liver Tregs from both dietary models acquired the characteristic transcriptional features of nonlymphoid-tissue Tregs (SI Appendix, Fig. S2A, B).”

Using Pparg-Tdt reporter mice, PPARγ+ Tregs are confirmed to be enriched in the liver after CDAHFD feeding.

“Using Pparg-Tdt reporter mice, we confirmed the enrichment of PPARγ+ Tregs in the liver after CDAHFD feeding (Fig. 4A).”

Pparg deletion in Tregs has no significant effect on the frequency or absolute number of liver Tregs in either healthy mice or mice with MASH.

“Pparg gene deletion in Tregs had no significant effect on the frequency or absolute number of liver Tregs in either healthy mice or those with MASH (Fig. 4B).”

Gene-set enrichment analysis shows a strong increase in inflammatory pathways, such as TNFα and IFN responses, in Treg-deficient livers.

“Gene-set enrichment analysis showed a strong increase in inflammatory pathways, such as responses to TNFα and IFN (Fig. 6B), consistent with our flow cytometric findings (Fig. 5E, K).”

Similar results are obtained when Tregs are depleted at 8 weeks of CDAHFD feeding.

“Similar results were obtained when Tregs were depleted at 8 weeks of CDAHFD feeding (SI Appendix, Fig. S4C–G).”

Treg-activation-associated programs are significantly elevated in liver Tregs from both CDAHFD- and AMLN-fed mice.

“Treg-activation-associated programs were significantly elevated in Tregs from both CDAHFD- and AMLN-diet-fed mice (Fig. 2G, H), further confirming that MASH drives the hepatic Treg compartment toward an activated, eVAT-Treg-like state.”

The PPARγ− reparative Treg subtype expresses higher levels of Gzmb, Ccr2, Ccr3, and Icos.

“While the PPARγ− reparative subtype expressed higher levels of Gzmb, Ccr2, Ccr3 and Icos, its PPARγ+ counterpart was characterized by Pparg and Gata3 expression, as well as high levels of transcripts encoding co-stimulatory or co-inhibitory receptors (Pdcd1, Tnfrsf4, Tnfrsf9, Tnfrsf18) and TCR signaling molecules (Dusp1, Fosl2, Junb, Nr4a1–3, Fos, Jun), indicative of recent antigenic stimulation (”

In CDAHFD-fed mice, hepatic CD45+ immune cells accumulate within the first week of the enriched diet.

“Longitudinal profiling of CDAHFD-fed mice revealed an accumulation of hepatic immunocytes (CD45+ cells) already within the first week after introducing the enriched diet (SI Appendix, Fig. S1A).”

After Treg depletion, CD4+ Tconv cells produce the corresponding cytokines IFN-γ, IL-4/5/13, and IL-17A.

“They also produced the corresponding cytokines IFN-γ, IL-4/5/13 and IL-17A (Fig. 5E).”

Bulk RNA-sequencing of livers from CDAHFD-fed mice after Treg depletion shows 380 genes significantly modulated compared with Treg-positive controls.

“380 genes were significantly modulated (Padj <0.05, |FC| >2) in Treg-deficient livers compared with Treg-positive controls (Fig. 6A).”

After Treg depletion, CD4+ Tconv cells show elevated expression of Tbet, GATA3, and RORγt.

“Post-depletion, CD4+ Tconv cells exhibited elevated expression of Tbet, GATA3 and RORγt, supreme regulators of Th1, Th2 and Th17 cell differentiation (Fig. 5D).”

CDAHFD-fed, Treg-deficient mice have elevated plasma ALT and AST.

“These molecular changes were accompanied by elevated measures of plasma Alanine Aminotransferase (ALT) and Aspartate Aminotransferase (AST) in CDAHFD-fed, Treg-deficient mice (Fig. 6D, SI Appendix, Fig. S5C).”

After Treg depletion, CD8+ T cells display higher levels of PD-1 and CXCR6.

“CD8+ T cells displayed higher levels of PD-1 and CXCR6 (Fig. 5F), a phenotype that has been linked to autoaggressive cytotoxic lymphocytes implicated in hepatocyte injury during MASH development (14, 23).”

Upregulation of TCR signaling genes in Tregs from MASH livers suggests that disease-associated antigens might drive their increased accumulation.

“The upregulation of TCR signaling genes in Tregs from MASH livers suggests that disease-associated antigens might drive their increased accumulation.”

Tregs substantially accumulate and adapt during MASH in both mice and humans.

“Our studies have uncovered a substantial accumulation and adaptation of Tregs during the course of MASH in both mice and humans.”

PPARγ is largely dispensable for liver Treg accumulation, even under MASH-associated inflammatory stress.

“Thus, although PPARγ marks a transcriptionally distinct subset of liver Tregs, it is largely dispensable for their accumulation in the liver, even in the context of MASH-associated inflammatory stress.”

The effects of acute Treg depletion are not attributable to diphtheria toxin itself, because DT-treated Foxp3.Dtr− littermates show no such changes.

“These effects were not attributable to DT itself, as DT-treated Foxp3.Dtr− littermates showed no such changes (SI Appendix, Fig. S4A, B).”

The murine liver Treg compartment in MASH adopts an activated, nonlymphoid-tissue-associated phenotype resembling that of eVAT Tregs.

“Beyond this increase, the murine liver Treg compartment adopted an activated, nonlymphoid-tissue-associated phenotype resembling that of eVAT Tregs.”

Hepatic Tregs play a non-redundant role in restraining activation of both innate and adaptive immune cells.

“Functionally, hepatic Tregs played a non-redundant role in restraining the activation of both innate and adaptive immunocytes.”

Tregs in the liver emerge as central regulators of immunologic tone in MASH.

“Alongside their established roles in eVAT, exercised skeletal muscle and systemic metabolic homeostasis (5, 25, 31–33), Tregs in the liver emerge as central regulators of the immunologic tone in MASH.”

PPARγ is a phenotypic rather than functional marker of the PPARγ+ liver Treg population.

“Although hepatic Tregs in mice with MASH showed increased PPARγ expression, genetic perturbation studies indicated that it was largely dispensable for their accumulation, indicating that PPARγ is a phenotypic rather than functional marker of this population.”