Exploring the mechanism of berberine-mediated T_fh cell immunosuppression

Abstract

Background:

Our previous research revealed a novel function of berberine (BBR), a clinically relevant plant-derived alkaloid, as a suppressor of follicular T helper (Tfh) cell proliferation in secondary lymphoid organs of BBR-treated mice that underwent immunization for collagen-induced arthritis (CIA) in DBA1/J mice. Due to the importance of T_fh cell and B cell interactions in the generation of T cell-dependent humoral responses, the suppression of T_fh cell activity may have implications for the general safety of BBR as a prophylactic dietary supplement, and its potential use in antibody-driven autoimmune and hypersensitivity disorders.

Purpose:

This research aims to characterize BBR’s impact on the activation, differentiation, and proliferation of T_fh cells by examining the expression of key extracellular signaling molecules, as well as the activity of intracellular signaling molecules involved in the Ca²⁺-calcineurin-NFAT pathway and STAT3 phosphorylation, following activation.

Study Design:

In vitro experimental study using primary tissues.

Methods:

To explore the direct effects of BBR on the proliferation and differentiation of Tfh cells, isolated naïve CD4⁺ T cells (>95% pure) were activated and differentiated into pre- Tfh cells in the presence or absence of BBR. The resulting Tfh cell populations and the expression of the key extracellular molecules CXCR5, ICOS, and PD-1 were measured. In addition, we examined the impact of BBR treatment on the activity of key intracellular signaling molecules involved in T_fh cell activation and differentiation following TCR ligation and/or CD28 signaling (p-ZAP-70, p-Lck, p-PLCγ1, NFATc1 and intracellular calcium, Ca²⁺, concentrations), as well as IL-6 signaling (p-STAT3).

Results:

Treatment with BBR significantly reduced the expression of both CXCR5 (p <0.01) and ICOS (p < 0.005), but not PD-1, and reduced the percentage of Tfh cells within the total CD4⁺ T cell population. BBR treatment also led to a reduction in intracellular Ca²⁺ flux, activation of p-STAT3, and IL-21production.

Conclusion:

Our observations provide insight into the mechanism of BBR-mediated T_fh cell suppression and suggest that BBR treatment can directly inhibit T_fh cell activity, perhaps through interfering with cytokine receptor or downstream signaling.

Introduction

T Follicular Helper Cells

T follicular helper cells (T_fh cells) are a specialized subset of CD4⁺ T_h cells that reside within secondary lymphoid organs and are typically characterized as CXCR5⁺ICOS⁺PD-1⁺ cells within the CD3⁺CD4⁺ T cell population (Crotty, 2019, 2014a; Morita et al., 2011). These cells play a critical role in the generation of humoral immunity by aiding in germinal center formation, B cell affinity maturation, antibody isotype class switching, and the differentiation of B cells into long-lived plasma cells and memory B cells (Crotty, 2014a). Under normal physiological conditions, this is particularly important for generating specific, high affinity antibodies during primary immune responses that take place due to infection and/or vaccination, as well as for generating long-term humoral immunological memory (Crotty, 2019, 2014a). T_fh cells’ effector function of facilitating high-affinity antibody production can also be pathogenic, however. When exacerbated, this can be a driving force in some antibody-mediated autoimmune pathologies and graft rejection, and when suppressed, can diminish the body’s ability to safely and effectively respond to pathogens and/or vaccines.

Key Molecules Driving T_fh Cell Differentiation and Effector Function

The commitment of CD4⁺ T_h cells to the T_fh cell lineage versus other T_h effector cell phenotypes (e.g., T_h1, T_h2, T_h17, etc.) during activation and differentiation is thought to be predominantly driven early on by the presence of either IL-6 (in mice) or IL-6/IL-12 (in humans), ICOS signaling, and low levels of the T_fh cell-suppressing cytokine IL-2 (Crotty, 2019). Signaling through the IL-6 receptor and ICOS induces BCL6, a transcription factor which commits CD4⁺ T_h cells to the T_fh cell lineage by repressing transcriptional activators of other T_h effector cell phenotypes (e.g., T-bet, GATA-3, RORγt) (Choi et al., 2011; Crotty, 2019). Activation of the BCL6 transcription factor leads to the expression of CXCR5, a chemokine receptor for CXCL13, as well as increased expression of ICOS and PD-1. These cells also begin to secrete IL-21, which provides autocrine signals to support the continued expression of key T_fh cell-surface signaling molecules (Nurieva et al., 2008). These molecules, however, are not yet expressed in high quantities by these newly differentiated “pre-T_fh” or transitional T_fh cells, a fact that is important to note as it distinguishes the pre-T_fh cell phenotype from fully mature germinal center (GC) T_fh cells which have a higher expression of each (Crotty, 2019).

Maturation of pre-T_fh cells into fully functional GC T_fh cells depends on a continued stable interaction with, and antigen presentation by, cognate B cells, which pre-T_fh cells can initially receive upon CXCR5 and ICOS co-mediated migration to the T-B cell border and eventually into B cell follicles (Crotty, 2019, 2014b). While CXCR5 predominantly contributes to the directionality of T_fh cell migration to the B cell follicles, ICOS appears to predominantly contribute to the motility of T_fh cells via PI3K-mediated actin rearrangement and pseudopod activity (Xu et al., 2013). Signaling through the ICOS receptor is also thought to “override” the strong inhibitory PD-1 signaling that begins to occur in T_fh cells (Shi et al., 2018) as they increasingly interact with both cognate and bystander B cells. In this way, the PD-1 signaling in T_fh cells is a form of negative selection, ensuring that only T_fh cells expressing high quantities of ICOS remain active and localized to the B cell follicles (Shi et al., 2018). As ICOS signaling promotes the expression of key T_fh cell molecules that are involved in germinal center activities, such as CD40L, IL-21 and IL-4 (Liu et al., 2015; Panneton et al., 2019), as well as the maintenance of the T_fh cell phenotype, this PD-1-mediated selection for ICOS high expressing T_fh cells is particularly important.

