Relationship between specific microRNA expression and radioresistant conditions in HL60 acute myeloid leukemia cells

Introduction

Leukemia disrupts the homeostasis of the hematopoietic system and causes abnormal proliferation of blast cells (1). Acute myeloid leukemia (AML), including its M2 subtype, reduces survival in adults; therefore, establishing effective treatment strategies is critical for improving prognosis (2,3). Chemotherapy, such as the ‘7+3’ regimen, remains the standard first-line treatment for AML. However, in some relapsed or refractory cases, total body irradiation (TBI) is used as part of conditioning regimens before hematopoietic stem cell transplantation to eradicate residual leukemic cells (4,5). Repeated exposure to radiation in such settings may, in rare cases, contribute to the emergence of leukemic cells with reduced radiosensitivity, potentially impeding complete remission. Although the concept of ‘radiation-resistant AML’ is not widely recognized in clinical settings, where resistance is more commonly associated with chemotherapy, TBI has been implicated as a contributing factor in therapy-related AML (6). Furthermore, a phase II clinical trial evaluating treosulfan, fludarabine, and low-dose TBI in pediatric and young adult AML/MDS patients highlighted the clinical relevance and complex outcomes of TBI-based regimens, including relapse and non-relapse mortality (7). While radiation resistance is not a routine concern in AML treatment, these findings support the relevance of exploring the molecular basis of radiation response.

In the present study, we employed an in vitro AML model to investigate the acquisition of radioresistance. Specifically, we previously established a radioresistant AML cell line (Res-HL60) derived from HL60 cells, an AML M2 subtype cell line of human origin, and investigated the molecular mechanisms underlying radioresistance. Through repeated irradiation, the Res-HL60 cells acquired smaller morphology, increased proliferative potency, and showed elevated expression of CD38 cell surface antigen and several mRNAs (8–12). However, the upstream regulatory mechanism governing these gene expression changes remains unclear, necessitating further investigation.

MicroRNAs (miRNAs), a class of noncoding RNAs that regulate mRNA translation, have been implicated in cancer progression and cell behavior, and their expression patterns are increasingly recognized as diagnostic and prognostic biomarkers (13,14). Additionally, miRNAs have emerged as potential therapeutic targets in cancer. Despite this, the profile of miRNA expression in radioresistant AML cells such as Res-HL60 has not been previously reported. Given that chromosomal abnormalities, epigenetic dysregulation, and impaired miRNA biogenesis, potentially triggered by frequent radiation exposure, can all influence miRNA expression (15), investigating these changes may help clarify the molecular basis of acquired radioresistance. Therefore, in this study, we performed transcriptome analysis focusing on miRNA expression in Res-HL60 cells to explore mechanisms associated with radioresistance in AML.

Materials and methods

Leukemia cell line and culture

The human AML cell line HL60 (wild-type: Wt-HL60) was obtained from the RIKEN BioResource Center. The Res-HL60 cell line was generated by exposing Wi-HL60 cells to 4-Gy irradiation per week for 4 weeks. Wt-HL60 and Res-HL60 cells were cultured in RPMI-1640 medium (Thermo-Fisher Scientific, Inc.) supplemented with 10% heat-inactivated fetal bovine serum (Japan Bioserum, Inc.) and 1% penicillin/streptomycin (Thermo-Fisher Scientific, Inc.) in a humidified atmosphere containing 5% CO2 at 37°C. In this study, the control group was defined as non-irradiated Wt-HL60 cells cultured under these standard conditions (5% CO2 and 95% air). The characteristics of Res-HL60 cells, including higher proliferative capacity, smaller cell size and altered chemical reactivity, have been reported previously by our group (8–12). In the present study, a time-course analysis of morphological changes in Res-HL60 cells was conducted (Fig. 1). Cells were initially seeded at a density of 2×105 cells/ml on Day-1. X-ray irradiation was performed on Day 0. On Day 2 (i.e., the third day after initial seeding), cells were passaged and reseeded at 2×105 cells/ml. When the cell density remained below 2×105 cells/ml, only medium replacement was performed. No additional supplements (e.g., non-essential amino acids, sodium pyruvate) were added. To calculate the cumulative total viable cell number including all passages, the following formula was used: Cumulative cell number (/ml)=A × [(2×105) + B]/(2×105), where ‘A’ is the viable cell number (cells/ml) measured up to Day 2 (before passaging), and ‘B’ is the net increase in viable cell number (cells/ml) from Day 3 onward, relative to the reseeded density (2×105 cells/ml). B=0 for Days 0–2. This method allowed us to reflect the overall proliferation dynamics across reseeding and media replacement steps. The updated y-axis label in Fig. 2A and B reflects this definition. To confirm the identity and authenticity of the cell lines used in this study, short tandem repeat (STR) profiling was performed. The STR profiles of both Wt-HL60 and Res-HL60 showed a perfect match with the reference profile for HL60 (Cellosaurus ID: CVCL_0002, URL: http://www.cellosaurus.org/CVCL_0002), as provided by the RIKEN BioResource Center, indicating no evidence of contamination or misidentification. The STR profiles are provided in Fig. S1 and Table SI.

