Written informed consent was obtained from patients prior to blood donation for research purposes. All experiments conducted on human material were approved by the Ethics Committee of the University of Freiburg (55/14; 11.02.2014). All performed methods are compliant with the regulations of the Ethics Committee of the University of Freiburg.

Preparation and cultivation of human peripheral lymphocytes were performed as indicated (Zimmermann-Klemd et al., 2019). Briefly, peripheral blood mononuclear cells (PBMCs) were isolated from the blood of healthy adult donors, which was obtained from a blood transfusion center (University Medical Center, Freiburg, Germany). Venous blood was centrifuged on a LymphoPrep™ gradient (1.077 g/cm³, 20 min, 500 × g, 20°C; Progen, Heidelberg, Germany). After centrifugation cells were washed twice with phosphate buffered saline (PBS) and subsequently cultured in Roswell Park Memorial Institute medium (RPMI) 1640 medium supplemented with 10% heat-inactivated fetal calf serum, 2 mM l-glutamine, 100 U/ml penicillin, and 100 U/ml streptomycin (all from Life Technologies, Paisley, UK). Cells were cultured at 37°C in a humidified incubator with a 5% CO₂/95% air atmosphere.

Lymphocytes were activated with anti-CD3 (clone OKT3) and anti-CD28 (clone 28.2) mAbs (each 100 ng/ml; both from eBioscience, Frankfurt, Germany) in the presence of medium; cyclosporine A (CsA; 4.16 µM; Sandimmun 50 mg/ml, Novartis Pharma, Basel, Switzerland); camptothecin (CPT; 300 µM: Tocris, Bristol, UK); 0.5% Triton-X 100; plant extract; or isolated compounds from A. argyi, as described previously (Zimmermann-Klemd et al., 2019). After cultivation, the cells were used in biological tests.

Determination of apoptosis and necrosis of T cells was performed as described previously (Zimmermann-Klemd et al., 2019). Cells were treated for 48 h. Cultured cells were washed with PBS and stained with Annexin V-FITC using the apoptosis detection kit (eBioscience, Frankfurt, Germany) according to instructions of the manufacturer. Propidium iodide (PI; eBioscience, Frankfurt, Germany) was added, and cells were stained for 15 min at room temperature in the dark. The proportion of apoptotic/necrotic lymphocytes was determined by flow cytometric analysis (FACSCalibur instrument; BD Biosciences, Franklin Lakes, NJ).

The proliferation of T lymphocytes was determined using carboxyfluorescein diacetate succinimidyl ester (CFSE) staining, as described previously (Parish et al., 2009; Gründemann et al., 2012). Lymphocytes were isolated, washed twice in cold PBS, and resuspended in PBS at a concentration of 5 × 10⁶ cells/ml. Cells were stained for 10 min at 37°C with CFSE (5 µM; Sigma-Aldrich, St. Louis, MO). The staining reaction was stopped by washing twice with a complete medium. Stained cells were treated for 72 h. The progress of cell division was determined by flow cytometric analysis (FACSCalibur instrument; BD Biosciences, Franklin Lakes, NJ).

The activation state of T lymphocytes was determined via cell-surface analysis of CD25 and CD69, as previously reported (Gründemann et al., 2014). Briefly, cells were treated for 24 h. Then, cells were washed with PBS and stained with PE-labeled, anti-CD25 mAbs; FITC-labeled, anti-CD69; and, for the differentiation of CD4⁺ and CD4^– T cells, with APC-labeled, anti-CD4 mAbs (all from eBioscience, Frankfurt, Germany) for 20 min at 4°C. Afterward, cells were washed twice with PBS, resuspended, and transferred into FACS vials. The expression of CD25 and CD69 was measured for CD4⁺ and CD4^– T cells, respectively, by flow cytometric analysis (FACSCalibur instrument; BD Biosciences, Franklin Lakes, NJ).

After 20 h of treatment, cells were restimulated with PMA (50 ng/ml; Sigma-Aldrich, Taufkirchen, Deutschland) and ionomycin (500 ng/ml; Sigma-Aldrich, Taufkirchen, Deutschland) for 4 h. Supernatants were stored at –20°C. The amount of cytokines was quantified using ELISA technique according to manufacturer's instructions (Affymetrix, Frankfurt, Germany).

