Wnt Signaling Downregulation Mediates T Cell Apoptosis Following OKT3‐Induced T Cell Activation in Preclinical Models

ABSTRACT

OKT3 is a murine monoclonal antibody (mAb) directed against human CD3 expressed on T cells and is known to induce T cell activation followed by activation‐induced cell death. This study investigated the role of canonical Wnt signaling in OKT3‐mediated T cell activation and apoptosis using preclinical models. In primary human PBMCs, OKT3 treatment downregulated the expression of canonical Wnt signaling‐related genes, whereas the Wnt pathway activator CHIR99021 upregulated their expression and reversed the OKT3‐induced suppression. OKT3 stimulation upregulated T cell activation markers (CD69, CD25), particularly in CD8⁺ T cells, and induced apoptosis as assessed by flow cytometry. Co‐treatment with CHIR99021 attenuated OKT3‐induced CD25 expression and late apoptosis, suggesting that Wnt signaling modulates T cell activation and cell death. In a humanized mouse model, OKT3 administration led to marked depletion of hCD3⁺, hCD4⁺, and hCD8⁺ T cells in the peripheral blood. Transcriptomic profiling using RNA‐seq of liver‐infiltrating human immune cells revealed significant upregulation of genes involved in T cell activation, apoptosis, cytokine signaling, and oxidative stress, accompanied by widespread downregulation of Wnt signaling genes (CTNNB1, LRP6, DVL1, FZD family). Therefore, these findings suggest that OKT3 induces T cell responses characterized by activation and subsequent apoptosis, in association with Wnt signaling. This study provides mechanistic insight into OKT3‐mediated immunomodulation and supports preclinical evaluation of T cell‐targeting monoclonal antibodies.

In OKT3‐treated human PBMCs, Wnt signaling was downregulated, and co‐treatment with the Wnt activator CHIR99021 reduced T cell activation and apoptosis. In humanized mice, transcriptomic profiling revealed upregulation of genes associated with T cell activation, cytokine and ROS production, and apoptosis, with downregulation of canonical Wnt signaling. These findings highlight the immunomodulatory role of Wnt signaling in OKT3‐induced T cell responses and support the use of integrated in vitro and in vivo models for evaluating antibody therapeutics.

1. Introduction

Muromonab (OKT3), a mouse monoclonal antibody (mAb) targeting the human CD3ε chain of the T cell receptor (TCR) complex, was developed to reduce the incidence and severity of graft‐versus‐host disease (GvHD) [1]. OKT3 demonstrated potent immunosuppressive activity through the disappearance of the CD3/TCR complex from the cell surface via receptor shedding or internalization [2]. However, it was withdrawn from clinical use due to unexpected adverse effects, including severe cytokine release syndrome (CRS) triggered by rapid T‐cell activation [2].

T cell activation induced by OKT3 not only promotes proliferation and effector functions but also triggers apoptosis via a process known as activation‐induced cell death (AICD). Upon TCR stimulation, T cells upregulate both Fas (CD95) and its ligand FasL, and the interaction between these molecules initiates apoptotic signaling cascades involving the activation of caspase‐8 and caspase‐3 [2, 3]. Although this Fas/FasL‐mediated pathway is well established, the involvement of alternative signaling mechanisms that contribute to OKT3‐induced T cell apoptosis remains incompletely understood.

The canonical Wnt/β‐catenin pathway plays a pivotal role in cell proliferation, survival, and differentiation in human T cells as well as in tumors [4, 5]. Canonical Wnt signaling is initiated by the binding of Wnt ligands to Frizzled (FZD) receptors and co‐receptors LRP5/6, which inactivates the β‐catenin destruction complex composed of axis inhibition protein (AXIN), adenomatous polyposis coli (APC), glycogen synthase kinase‐3β (GSK‐3β), and casein kinase Ι (CK1). This leads to cytoplasmic accumulation and nuclear translocation of β‐catenin, where it forms a transcriptional complex with T‐cell factor/lymphoid enhancer factor (TCF/LEF) to regulate Wnt target genes such as TCF7 and AXIN2 [4].

CHIR99021, a selective inhibitor of glycogen synthase kinase‐3β (GSK‐3β), is widely used as a canonical Wnt/β‐catenin signaling pathway activator. Inhibition of GSK‐3β by CHIR99021 prevents the phosphorylation and subsequent degradation of β‐catenin, leading to its nuclear accumulation and transcriptional activation of Wnt target genes [6]. GSK‐3β inhibitors, including CHIR99021, have been used to investigate the effect of Wnt signaling on human T cells regarding T cell receptor (TCR) signaling, cellular differentiation, clonal expansion, and activation [7, 8, 9, 10].

Non‐clinical evaluation of mAbs targeting human antigens has traditionally been performed using human‐derived cell lines or peripheral blood mononuclear cells (PBMCs) in vitro [11, 12]. Humanized mouse models have been considered powerful in vivo tools for assessing the immunological effects of mAbs [13]. These mouse models have been established in immunodeficient mice engrafted with human PBMCs or hematopoietic stem cells (HSCs), enabling partial reconstitution of the human immune system [14]. In particular, human PBMC‐engrafted humanized mice, which primarily support T cell reconstitution, have been widely used to evaluate immune responses elicited by mAbs targeting T cell antigens such as CD3 and CD28 [13].