Figure 1. Characterization of Res-HL60 cell. (A) Cellular morphology and subpopulation characteristics were analyzed using flow cytometry. The horizontal and lateral axis indicate the mean size of the cell (EV) and the intracellular complexity (SS), respectively. (B) The EV histogram of each type of cell. (C) The time course of the CD38 expression intensity and (D) dot plots on the CD38-CD45 cell surface antigen after the exposure of 4-Gy X-ray is shown. Non-irradiated HL60 cells cultured under standard normoxic conditions (5% CO2 and 95% air) were used as the control group in all analyses. Wt, wild type; Res, resistant; EV, electronic volume; SS, side scatter.

Figure 2. Cell proliferation curve and DNA fragmentation analysis. (A) Cumulative proliferation of Wt-HL60 and Res-HL60 cells under non-irradiated conditions. Cells were seeded at 2×105 cells/ml on Day-1 and cultured without irradiation. On Day 2, the cells were passaged to maintain the same seeding density. If cell density remained below this threshold, only the medium was replaced without passaging. The cumulative number of viable cells was calculated using the formula: Cumulative cell number (/ml)=A × [(2×105) + B]/(2×105), where ‘A’ is the viable cell number up to Day 2, and ‘B’ is the net increase from Day 3 onward relative to the reseeding density. (B) Cumulative proliferation of Wt-HL60 and Res-HL60 cells under 4-Gy irradiated conditions. The same seeding and passaging protocol as in (A) was followed, except that the cells were exposed to 4 Gy of X-ray irradiation on Day 0. Cell proliferation was then monitored to assess the impact of irradiation on the growth of wild-type and resistant HL60 cells. Cumulative viable cell numbers were calculated using the same formula as in (A). (C) Representative images of the comet assay showing DNA damage 24 h after X-ray exposure at doses of 1–4 Gy. (D) Quantification of DNA damage based on tail moment calculated from the comet assay shown in (C). Data are shown as the mean ± standard error of the mean from four independent experiments. *P<0.05, **P<0.01 vs. Wt-HL60, determined by unpaired Student's t-test. Wt, wild type; Res, resistant.

Irradiation

X-ray irradiation (150 kVp, 20 mA with 0.5-mm aluminum and 0.3-mm copper filters) was performed using an X-ray generator (MBR-1520R-3; Hitachi Medical Co., Ltd.) at a distance of 45 cm between the focus and target. The dose was monitored during irradiation using a thimble ionization chamber set next to the sample. The dose rate was 1 Gy/min. The cultured cells were exposed to X-rays as previously described (8–12).

Flow cytometry analysis

Flow cytometry was performed using Cell Lab Quanta™ Sc MPL system (Beckman Colter Immunotech) to assess cell size, granularity, and CD38 surface expression. Unlike conventional flow cytometers that use forward scatter to estimate cell size, the Cell Lab Quanta utilizes Electric Volume (EV), which is based on the Coulter principle, to directly measure cell size via changes in electrical resistance as cells pass through an aperture. This method allows for accurate assessment of cell volume independent of optical scattering. Granularity, representing internal cellular complexity, was evaluated based on side scatter (SSC) intensity, as is standard in flow cytometry. CD38 expression was also assessed due to its dual relevance as both a hematopoietic differentiation marker and a potential indicator of acquired radioresistance. CD38 is a well-characterized surface antigen known to be markedly upregulated during granulocytic differentiation of HL60 cells (16). Furthermore, our previous studies demonstrated sustained CD38 expression in Res-HL60 cells even in the absence of differentiation-inducing agents, suggesting a role in the radioresistant phenotype (8–12). Based on these findings, CD38 was utilized in this study both as a marker of differentiation and as a surrogate indicator of radiation resistance.

