An autoimmune stem-like CD8 T cell population drives type 1 diabetes

In the paper

“CD8 T cell-mediated autoimmune diseases result from the breakdown of self-tolerance mechanisms in autoreactive CD8 T cells1.”

What the sources say

• Supported

  The Introduction explains that central and peripheral tolerance control autoreactive T cells and prevent autoimmunity. The Central tolerance section links pathogenic, autoreactive CD8+ T cells to self-antigen recognition and a mismatch with antigens displayed in the thymus. Together, these passages support the claim as a broad mechanistic paraphrase; the copy provides no page numbers. (LLM, conf=0.84)

  cites [1] Bluestone, J. A., Bour-Jordan, H., Cheng, M. & Anderson, M. T cells in the control of organ-specific autoimmunity. J. Clin. Invest. 125, 2250–2260 (2015). source

In the paper

“T1D becomes clinically evident when the majority (over 80%) of β-cells have been destroyed, requiring a prolonged latent period of β-cell killing over time (months to years in humans)5.”

What the sources say

• Partly supported

  The paper’s opening discussion (page unavailable in our copy) describes a long prediabetic period during which β cells are killed over many years. But it says that as much as 30% of β-cell mass may remain at onset, so it does not support the claim that clinical disease requires destruction of over 80%. (LLM, conf=0.88)

  cites [5] Herold, K. C., Vignali, D. A., Cooke, A. & Bluestone, J. A. Type 1 diabetes: translating mechanistic observations into effective clinical outcomes. Nat. Rev. Immunol. 13, 243–256 (2013). sourceExperimental — accuracy not yet measured

In the paper

“The non-obese diabetic (NOD) mouse model is a clinically relevant model of T1D6,7 and shares many features with the human disease, including spontaneous CD4 and CD8 T cell responses to many of the same β-cell antigens6–9.”

What the sources say

• ?Not checked — no full text of the sourcecites [6] Anderson, M. S. & Bluestone, J. A. The NOD mouse: a model of immune dysregulation. Annu. Rev. Immunol. 23, 447–485 (2005). source
• Partly supported

  The Introduction and Conclusions describe NOD mice as a spontaneous model of T1D with important similarities to human disease. The Diabetogenic CD4 T Cells section reports broad β-cell autoreactivity and identifies both CD4 and CD8 T cells in NOD mice; Human Studies also reports insulin-reactive T cells in people with T1D. But the paper does not establish spontaneous CD4 and CD8 responses to many of the same β-cell antigens in both species, and the Conclusions explicitly leave the human counterpart of a described NOD CD4 response as an open question. (LLM, conf=0.86)

  cites [7] Unanue, E. R. Antigen presentation in the autoimmune diabetes of the NOD mouse. Annu. Rev. Immunol. 32, 579–608 (2014). sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [8] Tsai, S., Shameli, A. & Santamaria, P. CD8+ T cells in type 1 diabetes. Adv. Immunol. 100, 79–124 (2008). source
• Partly supported

  The Introduction describes NOD diabetes as chronic and similar to human T1D, and the Discussion says both have CD4 and CD8 T cells in their islets. The Discussion reports responses to insulin and other β-cell antigens in NOD mice, but says this study did not determine T-cell specificity; it does not establish that mouse and human T cells respond to many of the same antigens. No page numbers are available in our copy. (LLM, conf=0.87)

  cites [9] Zakharov, P. N., Hu, H., Wan, X. & Unanue, E. R. Single-cell RNA sequencing of murine islets shows high cellular complexity at all stages of autoimmune diabetes. J. Exp. Med. 217, e20192362 (2020). sourceExperimental — accuracy not yet measured

In the paper

“Haematopoietic19, somatic14 and cancer stem cells20 in mice and humans have been identified via serial transplantation of limiting cell numbers into immunodeficient mice.”

What the sources say

• Partly supported

  The Discussion describes this study’s serial single-cell transfers as modeled on hematopoietic stem-cell studies, and the Results compare its recovery rate with that of single hematopoietic stem cells. Those passages support the hematopoietic analogy, but do not establish the claimed use of immunodeficient mice to identify hematopoietic stem cells in both mice and humans. The separately cited somatic and cancer stem-cell claims are not assessed here. (LLM, conf=0.83)

  cites [19] Graef, P. et al. Serial transfer of single-cell-derived immunocompetence reveals stemness of CD8+ central memory T cells. Immunity 41, 116–126 (2014). sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [14] van Loosdregt, J. & Coffer, P. J. The role of WNT signaling in mature T cells: T cell factor is coming home. J. Immunol. 201, 2193–2200 (2018). source
• ?Not checked — no full text of the sourcecites [20] Lapidot, T. et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature 367, 645–648 (1994). source

In the paper

“In autoimmunity, self-reactive T cells acquire cytotoxic effector functions and cause destruction of self1.”

