Minoxidil ameliorates myelodysplastic syndrome by targeting Wnt4 while sparing normal hematopoiesis

Abstract

Background

Hematopoietic stem cells (HSCs) maintain blood production via tightly regulated differentiation. Disruptions at this level can lead to myelodysplastic syndromes (MDS), characterized by ineffective hematopoiesis and marrow failure. Despite its clinical use, the antihypertensive and hair-growth agent minoxidil has been linked to hematologic side effects, yet its mechanism remains unknown.

Methods

Compounds affecting hematopoiesis were identified by a small-molecule screen using Tg(mpl:eGFP) zebrafish embryos and validated by immunofluorescent antibody staining. The impact of minoxidil on different blood cell types was determined by whole-mount in situ hybridization. HSPCs proliferation and apoptosis were assessed in Tg(cd41:eGFP) embryos using bromodeoxyuridine (BrdU) incorporation and TUNEL assays. Gene expression changes were profiled by RNA sequencing. Functional relevance of wnt4 was assessed through overexpression and F0 knockout experiments. At suitable concentrations that avoided notable developmental delay, minoxidil demonstrated its dual effects—therapeutic efficacy and hematopoietic toxicity—across larval and adult MDS-like zebrafish and in wild-type mice, and further exerted antiproliferative effects in human malignant hematopoietic cells in vitro.

Results

Minoxidil significantly reduced hematopoietic stem and progenitor cell numbers in zebrafish embryos, leading to broad suppression of multiple blood lineages. Transcriptomic profiling revealed that minoxidil downregulated wnt4 expression. Functional validation demonstrated that wnt4 directly modulates HSPC abundance: knockout of wnt4 recapitulated the hematopoietic suppression seen with minoxidil, while overexpression restored HSPC levels. In c-myb^hyper MDS-like zebrafish, minoxidil treatment alleviated myeloid hyperplasia at appropriate doses without impairing lymphoid or erythroid lineages. Consistently, minoxidil showed inhibitory effects on human malignant hematopoietic cells in vitro, supporting its conserved suppressive effect on myeloid and progenitor expansion. In both adult zebrafish and wild-type mice, low or intermittent minoxidil dosing preserved hematopoietic integrity, whereas continuous high-dose treatment resulted in multilineage cytopenia.

Conclusion

Our findings demonstrate that minoxidil modulates hematopoiesis through wnt4 downregulation, resulting in both HSPC suppression and therapeutic alleviation of MDS-like phenotypes. At optimized dosing, minoxidil exhibits hematologic safety in vivo. This study identifies wnt4 as a regulatory node linking pharmacologic intervention to HSPC homeostasis and highlights its therapeutic potential in MDS.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12964-025-02615-z.

Background

Hematopoietic stem cells (HSCs) give rise to all blood cell types through hematopoiesis [1]. Cytogenetic alterations at the HSCs level lead to the development of myelodysplastic syndromes (MDS) [2], a group of clonal hematopoietic malignancies characterized by ineffective hematopoiesis and frequent progression to acute myeloid leukemia (AML) [3]. The symptoms of MDS primarily result from abnormal clonal expansion of hematopoietic stem and progenitor cells (HSPCs), leading to bone marrow failure, where cells fail to mature and develop into healthy blood cells [3]. As a result, bone marrow-suppressive agents such as cytarabine are commonly used in the treatment of MDS and AML [4]. Therefore, existing therapies for MDS and AML remain limited by poor efficacy, high relapse rates, and severe complications (e.g., transfusion dependence), driving urgent demand for novel therapies targeting cancer cell death pathways, abnormal blood cell maturation, and drug resistance mechanisms.

Over the last two decades, zebrafish has been a powerful vertebrate model system to study hematopoiesis. Their transparent embryos, rapid growth, and genetic similarities to humans have also established these organisms as a critical platform for advancing drug discovery research. The developmental processes and molecular mechanisms involved in hematopoiesis in zebrafish are highly conserved with those of higher vertebrates [5–7]. Hematopoietic defects in zebrafish can reliably recapitulate the phenotypes of human blood disorders [5]. Zebrafish-based experiments are technically well established and easy to perform. Furthermore, zebrafish serve as a powerful platform for small-molecule screening and the design of therapeutic strategies [5, 8]. In our previous work, we screened a series of compounds using this platform. In addition to their primary pharmacological effects, these compounds were also found to influence the development of hematopoietic system, including paclitaxel, gemcitabine, and famciclovir. Paclitaxel was shown to increase the number of zebrafish thrombocytes through a mechanism independent of the classical thrombopoietin receptor (Mpl) pathway. Gemcitabine was found to reduce both zebrafish thrombocyte and neutrophil numbers. Famciclovir was identified to exert a broad inhibitory effect on hematopoiesis across multiple blood lineages [9–11]. More recently, we discovered that minoxidil, a commonly used hair growth treatment, inhibits the development of zebrafish HSPCs.

Minoxidil is an antihypertensive agent and vasodilator that was originally introduced in the 1970s for the treatment of severe and refractory hypertension [12]. However, prolonged oral administration of minoxidil (beyond two weeks) was found to induce hypertrichosis in patients [13]. The discovery of minoxidil’s hair growth-promoting effect originated from clinical observations of increased hair growth in hypertensive patients treated with oral minoxidil. Since then, topical minoxidil has been widely used for the treatment of hair loss and has proven effective in promoting hair growth in individuals with androgenic alopecia, regardless of sex [14]. Minoxidil sulfate (also referred to as minoxidil N–O-sulfate or minoxidil sulfate ester) is the pharmacologically active metabolite of minoxidil [15]. In vivo, it is generated from minoxidil via sulfotransferase enzymes. It exerts vasodilatory effects by activating potassium channels (K-ATP channels) in vascular smooth muscle cell membranes, thereby reducing peripheral resistance and lowering blood pressure [15, 16]. Similar K-ATP channels have also been identified in human hair follicle dermal papilla cells, one of which is responsive to minoxidil [17]. Activation of these channels may enhance the delivery of oxygen and nutrients to hair follicles, providing a mechanistic basis for minoxidil-induced hypertrichosis. There are various explanations for the molecular mechanism of minoxidil’s hair growth effect, several other potential pathways have also been proposed. Subsequent studies have suggested that minoxidil exerts its effects on hair primarily by modulating the hair cycle and possibly increasing hair shaft diameter [18, 19]. Furthermore, minoxidil has been shown to extend the survival and delay the senescence of cultured human keratinocytes, as well as to prolong the anagen phase of follicular matrix cells. Emerging evidence also implicates the activation of the VEGF-associated β-catenin signaling pathway in minoxidil-mediated hair growth [20].

