The mouse experiments were approved by the Animal Ethics Committee of the Institute of Zhengzhou University (ZZUIRB 2023–243). Wild-type C57BL/6 mice (8 weeks old, weighing 20 ± 2 g) were purchased from Charles River. The mice were randomly assigned to either the control group or the BI6727 treatment group. BI6727 was administered at a dose of 25 mg/kg via intraperitoneal injection once a week for 2 weeks. The vehicle used was a mixture of PEG 300, Tween 80, and saline. Control group mice received only the vehicle. Mice were euthanized on Day 14 for further analysis.

CD34⁺ cells were isolated from human cord blood using magnetic activated cell sorting (MACS) per the manufacturer’s instructions, with a purity of 95% to 98%. Cells were cultured in Iscove’s Modified Dulbecco’s Medium (IMDM) supplemented with 8% fetal bovine serum (FBS), insulin (10 μg/mL), heparin (3 IU/mL), SCF (10 μg/mL), IL-3 (1 μg/mL), EPO (3 IU/mL) and penicillin-streptomycin (1%). Cultures were maintained at 37°Cin 5% CO₂. The culture process was divided into four phases. From Day 0 to Day 7, cells were cultured in full medium at 1 × 10⁵/mL. IL-3 was removed from day 7 to day 11, SCF and IL-3 were excluded from day 11 to day 15, and EPO was omitted in the final phase. K562 cell line were cultured in 1,640 medium containing 15% FBS.

For mouse BM, cells were flushed with PBS containing 2% FBS and 2 mM EDTA, then filtered through a cell strainer into a centrifuge tube. Lineage-positive cells were depleted, and flow cytometry was conducted to analyze hematopoietic stem and progenitor cell (HSPC) development and erythropoiesis, following established methods. In vitro erythroid cells were collected from culture and analyzed for surface markers including glycophorin A (GPA), IL-3R, CD34, CD36 and CD105, along with Hoechst 33,342 to assess early and terminal erythroid differentiation and enucleation. Antibody protocols followed previous publications (Xu et al., 2022; Zhang et al., 2021). Apoptosis was assessed using an Annexin V kit (eBioscience, #88–8,103-74).

Cell cycle analysis was performed using an EdU staining kit (Invitrogen, C10632) following the manufacturer’s instructions. Briefly, one million cells were cultured with 10 μM EdU for 2 h to incorporate EdU into newly synthesized DNA during active DNA replication. After incubation, the cells were harvested and fixed with 4% paraformaldehyde for 15 min at room temperature. The Click-iT reaction mixture was then prepared and added to the cells, enabling the fluorescent labeling of EdU. After incubation in the dark for 30 min, the cells were washed with PBS, resuspended in buffer, and analyzed by flow cytometry to assess cell cycle progression.

Cells were harvested and placed onto glass slides using a cytocentrifuge. The slides were first stained with May-Grunwald solution (Sigma) for 5 min, then transferred to freshly prepared 10% Giemsa solution for 10 min. After staining, the slides were gently rinsed twice with deionized water, air-dried, and sealed with mounting medium. The stained slides were examined using an optical microscope, and images were captured for further analysis. Cell differentiation stages were determined based on cell size using Image-Pro software for image analysis.

For the colony formation assay, in vitro cultured cells were plated in triplicate at a density of 200 cells per well in 1 mL of MethoCult H4434 (Stemcell Technologies, #04434) medium for BFU-E colony analysis, and in MethoCult H4330 (Stemcell Technologies, #04330) for CFU-E assays. The plates were incubated at 37°C with 5% CO₂ in a humidified incubator. Colony counts for CFU-E and BFU-E were performed on days 7 and 14, respectively. Mouse BM cells were also plated in triplicate at a density of 50,000 cells per well in 1 mL of MethoCult M3334 (Stemcell Technologies, #03334) or M3434 (Stemcell Technologies, #03434) (Han et al., 2024). These plates were incubated under identical conditions, with CFU-E and BFU-E colonies counted on days 3 and 7, respectively.

Mice blood samples were collected from the suborbital vein and then examined by ADVIA^®2120i hematology system (Siemens).

All experiments were conducted in a randomized and blinded manner. Flow cytometry data were analyzed using FlowJo software, while ImageJ was employed for analyzing band signal intensities. Statistical analyses were performed with GraphPad Prism 8.0. Data are presented as mean ± SD, with statistical significance set at *p < 0.05, **p < 0.01, ***p < 0.001.

In order to investigate the role of PLK1 in erythropoiesis, we examined their expression during human erythroid differentiation. Figure 1A shows our RNA-sequencing data of erythroblasts cultured from CD34⁺ cells at distinct stages of development, which reveals abundant expression of PLK1 with distinct expression patterns, with PLK1 upregulated in late-stage erythroblasts. We differentiated cord blood-derived CD34⁺ cells to erythroid cells using a three-step culture erythroid system, as shown in Figure 1B. The expression patterns of PLK1 were further confirmed by quantitative real time polymerase chain reaction (qRT-PCR) analysis (Figure 1C). The constant and abundant expression of PLK1 suggest a potentially important role during human erythropoiesis.