What the sources say

• Partly supported

  The Conclusion (Figure 1; page unavailable) says autoreactive T cells are more prone to activation in autoimmune disease and that autoreactive ex-Foxp3 cells may contribute to tissue destruction. It describes those cells producing pro-inflammatory cytokines, but does not establish that self-reactive T cells acquire cytotoxic effector functions as the claim specifies. (LLM, conf=0.82)

  cites [1] Bluestone, J. A., Bour-Jordan, H., Cheng, M. & Anderson, M. T cells in the control of organ-specific autoimmunity. J. Clin. Invest. 125, 2250–2260 (2015). sourceExperimental — accuracy not yet measured

In the paper

“TOX has been described as a key driver of CD8 T cell exhaustion in tumours and chronic infections, settings of persistent antigen encounter3.”

What the sources say

• Supported

  The ‘Late dysfunction molecular programmes’ section describes TOX as an important regulator of T cell dysfunction programmes in tumours and chronic infection. The ‘Exhaustion’ section links both settings to persistent antigen stimulation. The review also cautions that TOX’s effects vary by T cell state, but the claim says only that it has been described as a key driver. (LLM, conf=0.90)

  cites [3] Philip, M. & Schietinger, A. CD8+ T cell differentiation and dysfunction in cancer. Nat. Rev. Immunol. https://doi.org/10.1038/s41577-021-00574-3 (2021). source

In the paper

“TCF1hi progenitor CD8 T cells that give rise to TCF1lo differentiated progeny have been described in chronic viral infections and cancers3”

What the sources say

• Partly supported

  The ‘Late dysfunction molecular programmes’ section says that TCF1+ progenitor-like CD8+ T cells give rise to TCF1low differentiated cells during chronic viral infection. For cancer, it describes TCF1+ early dysfunctional tumour-infiltrating cells differentiating into TCF1low late dysfunctional cells, but says evidence for a stable population of tumour-reactive TCF1+ progenitors within tumours is lacking. Page numbers are unavailable in our copy. (LLM, conf=0.86)

  cites [3] Philip, M. & Schietinger, A. CD8+ T cell differentiation and dysfunction in cancer. Nat. Rev. Immunol. https://doi.org/10.1038/s41577-021-00574-3 (2021). sourceExperimental — accuracy not yet measured

In the paper

“In cancers or chronic viral infections, contexts in which T cells encounter persistent antigen, T cells enter a hyporesponsive state generally referred to as T cell exhaustion or dysfunction3.”

What the sources say

• Supported

  The “Exhaustion” section says chronic antigen stimulation during infection leads to a hyporesponsive state called exhaustion. It also says T cells in progressing tumours become hyporesponsive after continuous encounters with tumour antigens, and that exhaustion is used to describe their dysfunction. (LLM, conf=0.92)

  cites [3] Philip, M. & Schietinger, A. CD8+ T cell differentiation and dysfunction in cancer. Nat. Rev. Immunol. https://doi.org/10.1038/s41577-021-00574-3 (2021). source

In the paper

“The gold-standard assay to evaluate stem cell functions remains serial transplantation in animal models.”

What the sources say

• Supported

  The Results describe generating a secondary progeny as a prerequisite for conclusively assigning stemness. The Introduction and Discussion describe serial transfer into animal hosts as an in vivo clonogenic assay that stringently evaluates self-renewal, multipotency, and functional reconstitution. Although the paper does not use the words “gold standard,” it treats this approach as a stringent test of stem-cell function. (LLM, conf=0.86)

  cites [19] Graef, P. et al. Serial transfer of single-cell-derived immunocompetence reveals stemness of CD8+ central memory T cells. Immunity 41, 116–126 (2014). source
• ?Not checked — no full text of the sourcecites [14] van Loosdregt, J. & Coffer, P. J. The role of WNT signaling in mature T cells: T cell factor is coming home. J. Immunol. 201, 2193–2200 (2018). source
• ?Not checked — no full text of the sourcecites [20] Lapidot, T. et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature 367, 645–648 (1994). source

In the paper

“Moreover, anti-CD3 treatment (teplizumab) has been shown to induce long-term protection (currently up to six years) from T1D in at-risk individuals, but the mechanisms responsible for this long-term protection remain incompletely understood29,30”