Despite these advances in hair growth, the systemic effects of minoxidil on non-cutaneous tissues remain underexplored. During consultations in pediatrics, dermatology, and cardiology at Virgen de las Nieves University Hospital, the side effects of minoxidil were assessed through clinical examination, questionnaires, and relevant laboratory tests, including blood work, electrocardiogram (ECG), and echocardiography. Leukopenia and thrombocytopenia have been reported as systemic side effects of accidental high-dose oral minoxidil exposure in children due to contaminated omeprazole preparations [21]. However, the mechanisms underlying these hematologic effects remain unclear. Our study may help address how minoxidil induces these hematologic effects in the hematopoietic system.

Transcriptomic analysis in our study revealed a significant downregulation of wnt4 in minoxidil-treated HSPCs, which prompted us to further investigate the Wnt signaling pathway and its established roles in hematopoietic regulation. As a central regulator of hematopoietic development, Wnt signaling operates through multiple pathways, including the canonical Wnt/β-catenin pathway [22, 23], the non-canonical Wnt/Ca²⁺ pathway [24], and the planar cell polarity (Wnt-PCP) pathway [25]. The Wnt family comprises numerous ligands, among which Wnt1 and Wnt3a primarily activate the canonical β-catenin pathway [22]. Wnt3a promotes the self-renewal of HSCs and is required for myeloid lineage development at the progenitor level [26]. In contrast, Wnt5a, which signals predominantly through the non-canonical Wnt/Ca²⁺ pathway, plays a critical role in maintaining long-term repopulating HSCs [27]. Wnt4 is capable of engaging both canonical and non-canonical pathways—including β-catenin, PCP/JNK, and Ca²⁺/CaMK signaling—and regulates diverse biological processes such as osteogenesis, β-cell maturation, decidualization, neurogenesis, and immune cell differentiation [28]. Also, emerging evidence suggests that Wnt4 contributes to the self-renewal of HSPCs [29]. Moreover, overexpression of Wnt4 in mice increases the number of lymphoid-primed multipotent progenitors (LMPPs; Flt3⁺ LSKs) via a non-canonical, β-catenin-independent planar cell polarity-like pathway [30, 31]. While Wnt4 has been studied in the context of hematopoietic development, its potential as a regulatory factor or therapeutic target in disorders of HSPCS dysregulation, such as MDS, remains unexplored. We therefore sought to determine whether pharmacologic modulation of Wnt4 affects hematopoiesis and progression in vivo.

Using zebrafish models, we investigated the effects of minoxidil on the hematopoietic system in vivo. Minoxidil treatment was found to suppress hematopoiesis by inhibiting the proliferation of HSPCs, without inducing apoptosis. Transcriptomic analysis revealed that minoxidil downregulates wnt4, suggesting a mechanism through which HSPCs’ proliferation is impaired. Notably, we identified wnt4 as a positive regulator of HSCs expansion. Furthermore, minoxidil alleviated the myeloid malignancy phenotype in both larval and adult c-myb^hyper zebrafish models of MDS, mitigating MDS-like symptoms while minimizing adverse effects, indicating its therapeutic potential. The c-myb^hyper zebrafish has been established as a validated MDS-like model rather than a purely myeloproliferative one. In this model, hyperactivation of c-myb leads to upregulation of early myeloid markers (c/ebpa, lcp) and to granulocytic dysplasia characterized by enlarged cell size and increased granularity, reflecting both progenitor expansion and functional abnormality [32]. Also, through dosage optimization, we identified a concentration of minoxidil that retains its therapeutic efficacy in alleviating MDS-like symptoms while minimizing hematopoietic toxicity. Consistent with the in vivo findings, minoxidil also suppressed the proliferation of human malignant hematopoietic cells in vitro, supporting the conserved inhibitory effect of minoxidil on myeloid and progenitor expansion. In wild-type (WT) mice, high-dose minoxidil treatment led to significant suppression of multiple blood lineages, confirming its myelosuppressive effects. However, this adverse effect was not observed under a lower-dose regimen, suggesting that minoxidil’s therapeutic benefits can be achieved with minimal off-target hematopoietic toxicity.

These findings highlight wnt4 as a novel molecular target and suggest that its suppression may offer a new therapeutic approach for MDS. Given that minoxidil exerts its MDS-alleviating effects, at least in part, through downregulation of wnt4, our results further support the potential repurposing of minoxidil as a candidate therapeutic targeting wnt4-mediated hematopoietic dysregulation.

Materials and methods

Model animal strains and maintenance

All experiments involving zebrafish were approved by the Institutional Animal Care and Use Committee of Southern Medical University and South China University of Technology. As previously described, zebrafish were maintained under standard conditions (28.5 ◦C, 14 h light; 10 h dark) [33]. The following zebrafish strains were used: Wild-type (WT) AB strain, Tg(c-myb:GFP)(c-myb^hyper) [32, 34], and transgenic strains Tg(mpl:eGFP) [35], Tg(cd41:eGFP) [36]. WT mice were maintained as previously described [37].

