Mayo Clinic College of Medicine and Science Rochester, MN, United States
Disclosure information not submitted.
Shozo Ohtsuki1, Jose Morales2, Yuki Sato2, Chenyao Wang2, Matthew Koster3, Kenneth Warrington3, Gerald J. Berry4, Jorg Goronzy3 and Cornelia M. Weyand5, 1Mayo Clinic College of Medicine and Science, Stanford University School of Medicine, Rochester, MN, 2Mayo Clinic College of Medicine and Science, Rochester, MN, 3Mayo Clinic, Rochester, MN, 4Stanford University School of Medicine, Stanford, CA, 5Mayo Clinic School of Medicine and Stanford University, Rochester, MN
Background/Purpose: In Giant Cell Arteritis (GCA), granulomatous infiltrates occupy the vessel wall and elicit maladaptive vascular remodeling with intimal hyperplasia. The major cell types of the granulomatous lesions are CD4+ T cells and macrophages, some of which differentiate into multinucleated giant cells. Lesional CD4+ T cells undergo clonal expansion and vasculitic arteries contain mRNA and protein of multiple T cell effector cytokines, but it is unclear whether GCA patients possess specialized T cell subsets that promote macrophage multinucleation and granuloma formation in the vessel wall.
Methods: Patients with a positive temporal artery biopsy or unequivocal evidence for GCA aortitis were enrolled into the study. Patients with granulomatosis with polyangiitis served as disease controls and age-matched healthy controls were recruited through the Biobank. Tissue lesions were analyzed by immunofluorescence staining of temporal artery sections. T cell phenotyping relied on multiparametric flow cytometry and T cell reactivity was tested against anti-CD3-loaded antigen-presenting cells. The functional relevance of CD4+ T cell subsets was examined in immunodeficient mice engrafted with human arteries and immuno-reconstituted with immune cell populations from GCA patients (human artery-SCID chimeric mice).
Results: Memory CD4+ T cells isolated from GCA patients and age-matched controls fell into 10 clusters based on the combinatorial expression of 8 immunoreceptors (CD45RA, CCR7, PD1, LAG3, CD226, CD96, TIGIT, TIM3). GCA patients selectively expanded CD4+CD96+ memory T cells (10.9% control, 16.9% GCA), while CD4+TIGIThigh populations were reduced (24.5% control, 17.4% GCA). CD4+CD96low T cells, generated by siRNA transfection, induced vascular inflammation in artery-SCID chimeric mice (p=0.0044), indicating that CD96 delivers a negative signal and opposes T cell activation. The expansion of CD4+CD96+ T cells was dependent on interaction of CD96 with its ligand CD155 on the surface of antigen-presenting cells. Maldifferentiation of CD4+ T cells in GCA patients was associated with the excessive production of three effector cytokines: IL-9 (p=0.02), IL-21 (p=0.028), and IFN-γ (p=0.03). In vivo testing identified IL-9 as a strong driver of vascular inflammation, associated with marked damage of the vessel wall smooth muscle cell layer. Anti-IL-9 treatment efficiently suppressed vascular inflammation (p=0.0025). In single cell RNA sequencing from tissue derived T cells, CD96 expression mapped to the T follicular helper cell population.
Conclusion: In GCA patients, the differentiation of CD4+ memory T cells is abnormal, leading to the selective expansion of immature and multifunctional T cells, while the transition into effector T cells is decelerated. The underlying defect lies in antigen-presenting cells that withdraw opposing signals as T cells progress through their differentiation cycle. Resulting CD4+ memory T cells hyperproduce IL-9, IL-21, and IFN-γ. Blocking T cell effector functions in GCA will therefore require targeting an array of cytokines.
S. Ohtsuki: None; J. Morales: None; Y. Sato: None; C. Wang: None; M. Koster: None; K. Warrington: Bristol-Myers Squibb(BMS), 5, Chemocentryx, 1, 6, Eli Lilly, 5, kiniksa, 5; G. Berry: None; J. Goronzy: AbbVie/Abbott, 1, Bristol-Myers Squibb(BMS), 1, Gilead, 1; C. Weyand: AbbVie/Abbott, 1, Bristol-Myers Squibb(BMS), 1, Gilead, 1.