In this study, the immunomodulatory effects of OKT3 were assessed in both in vitro and in vivo models. In the in vitro model, primary human PBMCs were used to examine the immunological effects of OKT3 on T cell activation and apoptosis. In the in vivo model, a humanized mouse model was employed to evaluate transcriptomic changes in human immune cells following OKT3 administration. In this study, we analyzed human immune cells infiltrating the liver, which serves as a primary target organ in OKT3‐treated GvHD and plays a central role in the pharmacokinetics of monoclonal antibodies [15]. This study may provide fundamental insights into OKT3‐induced immune responses in T cells by integrating immunological assessment with transcriptomic profiling in preclinical studies.

2. Materials and Methods

2.1. PBMC Isolation and T Cell Transfection

PBMCs were isolated from fresh human peripheral blood (The Korean Red Cross Daejeon, Sejong and Chungnam Blood Center) using density gradient centrifugation with Ficoll‐based separation medium (Histopaque‐1077, Sigma‐Aldrich, U.S.) in SepMate tubes (STEMCELL Technologies, U.S.). Whole blood was diluted 1:1 with PBS containing 2% FBS and carefully layered over the Histopaque‐1077 in SepMate tubes. Samples were centrifuged at 1200 g for 10 min at room temperature with the brake on. The mononuclear cell layer was washed twice with PBS containing 2% fetal bovine serum (FBS), and resuspended in RPMI‐1640 medium supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. Jurkat T cells (ATCC, USA) were maintained in RPMI‐1640 medium supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin. Cells were transfected with either the empty pcDNA3.1(+) vector or the pcDNA3.1(+)‐CTNNB1(S33Y) (Korea Institute of Science and Technology, Korea) expression construct and subsequently selected using G418 (Geneticin) to establish stable populations. Naïve Jurkat cells and G418‐selected transfected cells were exposed to OKT3 (5 μg/mL) for 72 h, after which cell viability and T‐cell activation levels were quantitatively assessed (Figure S1C,D).

2.2. Reverse Transcription(RT)‐PCR Analysis of Canonical Wnt Signaling Gene Expression

Human PBMCs were seeded at 3 × 10⁶ cells per well in 6‐well plates and cultured in RPMI‐1640 medium supplemented with 10% fetal bovine serum and antibiotics. PBMCs were stimulated with 5 μg/mL OKT3 (Cell Signaling Technology, U.S.) for 24 h, and total RNA was extracted according to the manufacturer's instructions (Qiagen, Japan). cDNA was synthesized from 0.5 μg of total RNA using a reverse transcription kit (BIONEER, Korea). RT‐PCR was performed using SYBR Green qPCR Mix (Abclonal, U.S.) on a QuantStudio 5 (Thermo Fisher Scientific, U.S.). Primers (Table S1) for canonical Wnt signaling [16, 17, 18, 19, 20] and apoptosis pathway components [21, 22] were used. The expression levels were normalized to GAPDH as an internal control. Relative gene expression was calculated using the 2^−(ΔΔCt) method.

2.3. Stimulation of Human PBMCs With OKT3 and CHIR99021

PBMCs were stimulated with 5 μg/mL OKT3 with or without 5 μM CHIR99021 (Sigma, U.S.), a selective GSK‐3β inhibitor that activates canonical Wnt signaling. After 24h stimulation, cells were harvested and stained for T cell activation markers using the following fluorochrome‐conjugated antibodies: Zombie Violet Fixable Viability Kit (BioLegend, U.S.), anti‐human CD8‐APC, anti‐human CD4‐BV510, anti‐human CD69‐FITC, and anti‐human CD25‐APC/Cy7 (BioLegend, U.S.). To assess apoptosis, cells were stained with APC‐conjugated Annexin V and propidium iodide (PI) using the Annexin V Apoptosis Detection Kit (BioLegend, U.S.) according to the manufacturer's instructions. In this study, all samples were analyzed using a Cytoflex S flow cytometer (Beckman Coulter, U.S.), and data were processed using Kaluza v. 2.2 (Beckman Coulter, U.S.).

2.4. Animals

Four‐week‐old female NOD.Cg‐Prkdc^scid Il2rg^tm1Wjl/SzJ (NSG) mice were obtained from The Jackson Laboratory (U.S.) and housed under standard specific pathogen‐free (SPF) conditions. Mice were acclimated for 10 days and randomly assigned to two experimental groups (CTRL and OKT3). Environmental parameters were maintained at a temperature of 22 ± 1 C, relative humidity of 55% ± 10%, and a 12h light/dark cycle. All animals, excluding those found dead prior to scheduled endpoints, were included in the final analysis. All procedures were approved by the Institutional Animal Care and Use Committee (IACUC Approval No. 2112–0079 and 2212–0017) and conducted in accordance with institutional guidelines for animal welfare. The study adhered to the ARRIVE guidelines to ensure transparent and ethical reporting of animal use.

2.5. Establishment of a Human PBMC‐Engrafted Humanized Mouse Model

A total of ten female NSG mice were randomly assigned into two groups (n = 4/group). All mice received intraperitoneal injections of busulfan (25 mg/kg; Otsuka America Pharmaceutical, Japan) once daily for two consecutive days. On the following day, human PBMCs (Lonza, Switzerland) were intravenously injected via the tail vein. Human PBMCs were thawed in a 37°C water bath and washed twice with phosphate‐buffered saline by centrifugation. Each mouse in the designated group received an intravenous injection via the tail vein of 2 × 10⁶ PBMCs suspended in RPMI 1640 medium without FBS or antibiotics. Mouse blood was collected weekly from the retro‐orbital vein to monitor human immune cell reconstitution. Red blood cells were lysed, and total leukocytes were stained with anti‐human CD45‐FITC (BioLegend, U.S.). Humanization was defined when the proportion of human CD45⁺ cells exceeded 25% of the total CD45⁺ leukocytes in mouse peripheral blood.