Drug screening and chemical treatment

As an exploratory experiment, the drug screening protocol followed our previously published method [9]. Tg(mpl:eGFP) is a thrombocyte lineage specific transgenic line established in our laboratory [35], enables rapid screening for drugs impacting thrombocyte development. Bone marrow failure is clinically defined by pancytopenia, a reduction in all major hematopoietic lineages. In this context, thrombocytes serve as an ideal indicator for drug screening due to their terminal differentiation status and low abundance, which allows for precise and straightforward quantification. This provides a highly reliable readout for our primary screen. This screening strategy can therefore identify compounds with broad effects on hematopoiesis, as well as those that target the thrombocyte lineage specifically. In the primary screen, Tg(mpl:eGFP) embryos with AB^SR background from the same clutch were randomly assigned to equal-sized groups, five dechorionated embryos were placed per well in 96-well plates containing 100 μM compounds (Microsource Discovery Systems, US Drug Collection). Treatment began at 1.5 days post-fertilization (dpf), and thrombocyte levels were assessed at 4 dpf by counting mpl:eGFP⁺ cells in the tail region. Compounds showing reproducible effects in > 3 independent assays were selected as candidates.

We treated 1 dpf zebrafish embryos by immersing them in solutions containing minoxidil (MedChemExpress, HY-B0112). The concentrations used in this study ranged from 0.4 mM to 8.0 mM. Dead and non-fluorescent individuals were excluded during sample collection. For adult zebrafish, Minoxidil (6 mM) was intraperitoneally injected once daily (50 µg/fish/day) for 6 days. On day 7, kidney marrow (KM) was harvested and analyzed for blood cell composition using flow cytometry. This dosing concentration was selected based on the solubility of minoxidil in PBS and the maximal intraperitoneal volume tolerated by zebrafish. No significant differences in body weight were observed between treated and control groups (Figure S2I), and no abnormal behaviors were noted. Mice, aged 6 to 8 weeks, were randomly divided into groups. Minoxidil was diluted in PBS to a concentration of 6 mM and administered via intraperitoneal injection at a dose of 50 mg/kg/day, based on a previous report [38–40]. The control group received an equivalent volume of PBS. Mice received intraperitoneal injections once daily either for 6 consecutive days (high dose) or for 6 days with a one-day interval after every two consecutive injection days (low dose), and blood samples were collected from the orbital sinus on the 7th day. Complete blood count (CBC) tests were performed to analyze blood cell composition. Throughout treatment, mice maintained normal feeding, drinking, activity, posture, and grooming behaviors, with no abnormal physical signs. Although the drug-treated cohorts showed an upward trend in body-weight fold-change (Figure S2G–H), the absolute changes were small (~ 2.7% vs. ~ 4.6% for high-dose; ~ 7.3% vs. ~ 4.6% for low-dose minoxidil versus controls). Treated mice displayed no visible edema, and these weight increases remained far below the ~ 10% threshold at which pitting edema typically becomes clinically apparent in humans [41]. As for cell experiments, minoxidil stock solution (1 M in PBS) was diluted to the indicated working concentrations before use. Based on mouse pharmacokinetic data showing that intraperitoneal injection of 50 mg/kg minoxidil produces a steady-state plasma concentration of 8–10 µg/mL (~ 40–48 µM), 50 µM was considered an approximate in vitro equivalent exposure. For 24-h treatment, cells were exposed to 50–600 µM (50, 100, 200, 500, and 600 µM). Because the short-term effects were limited, the concentration range was subsequently expanded for 48- and 72-h treatments (50, 100, 500, and 1000 µM) to evaluate potential time and dose-dependent responses. Cell proliferation was subsequently quantified with the CCK-8 assay (Beyotime Biotechnology, Shanghai, China). Cells (1 × 10⁵/mL) were seeded in 96-well plates, treated for 24–72 h with the indicated concentrations of minoxidil, and absorbance at 450 nm was measured to calculate relative proliferation. These procedures were adapted from published protocols [42, 43].

wnt4 gene overexpression

Total RNA was extracted from 4 dpf AB strain zebrafish larvae using TRIzol reagent (Invitrogen, 15596026), and reverse-transcribed into cDNA. The wnt4 coding sequence (CDS) was amplified from the cDNA and cloned downstream of a T7 promoter. wnt4 mRNA was synthesized using the mMESSAGE mMACHINE™ T7 ULTRA kit (Thermo Fisher Scientific), following the manufacturer’s instructions. Synthesized mRNA (250 ng/μL) was microinjected into one-cell stage embryos at a volume of 1 nL per embryo. After injection, embryos were maintained under standard conditions, as previously described. Embryos and larvae were collected at various developmental stages for subsequent analysis.

Generation of wnt4 zebrafish F0 knockouts (crispants)

Four target sequences specifically designed to disrupt wnt4 were listed in table S1. The corresponding full-length sgRNAs were synthesized by Tsingke (Beijing Tsingke Biotech Co., Ltd). One-cell stage zebrafish embryos were microinjected with a mixture of the four sgRNAs and Cas9 protein (EnGen Cas9 NLS, New England Biolabs, M0646T) at a final concentration of 20 µM, following established protocols for F0 CRISPR-based gene disruption in zebrafish [44]. To assess knockout efficiency, genomic DNA was extracted from F0 embryos at 2 dpf, and gene editing was validated by sequencing of PCR amplicons.

Flow cytometry analysis

The method for cell sorting in zebrafish by flow cytometry has been previously described [45, 46]. At 1 dpf, Tg(cd41:eGFP) embryos were treated with 4 mM minoxidil or vehicle control (egg water only) and collected at 3 dpf. Larvae were dissociated into single-cell suspensions in 0.9 × PBS supplemented with 5% FBS, passed through a 40-μm cell strainer, and analyzed by fluorescence-activated cell sorting (FACS) (Beckman Coulter Inc., Brea, CA, USA). Only cd41^low cells were sorted to represent HSPCs, as described previously [47].

