Arginase-1–Expressing Macrophages Suppress Th2 Cytokine–Driven Inflammation and Fibrosis

The textbook said Arginase-1 was the culprit. The enzyme that macrophages express during parasitic infections, the one that churns out the building blocks of scar tissue, was supposed to be making things worse. That story was wrong. Arginase-1 in macrophages is not the accelerant; it's the brake. And a two thousand nine paper by Pesce, Ramalingam, Mentink-Kane, Wilson, and colleagues at the National Institutes of Health, published in PLOS Pathogens, is what forced the correction. Here's the logic that needed dismantling. Arginase-1, or Arg1 for short, is an enzyme that converts the amino acid L-arginine into ornithine and urea. In the liver, that reaction is part of the urea cycle. But in macrophages, Arg1 is not constitutively expressed; it switches on in response to the Th2 cytokines interleukin 4 and interleukin 13. And once it's active, ornithine gets metabolized further into polyamines, glutamate, and crucially, L-proline. L-proline is the backbone of collagen. Collagen is scar tissue. So the chain seemed obvious: Th2 infection triggers interleukin 4 and interleukin 13, those cytokines activate Arg1 in macrophages, Arg1 produces proline, and proline feeds fibrosis. Macrophages expressing Arg1 were classified as alternatively activated macrophages, or AAMs, and they were cast as the fibrotic villains of chronic inflammation. Schistosoma mansoni made this question urgent. This parasitic flatworm deposits eggs in the liver and intestine, provoking a strong Th2 response dominated by interleukin 4, interleukin 5, interleukin 13, and interleukin 21. That response drives hepatic granulomas, clusters of immune cells walling off the eggs. Over time, the granulomas resolve into fibrosis, scarring the liver and backing up blood pressure in what's called portal hypertension. Arg1-expressing macrophages accumulate in and around those granulomas. So everything pointed in one direction. Arg1 in macrophages equals fibrosis. Except it didn't. To test causality rather than just correlation, the team built conditional knockout mice using the Cre-lox system, a technique that lets you delete a specific gene in a specific cell type. They flanked a critical portion of the Arg1 gene with loxP sites, then crossed those mice to animals expressing Cre recombinase from the lysozyme M locus, which is active in macrophages. The resulting Arg1-minus/flox;LysMcre mice delete Arg1 specifically in macrophages. As a second, broader model, they crossed the floxed Arg1 mice to Tie2-Cre animals, which delete the gene across all macrophage populations, achieving greater than ninety-nine percent Arg1 ablation in macrophages. The characterization confirmed that the tool worked. Thioglycollate-elicited macrophages from the LysM-Cre knockouts showed roughly a sevenfold reduction in arginase activity after Th2 cytokine stimulation compared to controls. Nitric oxide production, the output of nitric oxide synthase two, Arg1's metabolic rival for the same L-arginine substrate, remained intact, slightly elevated even, after interferon gamma plus lipopolysaccharide stimulation. Other markers of alternative activation, like the mannose receptor and Ym1, were induced normally in the knockout macrophages. So Arg1 deletion didn't break the AAM program; it removed one specific enzymatic step within it. Worm burdens and tissue egg counts were similar in knockout and control mice at nine, twelve, and twenty-two weeks post-infection, confirming that susceptibility to the parasite itself was unchanged. What happened to the mice when Arg1 was gone from their macrophages was the opposite of what the textbook predicted. The knockouts got sicker, faster. With a standard infection dose of thirty-five cercariae, the infectious larval form, roughly forty percent of Arg1-minus/flox;LysMcre mice had died by week twelve. Zero controls died in that interval. With a higher infectious dose, only twenty percent of the knockout mice survived beyond week ten, while controls hit fifty percent mortality only by week twelve. The animals that survived were in worse shape: livers enlarged, hydroxyproline content elevated — hydroxyproline is the biochemical marker of collagen deposition — and second harmonic confocal microscopy confirmed the fibrosis was real. Liver granulomas in the knockout mice were nearly threefold larger at the chronic stage than in controls, and they failed the normal downmodulation that granulomas undergo as infection progresses. Intestinal granulomas showed the same pattern. Egg shunting into the lungs increased, indicating that portal