OP0104 EXPANDED CD8+ T CELL CLONES FROM HLA-B*27-POSITIVE PATIENTS WITH SPONDYLOARTHRITIS SHOW SIGNS OF ANTIGEN-EXPERIENCE

Background

The pathogenesis of Spondyloarthritis (SpA) remains unknown but its strong association with some alleles of HLA-B*27 is peculiar. The arthritogenic antigen hypothesis assumes the existence of specific peptides presented by risk-conferring HLA-B*27 alleles to antigen-specific CD8+ T cells, which then initiate or sustain autoimmune reactions. Several studies analyzing T cell receptor (TCR) repertoire found preferred Variable TCR chains and motifs in the hypervariable complementary determining region (CDR) 3, but analyzed only TCR β-chains in bulk analyses1,2.

Objectives

To analyze full sequence information of TCR including matching α- and β-chains from single CD8+ T cells and characterize the transcriptomes of expanded and non-expanded clonotypes in synovial fluid (SF) of SpA patients.

Methods

We included 17 patients with active gonarthritis: 10 patients with HLA-B27 positive (B27pos) SpA, 4 with HLA-B27 negative (B27neg) SpA and 3 rheumatoid arthritis (RA) patients. Antigen-experienced CD8+ T cells were sorted out of SF by flow cytometry. Single cell sequencing was performed for all patients to analyze matching TCR α- and β-chains. For 7 patients (3 B27pos SpA, 2 B27neg SpA, 2 RA), additionally whole transcriptome analyses were performed.

Results

We found strong biases when analyzing α and β chains of TCR Variable regions and CDR1 and CDR2 sequences (Figure 1 a,b): AV21, AV12-2, and AV17 were highly enriched in B27pos SpA as compared to B27neg subjects. Amongst the highest expressed clones, we could confirm enrichment for previously described TRBV genes as BV19, BV5-1 and BV6-2. We examined TCR α/β combinations and focused on those detected in at least three different B27pos SpA but not in any of the B27neg patients (Figure 1 c-f). The combinations TRBV19/TRAV21 and TRBV6-2/TRAV21 were most likely specific for B27pos SpA and might reflect interaction of these TCR chains with HLA-B*27. Sequences of CDR3 loops, which predominantly interact with HLA-bound antigenic peptides, revealed striking common structural motifs in α- and β-chains. Focusing on the most prominent TRAV21 chains pairing with TRBV19, 5-1 and 6-2 chains, revealed identical sequences in different patients and striking common structural motifs in α- and β-CDR3 sequences in other patients. Such marked similarities in the antigen-recognition loops of the β-chains associated with TRAV21 suggest common or highly similar antigens. Gene expression levels provided evidence that expanded cell populations had tissue resident memory (TRM) phenotypes (elevated expression of activation, migration and tissue retention markers, downregulated genes characteristic for T cell egress), while this phenotype was not very pronounced in non-expanded cells. Furthermore, markers for T cell exhaustion and apoptosis were elevated in expanded cells of B27pos SpA patients.

Figure 1.

Distinct TCRαβ V chain usage in expanded clones from HLA-B27 positive SpA patients. A,B Mean number of all productive TRAV (A) and TRBV (B) genes used in expanded, antigen-experienced CD8 T cell clones (>1% of all cells) from SF of 10 B27pos SpA, 4 B27neg SpA and 3 B27neg RA patients. C-F TRAV chains paired with TRBV19 (C), TRBV5-1 (D), TRBV6-2 (E), or TRBV chains paired with TRAV21 (F) with corresponding TRAJ spanning partners in expanded cells (frequency ≥2) from all 10 B27pos SpA. Number of chains are 1250 (C), 886 (D), 1220 (E), and 4006 (F).

Conclusion

Analysis of single antigen experienced CD8+ T cells from SF of B27pos SpA patients revealed significant clonal expansions and common motifs in the CDR loops. Two of the four CDR1 and CDR2 loops were highly homologous suggesting that these loops interact with α-helices of HLA-B*27. Common motifs in CDR3 loops of expanded clonotypes suggest recognition of a limited set of antigenic peptides presented by HLA-B*27. Many of the expanded clonotypes showed a TRM phenotype, were exhausted and on the way to become apoptotic, which suggests that these clones had sustained contact to specific antigens.

