Treg-tissue cell interactions in repair and regeneration.

Regulatory T (Treg) cells are classically known for their critical immunosuppressive functions that support peripheral tolerance. More recent work has demonstrated that Treg cells produce pro-repair mediators independent of their immunosuppressive function, a process that is critical to repair and regeneration in response to numerous tissue insults. These factors act on resident parenchymal and structural cells to initiate repair in a tissue-specific context. This review examines interactions between Treg cells and tissue-resident non-immune cells-in the context of tissue repair, fibrosis, and cancer-and discusses areas for future exploration.

Amphiregulin from regulatory T cells promotes liver fibrosis and insulin resistance in non-alcoholic steatohepatitis.

Production of amphiregulin (Areg) by regulatory T (Treg) cells promotes repair after acute tissue injury. Here, we examined the function of Treg cells in non-alcoholic steatohepatitis (NASH), a setting of chronic liver injury. Areg-producing Treg cells were enriched in the livers of mice and humans with NASH. Deletion of Areg in Treg cells, but not in myeloid cells, reduced NASH-induced liver fibrosis. Chronic liver damage induced transcriptional changes associated with Treg cell activation. Mechanistically, Treg cell-derived Areg activated pro-fibrotic transcriptional programs in hepatic stellate cells via epidermal growth factor receptor (EGFR) signaling. Deletion of Areg in Treg cells protected mice from NASH-dependent glucose intolerance, which also was dependent on EGFR signaling on hepatic stellate cells. Areg from Treg cells promoted hepatocyte gluconeogenesis through hepatocyte detection of hepatic stellate cell-derived interleukin-6. Our findings reveal a maladaptive role for Treg cell-mediated tissue repair functions in chronic liver disease and link liver damage to NASH-dependent glucose intolerance.

Diagnostic reasoning prompts reveal the potential for large language model interpretability in medicine.

One of the major barriers to using large language models (LLMs) in medicine is the perception they use uninterpretable methods to make clinical decisions that are inherently different from the cognitive processes of clinicians. In this manuscript we develop diagnostic reasoning prompts to study whether LLMs can imitate clinical reasoning while accurately forming a diagnosis. We find that GPT-4 can be prompted to mimic the common clinical reasoning processes of clinicians without sacrificing diagnostic accuracy. This is significant because an LLM that can imitate clinical reasoning to provide an interpretable rationale offers physicians a means to evaluate whether an LLMs response is likely correct and can be trusted for patient care. Prompting methods that use diagnostic reasoning have the potential to mitigate the "black box" limitations of LLMs, bringing them one step closer to safe and effective use in medicine.