Along with CXCR5, ICOS, and PD-1, which are hallmarks of T_fh cell differentiation and effector function, the Ca^2+-calcineurin-NFAT pathway must be scrutinized as the key intracellular pathway contributing to T_fh cell activation following TCR ligation (Martinez et al., 2016). Without initial signaling through this pathway, none of the above-mentioned activities leading to T_fh cell differentiation would occur. With T_fh cell activation and differentiation specifically, studies suggest that both NFAT1,2 activity are critical for expression of the key molecules in T_fh cell activity, such as CXCR5, ICOS and PD-1 (Martinez et al., 2016), as well as IL-21 production by T_fh cells (Ray et al., 2015).

Berberine

Our previous research investigating the prophylactic potential of berberine (BBR), a clinically relevant plant-derived alkaloid, in a collagen induced arthritis mouse model (CIA; model of rheumatoid arthritis) revealed that BBR delayed onset of the disease, reduced production of autoantibodies, and suppressed CD4⁺ T helper (T_h) cell activity, including a novel function of suppressing the proliferation of T follicular helper (T_fh) cells specifically (Vita et al., 2021). Due to the importance of T_fh cell and B cell interactions in the generation of T cell-dependent humoral responses, the current research further investigates any direct effects BBR might have on T_fh cells.

Berberine (BBR) is a plant-derived isoquinoline alkaloid found in the roots, rhizomes, and stem bark of plants within a variety of genera, such as Berberis (its namesake), Mahonia, Hydrastis, and Coptis, among others. The full breadth of botanical sources, as well as the variety of extraction methods, are well-described in a recent review by Neag et al (2018) (Neag et al., 2018).

BBR has already proved to be of importance for a variety of disease states through successful clinical trials, such as polycystic ovary syndrome (An et al., 2014; Li et al., 2013), type II diabetes (Yin et al., 2008; Zhang et al., 2010), diarrhea-predominant irritable bowel syndrome (Chen et al., 2015), psoriasis (Janeczek et al., 2018), and osteoarthritis (Oben et al., 2009). Additionally, BBR has demonstrated the ability to inhibit the production of a variety of pro-inflammatory cytokines by various immune cells (Allijn et al., 2016; Bae and Cheon, 2016), and has been shown to successfully and strongly regulate the inflammatory responses involved in clinically apparent autoimmune diseases in vivo such as collagen-induced arthritis (Hu et al., 2011; Wang et al., 2017, 2014; Yue et al., 2017), type I diabetes mellitus (Li et al., 2014), ulcerative colitis (Li et al., 2016; Yan et al., 2012), and experimental autoimmune encephalomyelitis (Jiang et al., 2013). Regarding CD4⁺ T cell suppression specifically, BBR has been shown to act in vitro and in vivo through a number of suggested mechanisms that ultimately inhibit the phosphorylation of key signaling molecules and/or transcription factors (e.g., JAK/STAT, AMPK, MAPK) that regulate the expression of genes involved in cell activation, differentiation proliferation, and effector function (Cui et al., 2009; Liu et al., 2016; Takahara et al., 2019; Yue et al., 2017). The direct effect of BBR on T_fh cells specifically, however, is still unknown.

Research Objective

Despite BBR’s observed therapeutic roles in vitro, in vivo, and in select clinical trials as an immunosuppressant, studies have yet to elucidate a specific mechanism of action by which BBR is exerting an inhibitory effect on T_fh cells. While it is important to note that there is growing evidence that BBR’s systemic immunosuppressive effects may be due in part to interactions with the gastrointestinal microbiome (Habtemariam, 2020; Yue et al., 2019), previous research highlighting measurable levels of BBR in within the plasma and splenic tissues of rodents (Sun et al., 2018; Tan et al., 2013; Yan et al., 2009), among other locations, demonstrates the additional need to evaluate the direct impact of BBR on T_fh cells. Thus, this research aims to characterize BBR’s impact on the activation and differentiation of T_fh cells by examining the expression CXCR5, ICOS, and PD-1 on pre-T_fh cells, components of the Ca²⁺-calcineurin-NFAT pathway following TCR ligation, and STAT3 phosphorylation following IL-6 signaling. Since T_fh cell and B cell interactions are essential to adaptive immune function, specifically humoral immunity, the suppression of T_fh cell activity may have implications for the general safety of BBR as a prophylactic dietary supplement, and its potential use in antibody-driven autoimmune and hypersensitivity disorders.

Methods

General Reagents

The following specific materials were used for these experiments: DMSO (VWR, Radnor, PA, USA); 1X PBS, berberine hydrochloride (purity > 98%; Sigma-Aldrich, St. Louis, MO, USA); ACK lysis buffer (Quality Biological, Gaithersburg, MD); RPMI 1640 supplemented to 2 mM l-glutamine, 1% v/v penicillin/streptomycin, 1 mM sodium pyruvate, 10 mM HEPES (all from ThermoFisher, Waltham, MA, USA), 0.05 mM β-mercaptoethanol (Bio-Rad, Hercules, CA, USA), and 10% fetal bovine serum (VWR/Seradigm,), Fluo-4-AM Ester (ThermoFisher), Calcimycin (ThermoFisher), bovine serum albumin (BSA;VWR); and the following cytokines and antibodies were procured from BioLegend (San Diego, CA): IL-6, anti-IL4, anti-IFN-γ, UltraLeaf anti-CD28, UltraLeaf anti-CD3, biotinylated anti-CD3, biotinylated anti-CD28, and Streptavidin.

