Hematopoietic progenitor kinase 1 inhibitor BGB-15025 induces apoptosis in acute myeloid leukemia cells through the cell cycle pathway and mitogen-activated protein kinase/extracellular signal-regulated kinase pathway signaling axis

Abstract

Acute myeloid leukemia (AML) is a highly heterogeneous hematologic malignancy originating from the malignant clonal proliferation of hematopoietic stem/progenitor cells and is associated with a poor prognosis. Hematopoietic progenitor kinase 1 (HPK1, MAP4K1), a member of the MAP4K family, plays a critical role in immunomodulation and oncogenesis. Previous studies have highlighted its pro-oncogenic function in AML, suggesting its potential as both a prognostic marker and therapeutic target. This study aimed to investigate the anti-AML effects of the novel HPK1 inhibitor BGB-15025. We utilized preclinical models, including AML cell lines, primary patient-derived cells, and MV4-11 xenograft mice. Mechanistic investigations were conducted using RNA sequencing and Western blot analysis. BGB-15025 exerted potent cytotoxicity against AML cells and primary progenitors, inducing apoptosis and G0/G1 cell cycle arrest via downregulation of cyclin D1–cyclin-dependent kinase 4 and upregulation of P21. The inhibitor suppressed mitogen-activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK) signaling through reduced phosphorylation of P38 and ERK. In-vivo studies demonstrated a reduced leukemia burden in xenograft models. This study is the first to elucidate that BGB-15025 triggers AML apoptosis through cell cycle blockade and MAPK pathway inhibition, thereby proposing a novel precision therapeutic strategy with significant clinical translational value.

Introduction

The treatment landscape for acute myeloid leukemia (AML), the most prevalent form of adult leukemia, has experienced a significant transformation over the past decade. According to the WHO Classification System 2022, the molecular characterization of AML has been refined into a complex spectrum comprising 62 genetic abnormalities [1]. Despite these advancements, approximately 494 000 new cases are diagnosed each year globally [2], and the 5-year survival rate remains at only about 30%. This clinical challenge has motivated researchers to persist in exploring targeted therapeutic strategies.

The development of targeted agents has significantly altered the prognosis for certain subtypes of AML. A phase 1/2 clinical trial evaluating the isocitrate dehydrogenase (IDH)1/2 dual inhibitor enasidenib (AG2212-004) demonstrated that the median event-free survival for patients with relapsed/refractory AML harboring an IDH1/2 mutation was extended from 2.6 to 4.9 months compared with conventional chemotherapy [3,4]. In addition, a phase III study investigating the FLT3 receptor tyrosine kinase inhibitor quizartinib in FLT3-ITD mutant AML (QUAZAR-AML-001) confirmed a significant improvement in overall survival, reporting rates of 31.9 versus 15.1 months when compared with chemotherapy, with a P value less than 0.05 [5]. This agent has since been approved by the Food and Drug Administration for use in newly diagnosed patients. Furthermore, a randomized phase II trial assessing the XPO1 inhibitor selinexor in cases of nuclear translocation defect-associated AML revealed a higher remission rate, offering a new therapeutic option for patients resistant to demethylating agents [6].

However, the high heterogeneity of AML continues to pose a significant barrier to precision therapy. Whole-exome sequencing studies have demonstrated that over 50% of AML cases harbor two or more driver mutations [7]. Furthermore, epigenetic regulatory abnormalities – such as mutations in DNMT3A, TET2, and IDH1/2 – exhibit a notable reciprocal signature with other gene mutations [8]. This molecular heterogeneity imposes limitations on single-targeting strategies; for instance, patients with TP53 mutations exhibit an extremely poor response to conventional targeted therapies, resulting in a median survival of only 6.5 months [9,10]. Moreover, the therapeutic challenge is further complicated by the diversity of drug tolerance mechanisms. These include aberrant activation of the BCR-ABL1-like signaling pathway [11] and variations in CYP450 enzyme-mediated drug metabolism [12]. Notably, while the combination of the BCL-2 inhibitor venetoclax with demethylating agents has improved outcomes for some patients, secondary resistance – such as MCL-1 overexpression – still leads to treatment failure in approximately 30–40% of cases [13,14]. This pressing situation underscores the urgent need for developing new target-based therapeutic strategies aimed at overcoming the limitations inherent in existing treatments.

