HAX1 inhibits apoptosis and promotes maturation of neutrophils

Abstract

As the diverse functions of neutrophils continue to be uncovered, elucidating the molecular mechanisms that regulate their differentiation, development, and apoptosis has become crucial for overcoming limitations in the treatment of neutrophil-related diseases. Hematopoietic cell-specific protein 1-associated protein X 1 (HAX1), encoded by the primary pathogenic gene of autosomal recessive severe congenital neutropenia, serves as a key target for in-depth exploration of neutrophil function. In the Hax1 myeloid knockout C57BL/6J mice and stably transduced HL-60 cells with HAX1 knockdown that we constructed, our results showed that the differentiation of granulocyte-monocyte precursor cells (GMPs) and the maturation of neutrophils were inhibited, significantly reducing the proportion of myeloid cells and neutrophils in both bone marrow and peripheral blood. In addition, HAX1 deletion disrupted mitochondrial structure and mitochondrial membrane potential in neutrophils and increased the protein levels of B-cell lymphoma 2 (BCL-2) family members and cleaved Caspase-9. Through RNA sequencing and mRNA validation, we further demonstrated that HAX1 regulates neutrophil apoptosis and maturation via the mitochondrial-mediated classical apoptotic pathway and toll-like receptor 2 (TLR2)-mediated purine-rich box 1 (PU.1) signaling. This study elucidated the critical role of HAX1 in neutrophil differentiation, maturation, and apoptosis, providing new targets for research into neutrophil-related diseases.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12964-025-02353-2.

Plain language summary

HAX1 regulates neutrophil apoptosis through mitochondrial dependent pathway and multiple TLR2 mediated pathways, affecting neutrophil maturation by regulating PU.1 levels via TLR2.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12964-025-02353-2.

Introduction

Neutrophil maturation is a multi-step process involving granule generation. In the bone marrow (BM), hematopoietic stem cells (HSCs) first differentiate into common myeloid progenitors (CMPs), which are stimulated by granulocyte colony-stimulating factor (G-CSF) to produce GMPs. However, effective research on the regulatory mechanisms and functional connections between neutrophil maturation, and apoptosis remains limited [1]. Establishing more targeted and representative preclinical models is crucial for uncovering the molecular mechanisms underlying neutrophil dysfunction and diseases associated with myelodysplasia.

Congenital neutropenia (CN) is a hereditary and heterogeneous disorder characterized by reduced absolute neutrophil counts in peripheral blood (PB). Genetic studies have identified multiple pathogenic genes associated with CN, each linked to varying clinical manifestations and complex regulatory patterns. Major pathogenic genes include ELANE, G6PC3, TAZ, ROBLD3, WAS, GFI1, CXCR4, and HAX1 [2, 3]. Screening for genes involved in neutrophil differentiation and development in CN provides valuable insights into the mechanisms underlying neutrophil maturation. Currently, clinical management of CN primarily relies on G-CSF [4], which promotes neutrophil differentiation and reduces apoptosis to increase neutrophil counts. For patients with high-risk pathogenic mutations or G-CSF insensitivity, HSC transplantation [5, 6] and targeted therapies, such as neutrophil elastase inhibitors [7], have been explored. Furthermore, advancements in experimental techniques, including CRISPR/Cas9 gene editing [8, 9] and combination therapies with vitamin B3 [10, 11], are expanding the clinical options for CN treatment. The development of more physiologically relevant animal models for in vivo studies on neutrophil development driven by common CN-related genes is crucial for overcoming the challenges of low efficiency, high specificity, and limited stability in gene-editing therapies. Additionally, elucidating the upstream and downstream regulatory mechanisms of these pathogenic genes will aid in identifying novel therapeutic targets for CN and other hematological disorders.

Hematopoietic cell-specific protein 1-associated protein X 1 (HAX1), initially identified for its binding to hematopoietic cell-specific protein 1, is localized to the membranes of mitochondria, the endoplasmic reticulum, and the cytoplasm, where it plays a pivotal role in regulating cellular motility and transmembrane protein endocytosis [12, 13]. Additionally, due to its homology with the anti-apoptotic protein B-cell lymphoma 2 (BCL-2), HAX1 is crucial for controlling programmed cell death and promoting cell survival [14]. HAX1 interacts with various proteins, including Caspase-9 [15], serine peptidase 2 [16], protein kinase D2 [17], integrin-linked kinase family [18], microtubule end-binding protein 2 [13], and viral polymerase proteins [19]. It has been shown that HAX1 exerts its anti-apoptotic effects by alleviating mitochondrial and endoplasmic reticulum stress, thereby maintaining cellular homeostasis [17, 20, 21]. Moreover, HAX1 has been implicated in reducing myocardial and neuronal damage [22], promoting cancer invasion and metastasis [23], and inhibiting viral replication [24]. As a key pathogenic gene in autosomal recessive severe CN, mutations and deletions of HAX1 lead to early neuronal and lymphatic cell death, primary ovarian dysfunction with delayed puberty, and gonadal failure [21]. To date, research on HAX1 in neutrophil function has primarily focused on alleviating CN through induced differentiation in pluripotent stem cells to explore its pathogenic mechanisms [25] and its potential as a therapeutic target for CN in vitro [17, 26, 27]. However, comprehensive in vivo studies using systematic and mature mammalian models are still lacking. The impact of HAX1 on neutrophil maturation, apoptosis, and homeostasis remains to be fully elucidated. To address this gap, we constructed C57BL/6J mouse models with conditional Hax1 deletion in myeloid cells and established stable HL-60 cell lines with HAX1 knockdown and overexpression, aiming to investigate the molecular mechanisms of HAX1 in neutrophil maturation and apoptosis.

