Cytohesin-4/ARF6 facilitates the progression of acute myeloid leukemia through activating PIK3R5/PI3K/AKT pathway

Summary

In silico analysis revealed an elevated expression of cytohesin-4 (CYTH4) in acute myeloid leukemia (AML) cells, correlating with a poorer prognosis for AML patients. However, its role in AML is not fully understood. Our study using loss-of-function assays identified CYTH4 as an oncogene promoting leukemogenesis. Silencing CYTH4 in MV4-11 and THP-1 cells reduced cell proliferation and colony formation, and induced apoptosis and cell-cycle arrest at G0/G1, whereas overexpression had no significant impact. CYTH4 silencing also increased chemosensitivity to cytarabine. In a THP-1 xenograft model, CYTH4 silencing slowed AML progression and reduced leukemic cell homing and infiltration. Mechanistically, CYTH4 silencing inhibited PI3K/AKT pathway by lowering PIK3R5 and decreased ARF6-GTP levels, as confirmed by pull-down assays. Overexpression of PIK3R5 and AKT activation via SC-79 successfully countered the cellular dysfunctions from CYTH4 silencing. Thus, CYTH4 may play a role in AML progression, and targeting its pathway could be a promising anti-leukemic treatment strategy.

Graphical abstract

Highlights

• •CYTH4 silencing inhibits AML cell proliferation by inducing apoptosis and cell-cycle arrest
• •CYTH4 silencing delays AML progression and reduces the homing and infiltration of AML cells
• •CYTH4 demonstrated GEF activity with ARF6 but did not activate ARF1
• •CYTH4/ARF6 exerts its effects by activating the PIK3R5-mediated PI3K/AKT signaling pathway

Cancer; Cell biology

Introduction

Acute myeloid leukemia (AML) is a group of heterogeneous myeloid malignancies originating from myeloid progenitor cells with poor prognosis and few targeted therapies.^1,2 Over the past decades, the standard upfront treatment has remained a traditional 7 + 3 chemotherapy regimen involving daunorubicin and cytarabine (Ara-C), and the cure rates have not significantly improved.³ A significant proportion of AML patients do not achieve complete remission (CR) following chemotherapy, with rates exceeding 30%, and 40%–70% of the patients achieving CR relapse within 5 years⁴ To develop more efficacious treatment strategies for AML, it is imperative to identify the molecular pathways involved in the pathogenesis of this disease.

As members of the RAS superfamily, ADP-ribosylation factors (ARFs) regulate some critical signaling pathways that control various cellular processes, including membrane trafficking, transcriptional regulation, metabolism, cell division, motility, and apoptosis.⁵ Thus, it makes sense that ARFs and their regulators are implicated in tumorigenesis and progression.^6,7 The functions of ARFs depend on cycling between active GTP-bound and inactive GDP-bound forms. Guanine nucleotide exchange factors (GEFs) facilitate the conversion of ARF-GDP to ARF-GTP, whereas GTP-activating proteins accelerate the hydrolysis of ARF-GTP to ARF-GDP.^8,9 The ARF-GEFs are known to be involved in cancer progression by specifically activating ARFs.¹⁰ The blockade of GEFs activity or interaction between ARFs and GEFs should be considered as favorable therapeutic strategies for tumors.

The cytohesin (CYTH) proteins, consisting of CYTH-1, -2, -3, and -4, are a subfamily of ARF-GEFs that are important regulators of signal transduction.^7,11 A variety of studies have demonstrated that cytohesins facilitate carcinogenesis and cancer progression. CYTH1 has been reported to play a crucial role in adhesion-mediated leukemogenesis.¹² CYTH2 expression is upregulated in melanoma and facilitates tumor growth.¹³ The upregulation of CYTH3 has been observed in hepatocellular carcinoma, which promotes tumor growth and vascular invasion.¹⁴ CYTH4 has been associated with various neurodegenerative disorders, including schizophrenia, bipolar disorder, and Alzheimer’s disease.^15,16,17 Although reported to be associated with ovarian cancer, glioma, and AML,^5,18,19 the role of CYTH4 in AML and its underlying mechanisms have not been entirely elucidated.

In this study, we found that CYTH4 expression was upregulated in AML cells and reversely related with overall survival (OS) of the patients with AML. With a loss-of-function assay conducted on MV4-11 and THP-1 cells, we observed that CYTH4 silencing suppressed the growth and colony formation of AML cells and enhanced their sensitivity to Ara-C chemotherapy. Additionally, CYTH4 silencing was found to delay the progression of AML and the homing and extramedullary infiltration of leukemic cells. These effects were attained by suppressing the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway via the downregulation of the phosphoinositide 3-kinase regulatory subunit 5 (PIK3R5) protein. Both overexpression of PIK3R5 and treatment with SC-79, an AKT agonist, effectively counteracted the growth-inhibitory consequences of CYTH4 silencing. CYTH4 silencing resulted in a significant decrease in ARF6-GTP rather than ARF1-GTP levels. These results suggest that CYTH4 offers potential as a viable therapeutic target, with its inactivation potentially contributing to a blockage of AML progression.

Results

CYTH4 expression was notably elevated in AML cells and exhibited a negative correlation with the overall survival of AML patients

To screen out the key hub genes related to AML, we analyzed three gene expression data from two microarray datasets (GSE67936 and GSE65409) and a TCGA dataset of GEPIA2 (Gene Expression Profiling Interactive Analysis version 2) with online tools. A total of 33 overlapping differentially expressed genes (DEGs) were identified from three databases (Figure 1A; Tables S1 and S2), consisting of 8 downregulated DEGs and 25 upregulated DEGs. These DEGs were further subjected to OS assessments on the survival analysis websites (Figure S1). Interestingly, we found that CYTH4 was overexpressed in AML compared to healthy donors (p = 0.0116 in GSE67936, p = 0.0102 in GSE65409, and p < 0.05 in GEPIA2) (Figure 1B) and that the expression of CYTH4 was inversely correlated with the OS of AML patients (p = 0.0014 in GEPIA2, p = 0.0015 in TIMER [Tumor Immune Estimation Resource, Version 2], and p = 0.019 in UALCAN [The University of Alabama at Birmingham Cancer data analysis portal]) (Figure 1C). A review of the literature on PubMed revealed limited information regarding the function of this gene, except for its association with AML, ovarian cancer and glioma.^5,18,19 However, other members of the CYTH family have been documented to play roles in the proliferation and migration of tumor cells, including AML.^12,20,21,22 The mechanisms by which CYTH4 contributes to AML pathogenesis remain unclear. Consequently, CYTH4 was selected for further investigation. To validate the elevated expression of CYTH4 in AML, bone marrow samples were obtained from newly diagnosed adult AML patients and healthy donors for quantitative reverse-transcription PCR (RT-qPCR) and immunoblotting analyses. The findings indicated a notable upregulation of CYTH4 expression at both the mRNA and protein levels in AML patients compared to healthy controls (Figures 1D and 1E). Furthermore, MV4-11 and THP-1 cell lines exhibited significantly higher levels of CYTH4 mRNA expression compared to other AML cell lines and the mixed bone marrow mononuclear cells from five healthy donors (p < 0.001) (Figure 1F). Accordingly, the expressions of the CYTH4 protein in MV4-11 and THP-1 cells were upregulated compared to other AML cell lines (Figure 1G).