Hematopoietic progenitor kinase 1 (HPK1, MAP4K1), a member of the Ste20 family of serine/threonine kinases, was initially identified as a negative regulator of T-cell receptor signaling, specifically inhibiting T-cell activation through phosphorylation of SLP76 [15,16]. Numerous studies have demonstrated that HPK1 expression levels are negatively correlated with the survival duration of patients suffering from various tumors, including low-grade gliomas in the brain, clear cell carcinoma of the kidney, pancreatic cancer, and invasive breast cancer [17,18]. Notably, patients exhibiting low HPK1 expression tend to have longer survival times. In our previous research, we observed that both mRNA and protein levels of HPK1 in bone marrow samples from patients with AML were significantly elevated compared to those from healthy donors. Furthermore, high HPK1 expression was strongly associated with poor prognosis in these patients. In addition, our study confirmed that HPK1 is markedly overexpressed in homoharringtonine (HHT)-resistant cells; this overexpression enhances AML cell resistance to HHT treatment while knockdown of HPK1 increases sensitivity to therapy. These findings suggest a procancer role for HPK1 in AML and highlight its potential as both a prognostic marker and therapeutic target. Further experiments revealed that the HPK1 inhibitor sunitinib and the Janus kinase inhibitor SP600125 synergistically inhibited AML cell proliferation when combined with HHT treatment – particularly evident in drug-resistant cells [19]. However, it is important to note that sunitinib acts as a multitargeted kinase inhibitor, which may exert effects through non-HPK1-dependent pathways; thus, there is an urgent need for more specific inhibitors to validate these findings. Regarding BGB-15025 (Supplementary Figure S1, Supplemental digital content 1, https://links.lww.com/ACD/A626) discussed in this study, it represents a highly active and selective small molecule inhibitor targeting HPK1 developed independently by BeiGene Ltd (Beijing, China). [20]. This compound has now progressed into phase I clinical trials (NCT04649385), evaluating its efficacy both alone and in combination with the PD-1 mAb tirilizumab for treating patients with advanced solid tumors such as lung and esophageal cancers. However, to date, the effects of BGB-15025 on AML and the associated molecular mechanisms remain unknown.

This study aimed to elucidate the therapeutic potential of BGB-15025 in AML and its underlying mechanism of action. The antileukemic activity, in-vivo safety, and pharmacokinetic profile of BGB-15025 were systematically evaluated using a comprehensive array of experimental models, including in vitro AML cell lines, primary AML patient-derived cells, and xenograft mouse models. In addition, we employed RNA sequencing and Western blot to investigate the potential mechanisms through which BGB-15025 exerts its effects. Our findings indicate that the HPK1 inhibitor BGB-15025 induces a G0/G1 phase block and inhibits the proliferation of AML cells by targeting cell cycle regulation as well as the mitogen-activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK) signaling pathway. These results suggest that BGB-15025 represents a promising drug candidate with translational potential for clinical treatment of AML.

Materials and methods

Cell culture and primary acute myeloid leukemia samples

Human AML cell lines (MV4-11, MOLM-13, THP-1, U937, KG1, and KG1A) were kindly provided by Professor Jie Jin (The First Affiliated Hospital of Zhejiang University). Cells were cultured in complete KG1 medium (FuHeng Biology, Shanghai, China) or RPMI 1640 supplemented with 10% fetal bovine serum (FBS) (Vivacell Biotechnology, Shanghai, China) at 37 °C with 5% CO₂. Cell line authenticity was verified by short tandem repeat profiling, with mycoplasma contamination routinely excluded through standardized testing.

Bone marrow aspirates were obtained from newly diagnosed patients with AML following ethical approval by the institutional review board of Ningbo University People’s Hospital and written informed consent compliant with the Declaration of Helsinki. Primary AML blasts were isolated via Ficoll-Hypaque density gradient centrifugation and maintained in culture medium containing 20% FBS.

Chemical

BGB-15025 is provided by BeiGene Ltd.

Cell viability

AML cell lines (1–2 × 10⁵ cells/ml) and primary AML cells (1 × 10⁶ cells/ml) were seeded in 24-well plates at a final volume of 1 ml per well. Cells were exposed to graded concentrations of BGB-15025 and incubated for 48 or 72 h. Subsequently, 100 µl aliquots from each treatment group were transferred to 96-well plates in triplicate. Cell viability was quantified using the CellTiter 96 AQueous One Solution (Promega, Madison, Wisconsin, USA) according to the manufacturer’s protocol. The half-maximal inhibitory concentration (IC₅₀) was calculated by nonlinear regression analysis of dose–response curves. All experiments were performed in triplicate, with data pooled from three independent biological replicates.

Apoptosis assay

Apoptosis quantification was performed employing a commercial Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide detection kit (AP101; Multisciences Biotech, Hangzhou, China) following manufacturer specifications. AML cell lines and primary AML cells were seeded in 24-well plates and treated with graded concentrations of BGB-15025 for 48 h. Subsequent to PBS washing, cells were dual-stained with FITC-conjugated Annexin V and PI in a standardized binding buffer. Flow cytometric analysis was conducted using a CytoFLEX S instrument (Beckman Coulter, Indianapolis, Indiana, USA), with acquired data processed through CytExpert software (v2.4; Beckman Coulter).

Cell cycle assay

Cell cycle progression was analyzed in AML cell lines seeded in six-well plates and exposed to BGB-15025 for 48 h. Posttreatment cellular suspensions underwent PBS washing followed by fixation in chilled 70% ethanol at 4 °C for 12 h with intermittent vortexing. Fixed cells were resuspended in DNA-specific staining solution (CCS012; Multisciences Biotech, Hangzhou, China) and incubated under light-protected conditions at ambient temperature (25 °C) for 30 min. Cellular DNA content was quantified via flow cytometry (CytoFLEX S; Beckman Coulter), with cell cycle phase distribution analyzed using FlowJo software (v10.8.1; BD Biosciences).