Materials and methods

Mice

Hax1^fl/fl gene-edited mice were hybridized with Lyz2-Cre⁺mice specifically expressed in myeloid cells to construct conditional knockout (cKO) mouse models of Hax1^fl/fl; lyz2-Cre⁺, cooperating with Shanghai Model Organisms Center, Inc. In all studies, we randomly used male and female mice aged around 10 weeks.

Cell lines

HL-60 cells (CL-0110, Procell) were cultured in Iscove’s modified Dulbecco’s medium (IMDM) (12440053, Life Technologies) supplemented with 20% fetal bovine serum (A5670701, Life Technologies). HEK293T cells and Hela cells (CL-0005, Procell) were cultured in Dulbecco’s modified Eagle’s medium (SH30022.01, Cytiva) supplemented with 10% fetal bovine serum.

Expression plasmid production of HAX1-shRNA and over-expressed plasmids and retroviral infection

Two shRNA sequences targeting human HAX1 mRNA designed from NCBI website, were inserted into pSicoR (‌11579, Addgene) empty plasmids. The target sequences were listed in Table S1. The over-expression plasmids, PCDH-CMV-HAX1-3Flag-EF1a-CopGFP-T2A-Puro and PCDH-CMV-3Flag-EF1a-CopGFP-T2A-Puro were purchased from SyngenTech (China). Knockdown and over-expression vectors were inserted into retroviruses (12260 and 12259, Addgene) to infect HL-60 at a multiplicity of infection of 100-150 in the presence of 10 μg/mL polybrene (TR-1003, Sigma-Aldrich). Seventy-two hours after beginning the transduction, cells transduced with over-expression plasmids were screened at 4 μg/mL to 10 μg/mL puromycin (A1113803, Life Technologies), while cells transduced with the knockdown vectors were screened for FITC⁺ cells using flow cytometry.

Morphological and apoptotic analysis of cells

HL-60 was induced to differentiate by 0.75% dimethylsulfoxide (D8371, Solarbio) and 0.5 μM all-trans-retinoic acid (R2625, Sigma-Aldrich) in complete IMDM medium for 4 days, and then used for Wright-Giemsa stainingand flow cytometry analysis. After Wright-Giemsa staining, the proportion of cells with rod-shaped and lobulated nucleus in each group was percentage counted. Induced cells were stained with APC/Cy7-CD11b to analyze the proportion of neutrophils induced to differentiate, while stained with PE-Annexin V and BV421-DAPI to analyze the apoptosis. The antibody information is listed in Supplementary Table S2.

Flow cytometry analysis

Fresh single cell suspensions from PB and BM were analyzed and cell-sorted on FACSAria II Flow Cytometer (BD Biosciences, USA). The data were analyzed using FlowJo software. The antibody information is listed in Supplementary Table S2.