Cells were treated for 20 h, as described in section 2.3, and restimulated with PMA (50 ng/ml; Sigma-Aldrich, Taufkirchen, Deutschland) and ionomycin (500 ng/ml; Sigma-Aldrich, Taufkirchen, Germany) for an additional 4 h at 37°C. During this time, they were additionally treated with GolgiPlug (0.5 µl; BD Biosciences, Heidelberg, Germany). After incubation, cells of each sample were divided into three approaches: one for the determination of IL-2, one for the determination of TNFα, one for the determination of IFN-γ. The cells were fixed with 100 µl of 4% PFA (Morphisto, Frankfurt, Germany) for 10 min at room temperature and washed with PBS. Afterward, permeabilization was performed using 100 µl of 1× BD Perm/Wash Puffer (BD Biosciences, Heidelberg, Germany) per sample for 15 min at 4°C. Finally, cells were stained with 1 µl anti-IL-2 or anti-IFN-γ (both Affymetrix, Frankfurt, Germany) for 30 min at 4°C. After two washing steps samples were analyzed by flow cytometric analysis (FACS LSR Fortessa Instrument; BD Biosciences, Franklin Lakes, NJ).

A CD107a surface staining was performed, as described previously (Gründemann et al., 2013), to determine the T cell degranulation capacity. Cells were treated for 20 h and then restimulated for 4 h with PMA (50 ng/ml; Sigma-Aldrich, Taufkirchen, Deutschland) and ionomycin (500 ng/ml; Sigma-Aldrich, Taufkirchen, Deutschland). To each well containing 200 µl of cell suspension, 2.5 µl (~0.25 µg) of PE-conjugated, anti-CD107a mAbs (eBioscience, Frankfurt, Germany) was added. After incubation at 37°C for 1 h, 2 µl of 1/10 diluted GolgiStop (Becton Dickinson, Franklin Lakes, NJ) was added per well, and the cells were incubated for another 3 h. Samples were analyzed by flow cytometric analysis (FACSCalibur instrument; BD Biosciences, Franklin Lakes, NJ).

A 96 well F-bottom cell culture plate was coated with anti-human CD3 mAb (clone OKT3, 1 µg/ml, 50 µl/well) or PBS (unstim. control) at 4°C over-night. Reporter cells (Jutz et al., 2016) were harvested, washed twice with PBS, and seeded in a 5% FCS RPMI cell culture medium (0.15 × 10⁶ cells in 200 µl/well). Cells were treated with inhibitors (1 nM SP100030 for AP-1, 5 µg/ml cyclosporine A for NFAT, and 20 µM parthenolide for NF-κB), plant extract or isolated compounds from A. argyi or remained untreated (unstim. control, stim. control). Cells were incubated at 37°C for 8 h (AP-1) or 24 h (NFAT and NF-κB). Cells were washed twice with PBS and the expression of eGFP was determined by flow cytometric analysis (FACSCalibur instrument; BD Biosciences, Franklin Lakes, NJ).

Jurkat cells (0.5 × 10⁶) were stained in 200 µl RPMI medium, containing 1% FCS, 2.6 µM Fluo3 AM (Life Technologies, Carlsbad, California), 5.5 µM FuraRed AM (Invitrogen, Carlsbad, California), and 0.1% (w/v) Pluronic F-127 (Invitrogen, Carlsbad, California) in the presence of test substances (30 µg/ml A. argyi extract or 10 µg/ml compound/compound mix; merely for the determination of the calcium ER store depletion) for 45 min at 37°C. For differentiation of calcium ER store depletion and SOCE, the staining solution was further supplemented with 0.6 mM ethylenediaminetetraacetic acid (EDTA). Cell suspensions were gently mixed every 10 min. After staining cells were resuspended in 100 µl of 1% FCS RPMI medium containing the test substance (30 µg/ml A. argyi extract or 10 µg/ml compound/compound mix; merely for the determination of the calcium ER store depletion). For the measurement of calcium influx, 50 µl of cell suspension were prewarmed (37°C, 5 min) in 700 µl of 1% FCS RPMI medium supplemented with 0.6 mM EDTA (for differentiation of ER store depletion and SOCE) and the test substance (30 µg/ml A. argyi extract or 10 µg/ml compound/compound mix; merely for the determination of the calcium ER store depletion). Baseline calcium levels were determined by flow cytometric measurement (FACS CyAn ADP; Beckman Coulter, Brea, California) for 1 min. Afterward, calcium influx was induced via stimulation with anti-CD3 mAbs (clone OKT3, 1 µg/ml). After 2 min, the test substance (30 µg/ml A. argyi extract or 10 µg/ml compound/compound mix) was added, followed, 30 s later, by the addition of calcium dichloride (1 mM). Calcium influx was measured for 5 min.