Western blot

Standardized procedures were implemented for sequential cellular disruption, proteomic quantification, electrophoretic resolution, and nitrocellulose membrane immobilization [21]. Membranes were subsequently incubated overnight at 4 °C with specific primary antibodies: β-actin (#4970; 1 : 1000), HPK1 (#46510; 1 : 1000), cyclin D1 (#55506; 1 : 1000), P21 (#2947; 1 : 1000), ERK (#4696; 1 : 1000), phosphorylated ERK (p-ERK, #4370; 1 : 1000), P38 MAPK (#8690; 1 : 1000), and phosphorylated P38 MAPK (p-P38, #9211; 1 : 1000) (all from Cell Signaling Technology, Danvers, Massachusetts, USA). Following three 15-min TBST washes, membranes were incubated for 1 h at room temperature with species-matched horseradish peroxidase-conjugated secondary antibodies (goat anti-rabbit HS101-01 and goat anti-mouse HS201-01; TransGen Biotech, Beijing, China). Protein bands were visualized using the ChemiScope6100 imaging system (Clinx, Shanghai, China) following enhanced chemiluminescence detection. For target proteins with comparable molecular weights, membranes were regenerated using NCM Western blot stripping buffer (#WB6200; NCM Biotech, Suzhou, China) before sequential reprobing. Quantitative densitometric analysis was performed using ImageJ software (NIH, Bethesda, Maryland, USA).

Quantitative real-time PCR

Monocyte-derived total RNA was extracted using the TransZol Up Plus RNA kit (TransGen Biotech) following manufacturer-recommended protocols. First-strand cDNA synthesis was performed with the PrimeScript RT Master Mix (RR037Q; Takara Bio, Shiga, Japan) through reverse transcription of 1 μg total RNA. Quantitative PCR amplification was conducted in triplicate reactions (10 μl total volume containing 1 μl cDNA template) using TB Green Premix Ex Taq II (#639676; Takara Bio) on a 7500 Fast Real-Time PCR System (Applied Biosystems, Singapore, Singapore). Thermocycling parameters comprised: initial denaturation at 95 °C for 5 min; 40 cycles of denaturation (95 °C, 10 s), annealing (60 °C, 10 s), and extension (72 °C, 10 s). Relative quantification of target genes was normalized to GAPDH endogenous control using the comparative 2^−ΔΔCt method.

Primer sequences:

• HPK1-Forward: 5′-GTCGTGGACCCTGACATTTTC-3′

• HPK1-Reverse: 5′-CCTTAAAGACTTCCCCATACGTG-3′

• CCND1-Forward: 5′-TCGGTGTCCTACTTCAAATGTG-3′

• CCND1-Reverse: 5′-TCCTCCTCGCACTTCTGTTC-3′

• CDK4-Forward: 5′-ATGTAGACCAGGACCTAAGGAC-3′

• CDK4-Reverse: 5′-CCACTTGTCACCAGAATGTTC-3′

• P21-Forward: 5′-GGAGACAGACAACTCACTCGTC-3′

• P21-Reverse: 5′-CCTGAACAGAAGAAATCCCTG-3′

• GAPDH-Forward: 5′-TGTAAAACGACGGCCAGT-3′

• GAPDH-Reverse: 5′-CAGGAAACAGCTATGACC-3′

RNA sequencing

Total cellular RNA was isolated with the TransZol Up Plus RNA kit following manufacturer-recommended protocols. RNA quantification and purity measurements were obtained using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, Massachusetts, USA), with subsequent integrity verification performed via capillary electrophoresis on an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, California, USA). Sequencing libraries were prepared with the VAHTS Universal V6 RNA-seq Library Preparation Kit according to standard protocols. High-throughput sequencing was performed by OE Biotech Co., Ltd. (Shanghai, China) using the Illumina NovaSeq 6000 platform (San Diego, California, USA), generating 150 bp paired-end reads. Raw sequencing data underwent quality control using FastQC, followed by filtration of low-quality reads (Q score < 30). Differential gene expression analysis was conducted with DESeq2 (v1.34.0) applying Benjamini–Hochberg correction for multiple testing (false discovery rate–adjusted P < 0.05). Functional annotation of differentially expressed genes was achieved through Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis implemented in clusterProfiler R package (v3.2.0), employing a hypergeometric distribution algorithm with Bonferroni multiplicity adjustment.