Figure 1.

CYTH4 is highly expressed in AML cells and inversely correlated with the overall survival of AML patients

(A) After intersecting three datasets, 25 upregulated and 8 downregulated differentially expressed genes (DEGs) were identified.

(B) The significantly upregulated expression of CYTH4 in AML patients compared to healthy donors (HD) (left, GSE67936, ∗p = 0.0116; middle, GSE65409, ∗p = 0.0102; and right, GEPIA2, ∗p < 0.05).

(C) CYTH4 expression is reversely associated with the overall survival of AML patients (left, GEPIA2, p = 0.0014; middle, TIMER, p = 0.0015; and right, UALCAN, p = 0.019). Kaplan-Meier curve was used to analyze the survival distribution of the two groups.

(D) RT-qPCR assay for CYTH4 mRNA in bone marrow mononuclear cells (BMMCs) from AML patients (n = 35) and HD (n = 15). Data are represented as mean ± SD. Statistical significance was assessed by unpaired t test.

(E) Immunoblotting assay for CYTH4 protein in BMMC from AML patients (n = 6) and HD (n = 4).

(F) RT-qPCR assay for CYTH4 mRNA in human AML cell lines (n = 3). The mRNA level of CYTH4 in the mixed BMMCs from five HD was set as 1.0. Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(G) Immunoblotting assay for CYTH4 protein expression in human AML cell lines. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

CYTH4 silencing alleviated human AML cell proliferation and enhanced sensitivity to Ara-C

A loss-of-function analysis was performed to investigate the potential functions of CYTH4 in human AML cells. Lentivirus-encoded short hairpin RNAs (shRNAs) targeting CYTH4 were transduced into MV4-11 and THP-1 cells with high endogenous CYTH4 expressions. In MV4-11 cells, the most efficient mRNA knockdown was achieved by shCYTH4-1, followed by shCYTH4-4, while in THP-1 cells, shCYTH4-2 showed the greatest efficiency, followed by shCYTH4-1 (Figure 2A). Immunoblotting assay also demonstrated the successful silencing of CYTH4 protein in MV4-11 using shCYTH4-1 and shCYTH4-4 and THP-1 cells using shCYTH4-1 and shCYTH4-2, respectively (Figure 2B), indicating the establishment of CYTH4-knockdown stable cell lines. A subsequent evaluation of the impact of CYTH4 deficiency on cell proliferation and colony formation was conducted using these stable cell lines. The data presented in Figure 2C revealed a significant reduction in the in vitro proliferation of MV4-11 and THP-1 cells on day 4 following CYTH4 silencing (both p < 0.001). Additionally, the numbers of colony-forming units decreased significantly (both p < 0.001), and the clone sizes were smaller in shCYTH4-transduced MV4-11 and THP-1 cells, as the consequence of CYTH4 deficiency (Figure 2D).

Figure 2.

CYTH4 silencing alleviates the proliferation of human AML cells and increases the chemosensitivity of AML cells to Ara-C

(A) Validation of the downregulated expression of CYTH4 in four shCYTH4-transduced MV4-11 and THP-1 cells by RT-qPCR (n = 3). β-actin was used as an internal control. Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(B) Validation of the downregulated expression of CYTH4 in shCYTH4-transduced MV4-11 and THP-1 cells by immunoblotting. GAPDH was used as a loading control.

(C) CYTH4 silencing inhibits the proliferation of MV4-11 and THP-1 cells. CCK-8 assay (n = 3) was conducted to detect the viability of MV4-11 and THP1 cells at the indicated time points (0, 24, 48, 72, and 96 h). Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(D) Inhibitory effect of CYTH4 silencing on colony formation of AML cell was assessed by microscopically counting the numbers and sizes of colonies at 14 days after plating. The quantification is shown on the right (n = 3).Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(E) CYTH4 silencing in MV4-11 and THP-1 cells using shRNAs induces cell apoptosis. Following annexin V-APC and 7-AAD staining, the percentages of apoptotic cells were detected by flow cytometry. Representative flow cytometry plots (top) and quantification (bottom) are shown (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(F) CYTH4 silencing in MV4-11 and THP-1 cells using shRNAs induces cell-cycle arrest at the G0/G1 phase. Representative flow cytometry histogram plots (top) and quantification (bottom) are shown (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(G) Immunoblotting assay for c-MYC, BCL-2, BAX, CYCLIN B1, and CYCLIN D1 upon CYTH4 silencing in MV4-11 and THP-1 cells. GAPDH was used as a loading control.

(H) CYTH4 silencing sensitizes MV4-11 and THP-1 cells to Ara-C. The growth inhibitory effects of AML cells were evaluated through a CCK-8 assay following treatment with varying concentrations of Ara-C for a duration of 48 h. IC50 values were calculated utilizing GraphPad Prism 8.0 software based on the dose-response curve. Data are represent as mean ± standard deviation. Statistical significance was assessed by unpaired t test. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. Sc, scramble control shRNA; shCYTH4, CYTH4 shRNA.

To further investigate the mechanism underlying the CYTH4 silencing-mediated cell proliferation defect, we examined the cell apoptosis rates and cell cycle distribution. The findings revealed that CYTH4-deficient MV4-11/THP-1 cells exhibited notably higher rates of early and late apoptosis in comparison to those transduced with scramble control (Sc) shRNA (both p < 0.01) (Figure 2E). Specifically, The CYTH4-deficient MV4-11 cells resulted from shCYTH4-1 or shCYTH4-4 exhibited 11.2% and 10.3% early apoptosis, and 7.26% and 6.56% late apoptosis, respectively, whereas MV4-11 cells transduced with Sc shRNA showed 1.9% early and 1.76% late apoptosis. In THP-1 cells transduced with shCYTH4-1 or shCYTH4-2, 2.81% and 2.14% early apoptosis, as well as 6.79% and 7.79% late apoptosis were observed, in contrast to 1.14% early and 2.85% late apoptosis in THP-1 cells transduced with Sc shRNA. These findings collectively indicate a potential role for CYTH4 in the inhibition of apoptosis in AML cells. Furthermore, analysis of the cell cycle using flow cytometry revealed that CYTH4 deficiency led to a partial arrest at the G1/S transition, characterized by a notable increase in G0/G1 phase cells (p < 0.01) and a significant decrease in S (p < 0.01) and G2/M phase cells (p < 0.05) (Figure 2F), suggesting that CYTH4 may play a role in promoting the progression from G1 to S phase in AML cells. To investigate the molecular mechanisms involved, the expression levels of some key regulators associated with survival, apoptosis, and cell cycle were analyzed using western blot assay. The results revealed significant decreases in the protein expressions of cell proliferation-related c-Myc, anti-apoptotic Bcl-2, and cell cycle-related cyclin B1 and cyclin D1 following CYTH4 silencing (Figure 2G). Subsequent cytotoxicity testing with Ara-C at varying concentrations for 48 h allowed for the calculation of half-maximal inhibitory concentration (IC50) values. The results indicated that silencing of CYTH4 in MV4-11 using shCYTH4-1 and THP-1 cells using shCYTH4-2 resulted in the heightened sensitivity to Ara-C, as evidenced by 2.6- and 1.9-fold reductions in IC50 values, respectively (12.050 ± 0.842 μM vs. 4.676 ± 0.090 μM for MV4-11, p < 0.001; 20.050 ± 0.709 μM vs. 10.480 ± 0.099 μM for THP-1, p < 0.001) (Figure 2H).