RNA sequencing and transcriptomic data analysis

The RNA-sequencing protocol followed previously described methods [48, 49], cd41^low HSPCs were isolated via fluorescence-activated cell sorting, as described above, and subjected to mini-bulk RNA-seq. Libraries were sequenced using the DNBSEQ-T7 platform (MGI International Sales Co., Ltd.) with a depth of 10 Gb per sample. Three biological replicates were performed using cells collected from different individuals. Library preparation, sequencing, and initial data processing were conducted as previously reported [49]. Pearson correlation analysis was performed among replicates. Three control and three minoxidil-treated samples showing strong intra-group correlations were selected for downstream analysis. Pathway enrichment analysis was conducted using the Weishengxin website platform (Shanghai NewCore Biotechnology Co., Ltd.) [50], based on genes with log₂ fold change < 1 and baseMean > 50.

Cytological analysis

Kidney marrow (KM) from adult zebrafish intraperitoneally injected with minoxidil or PBS was resuspended in ice-cold PBS containing 5% FBS. The cell suspensions were then passed through a 40-μm nylon filter (Falcon). For flow cytometry analysis, the cell suspensions were subjected to forward and side scatter analysis using a flow cytometer (BD FACSDiscover™ S8, BD Biosciences, Franklin Lakes, NJ, USA), following the method described by Traver et al. [51].

Statistical analysis

Unpaired Student’s t-tests or Wilcoxon rank-sum tests were used to compare means between two independent groups. Chi-square or Fisher’s exact tests were applied for comparisons between categorical variables. One-way analysis of variance (ANOVA) was used for multiple comparisons of parametric data. A p-value < 0.05 was considered statistically significant. All data were analyzed using GraphPad Prism 9.5.0.

Results

Minoxidil inhibits hematopoiesis in zebrafish embryos

In our preliminary studies, we used the transgenic zebrafish line Tg(mpl:eGFP) to screen for compounds that influence platelet production. In the primary screen, five dechorionated embryos were placed per well in 96-well plates with 100 μM compounds (Microsource Discovery Systems, US Drug Collection). Treatment began at 1.5 days post-fertilization (dpf), and thrombocyte levels were assessed at 4 dpf by counting mpl:eGFP⁺ cells in the tail region. We found that paclitaxel increased platelet counts—a finding that has been previously published—while seven other compounds, including minoxidil, led to a decrease in platelet levels [9], which was evident from visibly reduced mpl:eGFP⁺ thrombocytes in the caudal hematopoietic region at 4 dpf. To better understand the effects of minoxidil on zebrafish hematopoiesis, and to determine an appropriate working concentration, we further examined its impact using the Tg(mpl:eGFP) line. In this model, thrombocyte lineage cells are specifically labeled with green fluorescent protein (GFP), allowing for clear visualization and quantification. 1-dpf WT embryos were treated with minoxidil at concentrations ranging from 0.04 to 8 mM for a period of 4 days. We observed the most significant decrease in thrombocyte levels at 4 mM minoxidil (Fig. 1A and B) and 8 mM Minoxidil caused deformity in most individuals and shortened body length, while other concentrations developed well (Figure S1 A and B). Based on these observations, 4 mM was chosen for the following analysis.

Fig. 1.

Minoxidil causes hematopoiesis failure in zebrafish embryos. A Representative images showing staining for mpl:eGFP protein in 4-dpf Tg(mpl:eGFP) larvae treated with egg water with PTU control and minoxidil. Scale bars: 50 μm. B Quantification of mpl:eGFP⁺ thrombocytes related to (A). Statistical significance was determined using a Mann–Whitney U-test. Data were combined from 4 biological replicates, mean ± SD, ns, not significant; *P < 0.05; ****P < 0.0001. C WISH of mpl expression in control (left panel) and minoxidil-treated (right panel) larvae at 4 dpf. D Quantification of mpl-positive cells in caudal hematopoietic tissue (CHT) (Student’s t-test; mean ± SD;*P < 0.05). E WISH of mpx expression in control (left panel) and minoxidil-treated (right panel) larvae at 4 dpf. F Quantification of mpx-positive cells in CHT (Mann–Whitney U test; mean ± SD; **P < 0.01). G WISH of mfap4 expression in control (left panel) and minoxidil-treated (right panel) larvae at 4 dpf. H Quantification of mfap4-positive cells in caudal hematopoietic tissue (CHT) (Student’s t-test; mean ± SD; ****P < 0.0001). I WISH of rag1 expression in the 4 dpf control (left panel) and minoxidil-treated (right panel) larvae. J Percentage of rag1-positive cells in caudal hematopoietic tissue (CHT) (Fisher’s exact test, ****P < 0.0001). K WISH of βe1-globin expression in the 4 dpf control (left panel) and minoxidil-treated (right panel) larvae. L Percentage of βe1-positive cells in caudal hematopoietic tissue (CHT) (Fisher’s exact test, ****P < 0.0001). M SB staining in control and minoxidil-treated larvae at 4 dpf. N Statistics result of SB-positive neutrophils in CHT region (Student’s t-test; mean ± SD; *P < 0.05). O WISH of c-myb expression in control (left panel) and minoxidil-treated (right panel) embryos at 60 hpf. P Quantification of c-myb-positive cells in CHT (Student’s t-test; mean ± SD; *P < 0.05). Q WISH of c-myb expression in control (left panel) and minoxidil-treated (right panel) embryos at 3 dpf. R Quantification of c-myb-positive cells in CHT (Student’s t-test; mean ± SD; ***P < 0.001)