Virus production and infection

Lentiviral particles were produced using vectors from GeneCopoeia (Guangzhou, China). The specific vector, psi-LVRU6GP, encoded either an HPK1-targeting short-hairpin RNA (shRNA) (GCCCTTCTCGTAAAGTTGTTC) or a scrambled control (GCTTCGCGCCGTAGTCTTA). HEK 293T cells were employed for virus packaging via transfection with a plasmid mixture containing 5 μg of the lentiviral construct, 3.3 μg of psPAX2, and 1.65 μg of pMD2.G. The medium was refreshed after 12 h, and viral supernatant was collected 60 h later. After filtration (0.45 μm), AML cells were infected with the supernatant and subsequently subjected to selection with 2 μg/ml puromycin to generate stable clones.

Animal studies

All animal procedures were executed in compliance with protocols approved by the Institutional Animal Care and Use Committee at Ningbo University School of Medicine (2024-107). To establish the AML xenograft model, female NOD-Prkdcem26Cd52Il2rgem26Cd22 (NCG) mice (strain T001475; GemPharmatech, Nanjing, China) aged 4–6 weeks underwent xenotransplantation via tail vein injection with 1 × 10⁷ luciferase-transduced MV4-11 cells on day 0. On day 7 postengraftment, mice were stratified into three experimental cohorts (n = 5/group) following bioluminescence verification via intraperitoneal D-luciferin administration (200 mg/kg; PerkinElmer, Waltham, Massachusetts, USA). BGB-15025 was administered intraperitoneally at 25 and 50 mg/kg every 48 h for 14 consecutive days. Daily monitoring of body weight and clinical signs was systematically conducted. Leukemic progression was quantified through serial bioluminescence imaging (IVIS Spectrum; PerkinElmer) at predefined intervals, with signal intensity analyzed using Living Image software (v4.3.1). Humane endpoints, including hindlimb paralysis, triggered immediate euthanasia. Survival duration was tracked until experimental termination.

Statistical analysis

Experimental data were expressed as mean ± SEM derived from three biological replicates. Intergroup comparisons were conducted through Student’s t-test or one-way analysis of variance with Tukey’s post hoc analysis implemented in GraphPad Prism (v9.0; GraphPad Software, San Diego, California, USA). Statistical significance was defined as *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. For survival analysis, Kaplan–Meier curves were generated and statistically compared via Mantel–Cox log-rank test.

Results

BGB-15025 targets and inhibits the expression of hematopoietic progenitor kinase 1 in acute myeloid leukemia cells

To elucidate the expression characteristics of HPK1 in AML, we systematically analyzed its transcriptional and translational profiles across six representative AML cell lines (MV4-11, MOLM-13, U937, THP-1, KG1, and KG1A). Through parallel quantification using qPCR and Western blot, the results revealed ubiquitous HPK1 expression with significant variability among cell types (Fig. 1a and b). Notably, THP-1 and KG1A cells exhibited higher mRNA levels compared with the group mean, a pattern consistently reflected at the protein level as evidenced by densitometric analysis. This intercellular heterogeneity in HPK1 expression provided critical experimental rationale. Given that pharmacodynamic responses are typically amplified in high-expression models, THP-1 and KG1A were strategically selected as optimal systems for subsequent inhibitor validation. To mechanistically confirm BGB-15025 as a bona fide HPK1 inhibitor, dose–response studies were conducted in the preselected high-expression models. THP-1 and KG1A cells were treated with increasing concentrations of BGB-15025. qPCR and Western blot analyses showed a dose-dependent decrease in both HPK1 mRNA transcripts and protein expression, demonstrating consistent suppression across transcriptional and translational levels (Fig. 1c–f). The strong concentration dependency confirmed target engagement fidelity, satisfying key pharmacologic criteria for HPK1-targeted inhibition.

Fig. 1.

Expression of HPK1 in various AML cell lines and alterations in levels following BGB-15025 treatment. (a) qRT-PCR was employed to assess the expression of HPK1 across six AML cell lines. (b) Following lysis of whole cell lysates, changes in HPK1 expression among six AML cell lines were evaluated using Western blot, with β-actin serving as an internal reference protein. (c and e) qRT-PCR analysis of HPK1 expression in THP-1 and KG1A cells after 48 h of drug treatment, with results presented as mean percentage ± SEM. Statistical significance was determined relative to the control group, where *P < 0.05, ***P < 0.001, and ****P < 0.0001. (d and f) Western blot analysis for HPK1 expression in THP-1 and KG1A cells posttreatment for 48 h. This experiment was conducted independently three times, and the most representative blot is depicted in the figure. (g) HPK1 expression was knocked down in THP-1 cells using shRNA (shHPK1), and protein levels were assessed by Western blot. (h) The proliferation of HPK1-knockdown AML cells (THP-1) was measured by MTS assay over 0–96 h. AML, acute myeloid leukemia; HPK1, hematopoietic progenitor kinase 1; MTS, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt; qRT-PCR, quantitative real-time PCR; shRNA, short-hairpin RNA.

To genetically validate HPK1 as a direct target, we performed knockdown experiments. Lentivirus-mediated delivery of shRNA against HPK1 (shHPK1) significantly reduced HPK1 protein expression in THP-1 cells compared with the nontargeting control (scrambled) (Fig. 1g). Consistently, this HPK1 knockdown markedly suppressed the proliferation of AML cells (Fig. 1h), indicating a critical role for HPK1 in AML cell survival. This successful model establishment enabled subsequent functional investigations.

