Design Principles of Life, Disease, and Aging

# Immune circuits, inflammation, and fibrosis as dynamical systems *Evidence guide: Mathematical circuit models, proposed surveillance, and primary mouse results. Model states do not establish universal clinical mechanisms.* After an injury, the tissue becomes inflamed: immune cells pour in, and the area becomes red, hot and swollen. Repair follows. Usually the wound closes and the tissue heals. But sometimes, especially after prolonged or repeated injury, scar tissue replaces healthy tissue. That is fibrosis. In the liver, lungs, kidney and heart, it can lead to organ failure. Two cell types are central. Macrophages: immune cells that eat debris and send signals. And myofibroblasts: activated fibroblasts that lay down collagen fibers and contract the wound. They form a circuit. Each secretes growth factors that help the other survive and divide, and myofibroblasts also stimulate themselves. In 2018, the labs of Ruslan Medzhitov and Uri Alon characterized this kind of circuit with cells grown together in culture. What stops such a circuit from growing without limit? Fibroblasts stop dividing when crowded, a carrying capacity, and macrophages consume the growth factor they respond to. The 2018 study found that features like these are needed for a stable two-cell circuit. Mutual support is positive feedback, and positive feedback can create bistability: two different stable states for the same system. Picture a ball on a landscape with two valleys. One valley is healing, with no activated cells. The other is fibrosis, a self-sustaining population of macrophages and myofibroblasts. Between them is a ridge: an unstable threshold. The same system, with the same parameters, ends up in one valley or the other depending on its history. A brief burst of inflammation stays on the healing side. A longer or repeated burst pushes the populations over the ridge, where they sustain each other even after the inflammation stops. In the 2020 model by Adler and colleagues, this offers an explanation for why there is a limited time window to stop inflammation, and why scars mature slowly: trajectories crawl near the unstable point. The model also predicts two kinds of fibrosis: hot fibrosis, with both macrophages and myofibroblasts, and cold fibrosis, with myofibroblasts alone. In this view, a fibrotic scar is not a dead end but a living steady state, with cells that keep turning over. That is consistent with observations that some fibrosis can regress, for example in the liver after hepatitis C is cured. This suggested a target. A 2025 study found that, in mice, heart attack led to cold fibrosis, and that antibodies against TIMP-1, a myofibroblast autocrine factor, reduced myofibroblast proliferation and fibrosis. Fibrosis in every organ need not follow exactly this circuit; it is a simplified model, tested in specific settings. Immune circuits may protect tissues in another way. Glands with a secrete-and-grow circuit face mutant cells that over-secrete and over-grow. In 2020, Korem Kohanim and colleagues proposed that T cells may remove such hypersecreting cells, by recognizing proteins from the hormone production pathway. The proposed cost is a fragility: in some people, this surveillance could escalate into autoimmune disease, such as type-one diabetes or Hashimoto's thyroiditis. This is a proposed systems-level mechanism, not an established explanation for all autoimmunity. Sources: [sysmed-notes](https://www.weizmann.ac.il/mcb/alon/courses/system-medicine-2022-2023), [adler2020](https://doi.org/10.1016/j.isci.2020.100841), [zhou2018](https://doi.org/10.1016/j.cell.2018.01.015), [miyara2025](https://doi.org/10.1016/j.cels.2025.101198), [korem2020](https://doi.org/10.1016/j.immuni.2020.04.022).