References

[1]Komech, et al. Rheumatology 2018

[2]Hanson, et al. A&R 2020

Acknowledgements

We thank all patients included in this study for their participation. This work was funded by the German Research Foundation (DFG) through grants DO 420/4 to KDo, PO 2124/2-1 to DP, and SyNergy (EXC 2145 SyNergy – ID 390857198) to KDo. Judith Rademacher and Katharina Deschler contributed equally. JR is participant in the BIH-Charité Clinician Scientist Program funded by the Charité –Universitätsmedizin Berlin and the Berlin Institute of Health. The authors would like to thank Martina Seipel for excellent technical assistance, Sabrina Sron for patient recruitment and study coordination, and Hildrun Haibel, Mikhail Protopopov, Fabian Proft, Valeria Rios Rodriguez and Laura Spiller for recruiting patients for this study.

Disclosure of Interests

None declared

2-Amino[1,2,4]triazolo[1,5-c]quinazolines and Derived Novel Heterocycles: Syntheses and Structure-Activity Relationships of Potent Adenosine Receptor Antagonists

2-Amino[1,2,4]triazolo[1,5-c]quinazolines were identified as potent adenosine receptor (AR) antagonists. Synthetic strategies were devised to gain access to a broad range of derivatives including novel polyheterocyclic compounds. Potent and selective A₃ AR antagonists were discovered, including 3,5-diphenyl[1,2,4]triazolo[4,3-c]quinazoline (17, K_i human A₃ AR 1.16 nm) and 5'-phenyl-1,2-dihydro-3'H-spiro[indole-3,2'-[1,2,4]triazolo[1,5-c]quinazolin]-2-one (20, K_i human A₃ AR 6.94 nm). In addition, multitarget antagonists were obtained, such as the dual A₁ /A₃ antagonist 2,5-diphenyl[1,2,4]triazolo[1,5-c]quinazoline (13 b, K_i human A₁ AR 51.6 nm, human A₃ AR 11.1 nm), and the balanced pan-AR antagonists 5-(2-thienyl)[1,2,4]triazolo[1,5-c]quinazolin-2-amine (11 c, K_i human A₁ AR 131 nm, A_2A AR 32.7 nm, A_2B AR 150 nm, A₃ AR 47.5 nm) and 9-bromo-5-phenyl[1,2,4]triazolo[1,5-c]quinazolin-2-amine (11 q, K_i human A₁ AR 67.7 nm, A_2A AR 13.6 nm, A_2B AR 75.0 nm, A₃ AR 703 nm). In many cases, significantly different affinities for human and rat receptors were observed, which emphasizes the need for caution in extrapolating conclusions between different species.

Interactions of Magnolia and Ziziphus extracts with selected central nervous system receptors

ETHNOPHARMACOLOGICAL RELEVANCE

Magnolia officinalis Rehder and Wilson [Magnoliaceae] bark and Ziziphus spinosa (Buhge) Hu ex. Chen. [Fam. Rhamnaceae] seed have a history of use in traditional Asian medicine for mild anxiety, nervousness and sleep-related problems.

AIM OF THE STUDY

To identify pharmacological targets, extracts of Magnolia officinalis (ME), Ziziphus spinosa (ZE), and a proprietary fixed combination (MZE) were tested for affinity with central nervous system receptors associated with relaxation and sleep.

METHODS

In vitro radioligand binding and cellular functional assays were conducted on: adenosine A(1), dopamine (transporter, D(1), D(2S), D(3), D(4.4) and D(5)), serotonin (transporter, 5-HT(1A), 5-HT(1B), 5-HT(4e), 5-HT(6) and 5-HT(7)) and the GABA benzodiazepine receptor.