BGB-15025 effectively induces apoptosis in acute myeloid leukemia cells and significantly inhibits their proliferation

Building upon the established HPK1 inhibitory activity, we systematically mapped the antileukemic efficacy of BGB-15025 using the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt (MTS) colorimetric assay. Initial MTS screening across six genetically diverse AML cell lines [MV4-11 (FLT3-ITD), MOLM-13 (MLL-rearranged), U937 (TP53mut), THP-1 (DNMT3A/ASXL1mut), KG1 (BCR-ABL1neg), KG1A (BCR-ABL1pos)] revealed mutation-dependent sensitivity gradients. The results indicated that AML cells with different mutational backgrounds exhibited distinct sensitivities to the drug (Fig. 2a). Compared with the control group, the cell inhibition rate in the drug-treated group increased significantly; moreover, the inhibitory effect of BGB-15025 on AML cells was found to be concentration-dependent. This cytogenetic stratification prompted deeper temporal resolution. Subsequently, we evaluated the inhibitory effects on proliferation by selecting various concentrations of BGB-15025 across six AML cell lines treated for durations of 0, 24, 48, 72, and 96 h (Fig. 2b). Experimental findings demonstrated that among these six AML cell lines, the growth trend of AML cells in the treatment groups was significantly lower than that observed in control groups from 0 to 96 h. These results suggest that BGB-15025 has potential cytotoxic effects against AML cells and effectively inhibits their proliferation. The synchronized concentration-temporal dependency establishes BGB-15025 as a promising candidate for pulsatile dosing regimens in heterogeneous AML populations.

Fig. 2.

BGB-15025 demonstrates inhibitory effects on the proliferation of AML cell lines. Six AML cell lines were treated with varying concentrations of BGB-15025 for 48 and 72 h. Cell viability was assessed using the MTS assay. The data presented above represent the results from three independent replicate experiments, expressed as mean ± SEM. (b) Cell viability was assessed following treatment of AML cell lines (MV4-11, MOLM-13, U937, THP-1, KG1, KG1A) with varying concentrations of BGB-15025 over periods of 0, 24, 48, 72, and 96 h. The inhibition rate was calculated and visualized using GraphPad Prism software. (c) Western blot analysis confirmed successful HPK1 knockdown in THP-1 cells. (d) The dose-dependent cytotoxic effects of BGB-15025 were compared between scrambled and shHPK1 THP-1 cells. AML, acute myeloid leukemia; HPK1, hematopoietic progenitor kinase 1; MTS, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt.

Having established that HPK1 knockdown phenocopies the molecular effects of BGB-15025, we sought to determine whether HPK1 is essential for the drug’s cytotoxic activity. We treated both control (scrambled) and HPK1-knockdown (shHPK1) THP-1 cells with increasing concentrations of BGB-15025 and assessed cell viability. Strikingly, the cytotoxic effect of BGB-15025 was significantly blunted in shHPK1 cells (Fig. 2c and d). This attenuation of drug efficacy upon HPK1 depletion suggests that the inability of BGB-15025 to engage its target underlies the reduced effect, providing strong evidence that HPK1 is the functional target of BGB-15025.

BGB-15025 targeted primary acute myeloid leukemia blasts ex vivo

Having established the antileukemic activity in cell line models, we transitioned to clinically relevant systems using fresh bone marrow aspirates from 10 newly diagnosed patients with AML (Fig. 3b). This cohort intentionally included three ultra-high-risk cases (patients #7, #8, and #9) who succumbed within 72 h postadmission due to hyperleukocytosis (WBC > 100 × 10⁹/L) and disseminated intravascular coagulation. After 48 h of incubation, BGB-15025 was found to effectively reduce the number of viable cells in all ten primary AML samples in a concentration-dependent manner, though the samples’ responses varied slightly, with IC₅₀ values ranging from 461 to 2864 nM (Fig. 3a). These findings substantiate the translational advantages of BGB-15025 in addressing clinical emergencies and its potential to enhance efficacy in rapidly progressive subtypes of AML.

Fig. 3.

BGB-15025 exhibits an inhibitory effect on primary AML cells. (a) Cell viability was assessed using the MTS assay following treatment of 10 primary AML cell samples with varying concentrations of BGB-15025 for durations of 48 and 72 h. The inhibition rate was calculated and subsequently plotted utilizing GraphPad Prism software. (b) Calculated IC₅₀ for primary AML samples. AML, acute myeloid leukemia; IC₅₀, the concentration that resulted in 50% inhibition in cell viability; MTS, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt.