To test our hypothesis, we utilized two inhibitors to induce PLK1 dysfunction during erythropoiesis. Specifically, we treated CD34⁺ cells in an in vitro erythroid induction system with GSK461364 and BI6727, which are selective inhibitors of PLK1, starting on day 2 of culture. By day 6, we observed a significant, dose-dependent reduction in cell proliferation, with GSK461364 demonstrating stronger inhibitory effects than BI6727 at comparable concentrations (Figures 1D, E). Erythroid differentiation, particularly the transition from BFU-E to CFU-E, was assessed using surface markers glycophorin A (GPA), IL3-R, CD34, and CD36. Flow cytometry analysis on day 6 revealed that both inhibitors increased the percentage of BFU-E and reduced the percentage of CFU-E, suggesting that PLK1 inhibition arrested cells at the BFU-E stage and impeded their progression to CFU-E (Figures 1F–I). Furthermore, the inhibitory effect on erythroid progenitor proliferation was corroborated by a increased in the number of BFU-E colonies and a corresponding decrease in CFU-E colonies in PLK1 inhibitor-treated cells compared to controls on days 5 and 6 (Figures 1J, K). These findings indicate that PLK1 inhibition specifically disrupts erythroid progenitor differentiation.

We then examined the effects of PLK1 inhibition on terminal erythroid differentiation, the process during which morphologically recognizable proerythroblasts derived from CFU-E cells undergo four to five mitosis to generate enucleated reticulocytes. The differentiation of CFU-E cells to proerythroblasts is characterized by the surface expression of the erythroid specific marker GPA. As shown in Figure 2A, while more than 70% of control cells were GPA positive on day 7, less than 63% of both PLK1 inhibitors treated cells were GPA positive, implying PLK1 inhibition impairs differentiation of CFU-E to proerythroblasts. Differentiation of proerythroblasts to later stage erythroblasts is characterized by increased surface expression of GPA and decreased surface expression of CD105. As shown in Figure 2B, the upregulation of GPA and downregulation of CD105 were significantly delayed in PLK1 inhibition cells compared to control cells, implying delayed terminal erythroid differentiation.

We next examined whether PLK1 inhibition affects the proliferation of erythroid cells. Treatment with GSK461364 and BI6727 significantly impaired cell growth compared with the control group (Figures 2C, D). To explore the mechanisms responsible for impaired cell growth, we first examined the effects of PLK1 on cell cycle progression. Notably, PLK1 inhibition led to cell cycle arrest at the G2/M phase (Figure 2E). Furthermore, we analyzed the effects of PLK1 inhibition on apoptosis through Annexin V staining on days 7, 9, and 11 in both control and PLK1 inhibitor-treated groups. The results revealed that apoptosis became detectable by day 7, with a progressive increase over time, reaching rates of over 50% by day 11 in groups treated with the highest concentrations of GSK461364 and BI6727 (Figures 2F–I).

We further employed GSK461364 and BI6727 in human erythroleukemia K562 cells, selected for their ability to differentiate along the erythroid lineage upon hemin treatment, to verify the role of PLK1 in erythroid development. Both treatments with GSK461364 and BI6727 resulted in a significant reduction in cell numbers compared to control cells between days 3 and 4 post-hemin induction (Supplementary Figures S1A, B). Representative profiles of 7AAD and Annexin V staining from days 1 to 4 in K562 cells treated with GSK461364 and BI6727 following hemin induction are displayed in Supplementary Figures S1C, E. Quantitative analysis revealed a marked increase in apoptosis in PLK1 inhibitor-treated K562 cells on days 3 and 4 (Supplementary Figures S1D, F). These results indicate that PLK1 inhibition leads to reduced cell proliferation and increased apoptosis during erythroid differentiation in K562 cells, highlighting its critical role in erythropoiesis.

When examining the effects of PLK1 inhibition on morphology of erythroid cells, we surprisingly found that terminally differentiated erythroblasts exhibited abnormal nuclei following both GSK461364 and BI6727 treatment (Figure 3A). Quantitative analysis of data derived from three independent experiments showed that PLK1 deficiency significantly increased the generation of cells with abnormal nuclei, with approximately 22% and 15% on days 13 (Figure 3B). We used flow cytometry to examine the enucleation status of cells stained with Hoechst 33,342 and found that treatment with both GSK461364 and BI6727 resulted in a significant decrease in enucleation on day 15 (Figures 3C, D).

To further explore the role of PLK1 in erythroid development, we administered BI6727 injections to healthy C57BL/6 mice at a dose of 25 mg/kg, given once weekly for two consecutive weeks (Figure 4A). This treatment regimen resulted in no mortality but successfully induced a distinct hematological phenotype. Notably, PLK1 inhibitor-treated mice developed anemia, as evidenced by a rapid decline in peripheral red blood cell (RBC) parameters, hemoglobin (HGB) levels, and hematocrit (HCT) were significantly reduced following PLK1 inhibitor administration, whereas white blood cell (WBC) counts, mean corpuscular volume (MCV), and mean corpuscular hemoglobin (MCH) remained unaffected (Figure 4B). Upon skeletal dissection of the lower limbs, the BM cell suspension of PLK1 inhibitor -treated mice appeared markedly paler compared to control mice, a visual difference corroborated by the reduced BM cell counts (Figure 4C). Additionally, severe splenomegaly was observed in the PLK1 inhibitor-treated group, and flow cytometry analysis revealed a significant increase in erythroblast populations in the spleen (Figures 4D–F). In conclusion, PLK1-inhibitor treated mice exhibit pronounced anemia and splenomegaly, accompanied by extramedullary erythropoiesis.