CYTH4 silencing suppressed the activation of PI3K/AKT pathway by downregulating PIK3R5

Among the four shRNAs tested, shCYTH4-1 and shCYTH4-2 exhibited superior mRNA knockdown efficiencies in MV4-11 and THP-1 cells, respectively. Therefore, shCYTH4-1 was selected for ongoing mechanism studies with MV4-11 cells, while shCYTH4-2 was chosen for those involving THP-1 cells. To further elucidate the molecular mechanism underlying the inhibition of AML cell growth upon CYTH4 silencing, an RNA sequencing (RNA-seq) assay was conducted to identify the DEGs between Sc shRNA and shCYTH4-transduced AML cells. Volcano plots revealed 1,374 upregulated gens and 1,323 downregulated genes in shCYTH4-transduced MV4-11 cells compared to Sc shRNA-transduced MV4-11 cells, as well as 1,482 upregulated genes and 1,464 downregulated genes upon CYTH4 silencing in THP-1 cells (Figure 3A; Table S3). Among the top 5 significantly deregulated pathways according to the Kyoto Encyclopedia of Genes and Genome (KEGG) analyses, CYTH4 is implicated in the following pathways: PI3K-AKT signaling pathway, cytokine-cytokine receptor interaction, and cell adhesion molecules (Figure 3B). Immunoblotting assay revealed that the expressions of p-PI3K and p-AKT decrease upon CYTH4 silencing (Figure 3C), indicating that CYTH4 contributes to activating PI3K-AKT signaling pathway. Subsequently, the DEGs involved in the PI3K-AKT signaling pathway were identified from RNA-sequencing data, including 5 upregulated genes (ITGB1, BCL2L1, BCL2L11, JAK1, and PPP2CA) and 10 downregulated genes (ATF4, GNB1, HRAS, ITGAV, MAP2K1, mTOR, PIK3CD, PIK3R5, RAF1, and TSC2) (Figures 3D and 3E). RT-qPCR further confirmed that the expressions of MAP2K1, HRAS, TSC2, and PIK3R5 significantly decreased in shCYTH4-transduced both MV4-11 and THP-1 cells (Figure 3F), indicating a potential downstream relationship with CYTH4 in the PI3K/AKT pathway. Subsequent examination of gene expression correlations between CYTH4 and these four genes using GEPIA2 data revealed a strong positive correlation between CYTH4 and PIK3R5 (R = 0.58) (Figure 3G). We speculated that CYTH4 may enhance AML cell proliferation by activating the PIK3R5-mediated PI3K/AKT signaling pathway. The immunoblotting assay also confirmed that CYTH4 silencing resulted in a decrease in PIK3R5 expression (Figure 3H). To investigate the potential involvement of CYTH4 in the activation of ARFs, a GGA3-PBD pull-down assay was conducted in CYTH4-silenced AML cells to assess the levels of GTP-bound ARF1 and ARF6. The results revealed a significant reduction in GTP-ARF6 level, rather than GTP-ARF1 level, in shCYTH4-transduced MV4-11 and THP-1 cells (Figure 3I), suggesting that CYTH4, a GEF protein for ARFs, plays a crucial role in ARF6 activation in AML cells. We also examined the gene expression correlation between ARF6 and PIK3R5 using GEPIA2 data. The results revealed a significant positive association (R = 0.51) (Figure 3J), indicating that CYTH4 may modulate the expression of PIK3R5 through the activation of ARF6.

Figure 3.

CYTH4 silencing results in the deactivation of PI3K/AKT pathway via downregulating PIK3R5

(A) Volcano plots for the DEGs between Sc shRNA-transduced and shCYTH4-transduced MV4-11 and THP-1 cells.

(B) The KEGG pathway analysis utilizing mRNA-sequencing data from Sc shRNA-transduced and shCYTH4-transduced MV4-11 and THP-1 cells.

(C) Immunoblot analysis confirmed the inhibition of the PI3K/AKT signaling pathway upon CYTH4 silencing in MV4-11 and THP-1 cells, with GAPDH serving as a loading control.

(D) The intersection analysis of differential transcripts within the PI3K-AKT signaling pathway between MV4-11 and THP-1 cells. Five upregulated and 10 downregulated transcripts were identified at last.

(E) Clustering heatmap of differential transcripts in PI3K-AKT signaling pathway.

(F) Potential candidates involving in “PI3K-AKT signaling pathway” were examined in Sc shRNA-transduced and shCYTH4-transduced MV4-11 and THP-1 cells by RT-qPCR (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by unpaired t test.

(G) The gene expression correlation analyses between CYTH4 and MAP2K1, HRAS, TSC2, or PIK3R5 in AML cells were conducted using GEPIA2 data.

(H) Immunoblot confirmation for the downregulation of PIK3R5 following CYTH4 silencing in MV4-11 and THP-1 cells.

(I) The activities of ARF1 and ARF6 between Sc shRNA- and shCYTH4-transduced MV4-11/THP-1 cells were compared using a pull-down assay.

(J) The gene expression correlation analysis between ARF6 and PIK3R5 in AML cells was conducted using GEPIA2 data (R = 0.51). ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. Sc, scramble control shRNA; shCYTH4, CYTH4 shRNA.

The stable overexpression of CYTH4 did not impact the proliferation of AML cells

We performed a gain-of-function analysis to estimate the biological function of CYTH4 in AML cells by establishing CYTH4-overexpressing (OE) MV4-11 and THP1 cell lines. The overexpression of CYTH4 in MV4-11 and THP1 cells was validated at both the mRNA and protein levels (Figures 4A and 4B). According to the Cell Counting Kit-8 (CCK-8) assay, there was no significant difference in cell viability between the lenti-vector and lenti-CYTH4 groups in either MV4-11 or THP1 cells at any of the specified time points (All p > 0.05) (Figure 4C). Additionally, colony formation assays indicated that CYTH4 overexpression did not significantly affect the number or size of colonies (both p > 0.05) (Figure 4D). Flow cytometry analysis further demonstrated that CYTH4 overexpression in AML cells had no noticeable effect on cellular apoptosis (both p > 0.05) (Figure 4E).

Figure 4.

Overexpression of CYTH4 has no effect on AML cell proliferation

(A) The expression of CYTH4 was validated to be upregulated in CYTH4-transduced MV4-11 and THP-1 cells by RT-qPCR, with β-actin serving as the internal control (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by unpaired t test.

(B) Immunoblotting confirmed the upregulated expression of CYTH4 in MV4-11 and THP-1 cells transduced with CYTH4, with GAPDH as the loading control.