For statistical analysis, data were processed with Microsoft Excel and SPSS software (Version 22.0, IBM, Armonk, USA). Statistical significance was determined with the SPSS software by a one-way ANOVA followed by Dunnett's post hoc pairwise comparisons. Values are presented as mean ± standard deviation (SD) for the indicated number of independent experiments. The asterisks represent significant differences from controls (*p < .05).

The compounds used in this work were isolated from A. argyi ethyl acetate extract in a previous study (Reinhardt et al., 2019).

Previously, we evaluated the effect of 435 plant extracts from a focused extract library of TCM plants on the proliferative capacity of primary expanded human T lymphocytes in vitro. The ethyl acetate extract of A. argyi inhibited the proliferation of human T lymphocytes in a concentration-dependent manner, with a half maximal inhibitory concentration (IC₅₀) of 16.2 µg/ml (Reinhardt et al., 2019) (Figure 1A). The focused library also comprised extracts from other plants of the Asteraceae family (Supplement 2), but none of these exhibited notable activity in the assay.

To verify that the observed immunosuppressive activity was not due to cytotoxicity of the extract, flow cytometry analysis was performed to analyze the apoptosis and necrosis events of the cells. Using Annexin V/PI double staining, we found that the A. argyi extract had no toxic effects in the concentration range that was used for the mechanistic studies (Figures 1B, C).

Next, the effect of the A. argyi extract on the effector function of human T cells was examined. Stimulation of T lymphocytes via the T cell receptor (TCR) leads to the expression of the activation markers CD25 and CD69. We found that after TCR stimulation the treatment of T cells with the A. argyi extract significantly and concentration-dependently lowered CD25 and CD69 expressions (Figures 2A, B). This was the case in both the CD4⁺ and CD4^– T cells. In addition, we observed a significantly reduced CD25 expression of CD4^– T cells after treatments with low concentrations (3 and 10 µg/ml) of the A. argyi extract (Figure 2A).

Upon activation, T cells secrete IL-2, which is important for proliferation, and interferon-γ (IFN-γ), which defines T cell function. The A. argyi extract significantly suppressed IL-2 as well as the IFN-γ secretion capacity of activated T lymphocytes (Figures 2C, D).

To understand how the activation and proliferation of human T cells is inhibited by the A. argyi extract, we analyzed the effect of the A. argyi extract on the transcription factors AP-1, NFAT, and NF-κB. All three transcription factors regulate the transcription of the il-2 gene, which is a key regulator in these processes. Jurkat T cell reporter lines (Ratzinger et al., 2014; Jutz et al., 2016), in which eGFP is fused to the response elements of these transcription factors, were used for this purpose. After stimulation with anti-CD3 mAbs, activity of the transcription factors was quantified via flow cytometry. The treatment with the A. argyi extract did not change the anti-CD3-induced AP-1 activity in comparison to untreated activated AP-1 reporter cells (Figure 3A). In contrast, NFAT and NF-κB activity was reduced in a concentration-dependent manner by the A. argyi extract (Figures 3B, C). In summary, the reporter cell experiments pointed to a specific suppression of NFAT and NF-κB activities, but not to suppression of AP-1.

The previous experiments demonstrated that the A. argyi extract lowered activity of NFAT, which was induced by TCR triggering. Influx of calcium ions into the cytosol is upstream regulated of NFAT activation (Robert et al., 2011). Hence, we performed calcium flux experiments to determine whether the calcium influx to the cytosol was impaired. Inhibition of calcium influx would prevent the translocation of NFAT to the nucleus and, consequently suppress NFAT activity.

Jurkat cells were stained with the calcium indicators Fluo3 and FuraRed and treated with the A. argyi extract. Subsequently, calcium influx was induced by TCR stimulation using an anti-CD3 mAbs. The experiments demonstrated that the treatment of the cells with the A. argyi extract led to a complete suppression of the calcium influx (Figures 4A, B).

Next, we sought to characterize the inhibition of the calcium influx. Jurkat cells were treated with the A. argyi extract or the calcium ion chelator EDTA to determine whether the inhibition was mediated via chelation of calcium. As expected, a strong calcium influx was measured for untreated cells (control) after TCR stimulation (Figure 4C). This influx was prevented in EDTA-treated cells but restored by the addition of calcium dichloride to the medium (Figure 4C). In contrast, in cells treated with the A. argyi extract, the calcium influx capability could not be restored with the addition of increasing concentrations of calcium dichloride (Figure 4C). This suggested that the A. argyi extract did not inhibit the calcium influx through calcium chelating properties.