DNA fragmentation

The Comet Assay Single Cell Gel Electrophoresis Assay (cat. no. 4250-050-K; TREVIGEN, Inc., Gaithersburg, USA) was used to evaluate the DNA fragments of the cells irradiated with X-rays. One day after seeding in a 60-mmφ culture dish at a concentration of 1.0×105 cells/ml, the cells were irradiated with 1–4 Gy of X-rays, and the cells were collected after 24 h. The collected cells were fixed on glass slide (Matsunami Grass Inc., Osaka, Japan) using LM Agarose (TREVIGEN Inc.). The glass slide to which the sample was attached was immersed in an alkaline solution [200-mM NaOH, 1-mM ethylenediaminetetraacetic acid (EDTA)], and electrophoresis (21 V, 3 A, 30 min, 4°C) was performed. After washing the glass slide with distilled water and 70% EtOH and drying, it was stained with 100 µl of a staining solution containing a mixture of SYBR Green (TREVIGEN Inc.) and TE Buffer (10-mM Tris, 1-mM EDTA). After drying, the tail moment was observed using a fluorescence microscope (OLYMPUS Inc.). Images were taken using Comet Assay IV (Instem, Conshohocken, USA) for observation purposes. Tail moment was calculated using the formula: Tail Moment=Tail Length × % Tail DNA/100, using Comet Assay IV software. This calculation reflects both the extent and the intensity of DNA fragmentation in individual cells.

Extraction of total RNAs from cells

Total intracellular RNAs (four replicates performed in parallel) were extracted using ISOGEN II (Nippon Gene Co., Ltd.) according to the manufacturer's instructions. The purity and concentration of extracellular RNAs were assessed using a NanoDrop spectrophotometer (NanoDrop Technologies; Thermo-Fisher Scientific, Inc.). The RNA samples had 260/280-nm absorbance ratios of 1.8–2.0. Furthermore, the peaks of the small RNAs were confirmed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc.).

Microarray analysis of miRNAs

SurePrint G3 human miRNA 8×60-K Microarray (Agilent Technologies Inc.) was used to analyze intracellular microRNA expression. Cyanine 3 labeling was performed with 1 ng of intracellular microRNA using an miRNA Complete Labeling and Hybridization kit (Agilent Technologies Inc.), following the manufacturer's instructions. After hybridization and washing of the microarray slide, fluorescence image scanning was performed using a SureScan G 2600 D (Agilent Technologies Inc.). The expression data were processed using GeneSpring GX14.5 (Agilent Technologies, Inc.) to normalize the quantiles of all values in the respective microarrays. The obtained microarray data were registered with the Gene Expression Omnibus (GEO) (GSE285934).

Reverse transcription-quantitative PCR (RT-qPCR) analysis

First-strand cDNA was synthesized using Mir-X™ miRNA First-Strand Synthesis Kit (Takara Bio Inc., Otsu, Japan; Cat. Nos. 638313) according to the manufacturer's protocol (Protocol No. PT5001-1). MiRNA expression was assessed by qPCR using TB Green Advantage qPCR Premix (Takara Bio Inc., Cat. Nos. 639676) on a SmartCycler® II system (Takara Bio Inc.). The thermocycling conditions were as follows: 95°C for 10 sec, followed by 40 cycles of 95°C for 5 sec and 60°C for 20 sec. Relative miRNA expression levels were calculated using the 2−ΔΔCq method. U6 small nuclear RNA, included as an internal control primer set in the Mir-X™ First-Strand Synthesis Kit (Takara Bio Inc., Cat. Nos. 638313), was used for normalization. The exact primer sequences for U6 were not publicly disclosed, and direct inquiries to the manufacturer did not yield this information. However, the Mir-X™ system is widely used and has been independently validated in numerous peer-reviewed studies for reliable miRNA quantification. We adhered strictly to the manufacturer's standardized protocol to ensure data reproducibility and accuracy. For the analysis ΔCq was calculated as (Cq of target miRNA-Cq of U6). ΔΔCq was defined as (ΔCq of Res-HL60-ΔCq of Wt-HL60), and fold change was calculated as 2−ΔΔCq, with Wt-HL60 set as the baseline (fold change-1) (17). Sequences of the target miRNAs were obtained from the miRBase database (https://www.mirbase.org/). MiRNA-specific forward primers were synthesized by Eurofins Genomics Inc. and are listed in Table I. A universal reverse primer provided in the Mir-X™ kit was used for all miRNAs. The Mir-X™ qRT-PCR system employs a poly(A)-tailing strategy followed by reverse transcription with an adapter-anchored oligo(dT) primer. This approach selectively amplifies mature polyadenylated miRNAs, while excluding pre-miRNAs, long non-coding RNAs, and host transcripts lacking poly(A) tails. The specificity of the amplification was further supported by the observation of single, sharp melt curve peaks in all reactions.

Table I. Sequences of miRNA primers for reverse transcription-quantitative PCR.

Table I.

Sequences of miRNA primers for reverse transcription-quantitative PCR.