(C) CYTH4 overexpression did not influence the proliferation of MV4-11 and THP-1 cells. A CCK-8 assay was carried out to assess the viability of these cells at the indicated time points (0, 24, 48, and 72 h). Data are represented as mean ± SEM. Statistical significance was assessed by unpaired t test.

(D) The impact of CYTH4 overexpression on the colony formation of MV4-11 and THP-1 cells was evaluated by counting the number and size of colonies under a microscope at 14th day post-plating. The quantification is depicted on the right (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by unpaired t test.

(E) Overexpression of CYTH4 in MV4-11 and THP-1 cells did not significantly impact cell apoptosis. Apoptotic cell percentages were measured using flow cytometry after annexin V-APC and 7-AAD staining. Representative plots from flow cytometry are on the left, and their quantifications are on the right (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by unpaired t test. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. OE, overexpression.

CYTH4 silencing by shRNA suppressed tumor growth and extramedullary invasion in a human AML xenograft model

By the tail vein injection of THP-1 cells transduced with Sc shRNA or shCYTH4-2 into immunodeficient NOD-Prkdc^em26Cd52il2rg^em26Cd22/Nju (NCG) mice, we successfully established a xenogeneic AML model. The mice transplanted with shCYTH4-transduced THP-1 cells exhibited a significantly slower AML progression compared to the control group, which received Sc shRNA-transduced THP-1 cells (Figures 5A–5D). After transplantation for five weeks, the mice were euthanized and intact liver and spleen organs were obtained. As depicted in Figure 5A, the recipients receiving shCYTH4-transduced THP-1 cells exhibited notable reductions in both the size and weight of the livers and spleens in comparison to those receiving Sc shRNA-transduced THP-1 cells (both p < 0.001). The recipients receiving shCYTH4-transduced THP-1 cells also demonstrated a reduction in the disseminated tumor nodules in their livers. Engrafted human CD45⁺ cells were identified in the recipients through flow cytometry analysis. The proportions of human CD45⁺ cells in the peripheral blood, bone marrow, liver, and spleen of the recipients receiving shCYTH4-transduced THP-1 cells were 0.92%, 4.43%, 20.4%, and 0.69%, respectively, which were significantly less than those (5.23%, 26.70%, 78.6%, and 6.88%) of the recipients receiving Sc shRNA-transduced THP-1 cells (all p < 0.001) (Figure 5B). Hematoxylin and eosin staining of the femurs, livers, and spleens showed that CYTH4 silencing resulted in the decreased proliferation of THP-1 cells in bone marrows, the reduced dissemination of tumor nodules in livers, and the diminished infiltration of THP-1 cells in spleens (Figure 5C). Ten mice in each group were monitored for survival analysis. All mice in the Sc control group succumbed within 60 days post-transplantation, whereas those transplanted with shCYTH4-transduced THP-1 cells survived for up to 80 days (p < 0.001) (Figure 5D). The homing capabilities of CYTH4-silenced THP-1 cells in the xenograft model were investigated. The distribution of AML cells in the bone marrow and spleen was assessed 16 h following tail vein injection. Flow cytometry analysis revealed that CYTH4 silencing significantly inhibited the homing of AML cells to the bone marrow and spleen (both p < 0.001) (Figure 5E).

Figure 5.

CYTH4 silencing delays the progression and extramedullary invasion of AML in a xenografts model

(A) Representative images of the livers and spleens of the recipients following transplantation for five weeks. Quantifications of liver weight and spleen weight are shown on the right (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by unpaired t test.

(B) Human CD45 chimerism in the peripheral blood, bone marrow, liver, and spleen of recipients was assessed using flow cytometry following transplantation for five weeks (n = 3). THP-1 cells (4.0 × 10⁵ cells per mouse) were transplanted into NCG mice via tail vein injection. Following euthanizing, peripheral blood, femurs, livers, and spleens of the recipients were harvested for human CD45 chimerism analysis, with quantifications presented at the bottom. Data are represented as mean ± SEM. Statistical significance was assessed by unpaired t test.

(C) Representative hematoxylin/eosin staining images for the femurs, livers, and spleens. The infiltrated leukemic cells are indicated by black arrows.

(D) Kaplan-Meier survival curve for the mice transplanted with Sc shRNA/shCYTH4-transduced THP-1 cells (n = 10 per group).

(E) Flow cytometry analysis was conducted on the bone marrow and spleen of the xenograft recipient mice 16 h post tail vein injection for homing evaluation (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by unpaired t test. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. Sc, scramble control shRNA; shCYTH4, CYTH4 shRNA.

PIK3R5 overexpression successfully ameliorated the AML cell growth defect from CYTH4 silencing

To substantiate that CYTH4 silencing inhibits AML cell proliferation by downregulating PIK3R5/PI3K/AKT pathway, we conducted a rescue experiment by overexpressing PIK3R5 in CYTH4-silenced MV4-11 and THP-1 cells. GFP and mCherry double-positive cells were sorted by fluorescence-activated cell sorting (BD FACSAria III) and seeded for further expansion. The overexpression of PIK3R5 was confirmed at both the mRNA and protein levels (Figures 6A and 6B). This overexpression significantly counteracted the suppressive effects of CYTH4 silencing on cell proliferation and colony formation (all p < 0.01) (Figures 6C and 6D). Moreover, PIK3R5 overexpression reduced apoptosis induced by CYTH4 silencing in MV4-11 and THP-1 cells (all p < 0.01) (Figure 6E). A marked recovery of AKT phosphorylation at serine 473 was observed in CYTH4-silenced MV4-11 and THP-1 cells following PIK3R5 overexpression, indicating that PIK3R5 overexpression reinstates the AKT activity. Furthermore, PIK3R5 overexpression counteracted the reduction of anti-apoptotic proteins c-Myc and Bcl-2 caused by CYTH4 silencing in MV4-11 and THP-1 cells (Figure 6F).

Figure 6.

PIK3R5 overexpression effectively counteracts the AML cell growth defect resulting from CYTH4 silencing

(A) PIK3R5 overexpression in CYTH4-silenced MV4-11 and THP1 cells were confirmed at the mRNA levels by RT-qPCR (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(B) PIK3R5 overexpression in CYTH4-silenced MV4-11 and THP1 cells were confirmed at the protein levels by immunoblotting.

(C) CCK-8 assays showed that the impaired cell proliferations caused by CYTH4 silencing were reversed by PIK3R5 overexpression in MV4-11 and THP-1 cells (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(D) PIK3R5 overexpression could effectively reverse the decrease in clone formation caused by CYTH4 silencing (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(E) The apoptosis of cells in MV4-11 and THP1 lines transduced with shCYTH4 was examined by flow cytometry, and this could be alleviated by overexpressing PIK3R5 with a lentivirus packaging system (n = 3). Apoptosis was determined with flow cytometry following annexin V-APC and DAPI staining (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(F) Immunoblotting showing that overexpression of PIK3R5 in CYTH4-silenced MV4-11 and THP-1 cells counteracted the reduction of p-AKT, c-Myc, and Bcl-2. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. Sc, scramble control shRNA; shCYTH4, CYTH4 shRNA; pCDH, pCDH-CMV-MCS-EF1-mCherry-P2A-Puro vector.