Final Berberine Solutions Used in vitro

A stock solution of 10 mM berberine dissolved in DMSO was stored at −20 ° C when not in use. This stock solution was diluted in PBS to achieve the concentrations of 0.25 μM, 0.5 μM, and 1.0 μM to be used in treatment protocols. All final DMSO concentrations were <0.01%.

Fluorescent Antibodies

Fluorescent antibodies (all purchased from BioLegend, San Diego, CA, USA) were used for flow cytometric analysis and fluorescent cell sorting, and include APC anti-mouse CD4 (clone GK1.5), APC/Cy7 anti-mouse CD3ε (clone 145–2C11), Pacific Blue anti-mouse CD19 (clone 6D5), PE anti-mouse CD14 (clone M14–23), FITC anti-mouse CXCR5 (clone L138D7), Pacific Blue anti-mouse ICOS (clone C398.4A), PE anti-mouse PD-1 (clone 29F.1A12), PE anti-mouse phospho-ZAP70 (clone 1503319), PE anti-mouse phospho-Lck (clone A18002D), PE anti-mouse phospho-PLCγ1 (clone A17025A), PE anti-mouse NFATc1 (clone 7A6), and recommended isotype controls; viability and apoptosis were assessed by propidium iodide and annexin V (Alexa Fluor 488) Ready Flow reagents (ThermoFisher).

Isolation of CD4⁺CXCR5⁻ T Cells Via Fluorescent Cell Sorting

Spleens were harvested from C57BL/6 mice (6–8 weeks old) in accordance with IACUC protocol 2005C-NP-M-23 at the University of Northern Colorado. To create single cell suspensions, briefly, spleens were ground, washed with 3 mL of ACK lysis buffer for 5 min, and then strained into 35 mL of complete RPMI 1640. The resulting mixed splenocytes were then stained with fluorescent antibodies specific for CD4 (APC), CD14 (PE), and CXCR5 (FITC). To obtain a purified population of CD4⁺ T cells, the stained mixed splenocytes were isolated using a SONY SH800 by first sorting CD4⁺CD14⁺ monocytic cells out of the total CD4⁺ population; the remaining CD4⁺CD14⁻ were then sorted as CXCR5⁻. The resulting CD4⁺CD14⁻CXCR5⁻ population was expected to be naïve CD4⁺ T cells and confirmed via flow cytometry (>97% pure; Supplemental Figure 1).

Isolation of CD4⁺ T cells via Magnetic Cell Sorting

Spleens were harvested from C57BL/6 mice, 6–8 weeks old. CD4⁺ T cells were then sorted out of single cell suspensions of mixed splenocytes using a negative selection approach with the MojoSort Magnetic Cell Separation System (BioLegend), according to the manufacturer’s instructions. Briefly, splenic single cell suspensions were strained, centrifuged at 250×g, and resuspended in 1X MojoSort Buffer (BioLegend). Cells were incubated for 15 minutes at room temperature with MojoSort mouse “untouched” CD4 T cell biotin-antibody cocktail (BioLegend), followed by a 15-minute room temperature incubation with MojoSort SAV-nanobeads (BioLegend). The antibody and nanobead-loaded cells were then incubated at room temperature inside the MojoSort magnet for 10 minutes for cell sorting. Stromal, myeloid, and CD4⁻ T cells are captured by the antibody-conjugated beads and precipitated from suspension by incubation in the magnet. Following incubation inside of the magnet, the supernatant containing uncaptured CD4⁺ T cells was collected and used for cell protocols. Purity of cells was confirmed via flow cytometry to be ~95% pure (Supplemental Figure 2).

Pre-T Follicular Helper Activation and Differentiation in vitro

As this differentiation protocol models the initial interaction between naive CD4⁺ T cells and DCs, the resulting differentiated T_fh cell population had “pre-T_fh cell-like” qualities, as they would not become fully functional germinal center T_fh cell unless co-cultured with cognate B cells. Cells were sorted by fluorescence flow cytometry as described above, and the purified naive CD4⁺ T_h cells were then differentiated in a protocol adapted from Andris et al. (2017) (Andris et al., 2017). Briefly, CD4⁺ T_h cells (5×10⁵ cell/ml) were seeded into a 24-well plate pre-coated with anti-CD3 (5 μg/mL) and were incubated in the presence of soluble anti-CD28 (2.5 μg/mL), IL-6 (20 ng/mL), anti-IL4 (10 μg/mL), and anti-IFNγ monoclonal antibodies (10 μg/ml) for 4 days or 2 days. A 4-day incubation period was used for the detection of CXCR5, ICOS, PD-1 and total pre-T_fh cell population counts (Supplemental Figure 3), whereas a 2-day incubation period was used for the detection of NFATc1 and p-STAT3 during the early stages of differentiation (Supplemental Figure 4). The 2 day incubation period was chosen to capture pre-T_fh cells during the early stages of differentiation from CD4⁺ T_h cells, because T_fh cells have been detected as early as 2 days after initial activation in previous studies (Choi et al., ). Cells receiving experimental treatment were exposed to either 0.25 μM, 0.5 μM, and 1.0 μM BBR or a volume-matched vehicle control of <0.01% DMSO in PBS (labeled as “PBS 0.25 μM, 0.5 μM, or 1.0 μM” in results) at the beginning of the incubation period.

Berberine Concentration-Response Experiments

The final BBR concentrations used – 0.25 μM, 0.5 μM, and 1.0 μM – were chosen based on previously established in vitro work [24,39,50], our own calculations based on average oral bioavailability (<1%) [51] so the in vitro exposure of cells to BBR more accurately reflects in vivo exposure, and cell viability experiments. Cell viability of splenic T cells was measured following activation with CD3/CD28 for 72-hours in the presence or absence of BBR. Cell viability was then measured via flow cytometry using propidium iodide (necrotic cell detection) and annexin V (apoptotic cell detection) fluorescent staining (Example gating strategy in Supplemental Figure 7). The purpose of these experiments was solely to establish viable BBR concentrations, not to undertake a full study of mechanism of BBR killing as this would be irrelevant to the physiological conditions under which BBR would be used in vivo.