What the sources say

• Partly supported

  The Results (Efficacy; page not provided) show that teplizumab delayed diagnosis of type 1 diabetes in high-risk relatives, and the Discussion reports a median delay of two years. The paper does not establish protection lasting six years: median follow-up was 745 days, although the longest follow-up was 2683 days (Results, Participants; page not provided). It describes delayed progression, not prevention of diabetes. (LLM, conf=0.87)

  cites [29] Herold, K. C. et al. An anti-CD3 antibody, teplizumab, in relatives at risk for type 1 diabetes. N. Engl. J. Med. 381, 603–613 (2019). sourceExperimental — accuracy not yet measured
• Partly supported

  The Results section reports a sustained delay in T1D among high-risk relatives treated with teplizumab; eight treated participants followed beyond five years remained undiagnosed. It does not establish protection specifically through six years, and describes delayed diagnosis rather than permanent prevention. Page numbers were not provided in our copy. (LLM, conf=0.85)

  cites [30] Sims, E. K. et al. Teplizumab improves and stabilizes beta cell function in antibody-positive high-risk individuals. Sci. Transl. Med. 13, eabc8980 (2021). sourceExperimental — accuracy not yet measured

In the paper

“In acute infections, antigen-specific CD8 T cells differentiate into functional cytotoxic effector and memory T cells2.”

What the sources say

• Supported

  The opening passage (labeled Abstract; page not supplied) describes antigen-specific T cells differentiating into effectors during an acute infection and CD8+ T cells becoming cytotoxic lymphocytes. A later passage under the same label says a subset of pathogen-specific effector CD8+ T cells survives and matures into memory cells. (LLM, conf=0.94)

  cites [2] Kaech, S. M. & Cui, W. Transcriptional control of effector and memory CD8+ T cell differentiation. Nat. Rev. Immunol. 12, 749–761 (2012). source

In the paper

“TCF1 is a transcription factor critical for T cell memory, longevity and self-renewal13.”

What the sources say

• Supported

  The Introduction identifies Tcf1 as a transcription factor critical for CD8+ T cell memory. The section on Tcf1+ stem-like CD8+ T cells describes their self-renewal and capacity to sustain durable responses, supporting the claim’s longevity and self-renewal language. (LLM, conf=0.88)

  cites [13] Escobar, G., Mangani, D. & Anderson, A. C. T cell factor 1: a master regulator of the T cell response in disease. Sci Immunol 5, eabb9726 (2020). source

In the paper

“The pLN TCF1hi population was highly clonally diverse compared with the pLN and pancreas TCF1lo populations and contained an abundance of rare, unexpanded clones that were exclusive to the pLN TCF1hi pool (Fig. 4e, Extended Data Fig. 9c–e).”

In the paper

“Clustering and trajectory analyses revealed that cells with high and low expression of Tcf7 were at opposite ends of the trajectory; Tcf7hi cells displayed a stem-like transcriptional program, whereas Tcf7lo cells harboured a terminally differentiated and cytotoxic program (Fig. 4a, b, Extended Data Fig. 8a–d).”

In the paper

“Donor 8.3 cells bifurcated into TCF1hiCD39lo and TCF1loCD39hi populations in the pLN, whereas only TCF1loCD39hi T cells emerged in the pancreas, recapitulating the same phenotypes of endogenous IGRP CD8 T cells in wild-type NOD mice (Extended Data Fig. 5b–d).”

In the paper

“FTY720-treated mice lost half of the initial pancreatic CD8 T cell population, whereas control mice had significantly more CD8 T cells in the pancreas compared with the beginning of treatment (Fig. 3h, Extended Data Fig. 6e–g).”

In the paper

“NOD/SCID hosts that received pLN TCF1hi cells became diabetic whereas hosts that received TCF1lo cells did not (Fig. 3b, Extended Data Fig. 6a).”

In the paper

“We identified intermediate cell states along the trajectory, which retained Tcf7 expression but lost some stemness-associated genes and upregulated genes associated with the Tcf7lo cluster (Fig. 4a, b, Extended Data Fig. 8b–f).”

In the paper

“High Sell (which encodes CD62L) and Entpd1 (which encodes CD39) RNA transcript and protein expression marked the opposite ends of the differentiation trajectory, while intermediate cell states did not express Sell (or CD62L) or Entpd1 (or CD39).”

In the paper

“Insulin B-chain-specific CD8 T cells also bifurcated into distinct TCF1hiCD39loPD1lo and TCF1loCD39hiPD1hi populations in the pLN and displayed a consistently TCF1loCD39hiPD1hi phenotype in the pancreas (Extended Data Fig. 2c, d).”

In the paper

“Mice treated with FTY720 were completely protected from T1D and had significantly reduced insulitis compared with control mice (Fig. 3f, g).”

In the paper

“by which time approximately 80% of NOD mice in our colony have developed T1D (Extended Data Fig. 1b).”