BGB-15025 induces apoptosis and cell cycle arrest in acute myeloid leukemia cells

Building upon the observed growth inhibitory effects of BGB-15025, we first delineated its proapoptotic potential in AML cells. Using flow cytometry with Annexin V/propidium iodide double staining, THP-1 and KG1A cell lines were exposed to gradient concentrations of BGB-15025 for 48 h (Fig. 4a). Quantitative analysis revealed a striking concentration–response relationship: in THP-1 cells, apoptotic rates escalated from 5% at 2500 nM to 50.63% at 15 000 nM, whereas KG1A cells exhibited an increase from 11.38% (250 nM) to 40.33% (1000 nM). These data mechanistically demonstrate that BGB-15025 potently induces apoptosis in AML cells. Given the established crosstalk between apoptosis and cell cycle checkpoints, we next interrogated whether cell cycle modulation contributes to the anti-AML activity (Fig. 4b). Flow cytometric analysis demonstrated that BGB-15025 treatment caused a significant redistribution of cell cycle populations. Specifically, the proportion of G0/G1-phase cells increased in both THP-1 and KG1A, accompanied by a concomitant decrease in S-phase populations. This synchronized G0/G1 arrest implies that BGB-15025 may target key regulators of cell cycle entry, such as cyclin-dependent kinases (CDKs).

Fig. 4.

BGB-15025 induced apoptosis in AML cells and inhibited their progression through the G0/G1 phase. (a) THP-1 and KG1A cells were treated with BGB-15025 for 48 h, and the apoptosis rate was assessed using Annexin V/PI double staining followed by flow cytometry. The gradients of drug concentration exhibited a significant positive correlation with the levels of apoptosis. (b) To further investigate the potential mechanisms underlying apoptosis induction, we examined cell cycle distribution under identical treatment conditions. Treatment with BGB-15025 significantly increased the proportion of cells in the G0/G1 phase, and this effect was dose-dependent. Data were obtained from three independent replicate experiments (mean ± SEM). The differences between the experimental and control groups were statistically significant: **P < 0.01, ***P < 0.001, ****P < 0.0001. AML, acute myeloid leukemia; PI, propidium iodide.

BGB-15025 exerted anti-acute myeloid leukemia effects via modulation of cell cycle progression and MAPK/ERK signaling

To elucidate the molecular mechanism of BGB-15025-induced apoptosis in AML cells, we first performed transcriptomic profiling through RNA sequencing. Notably, KEGG pathway enrichment analysis pinpointed the cell cycle as the most significantly altered pathway (Fig. 5a). To quantitatively assess this observation, Gene Set Enrichment Analysis revealed a pronounced downregulation of cell cycle-related genes (normalized enrichment score = −4.55), suggesting global suppression of proliferative signaling (Fig. 5b). Based on the above computational predictions and previous experimental results, BGB-15025 induces dose-dependent G0/G1 phase arrest in AML cells, which functionally confirms the RNA sequencing findings. The common proteins regulating this cycle were identified as cyclin D1, CDK4, and P21. qPCR and Western blot analyses were conducted to detect changes in gene and protein levels after drug administration (Fig. 5c and d). The results indicated that the transcript and protein expression levels of CCND1 and CDK4 gradually decreased with increasing drug concentration, while those of P21 gradually increased, suggesting that BGB-15025 affects the AML cell cycle through the cyclin D1/CDK4/P21 signaling axis.

Fig. 5.

BGB-15025 inhibits the cell cycle and the MAPK/ERK signaling pathway in AML cells. (a) KEGG analysis revealed that differentially expressed genes were significantly enriched in relevant signaling pathways. (b) GSEA of differentially expressed genes in the treated group, compared with the control group, indicated a predominant enrichment in cell cycle-related pathways. (c) Two AML cell lines (KG1A and THP-1) were exposed to different concentrations of BGB-15025, and the expression levels of CCND1, CDK4, and P21 genes were quantified using qRT-PCR. (d) Various concentrations of BGB-15025 were administered to two AML cell lines (KG1A and THP-1), followed by the detection of cyclin D1, CDK4, and P21 protein expressions via Western blot analysis. (f) Different concentrations of BGB-15025 were administered to two AML cell lines, KG1A and THP-1. The expression levels of ERK, p-ERK, P38, and p-P38 proteins were assessed using Western blot analysis. Data presented are derived from at least three independent experiments. Statistical significance was determined as follows: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 when compared with the control group. (e and g) The effect of HPK1 knockdown on the expression of the above-mentioned proteins was assessed in AML (THP-1) cells. AML, acute myeloid leukemia; GSEA, Gene Set Enrichment Analysis; HPK1, hematopoietic progenitor kinase 1; KEGG, Kyoto Encyclopedia of Genes and Genomes; MAPK/ERK, mitogen-activated protein kinase/extracellular signal-regulated kinase; qRT-PCR, quantitative real-time PCR.