hypertension had driven the development of porto-systemic shunts, the liver's bypass valves when pressure backs up too far. The team checked every alternative explanation. Was it a Th1 flip, with macrophages without Arg1 pivoting to nitric oxide-driven hepatotoxicity? No. The inducible nitric oxide synthase inhibitor aminoguanidine had no effect on survival. Serum lipopolysaccharide levels were not elevated, ruling out endotoxemia. Liver enzymes rose after infection, but comparably in both groups. What did differ was the immune response itself. CD4-positive T cells from the livers of Arg1-deficient mice produced significantly more interleukin 4, interleukin 5, and interleukin 13. Bromodeoxyuridine incorporation showed greater CD4-positive T cell proliferation in the knockouts. The Th2 response, which was supposed to be fueled by Arg1, was actually being held in check by it. The in vitro experiments then mapped exactly how. Pesce and colleagues used the Cre-lox mice to run co-culture experiments with two antigen systems. In the OT-II and ovalbumin setup, macrophages from wild-type or knockout mice were pretreated for thirty-six hours with a cocktail of interleukin 4, interleukin 13, and granulocyte-macrophage colony-stimulating factor at one nanogram per milliliter each, enough to make them alternatively activated. Then CFSE-labeled OT-II CD4-positive T cells, which respond to ovalbumin peptide, were added. After ninety-six hours, wild-type AAMs completely failed to stimulate T cell proliferation. Knockout macrophages, those lacking Arg1, behaved like untreated macrophages and did not suppress proliferation at all. So the suppression required Arg1. But what was the mechanism? The paper ruled out the obvious candidates. Neutralizing antibodies to interleukin 10 and transforming growth factor beta one, alone or together, had no effect on the macrophage-mediated inhibition. These are the canonical immunosuppressive cytokines, and their irrelevance here was striking. What did matter was L-arginine itself. When the researchers added exogenous L-arginine back to wild-type AAM co-cultures, T cell proliferation was restored, and the dose required scaled with how much Arg1 activity had been induced. At high cytokine pretreatment doses of five nanograms per milliliter, large quantities of exogenous arginine, up to one thousand micromoles, were needed to rescue proliferation. At lower pretreatment doses, less was needed. The relationship was clean: the more Arg1 the macrophages expressed, the more arginine they consumed, and the more arginine depletion suppressed the T cells. A second antigen system, using soluble egg antigen from schistosomes, produced the same result in CD4-positive T cells from infected mice, confirming the effect was not an artifact of an artificial antigen. The picture that emerges is this. Arg1-expressing macrophages are not manufacturing the substrates of fibrosis and handing them to myofibroblasts. They are consuming the amino acid that CD4-positive T cells need to keep proliferating. When those T cells can't sustain their response, the Th2 cytokine signal — interleukin 4, interleukin 5, interleukin 13 — weakens. And it is that Th2 signal that drives the granulomatous inflammation and collagen deposition. The macrophages are limiting the very effector response that causes the scarring. This reconciles two things that seemed contradictory. AAMs expressing Arg1 can produce ornithine and proline, the collagen substrates, and in isolated biochemical terms, that pathway exists and is active. But at the system level, the dominant effect of macrophage Arg1 during a Th2 infection is suppression of the CD4-positive T cell response that drives fibrosis, not provision of building materials for it. The enzyme runs in one direction metabolically, and the immune consequence flows in the opposite direction from what was assumed. The clinical implication the authors flag is direct. Pharmacologic Arg1 inhibitors, compounds like S-(2-boronoethyl)-L-cysteine, are not macrophage-selective. They block the enzyme everywhere. If macrophage Arg1 is a brake on Th2-driven pathology, then a non-selective inhibitor removes that brake. You're not targeting the accelerant; you're disabling the regulatory circuit. In diseases driven by Th2 inflammation, and there are many beyond schistosomiasis including allergic disease and fibrotic conditions, that distinction matters enormously. The same molecule, in the same cell type, can be immunosuppressive or pathogenic depending on the immunological context it sits in. That's not a nuance; that's the finding. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.