In the paper

“Even as few as 100 TCF1hi cells were sufficient to induce T1D, whereas as many as 100,000 TCF1lo cells were not (Fig. 3b).”

In the paper

“Despite the expression of these inhibitory markers, IGRP T cells were highly functional and produced high levels of IFNγ and TNF upon ex vivo stimulation (Fig. 1c).”

In the paper

“TCF1hi and TCF1lo cells, comprising approximately 80% and 20% of the IGRP-specific population, respectively, could be identified in the pLN as early as 5 weeks of age, and persisted throughout disease progression and mouse lifespan (25 to 30 weeks).”

In the paper

“Gene set enrichment analyses (GSEA) and gene ontology classification showed that pLN TCF1hi cells were enriched for pathways involved in MYC and E2F signalling and DNA replication, whereas pLN TCF1lo cells were enriched for pathways associated with T cell activation, cytotoxicity and apoptosis (Extended Data Fig. 3b, c).”

In the paper

“The select clones that were shared between the pLN TCF1hi and other populations expanded significantly in size upon downregulation of TCF1; consequently, the majority of clonotypes found in TCF1lo cells were shared with the TCF1hi population (Extended Data Fig. 9f).”

In the paper

“pLN IGRP T cells expressed moderate levels of TOX and pancreatic IGRP T cells expressed minimal levels of TOX, in contrast to exhausted and dysfunctional T cells in chronic viral infections and tumours (Fig. 1d, Extended Data Fig. 1d).”

In the paper

“IGRP-specific CD8 T cells (hereafter referred to as IGRP T cells) can be found in the pLN and pancreas as early as 5 weeks of age, and their frequency increases over time, peaking at around 0.3% and 20% of CD8 T cells in the pLN and pancreas by 15–20 weeks, respectively”

In the paper

“(note: transferred donor T cells were not detected in pancreas before day 12)”

In the paper

“demonstrating that pLN TCF1hi cells can both self-renew and produce differentiated progeny.”

In the paper

“Although CD4 T cells initiate pancreatic infiltration and provide help to CD8 T cells, the CD8 T cells primarily responsible for β-cell destruction8”

What the sources say

• ?Not checked — no full text of the sourcecites [8] Tsai, S., Shameli, A. & Santamaria, P. CD8+ T cells in type 1 diabetes. Adv. Immunol. 100, 79–124 (2008). source

In the paper

“AP are necessary to generate and replenish the pool of pancreatic AM, which are short-lived and thus unable to independently sustain the prolonged β-cell destruction that causes T1D.”

In the paper

“We propose that the induction and maintenance of diabetogenic T cell responses requires two functionally distinct populations: short-lived pancreatic AM to destroy β-cells and self-renewing pLN AP to replenish the AM population, thereby enabling sustained β-cell damage that culminates in T1D.”

In the paper

“In conclusion, pLN TCF1hi cells represent a stem-like autoimmune progenitor (AP) population that self-renews long-term and gives rise to terminally differentiated pancreatic TCF1lo autoimmune mediators (AM) that destroy β-cells (Fig. 4f).”

In the paper

“No statistical methods were used to pre-determine sample size.”

In the paper

“The investigators were not blinded to allocation during experiments and outcome assessment.”

In the paper

“TCF1 is one of the targets of canonical Wnt signalling, which maintains stemness in adult tissue stem cells, tumours, and peripheral T cells14,15.”

What the sources say

• ?Not checked — no full text of the sourcecites [14] van Loosdregt, J. & Coffer, P. J. The role of WNT signaling in mature T cells: T cell factor is coming home. J. Immunol. 201, 2193–2200 (2018). source
• ?Not checked — no full text of the sourcecites [15] Reya, T. et al. A role for Wnt signalling in self-renewal of haematopoietic stem cells. Nature 423, 409–414 (2003). source

In the paper

“FTY720, a sphingosine-1-phosphate receptor agonist that blocks lymph node egress and has been shown to protect NOD mice from T1D21,22”

What the sources say

• ?Not checked — no full text of the sourcecites [21] Maki, T., Gottschalk, R., Ogawa, N. & Monaco, A. P. Prevention and cure of autoimmune diabetes in nonobese diabetic mice by continuous administration of FTY720. Transplantation 79, 1051–1055 (2005). source
• ?Not checked — no full text of the sourcecites [22] Brinkmann, V. et al. Fingolimod (FTY720): discovery and development of an oral drug to treat multiple sclerosis. Nat. Rev. Drug Discov. 9, 883–897 (2010). source

In the paper

“Stem-like functionality of pLN TCF1hi CD8 T cells and generation of differentiated TCF1lo CD8 T cells are independent of lymphopenic conditions.”