期刊名称：Proceedings of the National Academy of Sciences
印刷版ISSN：0027-8424
电子版ISSN：1091-6490
出版年度：2020
卷号：117
期号：19
页码：10465-10475
DOI：10.1073/pnas.2003136117
出版社：The National Academy of Sciences of the United States of America
摘要：The antigen-presenting molecule MR1 presents riboflavin-based metabolites to Mucosal-Associated Invariant T (MAIT) cells. While MR1 egress to the cell surface is ligand-dependent, the ability of small-molecule ligands to impact on MR1 cellular trafficking remains unknown. Arising from an in silico screen of the MR1 ligand-binding pocket, we identify one ligand, 3-([2,6-dioxo-1,2,3,6-tetrahydropyrimidin-4-yl]formamido)propanoic acid, DB28, as well as an analog, methyl 3-([2,6-dioxo-1,2,3,6-tetrahydropyrimidin-4-yl]formamido)propanoate, NV18.1, that down-regulate MR1 from the cell surface and retain MR1 molecules in the endoplasmic reticulum (ER) in an immature form. DB28 and NV18.1 compete with the known MR1 ligands, 5-OP-RU and acetyl-6-FP, for MR1 binding and inhibit MR1-dependent MAIT cell activation. Crystal structures of the MAIT T cell receptor (TCR) complexed with MR1-DB28 and MR1-NV18.1, show that these two ligands reside within the A′-pocket of MR1. Neither ligand forms a Schiff base with MR1 molecules; both are nevertheless sequestered by a network of hydrophobic and polar contacts. Accordingly, we define a class of compounds that inhibits MR1 cellular trafficking. Mucosal-Associated Invariant T (MAIT) cells are a subset of evolutionarily conserved nonmajor histocompatibility complex (MHC)-restricted T cells, which are very abundant in human mucosal tissues, in peripheral blood, and in the liver ( 1 , 2 ). Similar to type I NKT cells, human MAIT cells express a semi-invariant T cell receptor (TCR) composed of the Vα7.2 chain rearranged mainly to Jα33 and paired with a limited number of Vβ chains, mostly TRBV6, TRBV13, and TRBV20 ( 3 , 4 ). MAIT cells recognize small microbial metabolites presented by the monomorphic MHC class I-related molecule, MR1 ( 1 , 2 ). The physiological roles of MAIT cells remain unclear, but they are known to be involved in protective immunity ( 2 , 5 ⇓ – 7 ), possibly through modulation of innate and adaptive immune responses ( 8 , 9 ). Moreover, the role of MAIT cells in cancer ( 10 ) and inflammatory diseases, such as obesity ( 11 ), diabetes ( 12 ), multiple sclerosis ( 13 ), and inflammatory bowel disease ( 14 ), has been highlighted, and recent reports have suggested they may also play a role in tissue repair ( 15 , 16 ). Activation of MAIT cells induces the production of various proinflammatory cytokines, predominantly IFN-γ, TNF-α, IL-2, and IL-17 ( 17 , 18 ), and their potent cytolytic activity allows them to kill infected cells ( 19 ). Unlike MHC molecules, MR1 does not constitutively present antigens, but is found in the endoplasmic reticulum (ER) of all cells in a ligand-receptive conformation ( 20 ). The potency of known MAIT cell agonists appears to correlate with their ability to form a Schiff base with MR1 Lys43 located within the A′-pocket, thus allowing MR1 to egress to the cell surface, where the presence of a ribityl moiety in the covalently bound agonist allows for an interaction with the MAIT TCR ( 21 ⇓ – 23 ). To date, the strongest MAIT cell agonists are 5-(2-oxopropylideneamino)-6- D -ribitylaminouracil (5-OP-RU) and 5-(2-oxoethylideneamino)-6- D -ribitylaminouracil (5-OE-RU), both pyrimidine-based intermediates along the riboflavin biosynthetic pathway ( 24 ). Several bacterial and fungal species synthesize riboflavin ( 23 ), and MAIT cells have been shown to possess MR1-dependent antimicrobial activity against infected antigen-presenting cells ( 5 , 6 ). Conversely, vitamin B9 metabolites [including the folic acid derivative 6-formylpterin, 6-FP and its acetylated derivative Ac-6-FP ( 23 , 25 )] are strong MR1 binders and induce