Column chromatography was performed using a silica gel (200-300 mesh, Qingdao Marine Chemical Inc., China) and MCI gel HP-20 (75-150 μm, Mitsubishi Chemical Co., Ltd., Japan). Analytical high-performance liquid chromatography (HPLC) was performed on a Shimadzu LC-20AT instrument using a YMC C₁₈ column (250 × 4.6 mm, 5 μm) with a flow rate of 1.0 mL/min, and with the gradient program of MeCN/H₂O set to 10:90 (t = 0 min), 60:40 (t = 40 min), 60:40 (t = 45 min), 10:90 (t = 50 min), and 10:90 (t = 55 min). HPLC-grade acetonitrile, methanol, and other analytical grade reagents were purchased from Beijing Chemical Reagent Co., Ltd. (Beijing, China).

The plant Teucrium montanum L. was collected in Rtanj, Serbia (GPS: E 21.92658400, N 43.76136400), in August 2018. The plant material was identified by Prof. Z. Dajić Stevanović and taxonomically interpreted according to Euro+Med Plantbase and Flora of Serbia (Josifović 1970–1986) databases. An herbarium voucher (RS-030718-1) was deposited at the Department of Agricultural Botany in the Faculty of Agriculture, University of Belgrade.

The aerial parts of T. montanum (10 kg) plants were air-dried, chopped, and extracted three times (2 h each) with 95% EtOH under reflux conditions. The EtOH extract was evaporated to almost dryness in vacuo, and the resulting residue (1040 g) was redissolved in MeOH, and mixed with Si gel (1500 g, 200–300 mesh). After removing MeOH, the dry mixture was extracted in a Soxhletanalytical-grade extractor (100 × 15 cm) successively with petroleum ether (PE, 8.0 L), dichloromethane (DM, 8.0 L), ethyl acetate (EA, 8.0 L), and methanol (MT, 8.0 L) under reflux conditions. The PE, DM, EA, and MT extracts were then evaporated to dryness under reduced pressure. The yields of the extracts were: PE, m = 187 g; DM, m = 90 g; EA, m = 105 g, and MT, m = 400 g. The DM portion (90 g) was subjected to Si gel column chromatography (80 × 8 cm, 200–300 mesh) and eluted with a petroleum ether/acetone (10:–-1:1, v/v) gradient system to yield thirteen fractions, IVA-A1~A8, IVA-B, IVA-C, IVA-D, IVA-E, and IVA-F, on the basis of TLC analysis. Fraction IVA-B (10 g), eluted with PE-acetone (8:1), was further fractionated by MCI gel HP-20 chromatography and eluted with a step gradient of MeOH-H₂O (40:60, 60:40, 80:20, and 100:0 v/v) to give 10 subfractions, named IVA-B1~B10. Fraction IVA-C (9 g) was eluted with PE-acetone (6:1), fractionated by MCI gel HP-20 chromatography, and then eluted with a step gradient of MeOH-H₂O (40:60, 60:40, 80:20, and 100:0 v/v) and acetone to yield 17 subfractions, named IVA-C1~C17. Similarly, fraction IVA-D (8 g) was eluted with PE-acetone (5:1), fractionated by MCI gel HP-20 chromatography, and eluted with a step gradient of MeOH-H₂O (40:60, 60:40, 80:20, and 100:0 v/v) and acetone to yield 14 subfractions, named IVA-D1~D14. IVA-E (8.5 g) was eluted with PE-acetone (3:1), fractionated by MCI gel HP-20 chromatography, and eluted with a step gradient of MeOH-H₂O (40:60, 60:40, 80:20, and 100:0 v/v) and acetone to yield 14 subfractions, named IVA-E1~E14. The last fraction, IVA-F (3.3 g), was eluted with PE-acetone (2:1), fractionated by MCI gel HP-20 chromatography, and eluted with a step gradient of MeOH-H₂O (40:60, 80:20, and 100:0 v/v) to yield four subfractions, IVA-F1~F4 (Figure 1).

Primary splenocytes were isolated from 6–8-week-old C57BL/6 mice purchase from the Institute of Laboratory Animal Sciences (CAMS&PUMC, Beijing, China). Research was conducted in accordance with all institutional guidelines and ethics and approved by the Laboratories Institutional Animal Care and Use Committee of the Chinese Academy of Medical Sciences and Peking Union Medical College. All experimental protocols were reviewed and approved by the Institute of Materia Medica Animal Authorities (No. 00003375).