Expression of several hematopoietic markers was examined in zebrafish embryos with minoxidil treatment. Exposure to minoxidil led to a marked reduction in the expression of key blood cell lineage markers in developing embryos. These markers, which identify specific cell types such as thrombocytes (mpl), neutrophils (mpx, SB staining), macrophages (mfap4), lymphocytes (rag1), and erythrocytes (βe1-globin), collectively indicate a broad suppression of hematopoietic development (Fig. 1C-N). The observed decline across these diverse lineages suggested that minoxidil might impair the formation or maturation of multiple blood cell populations during early development. To investigate whether this effect stemmed from HSPCs, we analyzed c-myb and runx1, markers of HSPCs. Embryos treated with minoxidil showed significantly reduced c-myb and runx1 expression at both 60 hpf and 3 dpf (Fig. 1O-R and Figure S3A-D). To evaluate the impact of minoxidil on primitive hematopoiesis, we quantified pu1⁺, gata1⁺, and runx1⁺ cells at 16 and 26 hpf, the results revealed no significant differences between minoxidil-treated and control embryos (Figure S4A-F), indicating that the development of these primitive myeloid, erythroid, and progenitor populations was unaffected. These findings demonstrate that high-dose minoxidil disrupted HSPCs development, leading to a systemic failure of hematopoiesis.

Minoxidil affects HSPCs proliferation

To reveal the cellular basis underlying the defective hematopoiesis in minoxidil-treated embryos, we first investigated changes in HSPCs proliferation in embryos incubated with minoxidil. The Tg(cd41:eGFP) transgenic line was utilized for the assay because cd41:eGFP^low cells are recognized as HSPCs in the caudal hematopoietic tissue (CHT) of 3-dpf embryos [52]. To evaluate the impact of minoxidil on HSPCs proliferation, we conducted a BrdU incorporation assay, which detects DNA synthesis as an indicator of cell proliferation [53]. The results showed a significant decrease in the percentage of BrdU⁺/cd41:eGFP^low HSPCs in the CHT at 3 dpf following minoxidil treatment, while the percentage of BrdU⁺/cd41:eGFP^high cells showed no significant difference (Fig. 2A-C). These findings suggested that minoxidil inhibited HSPCs proliferation in zebrafish embryos. To investigate HSPCs apoptosis in minoxidil-exposed embryos, we performed TUNEL assay [53]. Results showed that after minoxidil treatment, there was no difference in the percentage of TUNEL⁺/cd41:eGFP^low HSPCs or the percentage of TUNEL⁺/cd41:eGFP^high cells in the CHT at 3 dpf (Fig. 2D-F), indicating that minoxidil does not increase HSPCs apoptosis. Together with the BrdU assay results, these findings suggest that the reduction in HSPCs following minoxidil treatment is primarily due to decreased cell proliferation, rather than increased cell death.

Fig. 2.

Minoxidil inhibits cd41:eGFP^low HSPCs proliferation in zebrafish embryos without triggering apoptosis. A Double staining of bromo deoxyuridine (BrdU) and cd41:eGFP cells in CHT region at 3 dpf. White arrowheads indicate cd41:eGFP^low/BrdU double positive cells, yellow arrowheads indicate cd41:eGFP^high/BrdU double positive cells. Scale bars: 25 μm. B Statistics result of the percentage of cd41:eGFP^low/BrdU double positive cells in total cd41:eGFP^low cells at 3 dpf (Student’s t-test, n = 14 and n = 8, respectively; mean ± SD; *P < 0.05). C Statistics result of the percentage of cd41:eGFP^high/BrdU double positive cells in total cd41:eGFP^high cells at 3 dpf (Student’s t-test, n = 14 and n = 8, respectively; mean ± SD; ns, not significant). D Double staining of TUNEL and cd41:eGFP cells in CHT region at 3 dpf. White arrowheads indicate cd41:eGFP^low/TUNEL double-positive cells. Scale bars: 25 μm. E Quantification of the percentage of cd41:eGFP^low/TUNEL double positive cells in total cd41:eGFP^low cells at 3 dpf (Student’s t-test, n = 13; mean ± SD; ns, not significant). F Quantification of the percentage of cd41:eGFP^high/TUNEL double positive cells in total cd41:eGFP^high cells at 3 dpf (Student’s t-test, n = 13; mean ± SD; ns, not significant)

Minoxidil may influence HSPCs proliferation by affecting wnt4 expression

To unveil the molecular mechanism of minoxidil’s effects on HSPCs proliferation, we compared the HSPC’s RNA profiles of the minoxidil-treated and control samples by collecting cd41:eGFP^low-labeled HSPCs for RNA-Seq. Following Pearson correlation analysis, samples exhibiting strong positive correlations underwent further investigation. Results revealed 143 genes were downregulated in the minoxidil-treated group. Gene Ontology/Kyoto Encyclopedia of Genes and Genomes pathway analysis identified growth pathways most closely associated with proliferation. Cluster analysis of these growth pathways highlighted nine downregulated genes (col14a1a, wnt4, npr2, col4a5, dcn, krt17, dvl2, ccn1, and cebpd). Differential analysis of these nine genes revealed that wnt4 exhibited the second-highest degree of change after col14a1a (Fig. 3A-D). And wnt4 has been reported to be associated with cell proliferation [28]. To further investigate the effect of wnt4 on HSPCs proliferation, wnt4 mRNA was overexpressed by microinjection in zebrafish embryos. WISH analysis of c-myb expression revealed that overexpression of wnt4 in minoxidil-treated embryos led to a noticeable recovery of c-myb levels compared to the drug-treated group alone, suggesting a potential return to normal expression. And the wnt4-overexprssed group showed a significant increase in c-myb expression (Fig. 3E and F). We further explored the effect of wnt4 overexpression on HSPCs in the WT embryos. By detecting the expression of HSPCs markers c-myb at 60 hpf and runx1 in CHT at 60 hpf and 3 dpf, we found a significant increase in the number of both c-myb-labeled and runx1-labeled HSPCs (Fig. 3G-J). These findings indicate that wnt4 positively regulated HSPCs proliferation. Minoxidil may impair this process by downregulating wnt4 expression.

Fig. 3.