Primer name	Accession no.	Sequence, 5′-3′
hsa-miR-3612	MIMAT0017989	AGGAGGCATCTTGAGAAATGGA
hsa-miR-324-5p	MIMAT0000761	CGCATCCCCTAGGGCATTGGTG
hsa-miR-146a-5p	MIMAT0000449	TGAGAACTGAATTCCATGGGTT
hsa-miR-1246	MIMAT0005898	AATGGATTTTTGGAGCAGG
hsa-miR-3190-3p	MIMAT0022839	TGTGGAAGGTAGACGGCCAGAGA
hsa-miR-3156-5p	MIMAT0015030	AAAGATCTGGAAGTGGGAGACA
hsa-miR-124-3p	MIMAT0000422	TAAGGCACGCGGTGAATGCCAA
hsa-miR-9-3p	MIMAT0000442	ATAAAGCTAGATAACCGAAAGT
hsa-miR-324-3p	MIMAT0000762	CCCACTGCCCCAGGTGCTGCTGG
hsa-let-7e-5p	MIMAT0000066	TGAGGTAGGAGGTTGTATAGTT
hsa-miR-3669	MIMAT0018092	TACGGAATATGTATACGGAATAT
hsa-miR-29b-3p	MIMAT0000100	TAGCACCATTTGAAATCAGTGTT
hsa-miR-30c-1-3p	MIMAT0004674	CTGGGAGAGGGTTGTTTACTCC
hsa-miR-671-5p	MIMAT0003880	AGGAAGCCCTGGAGGGGCTGGAG
hsa-miR-425-5p	MIMAT0003393	AATGACACGATCACTCCCGTTGA
hsa-miR-345-5p	MIMAT0000772	GCTGACTCCTAGTCCAGGGCTC
hsa-miR-610	MIMAT0003278	TGAGCTAAATGTGTGCTGGGA
hsa-miR-4260	MIMAT0016881	CTTGGGGCATGGAGTCCCA
hsa-miR-376c-3p	MIMAT0000720	AACATAGAGGAAATTCCACGT
hsa-miR-512-3p	MIMAT0002823	AAGTGCTGTCATAGCTGAGGTC
hsa-miR-744-3p	MIMAT0004946	CTGTTGCCACTAACCTCAACCT
hsa-miR-27b-3p	MIMAT0000419	TTCACAGTGGCTAAGTTCTGC
hsa-miR-7-2-3p	MIMAT0004554	CAACAAATCCCAGTCTACCTAA
hsa-miR-769-3p	MIMAT0003887	CTGGGATCTCCGGGGTCTTGGTT
hsa-miR-3675-5p	MIMAT0018098	TATGGGGCTTCTGTAGAGATTTC
hsa-miR-101-5p	MIMAT0004513	CAGTTATCACAGTGCTGATGCT
hsa-miR-20b-5p	MIMAT0001413	CAAAGTGCTCATAGTGCAGGTAG

[i] miR, microRNA.

Statistical analysis

Statistical analyses were performed using the Origin software package (Pro version 9.0; OriginLab Corporation, Northampton, MA, USA) and Microsoft Excel (Microsoft Office 365). To evaluate the statistical significance of differences between the control and experimental groups, an unpaired Student's t-test was used, as the data met assumptions for parametric testing. All results are presented as the mean ± SEM based on four independent experiments. P<0.05 was considered to indicate a statistically significant difference.

Results

Analysis of cell damage exposed to IR

The cell population exposed to 4-Gy radiation temporarily disappeared but reappeared on days 15–16 (Fig. 1A). However, it was confirmed that Res-HL60 had a decreased EV (Wt: 634.24, Res: 620.43) (Fig. 1B). Furthermore, the expression of the CD38 cell surface antigen in Res-HL60 cells was significantly increased after radiation compared with that in the wild-type cells (maximum on day 1: Res=7.55±0.53, Wt=3.54±0.89), confirming a similar previously reported response (Fig. 1C and D). In our previous study, we evaluated CD38 as a differentiation marker and found that its expression was consistently elevated in Res-HL60 cells (8). The Res-HL60 cells seeded at 1×105 cells/ml exhibited approximately 30-fold proliferation (3.1±0.4×105 cells/ml) on day 5, and the Wt-HL60 cells exhibited approximately 15-fold proliferation (1.6±0.1×105 cells/ml) (Fig. 2A). The 4-Gy irradiation condition of Res-HL60 cells was also maintained approximately 5-fold after exposure on day 4 (4.9±0.5×105 cells/ml); however, it remained at the level observed on day 0 in Wt-HL60 cells (1.9±0.1×105 cells/ml) (Fig. 2B). The extent of DNA damage caused by radiation exposure in these cells was determined using a comet assay (Fig. 2C). The tail moment of Res-HL60 cells irradiated with 1–4 Gy was similar to that observed in cells without irradiation, whereas the tail moment of Wt-HL60 cells exposed to irradiation was significantly increased (Fig. 2D).