The activation of AKT by SC-79 rescued the AML cell growth defect caused by CYTH4 silencing

Another rescue experiment was conducted using an AKT activator, SC-79. As illustrated in Figure 7A, the immunoblotting assay revealed a significant increase in AKT phosphorylation on serine 473 in shCYTH4-transduced MV4-11 and THP-1 cells following treatment with 10 μM SC-79 for 24 h, indicating a restoration of AKT activity. Furthermore, treatment with SC-79 effectively reversed the inhibitory effects of CYTH4 silencing on cell proliferation and colony formation (p < 0.001) (Figures 7B and 7C). Additionally, SC-79 treatment mitigated AML cell apoptosis induced by CYTH4 silencing (Figure 7D). Correspondingly, SC-79 treatment reversed the downregulation of anti-apoptotic proteins c-Myc and Bcl-2 induced by CYTH4 silencing in MV4-11 and THP-1 cells (Figure 7A).

Figure 7.

Activation of AKT by SC-79 rescues AML cell growth defects induced by CYTH4 silencing

(A) Immunoblot confirmation for restoration of AKT activity and the downregulated expression of c-Myc and BCL-2 induced by CYTH4 silencing following SC-79 treatment.

(B) SC-79 treatment rescued the growth defect phenotype induced by CYTH4 silencing in MV4-11 and THP-1 cells. The cells were treated with 10 μM SC-79 for 0, 24, 48, 72, and 96 h. CCK-8 assay was performed to evaluate cell viability (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(C) The colony formation defect caused by CYTH4 silencing was rescued by SC-79 in MV4-11 and THP-1 cells (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test.

(D) CYTH4 silencing-induced apoptosis were rescued by SC-79 in MV4-11 and THP-1 cells. Apoptosis was determined with flow cytometry following annexin V-APC and 7-AAD staining (n = 3). Data are represented as mean ± SEM. Statistical significance was assessed by ANOVA with Tukey’s multiple comparison test. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. Sc: scramble control shRNA; shCYTH4, CYTH4 shRNA.

Discussion

CYTH4, a member of the cytohesin family, was initially cloned from the human frontal cerebral cortex and has been shown to exhibit ARF-directed guanine-nucleotide-exchange protein (GEP) activity.²³ Sharing structural and functional similarities with other cytohesins, CYTH4 features an N-terminal coiled-coil domain that facilitates homodimerization, a central Sec7 catalytic domain with GEP activity, and a C-terminal pleckstrin homology domain responsible for membrane binding.^23,24 Recent studies have identified that cytohesin family members were overexpressed in certain tumor types and correlated with the poor prognosis of tumor patients. Specifically, CYTH1 has been found to be significantly upregulated in AML and independently associated with a poor prognosis.¹² Additionally, cytohesin-2 has been shown to exhibit high expression levels in human lung cancer, leading to shorter OS and disease-free survival.²⁵ Furthermore, upregulation of cytohesin-4 has been observed in glioma-associated M2 macrophages, contributing to pyroptosis and ultimately resulting in a poor prognosis.¹⁸ CYTH4 also exhibits the upregulated expression in ovarian cancer compared to normal ovary tissues, leading to a poorer OS and progression-free survival.⁵ To date, the role of CYTH4 in AML and its underlying molecular mechanisms have yet to be fully elucidated.

Here, in silico analysis suggests that CYTH4 is highly expressed in AML cells and associated with a worse prognosis (Figures 1B and 1C). Our RT-qPCR and immunoblotting assays further confirm the elevated expression of CYTH4 in bone marrow mononuclear cells from AML patients compared to those from healthy donors (Figures 1D and 1E), indicating an oncogenic role of CYTH4. Loss-of-function assays demonstrated that CYTH4 silencing inhibited the growth and colony formation of AML cells by inducing cell apoptosis and G0/G1 cell-cycle arrest (Figures 2C–2F). These findings align with a recently published literature.¹⁹ Additionally, our study revealed that CYTH4 silencing increased the sensitivity of MV4-11 and THP-1 cells to Ara-C (Figure 2H), indicating a potential benefit for AML patients receiving chemotherapy in combination with CYTH4 inactivation, as seen in the context of cytohesin-1 targeting to overcome resistance to ABT-199 by downregulating MCL1 expression.¹² A gain-of-function assay revealed that CYTH4 overexpression did not affect AML cell proliferation, colony formation and apoptosis in vitro (Figures 4C–4E). This may be because CYTH4’s endogenous expression is already saturated, making further overexpression ineffective in enhancing downstream signaling. This mechanism is seen in some cancers, such as lymphoma, where B cell receptor-associated kinases, despite being therapeutic targets, lose impact on tumor progression beyond a certain expression level.²⁶ The xenograft model also demonstrated that CYTH4 silencing delayed the progression of AML and extramedullary invasion in vivo (Figures 5A–5D). The mice transplanted with CYTH4-silenced THP1 cells exhibited a significant reduction in tumor burden and improved OS compared to those transplanted with Sc shRNA-transduced THP1 cells. Given that “cell adhesion molecules” were among the top five deregulated pathways and CYTH1 is involved into cell-adhesion-mediated leukemogenesis,¹² a cell homing assay was conducted. The results indicated that CYTH4 silencing led to a decreased homing of cells to the bone marrow and spleen (Figure 5E), suggesting that CYTH4-mediated cell adhesion may contribute to the observed in vivo phenotypes, such as homing, engraftment, progression, etc. The c-Myc protein plays a key role in regulating various cellular functions, such as cell cycle progression, apoptosis, proliferation, angiogenesis, and immortalization.^27,28 Cyclin D1 promotes cell proliferation by facilitating the G1/S phase transition,²⁹ while cyclin B1 is essential for regulating the G2/M phase transition.³⁰ Our findings demonstrated that CYTH4 silencing resulted in the decreased expressions of c-Myc, cyclin B1, cyclin D1, and BCL-2 proteins in AML cells (Figure 2G), suggesting that CYTH4 plays a role in promoting G1/S- and G2/M-phase transitions by upregulating the expression of cell cycle regulators, while also inhibiting cell apoptosis through the downregulation of BCL-2 in AML cells.

