TNF Receptor Superfamily: TNFR1 vs TNFR2
Blocking TNF cures rheumatoid arthritis in some patients and triggers demyelination in others. That is not a dosing problem or a patient-selection problem. It is the direct consequence of one fact this map is built around: TNF has two receptors that do opposite things. TNFR1 is ubiquitous, carries a death domain, and drives inflammation and apoptosis. TNFR2 is restricted to Tregs, myeloid cells and neurons, has no death domain, and drives survival, repair and suppression. Every anti-TNF antibody in the clinic blocks both. Once you see the superfamily this way — a shared ligand architecture split between death-domain and survival-only receptors — the rest of the family stops looking redundant. Click any protein for the matching Assay Genie ELISA kit or In Vivo antibody.
The death domain is the dividing line. Nineteen ligands and twenty-nine receptors sound unmanageable until you sort the receptors by one structural feature. Those with a cytoplasmic death domain — TNFR1, Fas, and the TRAIL receptors — can recruit TRADD and caspase-8 and kill the cell. Those without one — TNFR2, CD40, 4-1BB, OX40, CD27, RANK, BAFF-R, GITR — can only recruit TRAF adaptors and therefore only signal survival, proliferation and NF-κB. Nothing else about the family predicts function as cleanly. When a paper reports that a TNFRSF member is "pro-apoptotic" or "pro-survival", check the domain before believing the mechanism.
TNFR1 versus TNFR2 is the whole clinical story. TNFR1 is on nearly every cell and mediates the fever, the acute-phase response and the joint destruction. TNFR2 is restricted — highest on Tregs, where it is required for their expansion and suppressive function, and on oligodendrocyte precursors, where it drives remyelination. Non-selective anti-TNF therefore removes a brake at the same time as it removes an accelerator. That is the accepted explanation for why TNF blockade helps rheumatoid arthritis but worsens multiple sclerosis, and why it can precipitate paradoxical psoriasis. It is also why TNFR2-selective agonists are in development: the opposite intervention to the one that made the class famous. If you are modelling anti-TNF, blocking the ligand and blocking TNFR1 selectively are different experiments with different answers.
Signalling is a race between two complexes. TNF binding TNFR1 first assembles complex I at the membrane — TRADD, TRAF2, RIPK1 and cIAP2 — which ubiquitinates RIPK1 and activates NF-κB. Only if that complex fails does RIPK1 disengage and form the cytosolic complex II with caspase-8. So the default outcome of TNF is survival, and apoptosis is what happens when the survival arm is blocked, which is exactly why TNF alone kills so few cells but TNF plus cycloheximide kills readily. cIAP2 is the switch, and it is the reason IAP antagonists convert TNF from a growth signal into a death signal.
The rest of the family is the same architecture aimed elsewhere. RANKL-RANK is bone, with osteoprotegerin as a soluble decoy receptor — the RANKL:OPG ratio, not RANKL alone, sets resorption, and denosumab is simply a synthetic version of the decoy. BAFF and APRIL are B-cell survival. CD40, 4-1BB, OX40, CD27 and GITR are the costimulatory arm being developed as agonists, where the hard problem is that agonism needs receptor clustering rather than mere occupancy — which is why agonist antibodies are exquisitely sensitive to isotype and Fc engineering in a way blocking antibodies are not. The In Vivo tie-in: this is the deepest functional-grade layer in the library after the T-cell engager map — 14 of 37 nodes, and unusually it covers both sides of the TNFR1/TNFR2 question. TNF itself comes in low and ultra-low endotoxin grades, which matters more here than almost anywhere: LPS induces TNF directly, so a contaminated blocking antibody raises the very cytokine you are trying to remove. 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.