• I.Mouse PBMC and BM terminally differentiated cells were stained by a mouse antibody cocktail of FITC-CD4, PE-CD45, Percp/Cy5.5-CD3, PE/Cy7-B220, APC-CD8 and APC/Cy7-CD11b.
• II.Mouse BM cells for analysis of HSCs, lymphoid and myeloid differentiation were firstly stained with a cocktail of anti-mouse biotin-conjugated antibodies (B220, CD4, CD8a, CD11b, TER-119, Gr-1). Secondary labelling was performed using APC/Cy7-streptavidin with other antibodies (FITC-CD34, PE-Flt3, PE-CD16/32, Percp/Cy5.5-CD127, PE/Cy7-Sca1, APC-cKit).
• III.Mouse BM cells for analysis of neutrophilic maturity and differentiation were firstly blocked with purified antibody CD16/32, and then stained with mouse antibody cocktail combination (PE-CXCR2, Percp/Cy5.5-CD115, Percp/Cy5.5-CD3, PE/Cy7-CXCR4, APC-cKit, APC/Cy7-CD11b, BV421-Gr-1, FITC-CD90, FITC-B220, FITC-NK1.1). Cells were treated with absolute counting tubes (BD Biosciences, USA) for absolute counting of neutrophils.
• IV.The neutrophil mitochondria membrane potential (MMP) was determined using JC-1 Mitochondrial Membrane Potential Flow Cytometry Assay Kit (MedChemExpress, USA). Mouse BM cells were firstly incubated with JC-1 reagent solution at 37 °C for 30 min and rinsed with PBS. The cells were blocked with purified antibody CD16/32, and then stained with mouse antibody cocktail combination (Percp/Cy5.5-CD115, Percp/Cy5.5-CD3, Percp/Cy5.5-CD90.2, PE/Cy7-B220, PE/Cy7-NK1.1, APC/Cy7-CD11b, BV421-Gr-1).The stained cells were examined by the flow Cytometer.
• V.Mouse BM cells were firstly stained with a cocktail of anti-mouse biotin-conjugated antibodies (B220, CD4, CD8a, TER-119, Gr-1). Secondary labelling was performed using PE-streptavidin with other antibodies (PE/Cy7-Sca1, APC-cKit, APC/Cy7-CD11b). Myeloid progenitors (Lin⁻Sca1⁻cKit⁺) and terminally differentiated cells (CD11b⁺) were sorted for RNA sequencing.
• VI.The Blood routine test of PB in mice was completed by the ILAS Safety Evaluation Center (China).

Colony forming assays

One hundred thousand unseparated BM cells were evenly cultured in three wells of a twenty-four well plate with methylcellulose medium (03434, STEMCELL MethoCult™ GF) at 37℃, 5% carbon dioxide, and constant temperature and humidity. Colony-forming unit-granulocyte, macrophage (CFU-GM) and burst-forming unit-erythroid (BFU-E) were counted after 7 days of culture and colony-forming unit-granulocyte, erythroid, macrophage, megakaryocyte (CFU-GEMM) were counted after 12 days of culture.

RNA sequencing

Major construction of a RNA sequencing library were completed by Shanghai Majorbio Bio-pharm Technology Co.,Ltd (China). Detailed procedures of flow sorting are described in Flow cytometry analysis. The expression levels between sample groups were quantitatively analyzed using RSEM software. Simultaneously, DESeq2 software was employed to analyze the expression differences between sample groups. A classification strategy based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) database was used to conduct functional analysis of target genes. Fisher’s exact test, based on the KEGG database, was used to perform enrichment analysis on the gene sets obtained from KEGG functional annotation. The relationships of protein interactions expressed by the target genes were analyzed based on the STRING database.

Western blot

Cells and mouse tissues were lysed on ice for 30 min with a mixture of RIPA buffer (P0013B, Beyotime) and protease and phosphatase inhibitors (78440, Thermo Fisher Scientific). The supernatant was collected and prepared for protein loading. After SDS-PAGE electrophoresis separation, PVDF membrane transfer and 5% skim milk blocking, protein samples were sequentially incubated with primary and secondary antibodies. The membrane was imaged with the near-infrared imaging system Odyssey CLx (LI-COR Biosciences, USA) at 680 and 800 nm. All antibodies used for immunoblotting are listed in Supplementary Table S2. Image Studio Lite 5.2.5 software was used to analyze imprints and obtain relative protein abundance through the density analysis.

Reverse transcription and quantitative real-time PCR

RNA extracted and purified using TRIzol reagent (15596026CN, Thermo Fisher Scientific) was reversely transcribed into cDNA with PrimeScript™ RT Master Mix kit (RR036A, TaKaRa). TB Green PCR kit (RR820A, TaKaRa) was used to run qPCR programs on the StepOnePlus Real Time PCR System (Thermo Fisher Scientific, USA). The expression levels of target genes were standardized to the expression levels of internal reference genes. All sequence of qPCR primers are detailed in Supplementary Table S1.

Quantification and statistical analysis

We conducted multiple independent experiments using biological replicates to ensure the reproducibility of our findings. GraphPad Prism 8 (GraphPad Software Inc., La Jolla, CA, USA) was used for data analysis with *P < 0.05, **P < 0.01 and ***P < 0.001. The Shapiro–Wilk test was used to check the normality of the data when the F- test applied to test the variance homogeneity between two data groups. Unpaired t test was used for normal values, but nonparametric Mann–Whitney tests for non-normal ones. The Benjamini–Hochberg was applied to correct for multiple testing of P values in the RNA sequencing analysis. Genes with P value < 0.05 and |log₂FC|≥ 1 were considered to have significant differences for differential expression analysis. Corrected P value (P adjust) < 0.05 was considered to indicate significant enrichment of this function for KEGG functional enrichment analysis of the differentially expressed gene set.