HPK1, a serine/threonine kinase within the MAP4K family, functions as a critical upstream regulator of MAPK signaling – a cascade central to cell survival and proliferation. Given HPK1’s established role in oncogenic signaling and our prior observation of its overexpression in AML, we hypothesized that BGB-15025-mediated HPK1 inhibition might impair MAPK pathway activation. To test this hypothesis, we assessed the phosphorylation status of key MAPK effectors in BGB-15025-treated AML cells. Western blot analysis revealed dose-dependent reductions in total MAPK levels across the KG1A and THP-1 cell lines (Fig. 5f). More critically, p-ERK and P38 (p-P38) exhibited marked attenuation, indicating suppression of pathway activity beyond mere protein abundance. To contextualize these biochemical changes, we correlated MAPK inhibition with functional cytotoxicity. Collectively, these data establish that BGB-15025 exerts its cytotoxic activity in AML not only through cell cycle disruption but also via dual modulation of MAPK signaling – directly impairing kinase activation (phosphorylation) and indirectly reducing effector availability (total protein levels).

To determine if these molecular alterations were directly caused by HPK1 suppression, we analyzed the same pathways in HPK1-knockdown cells. Consistent with the pharmacological inhibition, genetic ablation of HPK1 recapitulated the key molecular changes: it downregulated cyclin D1 and CDK4, upregulated P21 (Fig. 5e), and markedly reduced the phosphorylation levels of ERK and P38 (Fig. 5g).

BGB-15025 prolongs animal survival in a murine acute myeloid leukemia model

To evaluate activity and safety in vivo, an AML xenograft model was established by inoculating NCG mice (denoted as day 0) with luciferase-labeled MV4-11 cells. On day 7, mice were randomly assigned to one of three groups and treated with a placebo or two different concentrations of BGB-15025 (25 and 50 mg/kg, qod, intraperitoneally). Fluorescence imaging revealed that mice receiving BGB-15025 exhibited significant attenuation of leukemia over time compared with mice in the placebo group (Fig. 6a and b). Notably, BGB-15025 effectively prolonged the survival time of the animals compared with the control group (Fig. 6d). Furthermore, no significant weight loss or other adverse events were observed throughout the course of treatment, suggesting that the BGB-15025 regimen was well-tolerated (Fig. 6c).

Fig. 6.

BGB-15025 demonstrates anti-AML effects in AML xenograft mouse models. (a) In-vivo imaging of AML xenograft mice following treatment with varying concentrations of BGB-15025. (b) Fluorescence intensity measurements from AML xenograft mice. (c) Body weight profiles of AML xenograft mice. (d) Survival curves for AML xenograft mice. *P < 0.05, **P < 0.01 compared with the control group. AML, acute myeloid leukemia.

Discussion

AML, characterized as a highly heterogeneous hematologic malignancy, presents a poor prognosis for the majority of patients [22]. In recent years, targeted therapeutic strategies aimed at kinase signaling pathways – such as FLT3 inhibitors and IDH inhibitors – have shown some progress. However, the complexity of molecular typing and the activation of compensatory pathways have significantly limited these advancements, necessitating urgent improvements in overall survival rates [23,24]. The present study reveals for the first time that the HPK1 inhibitor BGB-15025 induces apoptosis in AML cells by modulating cell cycle progression and regulating the MAPK/ERK signaling axis. This finding provides a new theoretical foundation for addressing existing therapeutic bottlenecks in AML treatment.

In recent years, HPK1, a member of the MAP4K family of mammalian Ste20-associated protein kinases, has emerged as a novel target for AML therapy because of its potential pro-oncogenic properties. Consequently, we propose that targeting HPK1 represents a promising therapeutic strategy. BGB-15025, identified as a targeted inhibitor of HPK1, has been validated in our preliminary results; notably, the expression levels of HPK1 exhibited a decreasing trend with increasing drug concentrations. Building on this foundation, further experiments demonstrated that the HPK1 inhibitor BGB-15025 exerted significant cytotoxic effects on AML cell lines and primary cells. This finding holds important implications for clinical practice: we observed that three AML patients (AML #7, #8, and #9), who succumbed within 72 h of admission, had progenitor cells displaying high sensitivity to BGB-15025 (IC₅₀ = 163.2–759.7 nM). This observation suggests that this drug may serve as a potentially salvageable therapeutic option for this cohort of very late-stage patients. Moreover, we further validated the antitumor activity and safety profile of BGB-15025 using an xenograft mouse model. When employing alternate-day dosing regimens at doses of 25 and 50 mg/kg, the drug significantly reduced tumor burden without causing notable weight loss in subjects. The observed dose-dependent efficacy profile of BGB-15025 suggests therapeutic applicability as a bridging pharmacological agent, particularly in acute clinical presentations requiring expedited management of neoplastic proliferation.