The RNA-seq assay has been extensively utilized to identify novel biomarkers and molecular mechanisms in cancer research.³¹ To better elucidate the molecular mechanism by which CYTH4 silencing inhibits AML cell growth, a comprehensive transcriptome analysis was conducted to identify genes with differential expression between shCYTH4- and Sc shRNA-transduced MV4-11/THP-1 cells, and to investigate the potential pathways affected by CYTH4. Interestingly, the deactivation of the PIK3R5-mediated PI3K-AKT signaling pathway upon CYTH4 silencing was identified through RNA-seq analysis and subsequently confirmed at the protein level via immunoblotting (Figures 3A–3H). This finding aligns with previous researches, such as the essential role of CNK1/cytohesin-2 in activating the insulin-mediated PI3K/AKT signaling cascade in various epithelial tumor cell lines, including HeLa, HepG2, and MCF7 cells.³² The silencing of cytohesin-1 by siRNA has been shown to decrease IGF1R/AKT signaling in prostate cancer.²¹ A recent research utilizing SecinH3 (a cytohesin-specific inhibitor) has demonstrated that the activity of cytohesin-ARFs is essential for the activation of the insulin-mediated AKT pathway.³³

Based on their similarities in genomic structures and amino acid sequences, ARFs are classified into three classes: class I (ARF1, ARF2, and ARF3), class II (ARF4 and ARF5), and class III (ARF6).³⁴ Similar to other small GTPases, ARFs function as molecular switches for signaling pathways by transitioning between active (GTP-bound) and inactive (GDP-bound) states.³⁵ Cytohesins, members of the GEF family for ARF small GTPases, facilitate the exchange of GDP for GTP to initiate signaling by ARFs, which subsequently recruit the downstream effectors.³⁶ The specific substrates of the cytohesin family remain a topic of ongoing discussion. While it is widely accepted that ARF1 is the most preferred substrate for cytohesins, recent accumulating evidences suggest that ARF6 also serves as the major substrate for cytohesins in various physiological and pathological contexts.¹¹ In our study, it was noted that CYTH4 exhibited GEF activity with ARF6 but failed to activate ARF1 in vitro (Figure 3I). Aberrant activation of ARF6 in human melanoma accelerated metastasis through the activation of PIK3R1/PI3K/AKT pathway.³⁷

In the rescue experiments, overexpression of PIK3R5 and treatment with SC-79, a potent AKT activator, markedly reversed the effects of CYTH4 silencing on the proliferation, colony formation, and apoptosis of MV4-11 and THP-1 cells (Figures 6 and 7). These findings suggest that CYTH4 promotes AML cell proliferation by activating the PIK3R5/PI3K/AKT pathway.

In conclusion, our research has demonstrated that CYTH4 is upregulated in AML cells and that CYTH4/ARF6 promotes AML cell viability by activating the PIK3R5-mediated PI3K/AKT pathway. These findings indicate that CYTH4 may represent a promising therapeutic target for the treatment of AML.

Limitations of the study

The current study acknowledges certain limitations and necessitates further research to elucidate the molecular mechanisms by which CYTH4/ARF6 contributes to the upregulation of PIK3R5. Additionally, this investigation does not explore the effects of CYTH4 on primary AML cells. Future research should incorporate an inducible gene knockdown mouse AML model to address these gaps.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Xiao-Li Chen (fsgzyy2805@ncu.edu.cn).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •RNA-seq data generated in this study have been deposited at the National Center for Biotechnology Information Sequence Read Archive data and are publicly available as of the date of publication. Accession numbers are listed in the key resources table. All data reported in this paper will be shared by the lead contact upon request.
• •This paper does not report the original code.
• •All other data reported in this paper will be shared by the lead contact upon request.

Acknowledgments

We also sincerely thank Ms. Ping Ye and Ms. Yulan Liu for their help with the FACS assays in this study. We also thank Xiaoming Feng for kindly providing NB4 cells. This work was supported by grants from the National Natural Science Foundation of China (81960035), Jiangxi Provincial Health Technology Project (202310096 and 202310097), Ganzhou Municipal Science and Technology Project (2023LNS17475, 2022--ZD1368, and GZ2024YLJ014), Tianjin Municipal Science and Technology Commission (21JCQNJC 01750).

Author contributions

X.-F.Q., X.-J.X., Q.N., and X.-L.C., conception of the project; X,-F.Q., C.-M.H., X.-L.C., and Q.N., methodology; X,-F.Q., X.-L.C., Y.Y., Y.-M.Z., M.-X.Z., and M.Z., investigation; X.-L.C., writing – original draft and writing – review & editing; X.-L.C., M.-X.Z., C.-M.H., M.Z., and Q.N., funding acquisition; X.-F.Q., visualization; X.-L.C. and J.-D.Z., supervision.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
HRP-conjugated Goat Anti-Mouse IgG(H + L)	Proteintech	Cat no. SA00001-1; RRID: AB_2722565
HRP-conjugated Goat Anti-Rabbit IgG (H + L)	Proteintech	Cat no. SA00001-2; RRID: AB_2722564
Rabbit polyclonal anti-CYTH4	Abcam	Cat no. ab227003
Mouse monoclonal anti-AKT (CloneNo. 2C5D1)	Proteintech	Cat no. 60203-2-Ig; RRID: AB_10912803
Mouse monoclonal anti-p-AKT(Ser473) (CloneNo. 1C10B8)	Proteintech	Cat no. 66444-1-Ig; RRID: AB_2782958
Rabbit polyclonal anti-p-PI3K p85 alpha (Tyr607)	Affinity	Cat no. AF3241; RRID: AB_2834667
Rabbit polyclonal anti-PI3K p85 alpha	Affinity	Cat no. AF6241; RRID: AB_2835340
Rabbit polyclonal anti-BAX	Proteintech	Cat no. 50599-2-Ig; RRID: AB_2061561
Rabbit polyclonal anti-c-MYC	Proteintech	Cat no. 10828-1-AP; RRID: AB_2148585
Rabbit polyclonal anti-CYCLIN B1	Proteintech	Cat no. 55004-1-AP; RRID: AB_10859790
Rabbit polyclonal anti-CYCLIN D1	Proteintech	Cat no. 26939-1-AP; RRID: AB_2880691
Rabbit polyclonal antii-BCL-2	Proteintech	Cat no. 12789-1-AP; RRID: AB_2227948
Rabbit polyclonal antii-ARF1	Proteintech	Cat no. 10790-1-AP; RRID: AB_2227609
Mouse monoclonal anti-ARF6	Cytoskeleton	Cat # ARF06
Rabbit polyclonal anti- PI3KR5 (p101)	Affinity	Cat no. AF9158; RRID: AB_2843348
Rabbit polyclonal anti-GAPDH	Proteintech.	Cat no. 10494-1-AP; RRID: AB_2263076
APC-conjugated mouse monoclonal anti-human CD45	eBioscience	Cat no. 17-9459-42; RRID: AB_10718532
Bacterial and virus strains
Trans5α Chemically Competent Cell	Transgen	Cat no. CD201-01
LV3 (H1/GFP & Puro)	This paper	N/A
pCDH-CMV-CYTH4-EF1-CopGFP-T2A-puro	This paper	N/A
pCDH-CMV-PIK3R5-EF1-mCherry-P2A-Puro	This paper	N/A
Biological samples
Bone marrow	This paper	N/A
Chemicals, peptides, and recombinant proteins
PI	Servicebio	Cat no. G1021-10ML
Trizol	Invitrogen	Cat no. 15596026CN
Ficoll Hypaque	Solarbio	Cat no. P8610
FBS	Gibco	Cat no. A5256701
Ara-C	Pfizer	Cat no. HY-13605
SC-79	GLPBIO	Cat no. GC11645
RPMI-640	Hyclone	Cat# SH30027.01
DMEM/high glucose	Hyclone	Cat# SH30023.01
IMDM	Hyclone	Cat# SH30228.01
Penicillin/streptomycin	Solarbio	Cat No. P1400
Polybrene	GLPBIO	Cat no. GC19206
Puromycin	GLPBIO	Cat no. GC32186
Polyethyleneimine	Polysciences	Cat# 24314
Low-melting-point agarose	Yeasen	Ca# 23966
Protease and phosphatase inhibitor cocktail	YamayBio	Cat no.BH5483
RIPA Buffer	Thermo Scientific	Cat no. 89901
CCK8	GLPBIO	Cat no. GK10001
Critical commercial assays
Annexin V-APC/7-AAD apoptosis detection kit	Elabscience	Cat no. E-CK-A218
Annexin V-APC/DAPI apoptosis detection kit	Elabscience	Cat no. E-CK-A258
MycoAlert mycoplasma detection kit	Lonza	Cat no. LT07–118
The cell cycle assay kit	Elabscience	Cat no. E-CK-A351; Cat no. E-CK-A353
QIAGEN Plasmid Plus Midi Kit	Qiagen	Cat no. 2699
Arf6 Pull-down Activation Assay Biochem Kit	Cytoskeleton	Cat no. BK033-S
BCA Protein Assay Kit	Thermo Fisher	Cat No. 23225
Deposited data
RNA sequencing data	This paper	GEO accession: GSE294040
Original western blot images	This paper
Experimental models: Cell lines
HEK-293T cells	CAS Cell Bank	N/A
MV4-11 cell line	CAS Cell Bank	N/A
THP-1 cell line	CAS Cell Bank	N/A
Experimental models: Organisms/strains
NOD-Prkdcem26Cd52il2rgem26Cd22/Nju (NCG) mice	GemPharmatech	N/A
Oligonucleotides
Primers for RT-qPCR	Table S4	N/A
Recombinant DNA
pMD2.G	Addgene	Cat#12259
psPAX2	Addgene	Cat#12260
LV3 (H1/GFP & Puro)-shCYTH4	GenePharma	N/A
pCDH-CMV-CYTH4-EF1-CopGFP-T2A-puro	WZBIO	N/A
pCDH-CMV-PIK3R5-EF1-mCherry-P2A-Puro	WZBIO	N/A
Software and algorithms
GraphPad Prism 8.0	GraphPad	http://www.graphpad.com/
FlowJo	v10	https://www.flowjo.com
Illustrator	Adobe	http://www.adobe.com/