Results

HAX1 influenced in the differentiation of HL-60 cells and regulated apoptosis

To investigate the effect of HAX1 on the differentiation of promyelocytes into neutrophils, we used the partially differentiated M2-type human acute myeloid leukemia (AML) cell line HL-60 as a cellular model. Compared to the control, HAX1 protein levels were significantly reduced in HAX1knockdown HL-60 cells, while the HAX1 overexpression group showed an opposite trend (Fig. 1A). HL-60 cells were induced to differentiate into neutrophils using dimethyl sulfoxide and all-trans retinoic acid [28–30]. Wright-Giemsa staining and flow cytometry analysis were performed post-induction to assess differentiation. Wright-Giemsa staining revealed a significant increase in neutrophils with rod-shaped or lobulated nuclei in the HAX1 overexpression group, whereas a marked decrease was observed in the knockdown group (Fig. 1B). Similarly, flow cytometry analysis of CD11b⁺ cells showed a distinct rightward shift in the HAX1 overexpression group, while the knockdown group exhibited a leftward shift (Fig. 1C). These findings, consistent with the morphological analysis, indicated that HAX1 promoted neutrophil differentiation.

Fig. 1.

HAX1 influenced in the differentiation of HL-60 cells and regulated apoptosis. A Western blot validation of protein expression of hematopoietic cell-specific protein 1 associated protein X 1 (HAX1). B Calculation of total cell percentage of neutrophils differentiated and produced within three random fields (400x) using Wright Giemsa staining solution, induced by 0.75% dimethylsulfoxide and 0.5 μM all-trans-retinoicacid for 4 days. Black arrows indicated undifferentiated HL-60 cells, while red arrows indicated differentiated HL-60 cells. C Flow cytometry analysis of the proportion of CD11b⁺cells. D Annexin V analysis of the apoptosis of knockdown (up) and over-expression (down) HL-60 cells. E, F Western blot analysis of protein expression of BCL-2-Associated X in HAX1 knockdown (E) and over-expression (F) HL-60 cells. Data were represented as mean ± SD. N = 3,*P < 0.05; **P < 0.01; ***P < 0.001, by Student's t test or Welch's t test. BAX: BCL-2-Associated X. sh-NC: HL-60 transduced with the control of HAX1 knockdown plasmid. sh-HAX1: HL-60 transduced with the HAX1 knockdown plasmid. PCDH-NC: HL-60 transduced with the control of HAX1 over-expression plasmid. PCDH-HAX1: HL-60 transduced with the HAX1 over-expression plasmid. VA: Viable apoptotic cells. VN: Viable nucleated cells. NVA: Non-viable apoptotic cells. NVN: Non-viable nucleated cells

In addition, Annexin V/DAPI flow cytometry apoptosis analysis revealed that HAX1 knockdown significantly increased the proportion of apoptotic cells while reducing the number of live cells. In contrast, HAX1 overexpression resulted in an increased number of live cells and a reduced apoptotic population (Fig. 1D). Given that HAX1 contains BCL-2 homologous domains [14], we measured the levels of key proteins involved in the classical apoptosis pathway regulated by the BCL-2 family to investigate the potential mechanism through which HAX1 modulates apoptosis. Our results showed that HAX1 knockdown led to a significant upregulation of BAX, whereas HAX1 overexpression had the opposite effect (Fig. 1E, F). These findings suggest that HAX1 downregulation promotes apoptosis.

Taken together, these results preliminarily demonstrate that HAX1 positively regulates the differentiation of promyelocytes into neutrophils and acts similarly to anti-apoptotic proteins in suppressing apoptosis during this process.

HAX1 deletion inhibited neutrophil maturation and myeloid differentiation

To systematically investigate the role of HAX1 in neutrophil maturation in vivo, we utilized CRISPR/Cas9 to generate Hax1 knockout mice. These mice displayed slow growth, weakness, and early mortality (3-9 weeks post-birth) due to severe fat vacuolization and cellular exhaustion in the BM (Figure S1). To overcome these limitations, we developed a conditional knockout (cKO) model by crossing Hax1^fl/fl mice with Lyz2-Cre⁺ mice, enabling myeloid cell-specific (monocytes, macrophages, and granulocytes) deletion of HAX1 (Fig. 2A). Genotype identification and Cre enzyme activation were confirmed at the DNA level (Fig. 2B), and HAX1 expression was found to be significantly reduced in the BM, compared to other major organs, at both the protein and mRNA levels (Fig. 2C-F).

Fig. 2.