Further studies have demonstrated that BGB-15025 induces a cell cycle block in AML cells. The precise regulation of the cell cycle is essential for maintaining normal cellular growth, proliferation, and differentiation [25]. However, tumor cells frequently disrupt this regulatory balance, resulting in uncontrolled proliferation; abnormal cell cycle regulation is one of the key factors contributing to tumorigenesis and development [26]. Cell cycle progression is primarily governed by a complex interplay between cyclins and CDKs [27]. During the G1 phase, cyclin D binds to CDK4/6 to form a complex that phosphorylates retinoblastoma protein [28], leading to the release of the transcription factor E2F from retinoblastoma protein’s binding site. This process activates the transcription of various genes associated with DNA synthesis, thereby propelling the cell into S phase. In our study, we observed that BGB-15025 effectively blocked AML cells in the G0/G1 phase in a dose-dependent manner. We hypothesize that this blockade may occur through inhibition of cyclin D-CDK4/6 activity or downregulation of cyclin D expression levels. Such actions would prevent normal phosphorylation of retinoblastoma protein and subsequent release of E2F. Multiple checkpoints exist within the cell cycle to ensure accuracy and integrity during progression. When cells encounter external stimuli or DNA damage, these checkpoints are activated to halt further advancement through the cell cycle. BGB-15025 may activate checkpoint mechanisms during G1 phase, potentially involving pathways such as P53-P21 signaling [29,30]. When applied to AML tumor cells, BGB-15025 could induce DNA damage or other stress signals that activate P53 proteins – key transcription factors involved in regulating cell cycle dynamics. The P53 protein upregulates P21 expression; P21 subsequently binds to and inhibits CDK–cyclin complexes’ activity, thereby reinforcing blockage at the G0/G1 phase [31,32].

At the level of hematopoietic system regulation, abnormal activation of the MAPK/ERK pathway – a critical hub that connects cell cycle progression and differentiation decision-making – can lead to dysregulated proliferation and differentiation disorders in hematopoietic cells [33,34]. This dysregulation may result in hematological conditions such as leukemia. Therapeutic strategies targeting the MAPK/ERK pathway have opened new avenues for treating AML, with advancements made in developing small-molecule inhibitors and exploring combination therapies [35]. However, challenges remain, including drug resistance and adverse drug reactions [36]. Research has demonstrated that activation of the MAPK/ERK pathway promotes AML cell proliferation through various mechanisms. For instance, ERK phosphorylates and activates transcription factors such as c-Myc and E2F, leading to upregulation of genes like cyclin D1 [37,38]. This process facilitates cellular transition from the G1 phase to the S phase, thereby accelerating cell cycle progression and promoting leukemic cell proliferation [39]. Our studies indicate that BGB-15025 inhibits phosphorylation levels of key proteins within the MAPK/ERK pathway. In AML cell lines treated with BGB-15025, Western blot assays revealed a significant reduction in phosphorylation levels of P38 and ERK. Given that ERK phosphorylation positively regulates cyclin D1 – and previous findings confirm that cyclin D1 can be inhibited by BGB-15025 – we hypothesize that this drug may impede the cell cycle by inhibiting ERK phosphorylation. Importantly, these findings suggest that BGB-15025 possesses multitarget regulatory advantages: it acts directly on components of the cell cycle machinery while also indirectly diminishing their driving effects through inhibition of the MAPK pathway. This dual action results in a synergistic antileukemic effect.

Despite the valuable findings of this study, several limitations warrant attention. First, there are inherent differences between in-vitro cellular experiments and animal models compared with the physiopathological environment encountered in human patients [40]. The complexity of the human immune system, the heterogeneity of the tumor microenvironment, and the metabolic processes of drugs in vivo were not fully replicated in our experimental models [41]. Consequently, the actual efficacy and safety profile of BGB-15025 in humans requires further validation through clinical trials. Second, this study primarily investigates the mechanism of action of BGB-15025 with a focus on cell cycle regulation and MAPK-related pathways. However, whether BGB-15025 influences other critical biological processes – such as cell migration, invasion, and modulation of the tumor microenvironment – remains inadequately explored.

In summary, through a systematic analysis of molecular mechanisms and validation in preclinical models, this study has confirmed that the HPK1 inhibitor BGB-15025 demonstrates significant advantages in the treatment of AML by targeting cell cycle checkpoints (specifically G0/G1 blockade) and MAPK/ERK signaling pathways. Its distinct dynamic regulatory properties, coupled with a favorable safety profile, provide a robust foundation for subsequent clinical translation. Future research will concentrate on developing biomarker-based precision therapeutic strategies as well as innovative combination regimens.

Acknowledgements

We would like to express our sincere gratitude to Prof. Jie Jin from the First Affiliated Hospital of Zhejiang University for generously providing us with acute myeloid leukemia cell lines for our experiments.

This study was supported by the Natural Science Foundation of Zhejiang Grants, China (LQ23H080001), the Natural Science Foundation of Ningbo (2022J028); and the Zhejiang Province Traditional Chinese Medicine Science and Technology Project (2025ZL124).

Writing – original draft, methodology, investigation, formal analysis: S.Y. Investigation, funding acquisition: F.L. Investigation: H.Z., D.C, R.P, S.L, and P.Y. Investigation, data curation: X.J., Y.Z. Writing – supervision, project administration, funding acquisition: Y.L.

Data will be made available on request.

Conflicts of interest

There are no conflicts of interest.

Supplementary Material

acd-37-329-s001.docx ^{(18.6KB, docx)}