Experimental model and study participant details

Animals

Xenograft models of THP-1 cells were performed with six-week-old female NOD-Prkdc^em26Cd52 il2rg^em26Cd22/Nju (NCG) mice (GemPharmatech, China) as described previously³⁸ in accordance with the approved protocol of the Ethical Committee of the Affiliated Ganzhou Hospital, Jiangxi Medical College, Nanchang University (No TY-DKY2023 -003-01). The mice were housed in a specific pathogen-free animal facility with a 12-h light/dark cycle, 40%–60% humidity, and a temperature of 22°C, with ad libitum access to food and water. THP-1 cells were injected into NCG mice at a concentration of 4.0 × 10⁵ cells per mouse via tail vein injection. The recipient mice were randomly allocated into two groups through a random selection process, consisting of the control group (Mice transplanted with THP-1 cells carrying Sc shRNA) and the shCYTH4 experimental group (Mice transplanted with THP-1 cells carrying shCYTH4-2). Changes in the body weight and disease symptoms of the mice were closely observed throughout the progression of the disease. Euthanasia of the mice was carried out through CO2 asphyxiation upon the display of terminal leukemic symptoms by any individual. The proportions of the engrafted cells in peripheral blood, spleen, liver, and bone marrow were determined using a flow cytometer with APC-conjugated anti-human CD45 antibody (Cat no. 17-9459-42, eBioscience). The development of extramedullary leukemia was assessed through histochemical analysis, utilizing freshly harvested mouse livers, spleens, and femurs that were fixed in 4% paraformaldehyde for over 24 h. Following fixation, the tissues underwent dehydration in ethanol and embedding in paraffin. Subsequently, the paraffin blocks were sectioned into 5 μm-thick slices and stained with hematoxylin-eosin. Two groups of mice, each consisting of ten individuals, were utilized for the survival assay. Mice were monitored until euthanasia and scored for OS.

Human cells

This study included 50 subjects, consisting of 35 de novo adult AML patients and 15 healthy donors with iron deficiency anemia and without abnormal bone marrow hematopoiesis. Diagnosis of AML patients was based on French-American-British criteria. This study received approval from the Ethical Committee of the Affiliated Ganzhou Hospital of Nanchang University. Prior to genetic research, all participants provided written informed consent. The laboratory and clinical characteristics of the patients are detailed in Table S5.

Cell culture

All cell lines were rigorously screened for mycoplasma contamination using the MycoAlert mycoplasma detection kit (Lonza) and maintained in a controlled environment at 37°C with 5% CO₂. Human embryonic kidney (HEK)-293T cells were cultured in High-glucose Dulbecco’s modified Eagle’s medium (Hyclone) with 10% (v/v) heat-inactivated fetal bovine serum (FBS, Gibco BRL), 100 U/mL penicillin, and 100 μg/mL streptomycin (Solarbio, China). MV4-11, NB4, THP-1, and U937 cells were grown in Roswell Park Memorial Institute (RPMI) −1640 medium (Hyclone) supplemented with 10% FBS, whereas KG1 and HL-60 cells were cultured in Iscove’s modified Dulbecco’s medium (Hyclone) containing 20% FBS.

Method details

Bioinformatics screening of acute myeloid leukemia survival-related genes

Three publicly available datasets related to AML were obtained from Gene Expression Profiling Interactive Analysis version 2 (GEPIA2) and Gene Expression Omnibus (GEO) with accession numbers GSE67936 and GSE65409. The identification of DEGs was conducted through the utilization of the "Differential Expression Analysis" section of the GEPIA2 Website (http://gepia2.cancer-pku.cn/#degenes) or the GEO2R tool (https://www.ncbi.nlm.nih.gov/geo/geo2r). DEGs in the GSE67936 and GSE65409 datasets were defined as genes with an adjusted p-value <0.05 and an absolute Log2FC > 1. In the GEPIA2 dataset, DEGs were determined as genes with an adjusted p-value <0.01 and an absolute Log2FC > 4. Survival analyses of hub genes were conducted using the online tools GEPIA2 (http://gepia2.cancer-pku.cn/#survival), UALCAN (http://ualcan.path.uab.edu) and TIMER (https://cistrome.shinyapps.io/timer/). Gene expression correlation analyses were performed using GEPIA2 (http://gepia2.cancer-pku.cn/#correlation).

Specimen collection

Bone marrow samples were collected using EDTA tubes before chemotherapy. Mononuclear cells were isolated through Ficoll Hypaque density gradient centrifugation (Solarbio) and stored at −80°C.