2.6. OKT3 Administration and Sample Collection in Humanized Mice

Humanized mice received either normal saline (CTRL group) or 10 mg/kg OKT3 via intravenous injection. Blood and liver were collected 4 h post administration. Total leukocytes were stained with fluorochrome‐conjugated antibodies specific for human markers: anti‐human CD45‐FITC, CD3‐PE, CD4‐Pacific Blue 450, and CD8‐APC (BioLegend, U.S.). The frequencies of human total leukocytes (hCD45⁺), cytotoxic T cells (CD3⁺CD8⁺), and helper T cells (CD3⁺CD4⁺) were evaluated within the total leukocyte population. For RNA‐seq analysis, liver tissues were immediately stored in TRIzol reagent (Invitrogen, U.S.) at −80°C until RNA extraction. The residual liver tissue was fixed in 10% neutral‐buffered formalin, embedded in paraffin, sectioned at 5 μm thickness, and stained with hematoxylin and eosin (H&E).

2.7. High‐Throughput Sequencing of Infiltrated Human PBMCs in Mouse Liver

Total RNA was extracted from human PBMC‐engrafted humanized mouse liver tissues treated with normal saline or 10 mg/kg OKT3 using the RNeasy Mini Kit (Qiagen, Hilden, Germany). The RNA‐seq library was generated using the TruSeq Stranded mRNA Library Prep Kit, according to the TruSeq Stranded mRNA Reference Guide # 1000000040498 v00 (Illumina, San Diego, CA, USA). Using the Illumina NovaSeq 6000, 101 bp paired‐end sequencing reads were obtained. The processed reads, using Trimmomatic version 0.38, were aligned to Genome Reference Consortium Human Build 38 (GRCh38) reference genome sequences with hierarchical indexing for spliced alignment of transcripts (HiSAT) version 2.1.0 [23, 24]. Then, the aligned reads were assembled using StringTie version 2.1.3b with reference annotations of Homo sapiens (NCBI_109.20200522) [25] After assembly, the abundance of the transcripts was estimated using read counts and Fragments Per Kilobase of transcript per Million mapped reads (FPKM). After normalization as trimmed mean of M‐values (TMM) normalization, differentially expressed genes (DEGs) were selected from the experimental conditions using a change greater than or less than two‐fold and a Benjamini‐Hochberg p value (BH p value) of less than 0.05 using edgeR [26]. The information of DEGs is described in Table S2. To represent the gene expression profiles, the multidimensional scaling (MDS) method was used. The dissimilarity between the four samples based on the FPKM values of DEGs was used to generate MDS plots, after calculating the Euclidean distance using the cmdscale function in the stats R package.

2.8. Gene Set Enrichment Analysis and Heatmap of Significant Pathways

To analyze the mechanism of OKT3 in the human PBMC‐engrafted humanized mouse model, a gene set enrichment analysis (GSEA) was performed using the Metascape web portal [27]. In the GSEA analysis, significant biological pathways within the Biological Process (BP) in the Gene Ontology (GO) database were selected. In the GSEA performed using the Metascape, a cumulative hypergeometric test was applied to calculate the statistical significance (p value) for each biological pathway, which was then converted to –log (p value). Among the biological pathways ranked by –log (p value), we selected the top pathways for each condition, excluding duplicates. A total of 10 upregulated and 10 downregulated pathways were chosen. The information of pathways is described in Table S3. Some of these pathways were visualized using heatmaps and box plots with the heatmap.2 and ggplot2 R packages, respectively. In the box plot, the mean FPKM values in the Control and High groups were compared using a paired t‐test with the stat_compare_means function from the ggpubr R package. A p value < 0.05 was considered statistically significant.

2.9. Statistical Analysis

All data are presented as mean ± S.E.M. Statistical analyses were performed using GraphPad Prism version 10.4.1 (GraphPad Software, USA). Comparisons between the control and OKT3 groups (n = 3–5) were conducted using Student's t‐test, while one‐way or two‐way ANOVA was used for comparisons among four groups. p‐values less than 0.05 were considered statistically significant. Statistical significance was indicated as follows: p < 0.05 (*), p < 0.01 (**), p < 0.001 (****), and p < 0.0001 (****).

3. Results

3.1. Gene Expression Levels of Canonical Wnt Signaling in PBMCs Following OKT3 Treatment

To investigate the effects of OKT3 on canonical Wnt signaling, gene expression was analyzed in primary human PBMCs following 24h exposure to either normal saline (CTRL) or OKT3 (Figure 1A). Compared to the CTRL group, the expression of the canonical Wnt ligand WNT3 remained unchanged in the OKT3 group. However, transcripts of the Wnt receptor FZD5 and the Wnt destruction complex component APC were significantly downregulated. Notably, the expression of CTNNB1, encoding β‐catenin as a central mediator of canonical Wnt signaling, was also reduced. In line with these findings, the expression of downstream Wnt target genes, TCF7 and AXIN2, was markedly decreased in the OKT3 group, suggesting an overall suppression of canonical Wnt signaling activity. In addition, the expression of the downstream Wnt target genes was markedly decreased in the OKT3 group and restored by CHIR99021, suggesting that CHIR99021 reactivates the attenuated canonical Wnt signaling pathway in OKT3‐treated PBMCs (Figure S1A).

FIGURE 1.