Transcriptomic and functional validation of wnt4 as a minoxidil-regulated gene that promotes HSPCs accumulation in zebrafish. A Correlation analysis of 3 control groups and 3 minoxidil-treated groups. B Different expression genes were identified after minoxidil treatment. (q < 0.05, baseMean > 50, log2 fold change > 1). C GO/KEGG analysis of downregulated genes. D Heatmap showing the 8 genes in growth pathway. E WISH of c-myb expression in control, minoxidil-treated, wnt4-overexpressing, and co-treated embryos at 3 dpf. F Quantification of c-myb-positive cells in CHT (Student’s t-test; mean ± SD; ns, not significant; **P < 0.01; ****P < 0.0001). G WISH of c-myb expression in control (left panel) and wnt4-overexpression (right panel) embryos at 60 hpf. H Quantification of c-myb-positive cells in CHT (Student’s t-test; mean ± SD; ****P < 0.0001). I WISH of runx1 expression in control and wnt4-overexpressing embryos at 60 hpf (top) and 3 dpf (bottom); left panels show control, right panels show wnt4 overexpression. J Quantification of runx1-positive cells in CHT (Student’s t-test; mean ± SD; **P < 0.01; ****P < 0.0001; purple dots indicate 60 hpf, orange dots indicate 3 dpf)

Minoxidil alleviates MDS-like myeloid hyperplasia in zebrafish, maintains hematologic safety in vivo, and inhibits human malignant hematopoietic cells

Myelodysplastic syndrome (MDS) is a heterogeneous group of myeloid clonal diseases originating from hematopoietic stem cells, characterized by abnormal differentiation and development of myeloid cells. There are limited options available for the treatment of MDS (chemotherapy and bone marrow transplantation), pending the development of new drug treatment options. Therefore, it is necessary to further investigate whether minoxidil has a therapeutic effect on MDS. The c-myb-gfp transgenic zebrafish harbors c-Myb hyperactivity (named c-myb^hyper) [32]. The c-myb^hyper zebrafish displays a robust expansion of the granulocytic lineage caused by uncontrolled proliferation of myeloid progenitors and resembles human MDS. We wondered whether minoxidil could reduce expanded hematopoietic cells in the c-myb^hyper MDS-like zebrafish. Therefore, to obtain its effective concentration in c-myb^hyper, 1-dpf c-myb^hyper embryos were treated with 0.04 mM, 0.4 mM and 4 mM minoxidil for 4 days. Sudan Black B is a fat-soluble dye that stains intracellular lipids, appearing as brownish-black particles, and is commonly used to identify neutrophils [54]. In the original characterization of the c-myb^hyper model, increased numbers of SB-positive neutrophils were observed concomitantly with c-myb hyperactivity, cell-cycle gene upregulation, and organ infiltration [32]. SB-positive neutrophils in this model comprise a heterogeneous population, but are predominantly dysplastic and hyperproliferative rather than functionally mature neutrophils. Using Sudan Black B (SB) staining, we observed a significant reduction in the number of SB-positive neutrophils at both 0.4 mM and 4 mM minoxidil concentrations (Fig. 4A and B). Consistent with previous findings using the c-myb-targeting compound flavopiridol [32], reduction of SB-positive neutrophils correlates with alleviation of pathological myeloid expansion, providing a validated and clinically meaningful readout for therapeutic efficacy in this model. To evaluate the hematologic side effects of minoxidil treatment, we examined lymphocyte and erythrocyte counts in both WT and c-myb^hyper zebrafish following treatment with 0.4 mM and 4 mM minoxidil. At 4 mM, minoxidil significantly reduced lymphocyte and erythrocyte numbers in both genetic backgrounds, indicating notable hematopoietic toxicity despite its efficacy in alleviating MDS-like symptoms. In contrast, 0.4 mM minoxidil also decreased these cell populations in WT embryos, but had no significant impact on lymphocyte or erythrocyte levels in c-myb^hyper embryos (Fig. 4C-F). These data suggest that lower dose (0.4 mM) of minoxidil can ameliorate MDS-like myeloid hyperplasia with minimal additional hematopoietic suppression. To further evaluate the anti-myeloid expansion activity of minoxidil in adulthood, both MDS-like c-myb^hyper and WT adult fish were injected intraperitoneally with minoxidil (6 × 50 μg/fish/day) or PBS as control once per day for 6 days, and kidney marrow was collected on day 7 for quantification of myeloid cells by flow cytometry and by cytospots with May-Grünwald-Giemsa staining (Fig. 5A). Firstly, we analyzed the cell suspensions by flow cytometry. And the results showed that the myelomonocytes fraction in c-myb^hyper KM was present at 44.06% and 33.01% in the control and the Minoxidil group, respectively (Fig. 5B–C). To further define the affected myeloid subsets, we performed Giemsa-stained KM cytospin smear analysis and manually quantified four major hematopoietic cell types—precursors, lymphocytes, monocytes, and neutrophils—based on their morphological features. Consistent with the flow cytometry results, neutrophil counts were significantly reduced in c-myb^hyper fish following minoxidil treatment, whereas monocytes remained unchanged (Fig. 5D–E). In contrast, no significant differences were observed in wild-type siblings. These findings demonstrate that the selected dose of minoxidil (50 µg/fish/day) effectively alleviates the excessive myeloid expansion in c-myb^hyper zebrafish, mainly through a reduction in neutrophils, while exerting minimal hematotoxic effects in wild-type fish.

Fig. 4.

