This study was conducted from 2019 to 2024 and included adult patients (19 years or older) who underwent allogeneic SCT at Ajou University Medical Center for hematologic malignancies. Among the 154 patients who underwent allogeneic hematopoietic SCT during the study period, 40 patients diagnosed with acute GvHD were included in this study. The 22 patients in the control group who did not develop acute GvHD were selected based on matched clinical characteristics with the acute GvHD group to ensure comparability between the two groups. All Patients received peripheral blood stem cell transplants from matched sibling, matched unrelated, or haploidentical donors. For matched donor transplantation, the GvHD prophylaxis regimen consisted of a calcineurin inhibitor plus methotrexate and anti-thymocyte globulin [14]. For haploidentical donor transplantation, a calcineurin inhibitor and mycophenolate mofetil were administered in combination with post-transplant cyclophosphamide [15]. The severity of acute GvHD was assessed according to the Mount Sinai Acute GvHD International Consortium (MAGIC) criteria [16]. Detailed characteristics of 62 enrolled patients are described in Table 1. The study was approved by the Institutional Review Board of Ajou University Hospital (IRB number. AJOUIRB-OBS-2021-484), and all participants provided written informed consent. Additionally, all experimental procedures involving patient-derived specimens were conducted in accordance with the approved IRB protocol.

For patients diagnosed with acute GvHD, peripheral blood samples were collected at the time of diagnosis corresponding to the respective grades of acute GvHD. The median time of sampling was day 56 post-transplantation (range, 17 to 89 days). For patients in the control group who did not develop acute GvHD, blood samples collected on day 35 post-transplantation were used. Peripheral blood mononuclear cells (PBMCs) were isolated from whole blood by standard Ficoll-Paque (GE Healthcare, Uppsala, Sweden) density gradient centrifugation. Serum samples were also obtained for cytokine analysis.

The following fluorochrome-conjugated monoclonal antibodies used for multicolor flow cytometry: anti-CD3 (UCHT- 1), anti-CD4 (RPA-T4), anti-CD14 (M5E2), anti-CD19 (HIB19), anti-CD25 (M-A251), anti-CD45RA (HI100) (all from BD Biosciences, San Jose, CA, USA); anti-CD39 (eBioA1) (eBioscience, San Diego, CA, USA); anti-CD127 (40131) (R&D Systems, Minneapolis, MN, USA). Cryopreserved or freshly isolated lymphocytes were stained with fluorochrome-conjugated antibodies against surface markers for 30 minutes on ice and then washed. In all samples for flow cytometric analyses, dead cells were excluded by using Live/Dead fixable cell stain kit (Invitrogen, Carlsbad, CA, USA).

For intracellular staining, PE-conjugated anti-Foxp3 monoclonal antibody (PCH101) and anti-Helios (22F6) (all from eBioscience). Surface-stained cells were permeabilized using FoxP3 staining buffer kit (eBioscience) according to the manufacturer’s instruction and further stained for intracellular molecules.

Flow cytometry was performed on an LSR II instrument using FACSDiva software (BD Biosciences), and the data were analyzed using FlowJo software (Treestar, San Carlos, CA, USA).

Cryopreserved PBMCs were thawed, resuspended in RPMI 1640 containing 5% fetal bovine serum (FBS) and 2 mM L-glutamine, and rested overnight at 37°C. PBMCs were stimulated in the presence of anti-CD3 antibody (0.1 μg/mL) and anti-CD28 antibody (1 μg/mL), and brefeldin A (GolgiPlug; BD Biosciences) and monensin (GolgiStop; BD Biosciences) were added 1 hour later. After another 5 hours of incubation, PBMCs were first stained with Live/Dead fixable cell stain kit to exclude dead cells and then stained with fluorochrome-conjugated antibodies against surface markers. Stained cells were permeabilized using FoxP3 staining buffer kit and further stained for intracellular molecules with anti-FoxP3, either anti-IFN-γ (BD Biosciences) or anti-IL-17A (eBioscience) and anti-TNF-α (BD Biosciences).

The suppressive function of Treg cells was assessed using co-culture experiments with CFSE-labeled responder T cells as described elsewhere. Briefly, CFSE-labeled Treg-depleted responder CD4⁺ T cells (Tresp) were cocultured with magnetically isolated Treg cells in a range of ratios as indicated within RPMI 1640 containing 5% FBS and 2 mM L-glutamine. Treg cells are isolated from PBMCs using CD4⁺CD25⁺CD127^lo/- regulatory T cell isolation kit II (Miltenyi Biotec, Bergisch Gladbach, Germany) according to manufacturer’s protocol. At the end of coculture in the presence of soluble anti-CD3 anti-body (0.1 μg/mL; Beckman Coulter Immunotech, Marseille, France) and anti-CD28 antibody (1 μg/mL; BD Biosciences) stimulation for 96 hours, cells were analyzed for proliferation assessed by percentage of CFSE^lo dividing cells. Percentage of suppression was calculated as [1 - (% T cell proliferation with Treg cells / % T cell proliferation without Treg cells)] × 100.

