Regulatory T Cells and the IL-2 Paradox
Low-dose IL-2 expands regulatory T cells. High-dose IL-2 expands the cells they are meant to restrain. The same cytokine, the same receptor family, opposite clinical effects \u2014 and the explanation is a subunit that does no signalling at all. CD25 raises the receptor's affinity for IL-2 roughly a hundredfold while contributing nothing to the intracellular signal, so a Treg wins the competition for scarce IL-2 simply by carrying more of it. Raise the concentration and that advantage disappears: the intermediate-affinity CD122/\u03b3c receptor on memory CD8 and NK cells is engaged too, and the tolerogenic intent inverts. The second thing worth carrying away is that the lineage is not fixed. IL-6 opposes the same STAT5 signal that holds the FOXP3 enhancer demethylated, so a Treg can become an IL-17 producer \u2014 the cell at the top of this map becoming the cell at the bottom of it. Click any protein for the matching Assay Genie In Vivo antibody or ELISA kit.
The paradox is a subunit that does not signal. The IL-2 receptor exists in two functional forms. CD122 paired with the common gamma chain binds IL-2 with intermediate affinity and is carried constitutively by memory CD8 and NK cells. Add CD25 and affinity rises roughly a hundredfold \u2014 but CD25 has essentially no cytoplasmic tail and transduces nothing. It is a capture subunit. Regulatory T cells express it at high levels constitutively, so when IL-2 is scarce they take almost all of it; raise the dose and the intermediate-affinity receptor is saturated too, and the same cytokine now expands cytotoxic lymphocytes. That is the whole of the dose paradox, and it is why IL-2 muteins engineered away from CD122 are being pursued for autoimmunity while the opposite bias is pursued in oncology. It is also why the clone matters more than the target at the IL-2 node: JES6-1A12 occludes the CD122-binding face and so biases the complex towards CD25-high cells, which is a different experiment from neutralisation even though both are described as anti-IL-2.
Suppression by contact is removal, not inhibition. CTLA-4 is routinely described as an inhibitory receptor, and that description is misleading. Its dominant mechanism is trans-endocytosis: it physically strips CD80 and CD86 off the antigen-presenting cell and degrades them intracellularly. The Treg is not signalling into the APC \u2014 it is removing the costimulatory ligands that CD28 would otherwise use. This matters experimentally, because competition with CD28 and ligand removal predict different outcomes when CD80 and CD86 are blocked separately, and they are separately blockable here. The same caution applies to CD28 itself, which is required to generate Tregs in the thymus and is opposed by them in the periphery: one molecule, two opposite roles, depending on when you interrupt it. CD39 and CD73 add a contact-independent enzymatic route \u2014 ATP to AMP to adenosine \u2014 that needs no cytokine at all.
The TNF receptor family tunes it in both directions, and that is not a complication to be averaged out. TNFR2 signalling expands and stabilises Tregs, which is why anti-TNF therapy removes a Treg-supporting signal at the same time as an inflammatory one. OX40 is constitutive on Tregs and its agonism opposes FoxP3 \u2014 the reverse of its effect on conventional T cells, so an OX40 agonist has two mechanisms running simultaneously. GITR is the most misread node on the map: agonism expands Tregs and simultaneously makes responders resistant to them, so published net effects run in both directions depending on which population dominates. None of these is a discrepancy in the literature; they are consequences of the same receptor being present on both sides of the interaction, which is the design problem any experiment here has to solve first.
The In Vivo tie-in, and where it stops. 27 of the 38 nodes carry a functional-grade antibody \u2014 the deepest such layer of any map in this library, and the reason this pathway is worth drawing at all. CD25 is blockable in both species, mouse and human clone 7G7B6, and the mouse side carries the depleting-versus-blocking distinction that dominates this field: PC61 depletes, and a great many results attributed to CD25 blockade are in fact depletion results. The same distinction is available at CTLA-4, where three functional-grade clones are stocked including the 9D9 and 9H10 pair. Add CD80 and CD86 separately, CD28, CD3, CD4, CD8, CD127, Thy-1, CD62L, CD44, LAG-3, TIGIT in both species, CD155, PD-1, PD-L1, IL-10, pan-TGF-\u03b2, IFN-\u03b3, OX40, 4-1BB, ICOS and TNFR2, and most of the interactions on this map can be interrupted rather than only observed. Stated plainly, what cannot: CD122 and CD132 are research grade only \u2014 clones 5H4 and 3E12 \u2014 which is a real gap, since the intermediate-affinity receptor is half the paradox. CD39, CD73, GITR, IL-6R and IL-23 are biosimilar only; no functional-grade equivalent exists for any of them. And FoxP3, STAT5, JAK3 and mTOR are intracellular, so they are measurement nodes rather than intervention nodes \u2014 which is also why every Treg sorting strategy uses the CD25-high, CD127-low surface surrogate instead of the transcription factor that actually defines the cell. For research use only; not for use in diagnostic or therapeutic procedures.
Every protein node links to a product — ELISA kit, In Vivo antibody or research antibody.