Virus production and cell infection

Virus production and cell infection were conducted in accordance with previously published protocols.³⁹ Briefly, Lentiviral vectors encoding shRNAs or cDNAs (GenBank: NM_013385.5 for human CYTH4, GenBank: NM_001142633.3 for human PIK3R5) were co-transfected with packaging vector pSPAX2 (Addgene) and pMD2.G (Addgene) into HEK-293T cells using PEI transfection reagent (Polysciences, USA). MV4-11 and THP-1 cells underwent spin-infection twice with the viral supernatant containing 4 μg/mL polybrene (GLPBIO), followed by puromycin selection (2.5 μg/mL, GLPBIO) for 7 days.

Cell proliferation assay

Cell viability was assessed using the cell counting kit 8 reagent (CCK-8, GLPBIO). MV4-11 and THP-1 cells in the logarithmic growth phase were seeded into 96-well plates at a density of 2 × 10⁴ cells per well and cultured for 1–4 days. Following incubation, 10 μL of CCK-8 solution was added to each well and the cells were further incubated for 2 h. Subsequently, the plates were analyzed using a microplate reader (Allsheng Instruments Co., Ltd., Hangzhou, China) at 450 nm. The inhibitory effects of CYTH4 on the proliferation of AML cells were assessed using a CCK-8 assay after exposure to different concentrations of Ara-C over a 48-hour period. The IC 50 values were determined based on the dose-response curve.

Colony formation assay

The colony formation assay was conducted utilizing a standard double-layer soft agar method in 6-well plates, with the bottom layer consisting of 2 mL of 0.7% (w/v) agarose and the top layer consisting of 1 mL of 0.35% (w/v) agarose. A total of 3,000 cells were suspended in 1 mL of RPMI-1640 medium containing 0.35% low-melting-point agarose (Yeasen, Shanghai, China) and 10% FBS, and subsequently plated on top of the solidified 0.7% agarose base layer. Following a 14-day incubation, the colonies were manually enumerated and captured using an Olympus IX73 inverted microscope (Olympus, Japan) equipped with an Olympus DP73 CCD camera.

Apoptosis and cell cycle assay

Apoptosis was quantified utilizing the Annexin V-allophycocyanin (APC)/7-aminoactinomycin (7-AAD) or DAPI apoptosis detection kit (Elabscience, China) in accordance with the manufacturer’s instructions. Specifically, 1.0 × 10⁵ cells were resuspended in 100 μL of Annexin V Binding Buffer, and then dual-stained with 2.5 μL of Annexin V-APC and 2.5 μL of 7-AAD or DAPI for 15 min at room temperature (25°C) in the absence of light. Subsequently, the stained cells were analyzed using FACSCanto II (Becton Dickinson) and data analysis was performed using FlowJo software (TreeStar).

The cell cycle was assessed utilizing the cell cycle assay kit (Elabscience, China) as per the manufacturer’s instructions. Briefly, 2 × 10⁵ cells were rinsed with phosphate-buffered saline and fixed in 80% ethanol overnight at −20°C. Subsequently, the cells were co-incubated with 100 μL of RNase A at 37°C for 30 min and Propidium Iodide (PI, 40 μg/mL) at 2°C–8°C for 30 min in the absence of light. The stained cells were then analyzed using FACSCanto II (Becton Dickinson), and data analysis was carried out using FlowJo software (TreeStar).

RNA sequencing

Construction of the cDNA library, RNA-seq, and subsequent Gene Ontology KEGG analyses were carried out by Lianchuan Biotechnology (Hangzhou, China) following the established protocols.⁴⁰ Candidate DEGs were identified based on an FDR-adjusted p-value <0.05 and an absolute log2FC > 1 using the R package edgeR.

Quantitative reverse-transcription polymerase chain reaction (RT-qPCR)

Total RNA was isolated using TRIzol (Invitrogen) followed by chloroform extraction, isopropanol precipitation, and ethanol wash. Subsequently, 1 μg of total RNA was reverse transcribed into first-strand cDNA using reverse transcriptase (AMV, TaKaRa, Japan) in a 20 μL reaction volume as per the manufacturer’s protocol. For mRNA expression quantification of the target genes, 20 ng of each cDNA template was subjected to triplicate amplification using gene-specific primers (10 μL Master Mix, 200 nM each primer) and SYBR Green PCR Master Mix (Transgen, Beijing, China). β-actin served as the internal control for normalization. The 2^−ΔΔCT method was employed for the analysis of PCR products.

Western blot

Total proteins were extracted from cells using radio-immunoprecipitation assay (RIPA) lysis buffer containing a protease and phosphatase inhibitor cocktail (GLPBIO). The protein concentration was determined using the Bicinchoninic Acid Protein Assay Kit (Pierce). Proteins were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels with a loading range of 20–40 μg per lane and transferred to polyvinylidene fluoride membranes (Millipore, USA). Blocking was performed using 5% skimmed milk in Tris-buffered saline with 0.5% Tween 20. The membranes were incubated with primary antibodies overnight at 4°C, followed by horseradish peroxidase-conjugated secondary antibodies for 2 h at room temperature. Proteins were detected using the SuperSignal chemiluminescence kit (Proteintech) and imaged with the Tanon 5200 Multi-intelligent imaging system (Tianneng, Shanghai, China).

GGA3-PBD pull down assay

The active form of ARFs can be isolated through interaction with the Golgi-associated, gamma adaptin-ear-containing, ARF binding protein 3 protein binding domains (GGA3-PBD, amino acids 1–316), with subsequent immunoblotting using anti-ARF6 or anti-ARF1 antibodies to confirm specificity for ARF6-GTP or ARF1-GTP. The activities of ARF6 and ARF1 were assessed using the Pull-Down Kit (Cytoskeleton, Denver, CO, USA) according to the manufacturer’s protocol. Briefly, cell lysates (500 μg) were incubated with 20 μL of GGA3-PBD beads at 4°C for 1 h with shaking, followed by centrifugation at 3,000g for 2 min at 4°C. The beads were then washed three times with 600 μL of wash buffer before mixed with 20 μL of 2 × sample buffer. The beads underwent a 2-min boiling, followed by an immunoblotting assay to assess the activities of ARF6 and ARF1.

In vivo homing assay

Homing assays were conducted as previously described.⁴¹ Briefly, a total number of 1 × 10⁷ THP1 cells with stable GFP expression were transplanted into NCG mice. Homing of the THP1 cells in the bone marrow and spleen were measured at 16 h after transplantation by flow cytometric analysis.

Quantification and statistical analysis

The experiments were performed in triplicate or more, and the results presented are representative. Statistical analyses were carried out using GraphPad Prism 8.0 software (GraphPad Software, Inc.) and SPSS 19.0 software (SPSS Inc., Chicago, USA). The study presents quantitative data in the form of mean values ±standard deviation (SD) or standard error (SEM). Kaplan-Meier curves were generated using the log rank test to compare the OS between the two groups. The normality of the data and homogeneity of variance were assessed using the Shapiro-Wilk test and Levene’s test, respectively. A Student’s unpaired t test (for two-group comparisons) and analysis of variance (ANOVA) with Tukey’s multiple comparison test (for multiple -group comparisons) were employed to determine statistical significance when the data met the assumptions of normality and homogeneity of variance. p < 0.05 was deemed as a significant difference.

Published: May 12, 2025