Detection of Intracellular Signaling Molecules at 5-, 15-, and 30-minute Time Intervals Post-Activation of CD4⁺ T Cells

CD4⁺ T cells were seeded into a 24-well plate and incubated at 37°C for 24 hours in complete RPMI medium. During this 24-hour incubation period, cells receiving BBR treatment were also incubated with either 0.25 μM, 0.5 μM, or 1.0 μM BBR or volume-matched vehicle control. Following incubation, cells were then transferred to a 24-well plate pre-coated with anti-CD3ε and soluble anti-CD28 at 5 μg/mL each in complete RPMI for T cell activation. The activation reaction was stopped at either 5, 15, or 30 minutes by pipetting cells into 2 mL conical vials on ice and then immediately centrifuging for 5 minutes at 250×g.

Flow Cytometry

Following the T_fh cell differentiation protocol and incubation period, the resulting cell populations were stained with fluorescent antibodies specific for: CD3⁺CD4⁺ T helper (T_h) cells, CD3⁺CD4⁺CXCR5⁺ T_fh cells, as well as the cell-surface molecules ICOS and PD-1 on CD3⁺CD4⁺CXCR5⁺ T_fh cells. Expression of cell surface molecules was determined by measuring the MFI and the population data were determined by measuring the percent positive cells. Example gating strategy can be found in Supplemental Figure 3. For staining of intracellular signaling molecules, the supernatant of cell suspensions was removed, and 500 mL of fixation buffer (BioLegend) was added to the cell pellet and incubated at room temperature in the dark for 20 minutes. A standard intracellular antibody staining protocol for either PE anti-phospho-ZAP-70, anti-phospho-Lck, anti-phospho-PLCγ1, anti-NFATc1, and anti-phospho-STAT3 (BioLegend) using 1X permeabilization buffer (BioLegend). For any cells also being stained extracellularly with APC anti-CD4 and FITC anti-CXCR5, a standard extracellular staining protocol was followed prior to intracellular staining. Cells for all experiments were then analyzed on an Attune NxT (ThermoFisher) flow cytometer as per the described gating strategies (Supplemental Figures 4 and 5).

Calcium Mobilization and Detection Assay Via Flow Cytometry

After the cell sorting protocol, cells were then seeded onto a 24-well plate and incubated at 37°C for 24 hours in complete RPMI medium. During this 24-hour incubation period, cells receiving BBR treatment were also incubated with either 0.25 μM, 0.5 μM, or 1.0 μM BBR or volume-matched vehicle control. Following incubation, cells were centrifuged for 5 minutes at 250×g and resuspended in calcium-free PBS supplemented with 0.5% of BSA for indicator loading. Fluo-4 AM Ester (ThermoFisher) was added to wells at 1μM per 1 × 10⁷ cells/mL for 45 minutes at 37°C. Cells were washed and resuspended in calcium free PBS supplemented with 0.5% of BSA and rested for 30 minutes at room temperature to allow for sufficient cleavage of AM esters, rendering the Fluo-4-AM cell impermeable. Cells were then treated with 5 μg/mL soluble biotinylated anti-CD3ε and 5 μg/mL soluble biotinylated anti-CD28 (both from BioLegend) for 20 minutes at room temperature. The baseline fluorescence of Fluo-4-AM (BL-1 channel) in CD4⁺ T was then acquired by flow cytometry on an Attune NxT cytometer (ThermoFisher) for 60 seconds without stimulus. The sample tube was briefly removed, and cells were stimulated with the addition of 20 μg/mL streptavidin to induce cross-linking of antibody bound CD3 and CD28 receptors. Stimulated samples were then recorded for an additional 240 seconds to detect any increases in Fluo-4-AM fluorescence. The sample tube was briefly removed, and cells were further stimulated with the addition of 1 μM Calcimycin as a positive control. All data for calcium mobilization were taken from the peak fluorescence values of a kinetics plot, gated as shown in Supplemental Figure 6.

Interleukin-21 ELISA

Following the T_fh cell differentiation protocol and incubation period, the cell supernatant was collected for the measurement of IL-21 concentrations via ELISA as per the manufacturer’s instructions and using manufacturer’s pre-made detection reagents (PeproTech). Briefly, plates were coated with capture antibody and incubated at room temperature for 24 hours. Plates were then washed in ELISA wash buffer, coated with blocking buffer, and incubated for 1 hour at room temperature. Plates were washed again in ELISA wash buffer and incubated at room temperature for 2 hours with the IL-21 standard serial dilutions and samples, followed by a wash and incubation at room temperature for 2 hours with detection (secondary) antibody, then followed by a wash and incubation with HRP-avidin conjugate for 30 minutes. Finally, plates were washed and the ABTS development substrate was added to wells. Plates were monitored every 5 minutes for color change, and optical densities were taken at 450 nm using a microplate reader.

Statistical Analysis

The assumption of normality was met for both cell population data and mean cell-surface molecule expression data. Thus, for all comparisons the ANOVA test with Tukey’s multiple comparisons test was used. All tests had an α = 0.05. All analyses were performed using Prism version 8 (GraphPad, San Diego, CA, USA).

Results

The Impact of Varying Concentrations of Berberine on Cell Viability-

To assess the impact of BBR on cell viability, the percent of apoptotic and necrotic cells was measured following BBR treatment. The use of BBR at 0.25 μM and 0.5 μM did not significantly contribute to necrotic or apoptotic events over the BBR 0 control (Figure 1). In contrast, the use of BBR at 1 mM led to, on average, about a 10% increase in apoptotic events and about a 1.5 % increase in necrotic events, when compared to baseline (BBR 0).