RESULTS

Interactions were demonstrated with the adenosine A(1) receptor, dopamine transporter and dopamine D(5) receptor (antagonist activity), serotonin receptors (5-HT(1B) and 5-HT(6) antagonist activity) and the GABA benzodiazepine receptor at a concentration of 100 microg/ml or lower. ME had an affinity with adenosine A(1) (K(i) of 9.2+/-1.1 microg/ml) and potentiated the GABA activated chloride current at the benzodiazepine subunits of the GABA receptor (maximum effect at 50 microg/ml). ME had a modest antagonist action with 5-HT(6) and ZE with the 5-HT(1B) receptor.

CONCLUSION

The interactions in the receptor binding models are consistent with the traditional anxiolytic and sleep-inducing activities of Magnolia officinalis bark and Ziziphus spinosa seed.

Adenosine regulates CD8 T-cell priming by inhibition of membrane-proximal T-cell receptor signalling

Adenosine is a well-described anti-inflammatory modulator of immune responses within peripheral tissues. Extracellular adenosine accumulates in inflamed and damaged tissues and inhibits the effector functions of various immune cell populations, including CD8 T cells. However, it remains unclear whether extracellular adenosine also regulates the initial activation of naïve CD8 T cells by professional and semi-professional antigen-presenting cells, which determines their differentiation into effector or tolerant CD8 T cells, respectively. We show that adenosine inhibited the initial activation of murine naïve CD8 T cells after alphaCD3/CD28-mediated stimulation. Adenosine caused inhibition of activation, cytokine production, metabolic activity, proliferation and ultimately effector differentiation of naïve CD8 T cells. Remarkably, adenosine interfered efficiently with CD8 T-cell priming by professional antigen-presenting cells (dendritic cells) and semi-professional antigen-presenting cells (liver sinusoidal endothelial cells). Further analysis of the underlying mechanisms demonstrated that adenosine prevented rapid tyrosine phosphorylation of the key kinase ZAP-70 as well as Akt and ERK1/2 in naïve alphaCD3/CD28-stimulated CD8 cells. Consequently, alphaCD3/CD28-induced calcium-influx into CD8 cells was reduced by exposure to adenosine. Our results support the notion that extracellular adenosine controls membrane-proximal T-cell receptor signalling and thereby also differentiation of naïve CD8 T cells. These data raise the possibility that extracellular adenosine has a physiological role in the regulation of CD8 T-cell priming and differentiation in peripheral organs.

Synthesis and preliminary evaluation of new 1- and 3-[1-(2-hydroxy-3-phenoxypropyl)]xanthines from 2-amino-2-oxazolines as potential A1 and A2A adenosine receptor antagonists

The development of potent and selective adenosine receptor ligands as potential drugs is an active area of research. Xanthines are one of the most important classes of adenosine receptor antagonists and have been widely developed in terms of affinity and selectivity for adenosine receptors. We recently developed new original pathways for the synthesis of xanthine analogues starting from 5-substituted-2-amino-2-oxazoline 5 as a synthon. These procedures allowed us to selectively introduce a large, functionalized and beta-adrenergic 2-hydroxy-3-phenoxypropyl pharmacophore at the 1- and 3-position of the xanthine moiety which allowed further structural modifications. In this study, we present a new synthetic access to racemic xanthine derivatives 1-4 from 5, and their evaluation as adenosine A1, A2A and A3 receptor ligands in radioligand binding studies. The 2-hydroxy-3-phenoxypropyl moiety was well tolerated in the 3-position of the xanthine core, while its introduction in the 1-position of the xanthine moiety led to a large decrease in adenosine receptor affinity. 1,7-Dimethyl-3-[1-(2-chloro-3-phenoxypropyl)]-8-(3,4,5-trimethoxystyryl)xanthine (2n) was the most potent and selective A2A antagonist of the present series (Ki=44 nM, >>200-fold selective vs A1). 1-Propyl-3-[1-(2-hydroxy-3-phenoxypropyl)]-8-noradamantylxanthine (3f) was identified as a potent (KiA1=21 nM) and highly selective (>>350-fold vs A2A and A3 receptor) adenosine A1 receptor antagonist.