Thread by @connor_rosen

Very cool pair of new preprints @biorxivpreprints from @barton_lab on TLR regulation by Unc93b1 - biorxiv.org/content/early/… and biorxiv.org/content/early/… Here's my summary and a couple thoughts... #TLR #CellBiology #Immunology #PreReview #ASAPBio

First one, on TLR9 regulation by Unc93b1. Triple-alanine scanning mutagenesis library revealed mutant of Unc93b1 (SKN) that abolishes TLR9 signaling, although leaves endosomal trafficking and maturation intact. Single serine mutant (S282A) is responsible for full SKN phenotype

SKN or S282A both reduce amount of TLR9 binding to CpG. Both also increase amount of TLR9 bound by Unc93b1. Other mutations in same region of Unc93b1 (loop 5) either enhance or block Unc93b1 binding but all impair TLR9 activation, with much less effect on TLR7

Replacement of juxtamembrane region of TLR9 with corresponding TLR3 sequence rescued loss of signaling in Unc93b1 S282A - suggesting juxtamembrane region binds to Unc93b1 loop 5, and altering this binding (too much or too little) blocks TLR9 signaling

Working model is TLR9 is released by Unc93b1 in endosomes to allow activation. Supporting this, there is more Unc93b1-TLR9 association in ER than endosomes, suggesting release through maturation. S282A increased association in all compartments - increased affinity for TLR9

Myd88 IP (for active signaling TLR9) failed to pull down Unc93b1, suggesting it has been released completely prior to signaling. Cysteine cross-linking through juxtamembrane mutations forced TLR9-Unc93b1 association and blocked TLR9 activation

Importantly, all these discovered mechanisms of regulation are TLR9-specific. TLR7 shows minimal to no effects with any of these Unc93b1 mutations, and does not dissociate from Unc93b1 (leading to second paper, which I’ll get to…)

Questions: What drives dissociation of Unc93b1 from TLR9 in endoscopes? Does another co-receptor displace Unc93b1, or is it membrane environment / pH driven? What does the biochemistry of the association look like, vs TLR7 that does not dissociate?

Moving on to next paper, on negative regulation of TLR7 by Syntenin-1. Same triple-alanine library on Unc93b1 reveals a C-terminal region regulating TLR7 (but not TLR3 or 9). Faster and more intense activation of downstream signaling in Unc93b1 mutant

Mutant Unc93b1 (“PKP”) does not alter TLR7 export / maturation (cleavage to active form) or localization to endosomes. Mass-spec on PKP (pull-down from phagosomes, again) identified Syntenin-1 as interacting partner of WT Unc93b1, but not PKP

Syntenin-1 is recruited specifically following TLR7, but not TLR9, stimulation. Recruitment depends on both serine phosphorylation of Unc93b1 as well as the PKP motif. Some (very rare) human SNPs in this C-terminal region of Unc93b1 result in enhanced TLR7 responses

Last, they generate PKP-knockin mice. PKP/PKP mice develop severe systemic inflammation in a TLR7-dependent manner - ANAs, hyperactivated T cells, failure to thrive, DCs and macrophages are hyper responsive to TLR7 stim. Hets have some milder phenotypes

Obvious first questions (some posed by authors) - what’s the kinase phosphorylating Unc93b1? Is there a corresponding phosphatase? How does Syntenin-1 regulate (block) Myddosome assembly? What drives Syntenin-1 dissociation?

Big question: there is some basal phosphorylation of Unc93b1 and association with Syntenin-1 even without TLR7. Does TLR7 stim result in increased phosphorylation? Or is there another mechanism enhancing Syntenin-1 recruitment (loss of another phospho-binding protein?)

Overall, two fantastic stories from @MajerOlivia and the @barton_lab! So much interesting cell biology and biochemistry of transmembrane receptor association, and I’m sure there’s even more complexity to unravel with TLR control by chaperones (Unc93b1 and more!)