Figure 1. The impact of BBR on cell viability.

CD4⁺ T cells were activated in the presence/absence of berberine (BBR; 0.25 μM, 0.5 μM, 1.0 μM) for 72-hours. Apoptotic cells (A) were measured as CD3⁺CD4⁺ T cells that were also positive for Annexin V staining (n = 6 per group). Necrotic cells (B) were measured as CD3⁺CD4⁺ T cells that were positive for propidium iodide (PI) staining (n = 6 per group). Statistical analysis was made with ANOVA and Tukey’s multiple comparisons tests (****p<0.0001).

Berberine Inhibits the Differentiation of Pre-T_fh Cells in vitro-

To determine if BBR has a direct suppressive effect on T_fh cells, the population of CD3⁺CD4⁺CXCR5⁺ T_fh cells (“pre-T_fh cells”) within the total CD3⁺CD4⁺ T cell population was measured following activation and differentiation of pre-T_fh cells in the presence or absence of BBR (0 μM, 0.25 μM, 0.5 μM, and 1.0μM), or a volume-matched vehicle control. BBR treatment at 0.25 μM, 0.5 μM, and 1.0 μM significantly reduced the population of pre-T_fh cells within the total CD3⁺CD4⁺ T cell population (Figure 2). When the resulting percent positive pre-T_fh cells of BBR-treated cells was compared with that of the volume-matched PBS vehicle control, it was confirmed that the vehicle of BBR delivery did not negatively impact the percent positive pre-T_fh cells.

Figure 2. Population of CD3⁺CD4⁺CXCR5⁺ pre-Tfh cells following activation and differentiation in the presence or absence of berberine.

(A.) Shown here is the percent of CD3⁺CD4⁺CXCR5⁺ T cells within the total CD3⁺CD4⁺ T cell population following activation and differentiation in the presence or absence of berberine (“BBR; 0.25 μM, 0.5 μM, 1.0 μM), or a volume matched 0.01% DMSO in PBS vehicle control (PBS); n = 8 per group. Statistical analysis was made with ANOVA and Tukey’s multiple comparisons tests (*p<0.05; **p<0.01; ***p<0.001; ****p<0.0001; ns for all p>0.05). (B.) Shown here are representative density plots displaying the percent of CD3⁺CD4⁺CXCR5⁺ T cells within the total CD3⁺CD4⁺ T cell population following activation and differentiation in the presence or absence of berberine (BBR; 0.25 μM, 0.5 μM, 1.0 μM).

Berberine Reduces Expression of Key Cell-Surface Molecules on Pre-T_fh Cells in vitro-

To further characterize the BBR-mediated suppression of T_fh cells, the expression of key cell surface molecules CXCR5, ICOS, and PD-1 was examined on CD3⁺CD4⁺CXCR5⁺ T_fh cells (“pre-T_fh cells”) following activation and differentiation of pre-T_fh cells in the presence or absence of BBR (0 μM, 0.25 μM, 0.5 μM, and 1.0 μM), or a volume-matched vehicle control (Figures 3 and 4). At all concentrations, BBR treatment significantly reduced expression of CXCR5 and ICOS (Figure 3A). However, there was no apparent impact on PD-1 expression except for a non-significant trend of increased expression at 1.0 μM. When the expression of CXCR5 and ICOS on BBR-treated cells was compared with that of the volume-matched PBS vehicle controls, it was confirmed that the vehicle of BBR delivery did not impact the expression of these molecules (Figure 3B).

Figure 3. Expression of cell surface molecules on CD3⁺CD4⁺CXCR5⁺ pre-Tfh cells following activation and differentiation in the presence or absence of berberine.

Naïve CD4⁺ T cells were isolated from mixed splenocytes and differentiated into a pre-T_fh cell phenotype in the presence or absence of herberine (BBR; 0.25 μM, 0.5 μM, 1.0 μM) (A), or a volume matched 0.01% DMSO in PBS vehicle control (PBS) (B); n = 8 per group. Shown are expression of CXCR5, ICOS, and PD-1 on CD3⁺CD4⁺CXCR5⁺ T cells as the fold change of mean fluorescence intensity (MFI). Statistical analysis of cell surface molecule expression made with ANOVA and Tukcy’s multiple comparisons tests (*p<0.05; **p<0.01; ***p<0.001; ****p<0.0001; ns for all p>0.05).

Figure 4. Representative density plots displaying expression of cell surface molecules on CD3⁺CD4⁺CXCR5⁺ pre-Tfh cells following activation and differentiation in the presence or absence of berberine.

Displayed are representative density plots showing expression of CXCR5, ICOS, and PD-1 on CD3⁺ CD4⁺ CXCR5⁺ T cells after differentiation of pre-T_fh cells from naïve CD4⁺ T cells in the presence or absence of berberine (BBR; 0.25 μM, 0.5 μM, 1.0 μM).

The Effects of Berberine On Key Intracellular Signaling Molecules During Early pre-T_fh Cell Differentiation

Intracellular concentrations of cytoplasmic NFATc1 and phosphorylated p-STAT3 were measured following the activation and differentiation of CD4⁺ T cells into pre-T_fh cells in the presence or absence of BBR (or a vehicle control); the cell culture occurred for 48 hours to capture early pre-T_fh cell differentiation. BBR treatment significantly reduced the concentration of p-STAT3 in early pre-T_fh cells (Figure 5A). However, BBR treatment did not significantly impact cytoplasmic NFATc1 protein concentrations at a physiologically relevant concentration (Figure 5B).

Figure 5. Concentration of key intracellular signaling molecules in CD3⁺CD4⁺CXCR5⁺ pre-Tfh cells following activation and differentiation in the presence or absence of BBR.