Minoxidil and wnt4 loss suppress hematopoiesis and partially reverse MDS-like myeloid expansion. A SB staining of 4 dpf c-myb^hyper embryos treated with increasing concentrations of minoxidil (0, 0.04, 0.4, and 4 mM). B Quantification of SB-positive neutrophils in the CHT region under the indicated conditions (mean ± SD; Student’s t-test; ns, not significant; ***P < 0.001; ****P < 0.0001). C, D WISH of rag1 expression at 4 dpf in WT and c-myb^hyper larvae treated with two concentrations of minoxidil (0.4 mM and 4 mM). Red arrowheads indicate rag1⁺ lymphocytes. Larvae were categorized into high, medium, or low expression groups based on staining intensity. (Chi-square test; replicates = 3, n > 3; ns, not significant; *P < 0.05; ****P < 0.0001). E–F WISH of βe1-globin expression at 4 dpf in WT and c-myb^hyper larvae treated with two concentrations of minoxidil (0.4 mM and 4 mM). Red arrowheads indicate βe1⁺ erythrocytes. Embryos were classified into high, medium, or low expression groups based on staining intensity. (Chi-square test, replicates = 3, n > 3; ns, not significant; ****P < 0.0001)

Fig. 5.

Minoxidil efficacy in c-myb^hyper MDS zebrafish and hematologic safety in adult zebrafish and mice. A Schematic of the experimental protocol. Adult zebrafish received daily intraperitoneal injections of minoxidil (50 μg/fish/day) for six consecutive days. Kidney marrow was collected on day 7 for cellular composition analysis. B Representative flow cytometry plots showing the distribution of lymphocytes, precursors, and myelomonocytes in kidney marrow (KM) of WT and c-mybhyper adult zebrafish treated with PBS or minoxidil. FSC was directly proportional to cell size and SSC was indicative of cellular granularity. C Quantification of myelomonocyte percentages in each group (Student’s t-test; mean ± SD; ns, not significant; *P < 0.05; **P < 0.01). D May-Grünwald–Giemsa staining of KM cells from WT and c-mybhyper fish with or without minoxidil treatment. The scale bar represents 50 μm. Arrows are colored as blue for precursors, yellow for lymphocytes, green for monocytes and red for neutrophils. E Quantification of precursors, lymphocytes, monocytes, and neutrophils in kidney marrow from WT and c-myb^hyper fish with or without minoxidil treatment, based on manual morphological (mean ± SD; Student’s t-test; ns, not significant; *P < 0.05). F Schematic of the experimental protocol. WT mice received intraperitoneal injections of minoxidil (50 mg/kg) or vehicle (PBS) on days 1, 2, 4, and 5. Peripheral blood (PB) was collected on day 7 for complete blood count analysis. G, H Quantification of peripheral blood components following minoxidil treatment, including: G white blood cells (WBC), and H neutrophils (Student’s t-test; mean ± SD; ns, not significant)

In order to further investigate the bone marrow suppression effects of minoxidil in different species, we conducted intraperitoneal drug injection experiments on wild-type mice. The results indicated that when injections were given every two days for a total of four days, there was no significant difference in the numbers of WBCs, neutrophils, or lymphocytes in PB (Fig. 5F-H and Figure S1C-E). However, after 6 consecutive days of injection (Figure S2A), the numbers of WBCs, neutrophils, lymphocytes, basophils, and RBCs decreased significantly (Figure S2B-F). Additionally, Wnt4 expression was significantly reduced in treated mice peripheral blood compared to control groups (Figure S2J). Consistent with these peripheral blood findings, Giemsa-stained bone marrow smears showed that high-dose minoxidil markedly reduced the proportion of precursors, whereas lymphocytes, neutrophils, and monocytes remained largely unchanged (Figure S5 F-J); in the low-dose group, none of these cell populations exhibited appreciable alterations (Figure S5 A-E). These findings indicate that the hematopoietic effect of minoxidil is dose-sensitive. While high-frequency or cumulative dosing can suppress multiple blood cell types in mice, lower doses remain well-tolerated across species. To further investigate the potential cellular basis of this effect, we performed in vitro drug treatments using MDS-L and K-562 hematopoietic cell lines. Cells were exposed to a range of minoxidil concentrations (50, 100, 200, 500, 600, 1000 μM) for varying durations to assess its impact on proliferation. After 24 h, MDS-L cells showed no significant change in proliferation across all tested concentrations. This non-responsive pattern persisted at 48 h, indicating that short-term exposure had minimal impact on cell growth. However, by 72 h, MDS-L cells displayed a clear, dose-dependent decrease in proliferation at higher concentrations (500 and 1000 μM), indicating a time-dependent inhibitory effect of minoxidil on these MDS-derived cells. In contrast, K-562 cells exhibited a marked reduction in proliferation as early as 24 h at concentrations of 50 μM and above, whereas the 100 μM group showed no significant change. At 48 h, the inhibitory effect became less pronounced, with only the 1000 μM group showing a significant reduction. By 72 h, the suppressive effect on K-562 cells had diminished, indicating that minoxidil exerts an early but transient inhibition on chronic myeloid leukemia cells (Figure S6A–F). In contrast, minoxidil exhibited a dose- and time-dependent inhibitory effect on MDS-L cells. Collectively, these in vitro and in vivo findings indicate that minoxidil acts in a dose- and time-sensitive manner to modulate proliferation in MDS-related hematopoietic malignancies across species.

Loss of wnt4 function phenocopies minoxidil-induced hematopoietic suppression

To assess whether wnt4 is functionally involved in mediating the hematopoietic effects of minoxidil, we generated wnt4 crispants (wnt4 KO) in both WT and c-myb^hyper zebrafish. Sudan Black B (SB) staining at 4 dpf revealed a significant reduction in SB-positive neutrophils and lymphocytes in wnt4 crispants across both genetic backgrounds (Fig. 6A and B, Figure S3E and F), while erythrocytes remained unchanged (Figure S3G and H). We next evaluated HSPCs levels following wnt4 crispants. Whole-mount in situ hybridization (WISH) for c-myb at 60 hpf and 3 dpf showed markedly reduced signal intensity in the caudal hematopoietic tissue (CHT) of both WT and c-myb^hyper embryos injected with wnt4 crispants, compared to their respective controls (Fig. 6C–F).

Fig. 6.