Upon TCR activation by antigen binding, the opening of a calcium channel in the membrane of the ER initiates calcium release from ER stores (ER store depletion). The depletion of these intracellular calcium stores causes the formation of calcium release-activated channels (CRAC) channels, which in turn leads to a strong influx of calcium from the extracellular space to the cytosol (store-operated calcium entry, SOCE) and allows replenishment of the calcium stores in the ER (Robert et al., 2011). To discriminate between these two options, we stained Jurkat cells with Fluo3 and FuraRed in a medium supplemented with EDTA. During the staining process, the cells were treated with the A. argyi extract in a calcium-free setting to ensure a calcium-free, extracellular compartment. Cells were stimulated via TCR to induce the calcium influx. Given that the SOCE was prevented, calcium store depletion could be measured. In comparison to the untreated cells (control), treatment with the extract lowered calcium ER store depletion (Figure 4D). To determine whether the SOCE was also inhibited by the A. argyi extract, or just prevented due to the lacking depletion of ER calcium store, we stained Jurkat cells with Fluo3 and FuraRed in a medium supplemented with EDTA, and induced the calcium ER store depletion by TCR stimulation. Next, the extract and calcium dichloride were added and the SOCE was measured. The medium of control cells was supplemented with calcium dichloride directly after depletion of the intracellular calcium stores. The results showed that the A. argyi extract also lowered the SOCE from the extracellular space (Figure 4D).

In summary, the strong reduction of calcium influx upon TCR stimulation after treatment with the A. argyi extract resulted from suppression of the ER calcium store depletion and a reduction of the SOCE.

We previously isolated a series of sesquiterpene lactones and flavones from the active extract (Figure 7A), some of which showed significant inhibitory effects on T lymphocyte proliferation (Reinhardt et al., 2019) (Figure 1A). Several of the most-active compounds were tested: two stereoisomeric guaianolides, artecanin and canin (1 and 5), artanomaloide (2), arteglasin A (3) differing from 1 and 2 in the decoration of the 7-membered ring, the moderately active flavone jaceosidin (4), and two stereoisomeric seco-guaianolides, seco-tanapartholides B and A (6 and 7) (Figure 5).

To better understand their contribution to the activity of the extract, their influence on activation, cytokine production, and degranulation capacity of human T lymphocytes was investigated.

The compounds 1, 2, 3, 5, and the compound mix significantly reduced the expression of CD25 in CD4^– and CD4⁺ T cells at 10 and/or 3 µg/ml. Compound 6 suppressed the CD25 expression in CD4⁺ T cells at all tested concentration levels, while compound 7 suppressed it at 10 and 3 µg/ml. Compounds 6 and 7 had no significant effect on CD4^– T cells. The isolated flavone 4 showed no effect on CD25 expression (Figure 6A).

All isolated compounds, with the exception of compound 4, significantly and strongly suppressed IL-2-producing cells at 10 µg/ml. While the effect was only miniscule for the flavone (4), compounds 1, 3, 5-7, and the mix also showed significant inhibition at 3 µg/ml. A similar pattern was observed for the suppression of tumor necrosis factor α-(TNFα) and IFN-γ-producing cells. The strongest suppression of IL-2- and IFN-γ-producing cells was observed from compound 6 (Figure 6B).

Upon release of perforin and granzymes, T lymphocytes express the lysosomal-associated membrane protein 1 (LAMP-1, CD107a) on their surface. Analysis of the LAMP-1 surface expression, showed concentration-dependent effects for the mixture of all compounds and the A. argyi extract (Figure 6C).

Jurkat-based NFAT and NF-κB reporter experiments were performed to shed light on the interaction of compounds by manipulation the T lymphocyte signaling. For the sesquiterpene lactones, but not for the flavone (4), a concentration-dependent, suppressive effect on the NFAT pathway was found, with IC₅₀ values < 10 µM (< 3 µg/ml) (Figures 7C, D). Likewise, NF-κB activity was significantly decreased by all compounds except compound 4 (Figures 7B, D). We here focused on the effects of the A. argyi extract and isolated compounds on the NFAT signaling pathway.

Thus we tested whether the isolated compounds were able to inhibit TCR-induced calcium signaling. To this end, we looked at both the ER calcium store depletion (Supplement 1, left panel) and the SOCE (Supplement 1, right panel), using the methodology, described above. Single compounds and a mixture of all isolated compounds were tested with the assumption that the mixture would, at least in part, mimic the activity of the whole extract. The A. argyi extract was used as a control. A slight inhibition of calcium influx from the ER combined with a significant time delay was observed for compound 5. A similar delay was observed for the compound mix and, to a lesser extent, for compounds 1, 6, and 7. Compounds 2 and 3 increased the intensity of the calcium influx. However, neither the single compounds nor the compound mix inhibited the calcium influx as effectively as did the A. argyi extract (Supplement 1).