Naïve CD4⁺ T cells were isolated from mixed splenocytes and differentiated into a pre-T_fh cell phenotype in the presence or absence of berberine (BBR; 0.25 μM, 0.5 μM, 1.0 μM), or a volume matched 0.01% DMSO in PBS vehicle control (PBS); n = 6 per group. Shown is the concentration of the phosphorylated form of p-STAT3 (A) and cytoplasmic NFATc1 (B) following differentiation. The concentration is represented as the fold change of mean fluorescence intensity (MFI). Statistical analysis of cell surface molecule expression made with the ANOVA and Tukey’s multiple comparisons tests (*p<0.05; **p<0.01, ***p<0.001, ****p<0.0001). (C) Also shown are representative density plots displaying p-STAT3 and cNFAT expression in CD3⁺CD4⁺CXCR5⁺ T cells following activation and differentiation in the presence or absence of berberine (BBR; 0.25 μM, 0.5 μM, 1.0 μM).

The Effects of Berberine On Key Intracellular Signaling Molecules Immediately Following T Cell Activation

Following pre-treatment with BBR for 24 hours, naïve CD4⁺ T cells were activated and the concentrations of p-ZAP-70, p-PLCγ1, and p-LCK were measured at 5-minute, 15-minute, and 30-minute timepoints; total cytoplasmic concentration of NFATc1 was also measured at these timepoints. Pre-treatment with BBR for 24 hours did not significantly impact the concentration of p-PLCγ, p-ZAP-70, or p-Lck (Figure 6A–C) at any of the time points in activating CD4⁺ T cells. Pre-treatment with BBR also did not have an impact on the cytoplasmic concentration of NFATc1 at any of the time-points (Figure 6D).

Figure 6. Concentration of key intracellular signaling molecules in CD3⁺CD4⁺ T cells at select time points following activation.

Naïve CD4⁺ T cells were isolated from mixed splenocytes (n = 6 per group), pre-treated for 24 hours with BBR (0.25 μM, 0.5 μM, 1.0 μM) OR a volume matched 0.01% DMSO in PBS vehicle control (PBS) and activated. Shown are concentrations of the phosphorylated forms of key signaling molecules p-PLCyl (A), p-ZAP-70 (B), and p-Lck (C), as well as cytoplasmic NFATc1 (D). The concentration is represented as the fold change of mean fluorescence intensity (MFI). No significant differences were observed. Statistical analysis of cell surface molecule expression made with the ANOVA and Tukey’s multiple comparisons tests.

The Effects of Berberine On Intracellular Calcium Mobilization

Following pre-treatment with BBR for 24 hours, naïve CD4⁺ T cells were activated, and intracellular calcium mobilization was measured by flow cytometry using Fluo-4 AM Ester (Figures 7 and 8). To quantitate mobilization, the peak value of intracellular Ca²⁺ concentration was measured while cells were in an unstimulated state and then subtracted from the peak value of stimulated (activated) cells. Calcimycin was used as a positive control following cell activation. Pre-treatment of naïve CD4⁺ T cells with 0.25 μM, 0.5 μM, 1 μM BBR elicited a significant reduction in intracellular Ca²⁺ concentration following T cell activation (Figure 7).

Figure 7. Calcium mobilization in CD4⁺ T cells following activation.

Naïve CD4⁺ T cells were isolated from mixed splenocytes and activated following pre-treatment for 24 hours with BBR (0.25 μM, 0.5 μM, 1.0 μM) OR a volume matched 0.01% DMSO in PBS vehicle control (PBS); n = 6 per group. Shown is the difference in the peak value of Ca²⁺ concentrations before and after stimulation, indicating the magnitude of Ca²⁺ mobilized after activation. The concentration is represented as the fold change from the BBR 0 group’s peak fluorescence values. Statistical analysis of cell surface molecule expression made with ANOVA and Tukey’s multiple comparisons tests (*p<0.05; **p<0.01, ***p<0.001, ****p<0.0001).

Figure 8. Representative graphs of calcium mobilization in CD4⁺ T cells following activation.

Naïve CD4⁺ T cells were isolated from mixed splenocytes and activated. Shown are representative graphs displaying Ca²⁺ concentrations before and after stimulation (n = 6 per group), indicating the magnitude of Ca²⁺ mobilized after activation.

Berberine Reduces IL-21 Production by Pre-T_fh Cells-

As IL-21 production by T_fh cells, in both the pre- T_fh stage and GC T_fh, is an important autocrine signal for T_fh cells as well as paracrine signal for nearby B cells, we chose to examine by ELISA if BBR influenced IL-21 production by pre-T_fh cells. We observed that 0.25 μM, 0.5 μM, and 1 μM BBR significantly reduced the production of IL-21 during the activation and differentiation of naïve CD4⁺ T_h cells into pre- T_fh cells (Figure 9).

Figure 9. IL-21 production by CD3⁺CD4⁺CXCR5⁺ pre-T_fh cells following activation and differentiation in the presence or absence of berberine.

Naïve CD4⁺ T cells were isolated from mixed splenocytes and differentiated into a pre-T_fh cell phenotype in the presence or absence of berberine (BBR; 0.25 μM, 0.5 μM, 1.0 μM) (A), or a volume matched PBS vehicle control (B). Change in concentration of IL-21 (pg/mL) in cell supernatant is shown as the fold change from BBR 0 or volume-matched control (n = 6 per group). Statistical analysis of cell surface molecule expression made with ANOVA and Tukey’s multiple comparisons tests (*p<0.05; **p<0.01; ***p<0.001; ****p<0.0001; ns for all p>0.05).