MR1 expression at the cell surface; however, the resulting complexes do not activate MAIT cells because they lack the ribityl moiety ( 22 ). Drug and drug-like molecules (including diclofenac and salicylates) also bind MR1 and either weakly activate or inhibit MAIT cells ( 26 ). However, it remains unknown whether there are other ligands that impact MR1-dependent antigen presentation. Through an in silico screen, we have identified additional MR1-binding ligands. We describe a ligand that down-regulates MR1 cell-surface expression and provide a molecular basis for its interactions with MR1. Results Identification of Nonmicrobial MAIT Cell Agonists. To identify MR1 binding ligands, we performed in silico screening using the crystal structures of the MAIT TCR in complex with MR1–antigen complexes [PDB codes 4L4V and 4LCC ( 22 , 27 )]. A total of 44,022 compounds were selected for docking runs, based on searches for fragment size substructures s1-s20 ( SI Appendix , Fig. S1 and Supplementary Methods ). Compound selection and constraints imposed during docking are detailed in the supplementary methods. Using this strategy, 80 commercial compounds were selected as potential MR1 ligands, of which 52 compounds were pulsed on MR1 overexpressing cells, alongside the canonical MAIT cell ligand 5-OP-RU, synthesized and validated in house ( SI Appendix , Fig. S2 A and B ). MAIT cell stimulatory activity was observed when THP1-MR1 cells ( Fig. 1 A and SI Appendix , Fig. S4 A ) were pulsed with compounds DB5, DB7, DB8, DB12, DB15, DB19, and DB23, whose chemical structures are shown in SI Appendix , Fig. S3 . Overall, these compounds were three to nine times less potent than 5-OP-RU ( SI Appendix , Fig. S4 A ). Unlike Ac-6-FP and 5-OP-RU ( 26 ), none of the tested compounds induced detectable up-regulation of cell-surface MR1, neither after 5 nor 22 h ( Fig. 1 B ). Presentation was MR1-dependent, as determined using the blocking anti-MR1 26.5 monoclonal antibody ( 28 ) ( Fig. 1 C ) and pulsing the compounds on MR1-KO THP1 cells ( SI Appendix , Fig. S4 B ). Consistent with their weaker potency, presentation by THP1 cells required a higher level of MR1 expression (THP1-MR1 WT), whereas THP1 cells nonoverexpressing MR1 were unable to present any of the DB compounds ( SI Appendix , Fig. S4 B ). In addition, presentation was reduced or abrogated when THP1 cells expressing GPI-linked molecules were used ( SI Appendix , Fig. S4 B ), suggesting internalization and possibly endo-lysosomal loading is required. However, we were unable to detect any MAIT cell activation by fixed THP1-MR1, even after pulsing with the potent agonist 5-OP-RU; therefore, we did not investigate intracellular trafficking further. Compounds DB5, DB12, and DB19 were also presented by monocyte-derived dendritic cells ( Fig. 1 D ). In this experimental setting, MAIT cell activation was also MR1-dependent ( SI Appendix , Fig. S4 C ). We next tested the compounds on unfractionated cells in whole blood and identified MAIT cells by Vα7.2 and CD161 co-staining. MAIT cell activation (measured by CD137 up-regulation) with compounds DB7, DB8, DB12, and DB19 was observed in some, but not all, of the seven donors tested ( Fig. 1 E ); this may reflect pairing of different ΤCR β-chains with the canonical MAIT TCR α-chain ( 3 ). Indeed, in one donor, we observed a lower response by MAIT cells expressing the TRBV13S2 chain as compared with the TRBV20.1 chain or neither of those two chains ( Fig. 1 F ). We also confirmed TCR-mediated recognition of some of the DB compounds using Jurkat cells transduced with a MAIT TCR composed of the canonical TCR α-chain paired with TRBV20.1 or 6.4 ( SI Appendix , Fig. S4 D and E ) ( 29 ). In conclusion, we have defined a series of compounds that bind MR1 and can activate, through MR1-TCR interaction, MAIT cells expressing a variety of TCR β chains.