Hax1-cKO mice were confirmed the significant reduction of Hax1 in the protein and mRNA levels. A Construction strategy of Hax1^fl/fl; Lyz2-Cre⁺ conditional knockout (cKO) mice. B Identification of genotype (left) and validation of Cre enzyme activation activity (right) of cKO mice by gel electrophoresis. C, D Western blot analysis of protein expression of HAX1 in bone marrow (BM) and other central tissues, N = 8. E, F Reverse transcription and real-time quantitative PCR analysis of transcription of Hax1 in BM (E), spleen and thymus (F), N = 3. Data were represented as mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001, by Student's t test or Welch's t test

After confirming that HAX1 disrupted neutrophil differentiation in vitro, we used the Hax1-cKO mouse model to assess its impact on myeloid differentiation in vivo. Blood routine tests and flow cytometry revealed a significant reduction in granulocytes in both PB and BM. In PB, both monocytes and neutrophils were reduced, with neutrophils showing a more pronounced decline (Fig. 3A, B). Further analysis of neutrophil developmental stages [31] revealed a significant decrease in overall neutrophils in the BM of Hax1-cKO mice (Fig. 3C, D), accompanied by an accumulation of immature neutrophils and a reduction in mature neutrophils, indicating that HAX1 regulates neutrophil maturation (Fig. 3C, E). To investigate the underlying cause of granulocyte reduction, we performed flow cytometry and colony-forming assays on neutrophil progenitors. While no significant differences were observed in myeloid, lymphoid, and HSCs (Lin⁻Sca1⁺c-kit⁺ cells), myeloid differentiation showed a downward trend. Specifically, GMPs were significantly reduced in Hax1-cKO mice, whereas megakaryocyte–erythroid progenitors (MEPs) and CMPs remained largely unchanged (Fig. 3F, G).

Fig. 3.

HAX1 deletion inhibited neutrophil maturation and myeloid differentiation. A, B Flow cytometry analysis of terminally differentiated cells in mouse BM and peripheral blood (PB). Blood routine test of terminally differentiated cells in mouse PB (B, down). C-E Flow cytometry analysis of different developmental levels of neutrophils in mouse BM. The absolute hindlimb neutrophil cellularity in mouse BM (D). F, G Flow cytometry analysis of progenitor cells and their downstream differentiation in mouse BM. Trials used mice aged 9–10 weeks, N = 5. Data were represented as mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001, by Student's t test or Welch's t test for normal values, but Mann–Whitney tests for non-normal ones. MP: Myeloid progenitor. LSK: Lin⁻Sca1⁺c-kit⁺. MPP: multipotent progenitor. CMP: common myeloid progenitor. GMP: granulocyte monocyte progenitor. MEP: megakaryocyte–erythroid progenitor. LYM: lymphocyte. MON: monocyte. NEU: neutrophil

Colony-forming assays revealed that BFU-E and CFU-GM colonies assessed on day 7 corresponded to MEP and GMP differentiation, respectively (Fig. 4A-D). By day 12, CFU-GEMM colonies reflected CMP differentiation (Fig. 4E, F). Compared to the control group, the number of CFU-GM colonies formed by BM cells from Hax1-cKO mice was significantly decreased (Fig. 4A), while BFU-E and CFU-GEMM colonies showed an increasing trend (Fig. 4B, F). These changes in colony formation were consistent with the flow cytometry results of progenitor cell populations, further confirming the observed differentiation patterns (Fig. 3F, G). Overall, these findings indicated that HAX1 deletion reduced the proportion of myeloid cells in both BM and PB by selectively inhibiting GMP differentiation, while MEP differentiation remained largely unaffected. Additionally, HAX1 deficiency impaired the maturation of neutrophils, leading to an accumulation of immature neutrophils, consistent with our in vitro findings.

Fig. 4.

These colony formation changes were consistent with the flow cytometry results of progenitor cell populations. A-D Counting analysis of burst-forming unit-erythroid (BFU-E) (A, B) and colony-forming unit-granulocyte, macrophage (CFU-GM) (C, D) colonies on the seventh day after cultivation. E, F Counting analysis of colony-forming unit-granulocyte, erythroid, macrophage, megakaryocyte (CFU-GEMM) colonies on the twentieth day after cultivation. Experiments adopted mice aged 9–10 weeks, N = 5. Data were represented as mean ± SD. *P < 0.05; **P < 0.01, by Student's t test or Welch's t test for normal values, but Mann–Whitney tests for non-normal ones

HAX1 regulated neutrophil apoptosis through mitochondrial mediated pathway

Based on our initial findings that HAX1 influenced apoptosis in vitro (Fig. 1E, F), we performed Western blot analysis to assess key apoptotic pathway proteins in mouse BM. Hax1-cKO mice exhibited significantly increased levels of cleaved Caspase-9 and BAX in the BM, accompanied by a marked reduction in the Caspase-9 level (Fig. 5A, B). These results suggest enhanced activation of the mitochondrial-mediated apoptotic pathway, involving the pro-apoptotic BCL-2 family and caspase enzymes. Mitochondrial-mediated apoptosis, a key mechanism of programmed cell death, is regulated by the balance between pro-survival and pro-apoptotic proteins, particularly those of the BCL-2 family. These proteins control the release of intermembrane space proteins, such as cytochrome C, from the mitochondria into the cytoplasm, thereby activating caspase enzymes and initiating apoptosis. In parallel with the elevated expression of pro-apoptotic proteins in the BM, transmission electron microscopy was employed to examine submicroscopic structural changes in neutrophils. Compared to the control group, neutrophils from Hax1-cKO mice showed pronounced mitochondrial vacuolization, structural disruption, and damage to mitochondrial cristae (Fig. 5C) and a significant decrease in neutrophil MMP in HAX1-deficient mice (Fig. 5D). Collectively, these findings suggest that HAX1 modulates neutrophil apoptosis via the mitochondrial-mediated apoptotic pathway.