Expression of canonical Wnt signaling components and T cell activation markers in human PBMCs. (A) Relative mRNA expression levels of canonical Wnt signaling‐related genes in PBMCs after 24h treatment with PBS (CTRL) or OKT3. (B–E) Expression of early (CD69⁺) and late (CD25⁺) T cell activation markers in PBMCs following 24h treatment with CTRL, OKT3, CHIR99021, or combination (OKT3 + CHIR99021). (B, C) Frequencies of CD69⁺ (B) and CD25⁺ (C) cells among CD4⁺ helper T cells. (D, E) CD69⁺ (D) and CD25⁺ (E) cell frequency among CD8⁺ cytotoxic T cells. All values represent mean ± S.E.M (n = 3–4). (A) *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. CTRL group. (B‐E) Different letters above the bars indicate statistically significant difference at p < 0.05.

3.2. Effect of CHIR99021 on OKT3‐Induced T Cell Activation

Stimulation with OKT3 significantly increased the frequency of CD69⁺ and CD25⁺cells within CD4⁺ T cells compared to the control (CTRL), whereas CHIR99021 treatment alone showed no difference relative to CTRL (Figure 1B,C). Co‐treatment with OKT3 and CHIR99021 resulted in a reduced frequency of CD69⁺ and CD25⁺ cells within CD4⁺ T cells compared to OKT3 treatment alone. In CD8⁺ T cells, OKT3 induced a marked increase (19.7‐fold) in the population of CD69⁺ cells (Figure 1D), while CHIR99021 treatment had no significant effect compared to CTRL. Co‐treatment with OKT3 and CHIR99021 did not alter the percentage of CD69⁺ cells among CD8⁺ T cells when compared to OKT3 alone. A significant increase (19.3‐fold) in CD25⁺ cell frequency within the CD8⁺ T cell subset was observed in the OKT3 group compared with CTRL (Figure 1E), whereas no difference was observed in CHIR99021. Combined treatment with OKT3 and CHIR99021 resulted in a significant decrease in the percentage of CD25⁺ cells within the CD8⁺ T cells compared to the OKT3 group.

3.3. Apoptosis Induction by OKT3 and Its Modulation by CHIR99021

To evaluate the apoptotic effects of OKT3 and CHIR99021, human PBMCs were treated with 5 μg/mL OKT3 with (OKT3 + CHIR99021 group) or without 5 μM CHIR99021 (OKT3 group) for 24 h. Apoptotic cell death in total PBMCs (Figure 2A–F) and T cell subsets (Figure 2G–J) was assessed by Annexin V and PI staining using flow cytometry.

FIGURE 2.

Early and late apoptosis in human PBMCs and T cell subsets. Apoptosis was assessed in total human PBMCs and T cell subsets following 24h treatment with PBS (CTRL), OKT3, CHIR99021, or combination (OKT3 + CHIR99021). Cells were stained with Annexin V and propidium iodide (PI), and analyzed by flow cytometry. (A, D) Representative dot plots show Annexin V⁺/PI⁻ early apoptotic and Annexin V⁺/PI⁺ late apoptotic/necrotic cells in each treatment group. (E, F) Percentages of early (E) and late (F) apoptotic cells in total PBMCs. (G, H) Percentages of early (G) and late (H) apoptotic cells in CD4⁺ helper T cells. (I, J) Percentages of early (I) and late (J) apoptotic cells in CD8⁺ cytotoxic T cells. All values represent mean ± S.E.M (n = 5). (E–J) Different letters above the bars indicate statistically significant difference at p < 0.05.

Flow cytometric analysis revealed a significant increase in early apoptotic cells (Annexin V⁺/PI⁻) in the OKT3‐treated group compared to the control group (CTRL) (Figure 2A,B,E). Late apoptotic/necrotic cells (Annexin V⁺/PI⁺) were also significantly elevated following OKT3 treatment (Figure 2F). CHIR99021 alone decreased early and late apoptotic cell proportions in PBMCs relative to the CTRL (Figure 2C). Notably, a further reduction in the percentage of early and late apoptotic cells in total PBMCs was observed in the co‐treatment group compared to the OKT3 group.

In CD4⁺ T cells, there was no difference in the proportion of early apoptotic cells between CTRL and OKT3 groups. However, CHIR99021 treatment alone significantly reduced the proportion of early apoptotic cells compared to the OKT3 group (Figure 2G). In contrast, OKT3 treatment induced a marked 3.7‐fold increase in late apoptosis in CD4+ T cells compared to the CTRL group (Figure 2H). Co‐treatment with CHIR99021 significantly decreased late apoptosis compared to OKT3 alone. In CD8⁺ T cells, no significant differences in early apoptosis were observed among the groups (Figure 2I). However, late apoptosis was significantly elevated in the OKT3 group relative to CTRL, and was markedly attenuated by co‐treatment with CHIR99021 (Figure 2J).

3.4. Immune Cell Population in Humanized Mice Given OKT3

No significant changes in body weight or abnormal clinical signs were observed in any group during the 3 weeks following PBMC engraftment (Figure 3A). The proportion of human CD45⁺ cells among total leukocytes in peripheral blood remained minimal during the first week but increased to 21% and 54% at 2 and 3 weeks post‐PBMC engraftment, respectively (Figure 3B).

FIGURE 3.