Expression profiles of miRs

To clarify the miRNA expression profiles of Res-HL60 cells, total RNAs were extracted from these cells. Using an Agilent bioanalyzer, the extracted total RNAs were confirmed to have an RIN value (max-10) and rRNA 18s and 28s peaks, which reveal whether small RNAs, including miRNAs, are of sufficient concentration and quality for analysis (Fig. 3). Total RNAs extracted from each cell exhibited an RIN >9 in all samples. Furthermore, the peak of small RNAs, 18s, and 28s in the histogram was cleared, and these samples were considered to have good quality for RNA analysis (Table II).

Figure 3. Histogram of total RNA detection from Res-HL60 and Wt-HL60 cells. Total RNA was detected in each cell using an Agilent 2100 bioanalyzer. Wt, wild type; Res, resistant.

Table II. Concentration and quality of total RNA extracted from Res-HL60 cells and Wt-HL60 cells.

Table II.

Concentration and quality of total RNA extracted from Res-HL60 cells and Wt-HL60 cells.

Sample list	Concentration of total RNA, µg/106 cells	RNA integrity number	Ratio of rRNA, 28s/18s
Wt
Replication 1	3.6	9.5	1.03
Replication 2	5.6	9.5	0.98
Replication 3	6.5	9.6	1.16
Replication 4	4.4	9.1	0.84
Mean ± SE	5.0±0.6	9.4±0.1	1.0±0.1
Res
Repliction 1	5.5	9.4	0.98
Replication 2	4.3	9.6	1.18
Replication 3	3.6	9.2	0.76
Replication 4	4.1	9.6	1.13
Mean ± SE	4.4±0.3	9.5±0.1	1.0±0.1

[i] rRNA, ribosomal RNA; Wt, wild type; Res, resistant; SE, standard error.

After miRNA microarray analysis, data were extracted after quantile normalization, excluding under background, and collection of unsaturated signals (Fig. 4A). After processing, the signal values obtained using multiple types of detection probes were averaged for each miRNA name, yielding 1,187 miRNA data. Among the miRNAs that significantly increased or decreased in Res compared with Wt, 10 miRNAs had a fold change (Res/Wt) >1.5, and 17 miRNAs had a fold change (Res/Wt) <0.66 (Fig. 4B-D). For the 27 miRNAs that exhibited expression changes in the microarray analysis, a functional enrichment analysis using the Reactome Pathway Database in OmicsNet 2.0 was performed to estimate the associated mRNA expression pathways. Cellular functional changes were observed in the following pathways: ‘Immune System (R-HAS-168256)’, ‘Signaling by PGDF (R-HAS-186797)’, ‘Adaptive Immune System (R-HAS-1280218)’, ‘Signaling by EGFR in Cancer (R-HAS-1643713)’, ‘Apoptosis (R-HAS-109581)’, and ‘Cell Cycle, Mitotic (R-HAS-69278)’ for tumor growth and/or invasion (Fig. 5A).

Figure 4. miRNA microarray analysis. (A) Statistical analysis for normalization, and the (B) relationship between fold change and P-value. (C) Significant upregulated miRNAs (P<0.05 by Student's t-test, >1.5 fold change), and (D) significant downregulated miRNAs (P<0.05 by Student's t-test, <0.66 fold change). The data are presented as the mean ± standard error of the mean of four experiments. miRNA/miR, microRNA; Wt, wild type; Res, resistant.

Figure 5. Top 20 enriched Reactome pathways associated with predicted target genes of differentially expressed miRNAs. Pathway enrichment analysis was conducted using the Reactome Pathway Database to categorize the predicted target genes of significantly altered miRNAs into curated biological pathways. Reactome pathway identifiers are shown in parentheses. Bar graphs show the top 20 enriched pathways ranked by statistical significance for the predicted target genes of (A) all 27 differentially expressed miRNAs, (B) 10 upregulated miRNAs and (C) 17 downregulated miRNAs. miRNA, microRNA.

Furthermore, when we focused on the functional analysis of the target mRNAs of the 10 upregulated miRNAs in Res-HL60 cells, these mRNAs were strongly related to the following functional pathways: ‘cell cycle’, ‘cell division’ and ‘apoptosis’ (Fig. 5B). In contrast, the target mRNAs of the 16 downregulated miRNAs in Res-HL60 cells were strongly related to the following functional pathways: ‘Signaling by PDGF’ and ‘Assembly of collagen fibrils and other multimeric structures (R-HAS-2022090)’ (Fig. 5C).