Fig. 5.

HAX1 regulated neutrophil apoptosis through mitochondrial mediated pathway. A, B Western blot analysis of expression of HAX1, BAX, cleaved Caspase-9 and Caspase-9 in BM (N = 3). C Transmission electron microscope scanning of the neutrophilic mitochondria in BM (15,000 × and 80,000 ×). The red arrow indicated the mitochondria. D The mitochondrial membrane potential (MMP) was assessed using a mitochondrial membrane potential detection kit. Data were represented as mean ± SD. *P < 0.05; **P < 0.01, by Student's t test or Welch's t test

Critical targets of HAX1 deficiency in attenuating neutrophil maturation and promoting cell apoptosis

To further investigate how HAX1 influences neutrophil apoptosis and maturation, we sorted myeloid progenitor cells and terminally differentiated cells, which were significantly reduced due to HAX1 deficiency, for RNA sequencing. The analysis revealed substantial differences in the enrichment of differentially expressed genes (DEGs) between control and Hax1-cKO mice. Specifically, compared to the control group, the cKO group showed a clear distinction between up-regulated and down-regulated genes, with 671 genes up-regulated and 766 genes down-regulated (Fig. 6A). KEGG functional annotation analysis of the DEGs indicated that these genes were primarily distributed across secondary categories of "Immune system", "Cell growth and death", and "Transport and catabolism", which fall under the broader primary categories of "Organismal systems" and "Cellular processes" (Fig. 6B). Further KEGG enrichment analysis of the DEGs in the “Immune system” and “Cell growth and death” categories revealed that the up-regulated genes were significantly enriched in the signaling pathways of “Apoptosis” and “Neutrophil extracellular trap formation” (Fig. 6C). In the “Transport and Catabolism” category, the up-regulated genes were significantly enriched in the “Mitophagy” pathway (Fig. 6D). In contrast, the significantly down-regulated genes showed a weaker association, with the top ten enriched pathways related only to "Mitophagy" but none of these were statistically significant (P < 0.05) (Figure S2).

Fig. 6.

Critical targets of HAX1 deficiency in attenuating neutrophil maturation and promoting cell apoptosis. A Volcanic (left) and heat maps (right) of differentially up-regulated and down-regulated genes. B KEGG annotation analysis of differentially up-regulated (up) and down-regulated (down) genes in "Cellular Processes" and "Organismal Systems" (showing the top ten secondary classifications of gene numbers). C, D KEGG functional enrichment analysis of differentially up-regulated genes in "Immune system" and "Cell growth and death" (C) or "Transport and catabolism" (D) (showing the top ten enrichment of pathways). E Protein interaction analysis of differentially up-regulated genes (P adjust < 0.05) in "Apoptosis" and "Neutrophil extracellular trap formation" (left) or "Mitophagy" (right). F Reverse transcription and real-time quantitative PCR analysis of mRNA levels of Bak1, Tlr2, Atg9a and PU.1 in BM. Experiments used 9–10 week old mice, N = 3. Data were represented as mean ± SD. *P < 0.05; **P < 0.01, by Student's t test or Welch's t test