Establishment and immune profiling of OKT3‐treated humanized mice. (A) Body weight monitoring during the 3 weeks post‐PBMC engraftment. (B) The proportion of human CD45⁺ cells in total lymphocytes in mouse peripheral blood during 3 weeks post‐PBMC engraftment. (C) Flow cytometry gating strategy to identify engrafted human leukocytes (hCD45⁺), total T cells (hCD3⁺), and T cell subsets (hCD4⁺ and hCD8⁺). (D–F) Proportions of hCD45⁺ cells (D), hCD3⁺ T cells among total leukocytes (E), and among hCD45⁺ cells (F) in the CTRL and OKT3 groups. (G, H) Percentages of hCD4⁺ helper T cells among total leukocytes (G), and among hCD45⁺ cells (H). (I, J) Percentages of hCD8⁺ cytotoxic T cells among total leukocytes (I), and among hCD45⁺ cells (J). (K, L) Histopathological evaluation of liver tissues with H&E staining. Immune cell infiltration around the portal vein (PV) was reduced in humanized mice treated with OKT3 (10 mg/kg, L) compared to the normal saline group (K). All values represent mean ± S.E.M (n = 4). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. CTRL group.

Mice at 3 weeks post‐PBMC engraftment were considered humanized and randomly assigned to receive either normal saline (CTRL group) or 10 mg/kg OKT3 (OKT3 group). To evaluate the effect of OKT3 on engrafted human immune cells, peripheral blood was collected and analyzed by flow cytometry. Live, single cells were gated using SSC‐H vs. FSC‐H and FSC‐A vs. FSC‐H. Engrafted human leukocytes (hCD45⁺), T cells (hCD3⁺), and T cell subsets (hCD4⁺ and hCD8⁺) were identified using a human‐specific gating strategy, as illustrated in Figure 3C.

In total leukocytes, the proportion of hCD45⁺ cells was 48.5% in the CTRL group and 10% in the OKT3 group (Figure 3D). The percentage of hCD3⁺ T cells was significantly lower in the OKT3 group than in the CTRL group within total leukocytes and among hCD45⁺ cells (Figure 3E,F). Furthermore, the proportion of hCD4⁺ helper T cells in total leukocytes was 10.7‐fold lower in the OKT3 group compared to CTRL, and was nearly absent within the hCD3⁺ T cell population (Figure 3G,H). Similarly, the percentage of hCD8⁺ cytotoxic T cells was 11.2‐fold lower in total leukocytes and extremely low among hCD3⁺ T cells in the OKT3 group compared to CTRL (Figure 3I,J). Infiltrated immune cells were observed around the portal vein in liver tissues of humanized mice administered with either normal saline (CTRL group) or 10 mg/kg OKT3. Immune cell infiltration in the liver was reduced in the OKT3 group compared to the CTRL.

3.5. OKT3‐Induced Transcriptional Changes in Human Immune Cells in Humanized Mice

RNA‐seq analysis was performed on the infiltrated immune cells in the liver obtained from normal saline‐treated (CTRL) and OKT3‐treated (OKT3) humanized mice to identify transcriptional profiles and signaling pathways underlying this immunomodulation.

To investigate gene expression changes upon OKT3 treatment, MDS using the Euclidean method was performed for the CTRL group (gray) and the OKT3‐treated group (yellow). The two groups were distinctly separated, indicating that OKT3 treatment alters the overall gene expression profile (Figure 4A). A total of 3307 DEGs were identified between the CTRL and OKT3 groups, including 1934 upregulated and 1373 downregulated DEGs (Figure 4B and Table S2). These DEGs were clearly distinct, showed strong statistical significance, and exhibited substantial fold changes. The expression differences were consistent across the range of average expression levels.

FIGURE 4.

Overview of the OKT3‐treated in vivo model. (A) The MDS plot was generated based on the gene expression levels of 3307 DEGs. The samples were grouped into control (gray) and OKT3‐treated (yellow) to visualize the overall differences in expression patterns between the two conditions. (B) Key information about the DEGs, including fold change, p values, FPKM, and the number of DEGs, was visualized using a volcano plot (left), an MA plot (middle), and a bar plot (right). Fold change and FPKM values were log2‐transformed for better visualization. (C) Bar plots were used to present the results of GSEA for both upregulated and downregulated DEGs.

Based on the DEGs between the CTRL and OKT3 groups, GSEA was performed to investigate which cellular pathways were most impacted (Figure 4C and Table S3). The functional classification of upregulated DEGs revealed activation of genes related to T cell activation and apoptosis (164 genes), cytokine production (90 genes), and ROS production (56 genes) (Figure 5).

FIGURE 5.

Upregulated pathways in the OKT3‐treated in vivo model. (A) The heatmap was generated using gene sets related to T cell activation and apoptosis with 164 genes, cytokine production with 90 genes, and ROS production with 56 genes. The samples were divided into two groups, with control samples shown in gray and OKT3‐treated samples shown in yellow, to compare differences in gene expression patterns. (B) Boxplots were created using the same gene sets used in the heatmap to visualize the distribution of gene expression and highlight statistically meaningful differences. *p < 0.05, **p < 0.01, ***p < 0.001 vs. CTRL group.

DEGs involved in T cell differentiation, positive regulation of T cell activation, T cell receptor signaling, apoptotic signaling pathway, and regulation of T cell apoptotic process were upregulated, indicating concurrent T cell activation and apoptosis (Figure 5A top and 5B top). T cell activation‐related genes such as CD3D and CD28, as well as LCK, which encodes a T cell‐specific protein‐tyrosine kinase, were significantly upregulated, indicating robust engagement of the T cell receptor complex and downstream signaling [28]. In parallel, genes involved in both extrinsic and intrinsic apoptotic pathways were upregulated. Upregulation of CASP2, CASP3, and CASP8AP2, together with TNFRSF14, ATF4, CDKN1B, and FOXN2, indicated activation of apoptosis via death receptor signaling and cellular stress–induced DNA damage responses [29, 30]. Together, these findings suggest that OKT3 induces a biphasic response in T cells, whereby initial activation is followed by programmed cell death via coordinated activation of both intrinsic and extrinsic apoptotic pathways.