Validation of the expression of 27 miRNAs

The significant expression of the 27 miRNAs obtained from these microarray analysis signals was reconfirmed by RT-qPCR. Thus, the expression of miRNA-146a-5p was significantly increased, whereas the expression of miRNA-30c-1-3p, miRNA-671-5p, miRNA-610, and miRNA-3675-5p was significantly decreased (Fig. 6).

Figure 6. Validation analysis of the 27 focused miRNAs by RT-qPCR. (A) Upregulated miRNA and (B) downregulated miRNA group were validated by RT-qPCR. The quantitation by fluorescence signal was performed using the ΔΔCq method. U6 small nuclear RNA was used as the internal control. Fold change was calculated by comparing Res-HL60 cells to Wt-HL60 cells after normalization to U6 expression. Expression values in Wt-HL60 were set to 1, and fold change values in Res-HL60 are shown relative to this baseline (horizontal line). The data are presented as the mean ± standard error of the mean of four experiments. **P<0.01, *P<0.05 by unpaired Student's t-test. RT-qPCR, reverse transcription-quantitative PCR; miRNA/miR, microRNA; Wt, wild type; Res, resistant.

Analysis of bioinformatics: The mRNA networks controlled by the five reproduced miRNAs were analyzed using OmicsNet 2.0. In this network, 28 mRNAs were associated with two or more miRNAs (Fig. 7). Of these 28 target mRNAs, seven (i.e., ALG10B, CCDC6, CCND2, CPM, CYBRD1, NACC1, and UMPS) were targeted by miRNA-146a-5p, which is an upregulated miRNA, and the other four miRNAs, which are downregulated. Furthermore, three miRNAs (i.e., miRNA-610, miRNA-671-5p, and miRNA-30c-1-3p) targeted superoxide dismutase 2 (SOD2). In the Target Scan analysis, three miRNAs (miRNA-610, miRNA-671-5p, and miRNA-30c-1-3p) were predicted to bind to the 3′-UTR region of SOD2, which was predicted as the target mRNA (Fig. 8). Furthermore, two miRNAs (i.e., miRNA-671-5p and miRNA-30c-1-3p) were predicted to bind to the 3′-UTR region of MYH9 mRNA. The putative target sites of miR-610 and miR-30c-1-3p for CD38 mRNA and miR-3675-5p, miR-671-5p, and miR-30c-1-3p for ATM mRNA were located in the 3′-UTR region.

Figure 7. Predictive network analysis between miRNA and its target mRNA using OmicsNet. OmicsNet was used to predict the target genes. This study used the Reactome database of OmicsNet to predict the target genes. Lines connect the miRNA target genes. The figure shows green for miRNAs, blue for target genes, and red for other related genes. miRNA/miR, microRNA.

Figure 8. Target scan analysis in humans. The images show a predictive region for adhesion by target mRNA on 3′UTR and miRNA. The data show (A) SOD2, (B) MYH9, (C) CD38 and (D) ATM. miRNA/miR, microRNA; SOD2, superoxide dismutase 2.

Discussion

In this study, we analyzed the miRNA expression profile of a radioresistant AML cell model (Res-HL60) derived from the HL60 cell line, which belongs to the FAB M2 subtype of AML. We reconfirmed that Res-HL60 cells, which were established by repeated irradiation of HL60 cells, maintained the morphological characteristics previously described, even when remade (Fig. 1). The HL60 cell line, which is derived from a patient with AML, exhibits inherently low CD38 expression, as previously reported (18). We have previously evaluated CD38 as a differentiation marker and found that its expression was consistently elevated in Res-HL60 cells (8). In this study, we observed transient increases in CD38 expression immediately after irradiation. This upregulation has been reported to be associated with cellular activation and proliferation (19). CD38 was selected as the primary marker in this study because it is a well-established differentiation-associated antigen in the HL60 model system; its expression increases markedly during granulocytic differentiation, as demonstrated by Partida-Sánchez et al (16). In our Res-HL60 cells, CD38 consistently showed strong upregulation both under baseline conditions and immediately after irradiation. These observations, combined with CD38′s known roles in NAD+-dependent calcium signaling and immune modulation, made it a biologically relevant candidate for investigating radioresistant phenotypes in AML. Although Breitman et al broadly discussed the HL60 differentiation model, CD38-specific functional insights were established in later work (20). In this context, ‘cellular activation’ refers to an increased responsiveness of cells to external stimuli, such as radiation, typically accompanied by changes in surface markers, including CD38 upregulation. Although direct functional activation markers were not assessed in this study, similar CD38 upregulation has been correlated with cell activation and immune modulation in leukemia models. Our previously established Res-HL60 cells demonstrated enhanced proliferative capacity compared to Wt-HL60 cells, as evidenced by shortened population doubling times and increased viable cell counts (8–12). Although these proliferation characteristics were not re-evaluated in the present study, the phenotypic stability and CD38 expression patterns observed here are consistent with those findings.