Discussion

In summary, BBR has a direct suppressive effect on T_fh cells, as seen through the reduced percentage of pre-T_fh cells within the total CD4⁺ T cell population (i.e., fewer CD4⁺ T_h cells differentiated into pre-T_fh cells) following activation and differentiation in the presence of BBR, as well as the decreased production of IL-21 and the expression of CXCR5 and ICOS, but not PD-1, on pre-T_fh cells. Moreover, BBR reduced the concentrations of phosphorylated STAT3 (p-STAT3), a key transcription factor involved in differentiating T_fh cell response to IL-6, as well as T_fh cell continued production of and response to IL-21 (Choi and Crotty, 2021; Choi et al., 2013a; Ray et al., 2014). Interestingly, while we observed a reduction in p-STAT3, we did not observe a decrease in the active forms of key signaling molecules involved in T cell activation following TCR and CD28 ligation – such as p-ZAP-70, p-Lck, p-PLCγ1—or cytoplasmic NFATc1. Thus, we hypothesize that BBR suppression of CD4⁺ T_h and T_fh cells is not mediated through direct interference with the TCR or CD28 receptors but is instead mediated through interference with cytokine receptor signaling and/or their downstream signaling components. As a consistently elevated expression of CXCR5 and ICOS, as well as IL-21 production, are required for the maintenance the T_fh cell lineage and effector function (Crotty, 2019, 2014b), it is likely BBR interferes with the development of a functional T_fh phenotype that is sufficient to facilitate robust GC responses. These results support our previous research which revealed that BBR delayed the onset of CIA, reduced autoantibody production, and suppressed CD4⁺ T_h cell activity, specifically T_fh cell activity (Vita et al., 2021).

The BBR-mediated reduction in p-STAT3 may be a possible mechanism for the decreased CXCR5 and ICOS expression and IL-21 production that was observed in our in vitro model. BBR has been shown to downregulate the expression of classic CD4⁺ T_h cell co-stimulatory molecules in multiple CD4⁺ T_h cell effector phenotypes, such as CD28 and CD154 (CD40L) (Vita et al., 2021). Our results add to this previously described role of BBR as an inhibitor of CD4⁺ T_h cell co-stimulatory molecule expression by showing that BBR also inhibits the expression of ICOS, a classic co-stimulatory molecule of T_fh cells, as well as the canonical T_fh cell marker and chemokine receptor CXCR5. However, the downregulation CXCR5 and ICOS, but not PD-1, is slightly confounding as many of the cell signaling pathways that lead to the upregulation of these factors during differentiation of T_fh cells are shared. For example, there is much evidence that IL-6-mediated STAT3 signaling during T_fh cell differentiation induces expression of Bcl6, the key transcription factor that promotes the T_fh cell lineage (Choi and Crotty, 2021; Choi et al., 2013a; Read et al., 2017). Il-6-mediated Bcl6 expression then indirectly promotes the expression of T_fh differentiation and effector genes by inhibiting the repressors of genes encoding for CXCR5, ICOS, PD-1, and IL-21, among others (Choi and Crotty, 2021). Once IL-21 is produced, it reinforces the T_fh lineage in much the same way as IL-6 – by signaling through STAT3 and reinforcing Bcl6 expression and the subsequent Bcl6-mediated indirect expression of key T_fh cell differentiation and effector proteins (Choi et al., 2013b).

However, while STAT3-mediated Bcl6 expression leads to the subsequent promotion of not only ICOS and CXCR5, but also PD-1, we question whether the dependance of ICOS and CXCR5 expression on STAT3-mediated Bcl6 activity during the early stages of T_fh cell differentiation is perhaps more important than that of PD-1. During initial CD4⁺ T cell activation and differentiation of the pre-T_fh cell phenotype, signaling through the IL-6 receptor in mice is largely responsible for the initial expression of ICOS, which then subsequently contributes to the expression of CXCR5 (Crotty, 2014b). While IL-6-mediated STAT3 signaling can certainly enhance the expression of PD-1, the fact that PD-1 expression is upregulated on all CD4⁺ T_h cells following activation in order to negatively regulate T cell expansion (Riley, 2009) indicates that its general expression might be more dependent on a pathway that is utilized in the activation of all CD4⁺ T_h effector cell lineages following TCR ligation, such as the Ca²⁺-calcineurin-NFAT pathway.

Indeed, NFAT1,2 co-expression has been implicated as a requirement for optimal PD-1 expression following TCR ligation (Martinez et al., 2016). Additionally, while JAK/STAT signaling in general is ubiquitous to all CD4⁺ T_h cell effector phenotypes, signaling through different cytokine receptors utilizes specific JAKs and STATs (Goswami and Kaplan, 2017), and so the phosphorylation of the specific types of JAK and STAT molecules themselves are largely dependent on the extracellular cytokine microenvironment. This variability is one of the multiple factors that contributes to the differentiation of naïve CD4⁺ T_h cell into different effector phenotypes (Goswami and Kaplan, 2017), and so perhaps it is important that PD-1, which is a classic T cell negative regulator (also known as a “checkpoint”), be upregulated during T cell activation regardless of the extracellular cytokine milieu.

BBR has previously been shown to inhibit JAK/STAT signaling in CD4⁺ T_h helper cells and impact the differentiation of CD4⁺ T_h cells into Th1 and Th17 effector phenotypes. Specifically, BBR has been shown to inhibit the phosphorylation of JAK1, JAK2, STAT1, and STAT4 involved in T_h1 differentiation, as well as STAT3 in T_h17 differentiation (Cui et al., 2009; Liu et al., 2016; Takahara et al., 2019; Yue et al., 2017). Thus, it is perhaps unsurprising that BBR should interfere with STAT3 phosphorylation in T_fh cells as well.