Alamar Blue (Cat. No. DAL1025) was purchased from Thermo (Waltham, USA) and antibodies for flow cytometry were purchased from BioLegend, eBioscience (San Diego, USA) or Signaling Technology (Danvers, USA) unless otherwise noted.

Splenocytes were re-suspended in RPMI medium (supplemented with 10% FBS, 1% penicillin and streptomycin, 10 mM HEPES, 1 mM glutamine, 1 mM sodium pyruvate and 55 μM β-mercaptoethanol) to prepare single cell suspensions and plated in 96-well U-button plates in concentrations of 2×10⁵ cells/well under standard conditions (5% CO2, 95% humidity, 37 ℃). At 41 h post-stimulation, the cells were incubated with 5% Alamar Blue (v/v)32 for 7 h prior to measuring the fluorescence intensities at 566 nm excitation and 586 nm emission using an Enspire Microplate Reader (Perkin Elmer).

Splenocytes were incubated with ACK lysis buffer for 5 min to remove erythrocytes. Cells cultured in RPMI-1640, supplemented with 10% fetal bovine serum, penicillin (100 U/mL), streptomycin (100 mg/mL), and IL-2 (100 U/mL) (Peprotech) at 37 ℃ in 5% CO₂, were stimulated with anti-CD3 (clone 145-2C11, 5 μg/ml, BioLegend) and anti-CD28 (clone 37.51, 2.5 μg/ml, BioLegend) mAbs for three days. Plant crude extracts or fractions were added to the culture at the beginning of experiments.

Splenocytes were labeled with CFSE (10 μM, Sigma, USA) at 37 °C for 5 min. After removing excess CFSE, cells were cultured in a 96-well, round-bottom plate with 2×10⁵ cells in complete RPMI 1640 medium supplemented with HEPES (10 mM), sodium pyruvate (1 mM), and 2-mercaptoethanol (55 μM) for 48 h before flow cytometry analysis.

Fluorescent dye-labeled antibodies against cell-surface markers CD8 (53-6.7), CD44 (IM7), CD62L (MEL-14), and CD69 (H1.2F3) were used for flow cytometry analysis. Prior to all flow cytometry staining, FcγIII/II receptors were blocked by incubating cells with anti-CD16/32 (2.4G2). All samples were acquired on a FACSVerse cytometer (Becton Dickson, USA) and data analysis was performed using FlowJo (TreeStar, USA).

Statistical analyses and a Pearson’s correlation test were performed using GraphPad Prism 8 software. Comparisons among means of more than two groups were conducted with through one-way ANOVA with Tukey’s post-hoc test. P values less than 0.05 were considered statistically significant.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Since T. montanum preparations usually consist of several dozen compounds, with few forming a sufficient majority to simply exert their effects singularly6, we tested different T. montanum extracts and fractions for their immunological effects on lymphocytes. As shown in the designed flowchart for the extraction and fractionation of T. montanum (Figure 1), four extracts of petroleum ether (IVA-PE), dichloromethane (IVA-DM), ethyl acetate (IVA-EA), and methanol (IVA-MT) from the aerial part of T. montanum were prepared (Figure 1A). Then, the lipid-soluble extract of IVA-DM was separated by silica gel column chromatography (Figure 1B) and fractionated by MCI gel HP-20 column chromatography to produce 67 fractions from all six crude fractions, IVA-A, IVA-B, IVA-C, IVA-D, IVA-E, and IVA-F (Figure 1C-G). The HPLC analyses of the bioactive extract IVA-DM and its chromatography fractions are shown in Figs. S1-S10.

To determine the immunoactivity of T. montanum, we used Alamar Blue to assess the growth of lymphocytes upon stimulation with the various pre-fractionated extracts of the plant. Increased lymphocyte numbers were observed after 48 h of stimulation with IVA-DM (Figure 1). The CFSE-based proliferation assay confirmed that cells proliferated in response to IVA-DM, but not to any other extracts, such as IVA-PE, IVA-EA, or IVA-MT. These results suggest the presence of pro-proliferative ingredients in the T. montanum extracts.

We then screened all 67 fractions generated from chromatographically fractionated IVA-DM for pro-proliferative activities using the Alamar Blue assay and identified nine fractions with dose-dependent cell growth-promoting activity, as indicated in Figure 3. These included fractions separated into petroleum ether/acetone at ratios of 8:1 (IVA-B8, -B9, and -B10), 6:1 (IVA-C13, -C14, -C15, and -C16), 5:1 (IVA-D13), and 3:1 (IVA-E13), respectively. None of the other fractions, such as the IVA-A, IVA-D, and IVA-F groups, were found to have the same effect. It is possible that several IVA-DM fractions interacted together and each contributed to part of the overall pro-proliferative activity for T. montanum.

