Myeloid Checkpoints: CD47-SIRPa, LILRB and Siglec-10
Phagocytosis is a subtraction, not a switch. A macrophage engulfs when its activating signals exceed its inhibitory ones, which means blocking a don't-eat-me receptor on a target cell that displays no eat-me signal removes a brake that was never opposing an accelerator. That is the most economical explanation for why anti-CD47 monotherapy underperformed and why combinations with opsonising antibodies did not. The second thing this map is built to show is that CD47 is not the only axis. LILRB1 reads MHC class I, LILRB2 reads HLA-G, and Siglec-10 reads CD24 \u2014 three further don't-eat-me systems running in parallel, any one of which can protect a cell that has lost the others. And engulfment is not the end point: MerTK-driven efferocytosis is actively tolerogenic, so raising phagocytosis by the wrong route raises tolerance rather than immunity. Click any protein for the matching Assay Genie In Vivo antibody or ELISA kit.
The arithmetic is the mechanism. A macrophage does not receive an instruction to engulf; it integrates activating input from Fc receptors and complement receptors against inhibitory input from ITIM-bearing receptors, and acts when the balance tips. Every consequence on this map follows from that. CD47 blockade alone does little to a cell that is not already opsonised or already displaying calreticulin, because there is nothing for the released brake to reveal. Conversely, an opsonising antibody works better when the brake is released \u2014 which is why the productive anti-CD47 clinical work was combination work with rituximab and with daratumumab, against CD20 and CD38, rather than monotherapy. The same arithmetic explains the on-target anaemia: CD47 is on every nucleated cell and on erythrocytes, so a systemic blockade removes the brake everywhere, and the antigen sink it creates is large enough to affect dosing.
There are at least four axes, not one. LILRB1 binds the conserved \u03b13 domain of MHC class I; LILRB2 binds HLA-G among others; Siglec-10 binds CD24; and CD200 signals through CD200R. Each is sufficient to protect a cell on its own, which is the most likely reason single-axis blockade disappoints, and it produces an awkward corollary at the MHC class I node: losing class I is the classic immune-escape mutation for CD8 T cells, but it simultaneously removes the LILRB1 brake on the macrophage. The same mutation makes a cell harder to kill by one mechanism and easier to eat by another. CD24-Siglec-10 was specifically proposed as the dominant axis in ovarian and breast cancer, where CD47 blockade has been least convincing.
Everything converges on two phosphatases, and one of them behaves backwards. All four inhibitory receptors recruit SHP-1 and SHP-2 to phosphorylated ITIMs, and it is SHP-1 that dephosphorylates non-muscle myosin IIA and the intermediates that would otherwise assemble the phagocytic cup. The motheaten mouse \u2014 SHP-1-null, and profoundly inflammatory \u2014 indicates how much runs through it. SHP-2 is the complication: in most receptor contexts it is activating, and here it is not, which is worth holding in mind when SHP-2 inhibitors developed for RAS-driven tumours are applied to a tumour whose myeloid compartment is part of the question. Downstream of all of it, the route matters: MerTK-mediated efferocytosis of apoptotic cells suppresses type I interferon and is tolerogenic, so a treatment that raises total engulfment without directing it can produce more tolerance, not less.
The In Vivo tie-in, and it is an honest one. 18 of the 36 nodes carry a functional-grade antibody \u2014 but the distribution matters more than the count, and it is the opposite of what this field's literature would lead you to expect. Of the four inhibitory receptors, exactly one is blockable in vivo: SIRP\u03b1, clone against mouse CD172a. That happens to be the better experiment anyway, since blocking the receptor sidesteps the antigen sink CD47's ubiquity creates \u2014 but it is availability rather than design, and it should be said as such. TIM-3 is the second receptor-side option. The real depth is on the ligand side \u2014 MHC class I in both mouse and human, CD24, CD200, PD-L1 \u2014 and on the phagocyte itself: Fc\u03b3RII/III, Fc\u03b3RIV (clone 9E9, which carries most antibody-dependent phagocytosis in mouse models), human CD16, CD11b, CD11c, CD18, CSF1R, CD68, Ly-6C, CD209b and CD8. Said plainly, what is not available: CD47 and LILRB2 are biosimilar only; LILRB1, calreticulin, MerTK and CD206 are research grade only; and Siglec-10 has no antibody at any grade \u2014 an ELISA kit is the whole of it, which means the CD24 side is the only interruptible end of that axis. One reagent is worth planning around regardless: the recombinant SIRP\u03b1-Fc decoy occupies CD47 without engaging a phagocyte receptor, and is the cleanest way to separate CD47 blockade from the Fc-mediated effects of an anti-CD47 antibody \u2014 a confound present in essentially every experiment in this area. 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.