Furthermore, DNA fragments did not increase in these cells due to radiation, and the response was confirmed to be similar to γ-H2AX, a marker for DNA double-strand breaks, and to previous DNA ladder analyses (9,21). In these damage responses in Res-HL60 cells, 27 miRNAs (10 upregulated and 17 downregulated) showed altered expression in the microarray analysis. Among them, five miRNAs (miR-146a-5p, miR-30c-1-3p, miR-671-5p, miR-610, and miR-3675-5p) consistently demonstrated significant changes in the same direction across both microarray and RT-qPCR validation experiments. This consistency suggests that these five miRNAs may play a regulatory role in modulating molecular pathways associated with the development of radiation resistance. These five miRNAs have been implicated in tumor proliferation, radiosensitivity, and hematopoiesis (22–30). miR-146a functions together with TRAF1 and IRAK1 as negative feedback factors for the two toll-like receptor pathway proteins (22). It is also involved in regulating the transcriptional activity of NF-κB, which is necessary for the differentiation of hematopoietic cells (23,24). In radiotherapy for meningioma, the expression of miR-30c-1-3p is suppressed when treatment resistance occurs (25). In another report, the suppression of miR-30c-1-3p expression was associated with the progression of prostate cancer (26). The overexpression of miR-671-5p has been shown to significantly increase apoptosis and DNA damage in human glioblastoma cells (27). In neuroblastoma cells, the overexpression of miR-610 induces apoptosis and reduces cell viability (28). miR-3651 promotes the proliferation of hepatocellular carcinoma and colorectal cancer cells (29,30). These reports from various cell types suggest that changes in miRNA expression in radioresistant AML are triggered by cell viability activity.

The analysis of linking target mRNA networks of five miRNAs with radioresistant AML is new information, and 11 target mRNAs of these miRs (i.e., ALG10B, CCDC6, CCND2, CPM, CYBRD1, NACC1, UMPS, SOD2, MYH9, CD38, and ATM) may regulate the expression of its associated proteins and transcription factors in a complex manner. SOD2 can scavenge active oxygen produced by radiation irradiation, and SOD2 has been reported to protect against cellular stress (31). Furthermore, MYH9 is involved in cell division, cell migration, and maintenance of cell morphology and acts as an antiapoptotic agent in head and neck cancers where MYH9 is highly expressed (32). Moreover, it also reduces intracellular reactive oxygen species levels through the activation of the MAPK signaling pathway and Nrf2/keap1 pathway and the promotion of the production of the antioxidant enzyme GCLC. It is also involved in radioresistance. This observation may also explain the morphological stability and resistance to radiation-induced apoptosis observed in Res-HL60 cells. To further investigate the clinical relevance of these target genes, we examined publicly available AML gene expression datasets from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO). Yu et al have reported that MYH9 overexpression was significantly associated with poor prognosis in patients with AML based on TCGA-LAML data (33). This finding is also supported by GSE12417 (GEO), a dataset originally published by Metzeler et al (34), which includes cytogenetically normal AML cases. In a recent analysis by Zhai et al (35), MYH9 was among the genes linked to adverse outcomes in this cohort. Expression of ATM, although not frequently mutated in AML, has been implicated in treatment responses; low ATM expression enhances sensitivity to FLT3 inhibitors and may affect DNA damage signaling (36). SOD2 has not been directly studied in AML; however, it plays a key role in scavenging reactive oxygen species. A meta-analysis of pan-cancer TCGA datasets by Noh et al (31) reported that SOD2 upregulation is associated with treatment resistance, suggesting its potential role in radioresistant phenotypes. We note that original survival analyses were not performed in this study. Instead, we relied on published results from previous analyses of these public datasets. This approach was taken due to the experimental focus of our study on the HL60-derived model. However, we fully acknowledge the importance of such analyses in increasing the translational relevance of our findings. While we agree with the reviewer's suggestion to perform updated survival analyses using the most recent TCGA and GEO datasets, this was beyond the scope of the current in vitro study. Nonetheless, incorporating such analyses is a clear objective for future work. We plan to conduct comprehensive bioinformatics evaluations of MYH9, ATM, and SOD2 expression in larger patient cohorts using integrated clinical and molecular data. Furthermore, the use of radiation-resistant AML patient samples to validate these gene targets was not feasible in the present study due to sample unavailability. However, this represents a crucial next step. Future studies will aim to collect and analyze patient-derived AML samples, particularly from individuals with known radiation treatment history, to directly confirm the relevance of MYH9, ATM, and SOD2 in clinical radioresistance. These findings complement our in vitro results and support the relevance of SOD2, ATM, and MYH9 in AML radioresistance.