Levels of soluble factors including IL-6 and IL-21 in sera of acute GvHD patients were quantified using CBA (BD Biosciences). Briefly, capture beads and PE-conjugated detection reagent for each soluble factor analyzed were incubated with serum of patients. Concentration in sera was determined by measuring fluorescence intensity of beads capturing soluble factors with LSRII flow cytometer and interpolating the value into standard equation.

Statistical analyses were performed using the Prism software version 5.0 (GraphPad, La Jolla, CA, USA). The Mann–Whitney U test was used for comparisons between two groups, and the Wilcoxon matched-pairs signed-rank test was used for paired values. A p value ≤ 0.05 was considered statistically significant.

We first determined the frequency of Tregs in the peripheral blood of patients with acute GvHD, which was collected at the time of diagnosis of GvHD. Flow cytometric analysis revealed a significant reduction in the percentage of CD25^hiCD127^lo/-FoxP3⁺ CD4⁺ Treg cells in patients with acute GvHD compared to non-acute GvHD control (Fig. 1A). We also investigated the association between quantitative and qualitative changes in Treg cells and clinical factors that could influence acute GvHD. However, no significant associations were identified. The human CD4⁺CD25^hiFoxp3⁺ T cell population is known to be functionally heterogenous and composed of three distinct subpopulations such as CD45RA⁺Foxp3^lo resting Tregs, CD45RA^-Foxp3^hi activated Tregs and cytokine-secreting CD45RA^-Foxp3^lo non-suppressive cells. When we analyzed for the composition of Treg cell subsets in more detail, we observed that the proportion of CD45RA^-FoxP3^hi (actively suppressive) Treg cells was decreased and the proportion of CD45RA^-FoxP3^lo (cytokine-producing non-suppressive) Treg cells was increased in patients with acute GvHD (Fig. 1B). Proportion of CD45R⁺FxoP3⁺ (naive) Treg cells was also decreased in patients with acute GvHD (Fig. 1B).

We assessed the production of inflammatory cytokines from Treg cells following TCR stimulation of peripheral blood lymphocytes with anti-CD3/CD28 antibody and intracellular cytokine staining. As a result, we found that a remarkable proportion of CD4⁺CD25^hiFoxP3⁺ Treg cells of acute GvHD patients produced TNF-α upon TCR stimulation (Fig. 2A). Notably, a significant proportion of TNF-α-producing Treg cells also secreted IL-17A simultaneously (Fig. 2A). These findings align with the increased CD45RA^-FoxP3^lo fraction observed among Treg cell subsets in aGvHD patients (Fig. 1B).

To further explore the clinical implications of this inflammatory Treg cell population, we analyzed their correlation with the progression and severity of acute GvHD. The frequency of TNF-α⁺ cells among circulating Treg cells significantly correlated with the severity of acute GvHD, with higher proportion of TNF-α⁺ Treg cells in patients with the higher grade of acute GvHD (Fig. 2B). This suggests that the pathological conversion of Treg cells to a pro-inflammatory state, characterized by TNF-α production, contributes to immune-mediated tissue injury during acute GvHD.

Next, we evaluated immunophenotypes known to be related to Treg cell suppressive function using multicolor flow cytometry. Our analysis revealed that TNF-α⁺ Treg cells of acute GvHD patients exhibited lower expression of FoxP3, a master transcriptional factor of Treg cell function, compared to the TNF-α^- counterpart (Fig. 3A). Similarly, the proportion of CD39⁺ cells, a marker associated with Treg cell suppressive activity, was reduced in TNF-α⁺ Treg cells relative to TNF-α^- cells (Fig. 3B). Additionally, we observed a decreased frequency of Helios-expressing Treg cells, a key determinant of Treg cell suppressive capacity, in TNF-α⁺ Treg cells from acute GvHD patients (Fig. 3B).

We further performed Treg cell suppression assay to quantify their functional capacity on a per-cell basis, using peripheral blood T cells of recipient of allogeneic SCT both at baseline and post-acute GvHD onset. Notably, the suppressive activity of Treg cells during acute GvHD was significantly diminished compared to pre-SCT controls, as evidenced by reduced inhibition of non-Treg CD4⁺ T cell proliferation in co-culture experiments (Fig. 3C). These findings indicate that the impaired suppressive function of Treg cells, coupled with their inflammatory conversion, may contribute to the immunopathogenesis of acute GvHD.

To investigate a potential mechanism underlying inflammatory changes in Treg cells, we quantified various soluble factors in systemic circulation before and after onset of acute GvHD. Our analysis demonstrated that serum levels of IL-6 (Fig. 4A) and IL-21 (Fig. 4B) were significantly elevated following the onset of acute GvHD compared to baseline levels. These findings suggest that these Th17-polarizing cytokines may play a pivotal role in driving the inflammatory conversion of Treg cells during the course of acute GvHD, thereby contributing to immune dysregulation and tissue injury.