CD8⁺ T cells play a central role in host immune response. Here, we evaluated how this type of cell responded to IVA-DM. CFSE-based analysis revealed that, compared to the CD8⁻ T cell population, CD8⁺ T cells proliferated selectively in response to IVA-DM, while other extracts, such as IVA-EA, showed no potential to stimulate cell proliferation (Figure 4A). The data indicate that the T. montanum extract IVA-DM effectively stimulated CD8⁺ T cells to proliferate. We then assessed whether IVA-DM is capable of activating CD8⁺ T cells by measuring the expression of CD69, an early activation marker on the surface of T cells. Flow cytometric analysis showed a significant dose-dependent increase in CD69 expression and in the percentage of CD69⁺ cells in the CD8⁺ population 48 h after IVA-DM stimulation, whereas there was no CD69 upregulation by IVA-EA (Figure 4B). As CD44 is a cell surface adhesion molecule that plays an important role in T-cell activation, we then examined whether IVA-DM could stimulate CD44 expression. Figure 4C demonstrated that the percentage of cells expressing CD44 markedly increased in CD8⁺ T cells following IVA-DM treatment, unlike IVA-EA, reaffirming the activity of T. montanum IVA-DM on T-cell activation.

We next explored whether the nine IVA-DM immunoreactive fractions could regulate the expression of CD44 and CD62L during the engagement of T-cell receptors. Among the nine fractions with proliferation-promoting activity, three of the 100% methanol-yielded fractions, IVA-B8, -B9, and -D13, and the only acetone fraction, IVA-C16, were able to upregulate CD44 expression dramatically (Figure 5A). Of the remaining five fractions collected from 100% methanol, IVA-B10 and -E13 failed to upregulate CD44 (Figure 5 A), while -C13, -C14 and -C15, which appeared to stimulate the highest cell proliferation, did not produce enough cells for analysis, presumably because of an inhibitory effect on cells activated by T-cell receptor engagement. Furthermore, we found that IVA-D13 induced a higher expression of CD62L, whereas the rest of the IVA-DM fractions did not demonstrate such effects (Figure 5 B).

CD44 and CD62L, as regulators of cell adhesion and migration, are essential in recruiting and activating effector and memory T cells. This prompted us to explore the role of these fractions in the differentiation of effector/effector memory (Teff/Tem) and central memory (Tcm) CD8⁺ T cells defined by CD44⁺CD62L⁻ and CD44⁺CD62L⁺, respectively. At 10 μg/ml, IVA-B8, -B9, and -C16 led to a significantly increased percentage of Teff/Tem, while IVA-B10 and -E13 demonstrated no significant changes to the Teff/Tem population (Figure 5 C).

It has been demonstrated that there is an intermediate-stage cell type, referred to as CD44⁻CD62L⁻ pre-effector T cells, that are involved in differentiation from naïve to Teff/Tem33. It seems very likely that the reduction of pre-effector T cells (Tpe) could be a consequence of enhanced Teff/Tem differentiation. We thus sought to investigate the changes of Tpe following stimulation. Compared to α3/28 only, all Teff/Tem-promoting fractions (IVA-B8, -B9, and -C16 at 10 μg/ml) were able to drastically lower the Tpe subset. Furthermore, IVA-B10 and -E13 also inhibited Tpe, but not as much as the above four fractions. Thus, it is possible that the Tpe population reduced by IVA-B8, -B9, and -C16 could be related to increased Teff/Tem differentiation (Figure 5 C).

In contrast to IVA-B8, -B9, and -C16, 10 μg/ml of IVA-D13 induced formation of more Tcm and a resulted in a substantially decreased proportion of Tpe (Figure 5 C). Thus, we reasoned that IVA-B10 and -E13 were unable to induce CD8⁺ T-cell differentiation at a lower dose, and therefore, that stimulation with a higher dose may be warranted. However, when the dose was increased to 30 μg/ml, obvious cytotoxicity was produced by IVA-B10 but not IVA-E13, which was then able to elevate the expression of both CD44 and CD62L (Figure 6 A). At this higher dose, IVA-E13 induced a significantly greater proportion of Tcm with a greatly reduced percentage of Tpe. Additionally, Tem was also substantially decreased (Figure 6 B). In sum, our findings demonstrate that the dichloromethane fractions IVA-B8, -B9, and -C16 were able to stimulate Tem differentiation, in contrast to IVA-D13 and -E13, which were capable of inducing Tcm formation.