DEGs related to positive regulation of cytokine production suggested that inflammatory responses and immune activation were enhanced (Figure 5A mid and 5B mid).

IRAK1, JAK2, and STAT1 were notably upregulated in OKT3‐treated humanized mice. IRAK1, one of the interleukin‐1 receptor‐associated kinases, plays an important role in IL1R/TLR‐mediated interferon responses [31]. JAK2 encodes a Janus kinase involved in multiple cytokine receptor signaling pathways, including those mediated by IL6 and interferons [32, 33]. STAT1, a key transcription factor activated by interferons, exhibits enhanced transcriptional activity upon CD3 stimulation, leading to amplified immune responses [34]. These results imply that OKT3 treatment led to T cell activation accompanied by the transcriptional induction of proinflammatory cytokine‐related signaling pathways.

DEGs associated with response to oxidative stress and regulation of oxidative stress (OS)‐induced intrinsic apoptotic signaling pathway were also upregulated, reflecting elevated intracellular oxidative stress levels (Figure 5A bottom and 5B bottom). Among the genes associated with ROS production, TXNIP, GPX1, and ATF4 showed marked upregulation in OKT3‐treated T cells. TXNIP, a negative regulator of thioredoxin, is known to promote oxidative stress when overexpressed [35]. GPX1, encoding glutathione peroxidase 1, is involved in the detoxification of hydrogen peroxide [36], and ATF4 is a transcription factor activated under oxidative stress conditions in mitochondria and endoplasmic reticulum [37]. The coordinated upregulation of these genes suggests that OKT3 treatment induces changes in redox‐regulatory gene expression.

Functional classification of the DEGs identified 44 downregulated genes associated with the canonical Wnt signaling pathway, accompanied by significant dysregulation of related biological processes (Figure 6A). In the OKT3 group, genes associated with the canonical Wnt signaling pathway were broadly downregulated, including CTNNB1, LRP6, DVL1, and several FZD. CTNNB1 encodes β‐catenin, a key effector of the canonical Wnt pathway that activates genes essential for T cell proliferation and survival [38]. Therefore, reduced expression of CTNNB1 may compromise these cellular functions. Likewise, the downregulation of LRP6 and DVL1, which are required for inactivating the β‐catenin destruction complex, implies impaired initiation of canonical Wnt signaling [39].

FIGURE 6.

Downregulated pathways in the OKT3‐treated in vivo model. (A) The heatmap and boxplot were generated using gene sets associated with the Wnt signaling pathway (canonical), which included 44 genes. In the heatmap, control samples were shown in gray color and OKT3‐treated samples in yellow color, allowing visual comparison between the two groups. The boxplot illustrates the distribution of gene expression, and statistical significance is indicated with asterisks. ***p < 0.001 vs. CTRL group. (B) Schematic representation of the canonical Wnt signaling pathway using DEGs. Each gene was colored according to the corresponding log2 fold change, with upregulated elements shaded in red color and downregulated elements in blue color, as indicated by the color scale bar. Solid arrows denote phosphorylation events, T‐bar lines indicate inhibitory interactions, and dotted arrows represent the sequestration or translocation of molecular complexes. This diagram integrated pathway topology with transcriptomic data to visualize expression dynamics under the studied conditions.

3.6. Dysregulation of Canonical Wnt Signaling by OKT3 Mediates T Cell Apoptosis

The canonical Wnt signaling cascade was summarized in the schematic diagram (Figure 6B). Under normal conditions, phosphorylation of LRP6 within the Frizzled and LRP6 co‐receptor complex allowed the docking of AXIN by binding to the phosphorylated LRP residues, and thereby AXIN was pulled away from the destruction complex composed of CK1, APC, AXIN, and GSK3β in the cytoplasm [39]. This dissociation suppressed the destruction complex, enabling β‐catenin to accumulate in the cytoplasm and translocate into the nucleus, where transcription of Wnt target genes was activated. However, treatment with OKT3 caused widespread suppression of receptor components and core destruction complex elements, which led to insufficient β‐catenin stabilization and continuous degradation, resulting in diminished transcriptional activation of Wnt‐responsive genes. Given that β‐catenin normally represses apoptotic genes such as CASP2 and CASP3, its reduction likely caused derepression and consequent upregulation of these genes, which suggested a shift in T cell fate toward terminal differentiation or programmed cell death [40, 41].

4. Discussion

mAbs have developed as immunotherapeutic agents for the treatment of cancer, autoimmune disorders, and transplant rejection [42]. However, safety concerns have been raised, particularly for mAbs targeting T cells. OKT3, like TGN1412, was discontinued due to adverse effects associated with CRS observed in healthy volunteers in clinical trials [43, 44, 45].

These results highlight the importance of preclinical models that closely mimic human immune responses for the development of safe monoclonal antibodies. While in vitro assays using hPBMCs remain a standard tool, humanized mouse models engrafted with hPBMCs have been utilized to evaluate the immune responses of mAbs [13, 43].