To explore the downstream targets in key regulatory pathways affected by HAX1 during neutrophil development and apoptosis, we constructed a protein interaction map for differentially up-regulated genes that were significantly enriched in the “Apoptosis”, “Neutrophil extracellular trap formation” and “Mitophagy” pathways. Several interactions were identified among genes regulating apoptosis and neutrophil function, including Akt2, Bak1, Bbc3, Tlr2, Ncf4, Ddit3, and Ctsd, as well as genes associated with mitophagy, such as Mul1 and Atg9a (Fig. 6E). In the Hax1-cKO BM, the transcriptional levels of the up-regulated genes Bak1 and Atg9a, which are involved in the mitochondrial-mediated apoptosis pathway, showed significant increases, consistent with the RNA sequencing data (Fig. 6F). BCL-2 antagonist/killer 1 (BAK1), encoded by Bak1, is a pro-apoptotic protein of the BCL-2 family that plays a crucial role in caspase activation [32, 33], further supporting HAX1's role in regulating the mitochondrial-mediated apoptosis pathway. In the enrichment analysis of neutrophil immune regulation, we identified Tlr2, which was significantly up-regulated at both the RNA sequencing and mRNA levels. Toll-like receptor 2 (TLR2), encoded by Tlr2, mediates various downstream signaling pathways in neutrophils [34–36], making it a critical target for understanding how HAX1 regulates neutrophil maturation. TLR2, a member of the Toll-like receptor family, is primarily located on the membranes of monocytes, macrophages, neutrophils, and mast cells. It forms heterodimers with TLR1 or TLR6, activating downstream signaling pathways upon ligand binding. TLR2 activation can reduce the abundance of the high-affinity IgE receptor on the cell membrane in a Purine-rich box 1 (PU.1)-dependent manner [37]. PU.1, an E26 family transcription factor encoded by the Spi1 gene, is highly expressed in mature neutrophils and regulates the transcription of genes essential for cell maturation. A decrease in PU.1 levels leads to stagnation and abnormal development of neutrophils [38, 39]. We hypothesize that HAX1 controls neutrophil maturation through a TLR2-dependent transcriptional pathway. In the Hax1-cKO BM, TLR2 and PU.1 exhibited significant upregulation and downregulation at the transcriptional level, respectively (Fig. 6F), consistent with trends reported in related studies [37]. Overall, these findings suggest that HAX1 influences neutrophil maturation by regulating PU.1 levels via TLR2, thereby halting the progression of neutrophils from immature to mature stages (Fig. 3C, E).

Discussion

In this study, we generated HL-60 cell lines with both knocked-down and overexpressed HAX1, as well as C57BL/6J mouse models with myeloid-specific Hax1 knockout, to investigate the impact of HAX1 on neutrophil function. Our results confirmed the significant role of HAX1 in neutrophil apoptosis and maturation. As a membrane-anchored protein containing a BCL-2 homology domain, HAX1 plays a key role in regulating cell migration and programmed cell death [13, 40]. We observed that the levels of representative pro-apoptotic proteins, including BAX and cleaved Caspase-9, increased at the protein level as HAX1 expression decreased (Figs. 1E and 5A, B). Furthermore, RNA sequencing revealed similar changes in BAK1 expression at the RNA level (Fig. 6F).

In the Hax1-cKO mice, neutrophil mitochondria showed significant vacuolization and a loss of normal cristae structure (Fig. 5C) with a significant decrease in MMP (Fig. 5D). This, coupled with the upregulation of the mitophagy-related gene Atg9a (Fig. 6F), underscored the disruption of normal mitochondrial structure and function due to HAX1 deficiency. Through comprehensive analysis, including protein-level assessment, mitochondrial morphological evaluation, MMP determination, RNA sequencing, and gene expression profiling, we demonstrated that changes in HAX1 levels could activate the mitochondrial-mediated apoptosis pathway, driven by the BCL-2 family proteins, leading to the activation of caspase enzymes. We hypothesize that HAX1, located on the mitochondrial membrane, interacts with pro-apoptotic BAX and/or BAK1, or potentially forms homodimers, similar to the interaction of anti-apoptotic BCL-2 proteins [41, 42]. As HAX1 levels increase, the formation of heterodimers between HAX1 and BAX/BAK1 becomes more abundant, enhancing the anti-apoptotic effect. Conversely, reduced HAX1 levels promote homodimerization of HAX1, which triggers the release and activation of BAX and BAK1, resulting in changes to mitochondrial membrane potential and permeability, and facilitating the release of cytochrome C. Cytochrome C further activates caspase family enzymes, thereby promoting apoptosis. Previous studies suggest that different transcripts of rat HAX1 regulate cell survival and death through homodimerization or heterodimerization, similar to the balance between anti-apoptotic and pro-apoptotic subunits of the BCL-2 family [43], further supporting our hypothesis (Fig. 7).

Fig. 7.

Schematic diagram of the pathway through which HAX regulates neutrophil maturation and apoptosis. HAX1 regulates neutrophil apoptosis through mitochondrial dependent pathway and multiple TLR2 mediated pathways, affecting neutrophil maturation by regulating PU.1 levels via TLR2. Created in https://BioRender.com. Created in https://BioRender.com