Most functions inferred from the Reactome analysis were related to anti-apoptotic processes and were associated with upregulated miR-146a-5p. Our original hypothesis was that miR-146a-5p promotes apoptosis by inhibiting antiapoptotic molecules, thereby enhancing radiosensitivity in Res-HL60 cells. However, the results of the inhibition of anti-apoptotic molecules did not support this hypothesis. Instead, these findings suggest that miR-146a-5p regulates an alternative pathway that could contribute to radiation resistance. Target-scan analysis revealed that the high expression of CD38 cell surface antigen protein, CD38 mRNA, and intracellular ATM is related to the regulation of the five miRNAs revealed here (Fig. 8). These miRNA regulations provide molecular biology information supporting previous cellular responses. This information enhances our understanding of the molecular mechanisms underlying radiation resistance in AML. To further elucidate the mechanisms linking CD38 expression and radiation resistance, our future studies will focus on dissecting the intracellular signaling pathways downstream of CD38. Specifically, we plan to investigate NAD+ metabolism and calcium signaling pathways, both of which are regulated by CD38 and are known to affect cell survival, DNA repair capacity, and oxidative stress responses. These pathways may provide crucial mechanistic links between CD38 expression and the observed radioresistant phenotype in AML cells. In addition, combining CD38 expression analysis with metabolic profiling may help clarify its functional relevance in therapeutic resistance. This approach may also reveal novel therapeutic targets to overcome resistance in CD38-expressing AML subtypes. Although CD38 was a useful marker in our initial evaluation, we recognized that other markers, such as CD34, CD123, and ALDH are also relevant in the context of leukemia stemness and therapy resistance. Exploring the role of these markers was not within the scope of this study. However, future studies incorporating a broader panel of surface markers are warranted to further elucidate the mechanisms of AML radioresistance.

One limitation of this study is the lack of a regulatory analysis for the expression of the five miRNAs in AML cells, including investigations of intracellular signaling pathways and knockdown of the 11 predicted target mRNAs. Another limitation is that all experiments were performed using a single cell line, HL60, and its derived radioresistant subline. Although HL60 is a widely used model for AML studies, it does not capture the full genetic and phenotypic heterogeneity of AML, which can significantly affect radiosensitivity and miRNA profiles. Therefore, caution should be exercised when generalizing these findings. Future studies should include additional AML cell lines representing different molecular subtypes, as well as patient-derived primary cells, to validate the observed miRNA and mRNA expression patterns. Furthermore, functional assays using miRNA mimics/inhibitors or siRNA-mediated knockdown of key targets such as MYH9, ATM, and SOD2 are necessary to directly confirm their roles in radiation resistance. Such validations will be crucial for establishing the broader applicability and clinical relevance of our findings. Clinically, the identified miRNAs and their targets (e.g., MYH9, ATM, SOD2) may serve as biomarkers for predicting radiotherapy resistance in AML. They may also provide novel therapeutic targets to overcome such resistance, enabling more personalized treatment strategies. Although our findings are currently preclinical, they lay important groundwork for future translational research aimed at improving AML patient outcomes.

In summary, further studies are warranted to clarify the role of these miRNA-mRNA networks in therapy-resistant leukemia. Future investigations employing integrative miRNA-mRNA analyses, as demonstrated here, may further elucidate the molecular basis of radiosensitivity in AML.

Supplementary Material

Supporting Data

Supporting Data

Acknowledgements

Not applicable.

Funding

This work was supported by JSPS KAKENHI, Grants-in-Aid for Scientific Research (B) (grant no. 21H02861/23K21419, to Satoru Monzen) and Takeda Science Foundation 2022 (to Satoru Monzen). The funders played no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Availability of data and materials

The sequencing data generated in the present study may be found in the GEO under accession number GSE285934 or at the following URL: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE285934. The other data generated in the present study may be requested from the corresponding author.

Authors' contributions

HS, YH and SM designed the study, prepared the manuscript draft and participated in the revision of the manuscript. HS, MK, MC and SM performed the biological analysis and collection of the experimental data. SM supervised the study, critically reviewed the manuscript and provided final approval for the version to be submitted and published. All authors have read and approved the final version of the manuscript. HS and SM confirm the authenticity of all the raw data.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References