Functional dissection of regulatory T cell plasticity and stability

Treg are crucial players in creating immune tolerance and maintaining immune homeostasis. Malfunction or insufficient numbers of Treg have been associated with diverse conditions, ranging from atopic disease, such as allergic asthma to autoimmune diseases. Treg are characterized by expression of the transcription factor Foxp3, which is essential for establishing and maintaining their suppressive transcriptional program. The importance of this transcription factor is demonstrated by the fact that loss of Foxp3 expression leads to the development of a lethal multi-organ autoimmune disease in both mice and human. Treg have generally been believed to be a terminally differentiated population with long-term stability in a healthy setting. However in recent years, several studies have suggested that signals in the microenvironment such as stress, danger or inflammation can trigger the loss of Foxp3 expression and acquisition of effector cytokines, resulting in so-called ex-Foxp3 cells.Different models have been proposed towards the origin and function of the ex-Foxp3 cell population. The generation of ex-Foxp3 cells may be a programmed response to acute inflammatory triggers, in order to aid in the initiation phase of the immune response, a model known as 'the acute response model'. However, evidence in vitro and in vivo have produced mixed results. Alternatively, it has been argued that genuine Treg are highly stable in vivo, and that "ex-Foxp3 cells" instead originate from an uncommitted population of Foxp3neg T cells, who transiently upregulate Foxp3. The nature of Treg stability therefore remains highly controversial. The question is therapeutically relevant, with increasing utilization of Treg in cell therapy. With latent autoimmune potential, retaining Foxp3 expression and suppressive status is key in any clinical use of Treg as a cell therapy. It has been proposed that certain Treg markers are associated with a higher potential for stability, however, it remains unclear as to whether these identify an unstable subset, or whether it represents an intermediate stage in fluctuating probabilistic deconversion.In this dissertation, we dissected the fundamental concepts of Treg instability. First, we formally tested whether Treg instability is a stochastic event or a subset property. We showed that microbial challenge does not result in an increased Treg instability, therefore finding no support for the acute infection model. Using an adoptive transfer model into a lymphopenic environment, we further dissected whether Treg instability is the result of probabilistic deconversion or a characteristic of a particular Treg subset. By transferring bona fide Treg and assessing their ability to become ex-Foxp3 cells over several sequential transfers, we demonstrated that only a fraction of Treg spontaneously deconvert to exFoxp3 cells.Further, we analysed the reconversion potential of exFoxp3 cells to the Treg identity. Due to their potential self-reactive nature, it is important to assess the reconversion potential of ex-Foxp3 cells. We demonstrated that ex-Foxp3 cells are refractory towards re-conversion, both in vitro and in vivo, indicating that instability is a terminal, rather than transient, state.Finally, using multi-omics approaches, we dissected the instability within the Treg population. We were able to identify the unstable Treg subset to be enriched for, though not limited to, Nrp1neg Treg bearing a naïve phenotype and thymic-derived RTE Treg. Therefore, we believe these results indicate Treg instability to be the result of a lack of commitment to the Treg identity.With increasing usage of Treg cell-based clinical trials, and the identified risk of autoreactive Treg becoming inflammatory ex-Foxp3 cells, the need to avoid Treg instability is paramount. The stochastic model would suggest that Treg for cell therapy should be treated in a manner to stabilize Foxp3 expression, limiting plasticity after injection, and indeed much effort has been made in this direction. Our results, demonstrating that Treg plasticity is a property restricted to an unstable subset, would suggest a radically different approach: rather than aid Treg stability, the 'purging' of the Treg population of unstable clones, by exposure to a destabilizing environment, may leave behind a residual highly stable Treg remnant population. A "destabilize and repurify" strategy may thus counter-intuitively result in a population more capable of fate-retention and safer for clinical use than the previous "stabilize and treat" approach. These models would, however, require validation in human Treg systems before incorporation into cell therapy trials.

Jaar van publicatie:2021

Toegankelijkheid:Open