Moreover, BBR has been shown to impact the activity of mechanistic target of rapamycin kinase (mTOR) and AMP-activated protein kinase (AMPK), although there is no specific evidence as of yet to indicate BBR directly binds to these molecules (Imenshahidi and Hosseinzadeh, 2016; Mao et al., 2018; Ming et al., 2014; Tabeshpour et al., 2017; Takahara et al., 2019). The kinase mTOR, which includes distinct complexes mTORC1 and mTORC2, regulates cell growth and metabolism by responding to diverse environmental signals such as available nutrients, ATP levels, various mitogenic signals, and stressors, and ultimately acts to promote glycolytic metabolism (Sabatini, 2017; Yang et al., 2014). The kinase AMPK serves an opposite function and responds to low intracellular ATP by inhibiting mTOR and promoting fatty acid oxidation (Ma et al., 2017). While STATs do not require interaction with mTOR in order to become phosphorylated, there is significant cross-talk between these two pathways, where mTOR signaling can enhance STAT activity and vice versa (Saleiro and Platanias, 2015). BBR has been shown to upregulate the activity of AMPK in IL-17-producing CD4⁺ T cells in a mouse model of inflammatory bowel disease, which was correlated to a decreased frequency of those cells and hypothesized to be a contributing factor to the amelioration of the disease (Takahara et al., 2019).Although it has yet to be explicitly described in regard to T_fh cells and should certainly be expanded upon in other CD4⁺ T_h subsets as well, BBR-mediated activation of AMPK and subsequent inhibition of mTOR could not only impact the phosphorylation of STAT3, but other STATs downstream of different cytokine receptors. Thus, our observation that BBR reduced STAT3 phosphorylation could be due in part to a BBR-mediated AMPK-driven inhibition of mTOR.

Due to the well-documented suppressive effect of BBR on various CD4⁺ T_h cell subsets, especially regarding activation and differentiation, including our own previous study which observed a BBR-mediated decrease in CD4⁺ T_h cell expansion and reduced expression of co-stimulatory molecules CD28 and CD154 (Vita et al., 2021), we also expected to see a downregulation in the activity of key molecules downstream of the TCR and CD28 co-stimulatory receptor. However, in this study, we observed that BBR did not impact the activity of cell signaling molecules p-ZAP-70, p-Lck, p-PLCγ1, and NFATc1 following CD3 and CD28 ligation at 5-, 15- and 30-minutes post-activation. Although we did observe a significant decrease in Ca²⁺ release, the maintained activity of NFATc1 indicates that the magnitude of this decrease is perhaps not great enough to impact the function of NFATc1. This may, however, contribute to the BBR-mediated decrease in IL-21 secretion by pre-T_fh cells that we observed, as Ca²⁺ is required for vesicle exocytosis.

While we are not aware of any previous studies examining the impact of BBR on p-ZAP-70 and p-Lck, there is evidence for the BBR-mediated suppression of PLCγ1 and NFATc1 activity in monocytic cell lines (Han and Kim, 2019; Ye et al., 2017). In those studies, BBR suppressed the activity of these molecules when cells were stimulated with either LPS or RANKL. It is important to note, however, that the cell lines and inflammatory stimuli used in these studies are different than our own. It is known that BBR competes with LPS for TLR4 binding (Chu et al., 2014), and so the lack of PLCγ1 and NFATc1 suppression observed in our own study may be due to the different pathway of cell activation (CD3/CD28). Thus, it is likely that BBR has different mechanisms of action in T cells versus other cell lines, and in response to different stimuli, based on the molecule(s) BBR is directly interacting with.

Conclusion

By examining signaling molecules that associate with the cytoplasmic tails of the TCR (e.g., ZAP-70), of CD4 and CD28 (e.g., Lck), and of the primary signaling cascade downstream of the TCR that is crucial for CD4⁺ T_h cell activation (e.g., PLCγ1- Ca²⁺-calcineurin-NFAT), as well as STAT3 downstream of the IL-6 receptor, we have provided more insight into exactly where in the “chain of command” BBR may be eliciting its effects in CD4⁺ T_h cells. Our observations support a hypothesis that BBR is not directly interfering with activation of the TCR or CD28 receptors themselves, but perhaps interferes with cytokine receptor signaling and/or cytokine-related signaling molecules directly in the cytoplasm. However, the ambiguity of exactly what BBR is interacting with in order to suppress the activity of these molecules highlights the need for further studies which utilize techniques that can directly determine what specific molecules BBR is directly binding to both on and inside of T cells.

Taken together, these results provide additional insight into the mechanism of BBR-mediated T_fh cell suppression and support a hypothesis that BBR treatment can directly suppress CD4⁺ T_h and T_fh cell activity. This could potentially impact the B cell helping capacity of these cells, thus providing less stable activation signals to and germinal center interactions with B cells. In other words, BBR could disrupt T cell-dependent humoral responses by having a direct suppressive effect on T_fh cells. While this may be beneficial for the treatment of antibody-mediated autoimmune diseases, this also raises concern that individuals taking BBR for non-immune related issues could have a diminished germinal center response to primary infections and/or vaccinations.

Abbreviations

AMPK

AMP-activated protein kinase

CXCR5

C-X-C chemokine receptor type 5

DMSO

Dimethyl Sulfoxide

ELISA

enzyme-linked immunoassay

ERK

Extracellular Regulated Protein Kinases

FBS

Fetal Bovine Serum

ICOS

Inducible T-cell co-stimulator

IL

Interleukin

JAK

Janus Kinase

Lck

lymphocyte-specific protein tyrosine kinase

MAPK

Mitogen-Activated Protein Kinase

mTOR

mammalian Target of Rapamycin

NFAT

Nuclear factor of activated T-cells

PBS

Phosphate Buffered Saline

PD-L1

Programmed death-ligand 1

PI3K

Phosphatidylinositol 3-Kinase

PLCγ1

Phospholipase C gamma 1

STAT

Signal Transducer and Activator of Transcription

Th

T helper cell

Tfh

T follicular helper cell

ZAP-70

Zeta-chain-associated protein kinase-70