Hematopoietic stem cells (HSCs) follow a meticulously regulated differentiation pathway, progressing into multipotent progenitors (MPPs). These MPPs then branch into common myeloid progenitors (CMPs), which further differentiate into megakaryocyte-erythrocyte progenitors (MEPs), granulocyte-monocyte progenitors (GMPs), and common lymphoid progenitors (CLPs), demonstrating the complex hierarchical structure of hematopoiesis. To explore whether the severe anemia observed in PLK1 inhibitor-treated mice stemmed from disruptions in HSC homeostasis or the committed differentiation of hematopoietic progenitors, we examined the impact of PLK1 on both HSCs and progenitor cell function. In our flow cytometry gating strategy, the LSK⁺ population was defined as lineage-negative (Lin⁻), Sca-1⁺, and c-Kit⁺, which represents HSCs (LSK⁺CD34⁻CD135⁻CD48⁻CD150⁺) and MPPs (LSK⁺CD34⁺CD135^-). Conversely, the LSK⁻ population, also lineage-negative, lacked Sca-1, defined as Lin-Sca-1⁻c-Kit⁺, which represents CMP (LSK⁻CD34⁺CD16/32^-), MEP (LSK⁻CD34⁻CD16/32^-) and GMP (LSK⁻CD34⁺CD16/32⁺). CLP was defined as LSK^medCD135⁺CD127⁺. Flow cytometry analysis of BM HSCs and progenitor populations revealed an increased percentage and number of HSCs and MPPs (Figures 5A, B), accompanied by a significant reduction in CMPs, GMPs, and MEPs compared to controls (Figures 5C, D). This indicates that PLK1 inhibitor treatment disrupts normal HSPC homeostasis within the BM. We further assessed the abundance of various terminal myeloid cells. Interestingly, following PLK1 inhibitor treatment, the percentage and number of neutrophils, granulocytes, macrophages, megakaryocytes, T cells, and B cells remained stable (Supplementary Figures S1A–H). These results suggest that while PLK1 inhibitor disrupts early hematopoietic progenitor cell homeostasis, it does not significantly affect the terminal differentiation or maintenance of mature myeloid and lymphoid cell populations.

We next investigated the influence of PLK1 on erythroid progenitors. Colony-forming unit assays were performed using total BM cells cultured in semisolid medium. Notably, a significant increase in BFU-E colonies was observed, while CFU-E colonies were markedly reduced following PLK1 inhibition (Figures 6A, B). These results indicate that PLK1 deficiency leads to an increase in BFU-E number, accompanied by a reduction in CFU-E number. Additionally, we found that after PLK1 inhibition, the morphology of CFU-E clones became significantly smaller, while no noticeable difference was observed in BFU-E (Figures 6C, D). To further explore these findings, we recently developed a flow cytometry-based method to isolate mouse BM BFU-E and CFU-E cells, defined as Lin⁻c-Kit⁺CD71⁻ and Lin⁻c-Kit⁺CD71^hi, respectively (Han et al., 2024). Using this method, we sorted BFU-E and CFU-E cells from the BM of both control and PLK1-inhibited mice to assess their colony-forming capacities and found that a reduction in the number of colonies formed by PLK1-inhibited BFU-E and CFU-E cells compared to control (Figures 6E, F). These findings suggest that the colony-forming capacity of PLK1-deficient erythroid progenitors was impaired. In parallel, we analyzed the terminal stages of erythroid differentiation by flow cytometry using the expression of Ter119 and CD44 as markers. The results demonstrated that the percentage and number of Ter119⁺ cells in PLK1-inhibited mice were significantly lower than in control mice, indicating a disruption in erythroid progenitor transformation within the BM. Following PLK1 inhibition, the proportion of erythroblasts was observed to increase, while the number of cells at terminal differentiation stages was decreased (Figures 6G–K).

To investigate whether apoptosis contributes to the observed impairments in erythropoiesis, we conducted flow cytometry analysis using 7AAD and Annexin V staining. Notably, administration of BI6727 increased apoptosis in erythroblasts compared to the control group (Figure 6L). Consistent with the elevated apoptosis, we also detected a reduction in mitochondrial membrane potential (MMP) during erythroid differentiation following PLK1 inhibition (Figure 6M). Moreover, a distinct arrest of the cell cycle at the G2/M phase was observed in BI6727-treated erythroblasts (Figure 6N). These findings suggest that the decrease in erythroid cell numbers is primarily driven by heightened apoptosis and cell cycle arrest resulting from PLK1 inhibition.