PU.1, a critical mediator of terminal neutrophil differentiation, plays a central role in the formation of CXC chemokine ligand 2 (CXCL2). CXCL2 is essential for neutrophil chemotaxis by binding to CXC chemokine receptor type 2 (CXCR2), which is highly expressed on mature neutrophils [1]. Given the strong conservation of gene expression between human and mouse neutrophils, including CXCR2 and SPI1 (which encodes PU.1), the regulatory influence of HAX1 on neutrophil maturation via the TLR2-mediated PU.1 transcriptional pathway in mice holds significant relevance for human biology [44]. Notably, TLR2 also contributes to apoptosis. It regulates apoptosis by activating Caspase-3 through the P38 MAPK pathway [45] and the TLR2/JNK/mitochondrial axis pathway [46]. Notably, TLR2 also plays a key role in apoptosis by activating Caspase-3 through the P38 MAPK pathway [45] and the TLR2/JNK/mitochondrial axis [46]. In HAX1 knockout models, both TLR2 and BAX levels are elevated (Figs. 6F, 5A and 1E). Previous studies have demonstrated that TLR2 activation leads to increased BAX expression [47, 48], suggesting that HAX1 not only regulates neutrophil maturation through TLR2 but also influences neutrophil apoptosis through multiple downstream apoptotic pathways (Fig. 7). However, whether there is an interaction between TLR2’s roles in apoptosis and myeloid differentiation remains unclear. One study found no significant interference between the TLR2-mediated pathways of myeloid differentiation and apoptosis [45], suggesting that TLR2 may function independently in these two processes.

Conclusion

In summary, our study demonstrated that HAX1 could regulate both neutrophil maturation and apoptosis, as demonstrated both in vitro and in vivo (Fig. 7). While previous research on HAX1 in neutrophils has primarily focused on its effects on G-CSF levels [25, 49], our findings offer new insights into its broader regulatory role. These results open up novel avenues for future research and highlight potential targets for therapeutic development, particularly in the context of immune diseases such as AML and disorders of neutrophil maturation. Further exploration of the upstream and downstream regulatory mechanisms of HAX1 will be crucial for improving the clinical application of gene-editing therapies targeting pathogenic genes, helping to overcome challenges such as low efficiency, high specificity, and instability.

Abbreviations

HAX1

Hematopoietic cell-specific protein 1-associated protein X 1

GMPs

Ggranulocyte-monocyte precursor cells

BCL-2

B-cell lymphoma 2

TLR2

Toll-like receptor 2

PU.1

Purine-rich box 1

BM

Bone marrow

HSCs

Hematopoietic stem cells

CMPs

Common myeloid progenitors

G-CSF

Granulocyte colony-stimulating factor

CN

Congenital neutropenia

PB

Peripheral blood

cKO

Conditionally knockout

IMDM

Iscove’s modified Dulbecco’s medium

MMP

Mitochondrial membrane potential

BFU-E

Burst-forming unit-erythroid

CFU-GM

Colony-forming unit-granulocyte, macrophage

CFU-GEMM

Colony-forming unit-granulocyte, erythroid, macrophage, megakaryocyte

KEGG

Kyoto Encyclopedia of Genes and Genomes

AML

Acute myeloid leukemia

sh-NC

HL-60 transduced with the control of HAX1 knockdown plasmid

sh-HAX1

HL-60 transduced with the HAX1 knockdown plasmid

PCDH-NC

HL-60 transduced with the control of HAX1 over-expression plasmid

PCDH-HAX1

HL-60 transduced with the HAX1 over-expression plasmid

VA

Viable apoptotic cells

VN

Viable nucleated cells

NVA

 Non-viable apoptotic cells

NVN

Non-viable nucleated cells

MEPs

Megakaryocyte–erythroid progenitors

MP

Myeloid progenitor

LSK

Lin-Sca1+c-kit+

MPP

Multipotent progenitor

LYM

Lymphocyte

MON

Monocyte

NEU

Neutrophil

DEGs

Differentially expressed genes

BAK1

BCL2-antagonist/killer 1

CXCL2

CXC chemokine ligand 2

CXCR2

CXC chemokine receptor type 2

Authors’ contributions

Hanwei Yue contributed as the first author. Hanwei Yue designed and performed research, analyzed data, and wrote the manuscript. Guiying Shi and Jiaming Tang performed flow cytometry experiments. Xinyue Li sorted out and analyzed the data on Hax1-KO mice. Lin Bai guided the project, reviewed the manuscript, critically revised the key intellectual content, and was ultimately responsible for the decision submitted for publication. All authors read and approved the final manuscript.

Funding

This work was supported by National Key Research and Development Project (2022YFA1103803 and 2023YFC2309000) and CAMS Innovation Fund for Medical Sciences (CIFMS) (2021-I2M-1-035).

Data availability

All data supporting the findings of this study are available within the paper and its Supplementary Information. Sequences and antibody information are provided in Supplementary Table 1 and Supplementary Table 2.

Declarations

Ethics approval and consent to participate

All methods were conducted in accordance with relevant guidelines and regulations. All mice were housed in the SPF grade vivarium at Institute of Laboratory Animal Sciences, CAMS & PUMC. All mouse procedures were approved the Institutional Animal Care and Use Committee of Institute of Laboratory Animal Sciences (NO. BL23001).

Consent for publication

The authors declare that they have no competing interests.

Competing interests

The authors declare no competing interests.