In this study, we investigated the modulatory role of canonical Wnt signaling in OKT3‐induced T cell activation and apoptosis using both in vitro and in vivo models. Stimulation of hPBMCs with OKT3 for 24 h resulted in downregulation of key components of the β‐catenin destruction complex, transcription factors, and the Wnt target genes. Consistent with these findings, Willinger et al. reported that cord blood‐derived CD8⁺ T cells stimulated with OKT3 exhibited a decrease in LEF1 transcript levels for up to 48 h, suggesting that TCR engagement suppresses canonical Wnt activity in human T cells [46].

All T cell subsets exhibited an increased proportion of CD69⁺ and CD25⁺ cells following OKT3 stimulation, indicating early and late activation, respectively. To investigate the role of Wnt signaling in T cells stimulated with OKT3, we used CHIR99021 to activate canonical Wnt pathways. GSK‐3β inhibition stabilizes β‐catenin, thereby inducing the transcription of canonical Wnt target genes, such as NLK, JUN, and TCF7, in human T cells [7]. The study demonstrated that Wnt pathway activation via GSK‐3β inhibition attenuates the expression of early and late activation markers (CD69⁺ and CD25⁺) in CD3⁺ T cells. However, the association of canonical Wnt signaling with AICD triggered by CD3 engagement has not been fully determined. In this study, CHIR99021 significantly alters the late activation levels of CD4⁺ or CD8⁺ T cells treated with OKT3, implying that Wnt activation can modulate T cell activation in the presence of TCR stimulation [7].

In this study, apoptosis induced by OKT3 was assessed in human PBMCs and T cell subsets. Consistently, anti‐CD3 monoclonal antibody‐induced cell death has previously been demonstrated in both CD4⁺ and CD8⁺ T cells using annexin V and propidium iodide staining [24]. Inhibition of the Wnt/β‐catenin pathway by ICG‐001 has been reported to reduce proliferation and survival in T‐cell acute lymphoblastic leukemia cells, implicating an association of Wnt signaling with apoptosis [25]. In the present study, the addition of CHIR99021 to OKT3‐stimulated cultures resulted in a marked reduction of late apoptotic cells within both CD4⁺ and CD8⁺ populations. Consistently, an upregulation of FASLG (log2 fold change: 4.3) was observed in OKT3‐treated humanized mice (Table S2), suggesting that Wnt activation may counteract Fas/FasL‐mediated AICD [47]. A previous study demonstrated that treatment with polyglycodendrimers downregulated WNT6 and WNT10A expression and induced both early and late apoptosis in a lymphocytic leukemia cell line [26]. The absence of early apoptotic cells and predominance of late apoptosis may reflect time‐ and dose‐dependent stimulation conditions. However, further studies are needed to investigate the kinetics of apoptosis in T cell subsets following OKT3 exposure Figure S1. Presents genetic evidence for the role of Wnt signaling in regulating T cell activation and viability. Jurkat‐pcDNA3.1(+)‐CTNNB1(S33Y) cells exhibited increased expression of AXIN2, a target gene of canonical Wnt signaling (Figure S2B). Compared with Jurkat‐pcDNA3.1(+) cells, the proportion of CD69⁺ and CD25⁺ cells was significantly reduced in Jurkat‐pcDNA3.1(+)‐CTNNB1(S33Y) cells under OKT3 exposure (Figure S1E,F). Moreover, cell viability was increased in Jurkat‐pcDNA3.1(+)‐CTNNB1(S33Y) cells relative to Jurkat‐pcDNA3.1(+) controls following OKT3 treatment (Figure S1G).

While no abnormal clinical signs or significant changes in body weight were observed during the engraftment period in the mice, flow cytometric analysis revealed a marked reduction in hCD45⁺ and hCD3⁺ cell populations in the peripheral blood following OKT3 administration. Notably, the proportions of CD4⁺ helper T cells and CD8⁺ cytotoxic T cells were reduced by more than ten‐fold in the OKT3 group compared to controls (CTRL), indicating a potent T cell‐depleting effect of OKT3 in the humanized mouse model. These findings are consistent with previous studies showing that anti‐CD3 monoclonal antibodies induce extensive T cell depletion through AICD [3, 43]. The observed decrease in hepatic immune cell infiltration in OKT3‐treated mice likely reflects the systemic T cell depletion in the liver.

Although hepatic T‐cell numbers were reduced following OKT3 treatment, our GSEA revealed several T cell pathways remained upregulated (Figure 4C), suggesting that immune and survival responses of the residual T cells can still be detected in this study. RNA‐seq analysis enabled the identification of novel pathways associated with proinflammatory responses and the characterization of AICD, highlighting the interplay between activation, apoptosis, and canonical Wnt signaling in OKT3‐treated humanized mice. The upregulation of T cell activation markers (CD3D, CD28, LCK) and apoptotic genes (CASP2, CASP3, CASP8AP2) suggests that OKT3 triggers cellular responses leading to T cell depletion, which is consistent with its known immunosuppressive effects observed in clinical settings [48]. These results correlated well with transcriptomic findings showing robust induction of T cell activation and proliferation‐related genes (IL2RA, GZMB, TNFRSF4), alongside the upregulation of pro‐apoptotic pathways such as FAS/FASL and PDCD1 following OKT3 stimulation in PBMC‐derived T cells [49].