Loss of wnt4 reduces HSPCs numbers and alleviates myeloid hyperplasia in zebrafish. A Sudan Black B (SB) staining of control and wnt4 crispants (wnt4 KO) in both WT and c-myb^hyper larvae at 4 dpf. B Quantification of SB-positive neutrophils in the caudal hematopoietic tissue (CHT) (Student’s t-test; mean ± SD; ****P < 0.0001). C WISH of control and wnt4 crispants (wnt4 KO) in both WT and c-myb^hyper embryos at 60 hpf. D Quantification of c-myb-positive cells in the CHT at 60 hpf (Student’s t-test; mean ± SD; *P < 0.05; ****P < 0.0001). E WISH of control and wnt4 crispants (wnt4 KO) in both WT and c-myb^hyper embryos at 3 dpf. F Quantification of c-myb-positive cells in the CHT at 3 dpf (Student’s t-test; mean ± SD; *P < 0.05; ****P < 0.0001)

Collectively, these results demonstrate that wnt4 crispants phenocopies the myelosuppressive phenotype observed in minoxidil-treated embryos, including reduction of HSPCs number and partial reversal of the MDS-like myeloid hyperplasia. These findings support wnt4 as a downstream effector of minoxidil and a potential therapeutic target in HSPCs-driven hematologic disorders such as MDS.

Discussion

Minoxidil has been previously studied for its effects on various biological processes, such as lowering blood pressure and promoting hair growth. In this study, we identified its myelosuppressive effect and found wnt4 as a new regulator of HSPCs proliferation. Additionally, using the c-myb^hyper zebrafish model of MDS, characterized by abnormal HSPC proliferation, we demonstrated that minoxidil alleviated MDS-like phenotypes. In addition to the in vivo evidence from zebrafish, minoxidil also suppressed the proliferation of human MDS cell line MDS-L in a dose- and time-dependent manner, indicating that its inhibitory effect on hematopoietic and myeloid expansion is conserved across species. These findings highlight the significant role of minoxidil in hematopoietic development and its potential therapeutic implications.

Minoxidil is a pethidine derivative that has been extensively studied for its antihypertensive and hair growth effects. Minoxidil achieves its vasodilation effect through K-ATP channels on vascular smooth muscle [15, 16], whereas there are many explanations for the hair growth effects of minoxidil, including K-ATP pumps in hair follicles [17], VEGF-β-catenin pathway, and activation of prostaglandin-endoperoxide synthase 1 (PTGS1) [55, 56]. In our study, we found that minoxidil may affect hematopoietic development by regulating the expression of wnt4. Previous studies have shown that mice Wnt4 promotes hematopoietic progenitor cell expansion through a β-catenin–independent, planar cell polarity–like pathway rather than through canonical Wnt/β-catenin signaling [30, 31]. These findings support that wnt4 functions independently of β-catenin to modulate progenitor proliferation, suggesting that minoxidil might influence hematopoiesis through this noncanonical wnt4 pathway. Further investigation is required to identify the specific mechanisms involved in this regulatory interaction.

Wnt4, one of the atypical ligands in the Wnt family, is involved in the development and differentiation of various cell types through β-catenin-dependent and β-catenin-independent pathways. It has been found that wnt4 is associated with heart development in zebrafish and that female sex development and reproductive duct formation in zebrafish depend on wnt4 [57, 58]. Increasing evidence suggests that wnt4 plays critical but distinct roles in hematopoietic development. In Wnt4-deficient mice, lymphangiogenesis is reduced, and Wnt4 promotes the expansion of hematopoietic progenitor cells [30, 31]. Consistent with these reports, our study demonstrated that Wnt4 overexpression expanded HSPCs, while wnt4 deficiency reduced their numbers and partially reversed myeloid hyperplasia in c-myb^hyper embryos. These findings underscore the essential role of Wnt4 in hematopoietic homeostasis.

Although direct Wnt4 mutations have rarely been documented in hematologic malignancies, the Wnt4 signaling network is evolutionarily conserved and involved in maintaining hematopoietic balance. Since both wnt4 deficiency and minoxidil treatment in wild-type zebrafish also lead to a reduction in HSPCs, we propose that the minoxidil–Wnt4 regulatory axis identified here is likely conserved beyond c-myb hyperactivation, potentially extending to other genetic contexts of myeloid dysregulation. In our future studies, we also plan to investigate how this regulatory axis operates across additional genetic and non-genetic models of MDS and AML to further elucidate its role in HSPC regulation. Finally, given that minoxidil is a clinically approved and well-tolerated small molecule, it could serve as a promising adjunct to current MDS/AML regimens by selectively modulating abnormal progenitor proliferation while sparing normal hematopoiesis.

Conclusion

In summary, this research has illustrated the significant role of minoxidil in hematopoietic development. Minoxidil inhibits the abnormal proliferation of hematopoietic stem progenitor cells by down-regulating the expression of wnt4, thereby ultimately alleviating the phenotype of MDS. Overall, our findings suggest that minoxidil may have potential therapeutic value for MDS, and targeting wnt4 may provide a promising therapeutic approach.

Authors’ contributions

X.S. and Y.L. wrote the manuscript, performed most of the experimental work, and conducted the statistical analysis. Y.L. and L.H. carried out the drug screening and some preliminary work. Y.O. and X.C. assisted with the mouse experiments. Q.L. revised the manuscript. Y.Z. and Q.L. designed the study and provided key suggestions and guidance.

Funding

This work was supported by grant from the National Natural Science Foundation of China (32370874, 32570967), National Natural Science Foundation of China and Research Grants Council of Hong Kong (32161160326), Guangdong S&T programme (2024TQ08A109) and Guangdong Medical Research Foundation (B2025166).

Data availability

The raw sequencing data have been deposited in the Genome Sequence Archive (GSA) at the BIG Data Center under accession number CRA026464.

Declarations

Competing interests

The authors declare no competing interests.