The observed upregulation of IRAK1, JAK2, and STAT1 following OKT3 stimulation indicates responses to proinflammatory cytokine signaling, particularly those mediated by IL‐6 and interferons. This is further supported by the RNA‐seq data showing significant induction of IL6 and IFNG transcripts in anti‐CD3‐treated T cells [49], implying one of the genetic signatures for CRS in preclinical settings [13]. The significant upregulation of oxidative stress‐related genes, including TXNIP, GPX1, and ATF4, highlighted the central role of ROS in OKT3‐mediated T cell responses. TXNIP promotes ROS accumulation by inhibiting thioredoxin [35], while GPX1 and ATF4 are activated as part of compensatory antioxidant and stress pathways [36, 37]. However, excessive ROS may override these protective mechanisms, leading to mitochondrial dysfunction and apoptosis. These findings suggest that oxidative stress contributes to OKT3‐induced T cell depletion through ROS‐mediated apoptotic signaling.

A striking finding of our study was the downregulation of canonical Wnt signaling components, including CTNNB1 (β‐catenin), LRP6, DVL1, and FZD, in humanized mice following OKT3 administration. In vitro studies have demonstrated that suppression of the Wnt/β‐catenin pathway is associated with reduced T cell survival and proliferation [38]. Moreover, TCR stimulation downregulated the Wnt target gene expression in isolated human T cells from healthy donors [46]. However, the effect of anti‐human CD3 monoclonal antibody‐induced Wnt pathway modulation has not been evaluated in vivo, and thus, this study aimed to investigate it using a humanized mouse model. Therefore, our study provides novel in vivo evidence that anti‐CD3 antibody therapy may regulate T cell responses through Wnt signaling suppression. These findings suggest a potential association with microRNAs such as miR‐125b in OKT3‐induced downregulation of Wnt signaling, although further investigation is needed to elucidate the underlying signaling cascade [50].

While this study evaluated OKT3‐induced T cell immune responses by integrating in vitro and in vivo analyses, several important limitations should be acknowledged. First, changes in specific T cell subsets, including Th1, Th2, effector, and memory populations, were not examined in PBMCs following OKT3 exposure. Second, the potential effects of the Wnt agonist CHIR99021 in OKT3‐treated humanized mice were not directly evaluated. Finally, subset‐specific transcriptional signatures of T cells in the animal model were not analyzed by RNA‐seq. These limitations highlight the need for further studies.

In conclusion, OKT3 treatment suppressed canonical Wnt signaling in human PBMCs, and co‐treatment with CHIR99021 reduced late apoptosis in CD8⁺ T cells, suggesting that Wnt signaling modulates OKT3‐induced T cell responses. This indicates a functional link between Wnt signaling and T cell activation and apoptosis. In humanized mice, OKT3 administration led to marked T cell depletion in the liver, reflecting clinically relevant effects. Transcriptomic analysis revealed the upregulation of genes related to T cell activation, apoptosis, cytokine signaling, and oxidative stress, alongside downregulation of Wnt signaling components. These findings highlight the interplay between Wnt signaling and OKT3‐induced AICD and support the importance of integrated in vitro and in vivo models to assess the immunomodulatory effects of monoclonal antibodies. Our findings suggest that the canonical Wnt/β‐catenin pathway is associated with OKT3‐induced T‐cell activation with subsequent apoptosis. This pathway may play an important role as a regulatory mechanism that balances therapeutic efficacy, manifested as T‐cell depletion, with treatment‐associated toxicity, notably cytokine release syndrome (CRS). These results provide fundamental data that may inform the rational design of next‐generation anti‐CD3 antibodies and guide strategies aimed at mitigating adverse immune activation in clinical applications.

Author Contributions

Myeongjin Choi and Duhyeon Park designed the study, performed formal analysis, and wrote the manuscript. Myeongjin Choi, Daeui Park, and Kyoung‐Sik Moon performed the review and editing. Da‐Hee Lee, Chang Hoon Choi, Minseong Jo, Seo Yule Jeong, and Seung Yeon Lim performed the animal experiments. Kyoung‐Sik Moon also acquired funding.

Funding

This work was supported by grants from the Korea Institute of Toxicology (KIT), KK‐2610‐01 and Ministry of Food and Drug Safety (MFDS), 25202MFDS002.

Ethics Statement

The studies involving mice were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC Approval No. 2112–0079 and 2212–0017, Korea Institute of Toxicology). This study was carried out in compliance with the ARRIVE guidelines.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1 List of primers used for RT‐PCR.

Table S2 Supplementary Table.

Table S3 Supplementary Table.

Figure S1: Assessment of Wnt‐related gene expression, T cell activation, and viability. (A) Relative mRNA expression levels of Wnt target genes in PBMCs following PBS (CTRL), OKT3, CHIR99021, or the combination of OKT3 and CHIR99021. (B) Relative mRNA expression levels of AXIN2 in parental Jurkat cells and transfected Jurkat cells (Jurkat‐pcDNA3.1(+) or Jurkat‐pcDNA3.1(+)‐CTNNB1(S33Y)) after 3 days of treatment with PBS or OKT3. (C–D) Representative fluorescence microscopy images showing mCherry expression in Jurkat‐pcDNA3.1(+) (C) and Jurkat‐pcDNA3.1(+)‐CTNNB1(S33Y) (D) cells. (E‐G) OKT3‐induced functional responses in parental and transfected Jurkat cells, including early T cell activation (E), and late activation (F), and cell viability (G). For all assays, values obtained from OKT3‐treated cells were normalized to those of PBS‐treated controls within each corresponding group. All values represent mean ± SEM (n = 3). **p < 0.01, ****p < 0.0001 vs. Jurkat‐pcDNA3.1(+).

Data Availability Statement

Stored in repository.