Bone marrow samples were collected from six patients treated at the Department of Hematology, Shengjing Hospital of China Medical University, (Shenyang, China), between September 2014 and December 2016. All patients were diagnosed with B-ALL and received the CCLG-ALL-2008 treatment plan (15). These patients included three boys and three girls (mean age, 5.8 years; age range, 4–8 years). Bone marrow samples were also collected from six individuals without B-ALL (control). These patients included three boys and three girls (mean age, 5.5 years; age range, 4–8 years). The inclusion criteria were as follows: Initial diagnosis was according to the 2008 World Health Organization classification standard of lymphoid and hematopoietic tissue tumors and bone marrow cytology, immunotyping, chromosome karyotype analysis and fusion gene analysis were performed (16). The exclusion criteria were as follows: i) Patients received anti-leukemia treatment prior to admission; ii) patients with mature B-ALL and mixed cell leukemia and iii) patients with secondary ALL or other tumors. The present study was approved by the China Medical University Ethics Committee (institution review board no. 2020028) and written informed consent was provided by all participants prior to the study start.

The human B-ALL cell line, Ball-1, and the human B lymphoblast cell line, HMy2.CIR, were purchased from the Type Culture Collection of The Chinese Academy of Sciences. Cells were maintained in RPMI-1640 medium (Hyclone; Cytiva) supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.), at 37°C with 5% CO₂.

The HMy2.CIR cell line was used as the control according to previous studies (17–20), which demonstrated that HMy2.CIR cells are comparable with human B lymphoblasts and have a lower malignancy potential than B-ALL cells.

Total RNA was extracted from Ball-1 and HMy2.CIR cells with TRIzol^® reagent (Beijing Solarbio Science & Technology Co., Ltd.). A total of three samples were taken from each cell line. circRNAs were enriched from 5 µg total RNA using the circRNA Enrichment kit (CloudSeq, www.cloud-seq.com.cn). Subsequently, strand-specific RNA-seq libraries were prepared using the TruSeq Stranded Total RNA Library Prep kit (Illumina, Inc.), which were subjected to deep sequencing using BioAnalyzer 2100 (Agilent Technologies, Inc.).

The RNA-seq FASTQ reads were mapped to a human reference genome (GRCh37/hg19) using TopHat2 (21). The number of spliced reads were used as the expression level of each circRNA. The total number of reads was used to standardize the samples, and log₂ transformation was performed to obtain the standardized number of reads. Using the standardized number of reads and the R software package DEGseq (22), differential expression of circRNAs between Ball-1 and HMy2.CIR cells were calculated. Log₂ (fold change) >2.0 (or <-2.0) and P<0.05 were considered to indicate significantly differentially expressed circRNAs. Hierarchical clustering analysis was also performed to generate an overview of circRNA expression profiles between the two cell lines.

Total RNA was extracted from Ball-1 cells, HMy2.CIR cells, B-ALL bone marrow and non-B-ALL bone marrow using TRIzol^® reagent, and reverse transcribed into cDNA using PrimeScript™ RT reagent Kit with gDNA Erase (Takara Biotechnology Co., Ltd.), according to the manufacturer's protocol. qPCR was subsequently performed using the SYBR-Green PCR Master Mix (Takara Biotechnology Co., Ltd.) within the Fast-Real-Time PCR System (CFX96™ Optics Module; Bio-Rad Co., Ltd.). The following primer sequences were used for qPCR: GAPDH forward, 5′-GGCCTCCAAGGAGTAAGACC-3′ and reverse, 5′-AGGGGAGATTCAGTGTGGTG-3′. Primer sequences for the circRNAs are listed in Table SI. The following thermocycling conditions were used for qPCR: 95°C for 30 sec, followed by 40 cycles of 95°C for 3 sec and 60°C for 30 sec and a final extension step at 60°C for 7 min. Relative expression levels were calculated using the 2^−ΔΔCq method (23) and normalized to the internal reference gene GAPDH.

The Database for Annotation, Visualization and Integrated Discovery (DAVID, http://david.ncifcrf.gov) was used to perform GO enrichment and KEGG pathway analyses of the differentially expressed circRNAs to determine the processes and pathways involved.

Under normal culture conditions, cells were starved in serum-free RPMI-1640 medium (Hyclone, Cytiva) for 2 h after reaching 60–70% confluence. OE-CIRC and OE-negative control (NC) were respectively transfected into HMy2.CIR cells using Lipofectamine^® 3000, according to the manufacturer's protocol. Cells were harvested for subsequent experimentation after 24–72 h. The concentration of plasmid OE-CIRC and OE-NC of hsa_circ_0000745 purchased from jtsbio company was 100 nM/l. The temperature and duration of transfection were 37°C and 24–72 h. Fluorescence intensity was detected via fluorescence microscopy (magnification, ×200; Nikon Corporation), according to the manufacturer's instructions.

After 24 h of transfection, cell suspensions of the experimental and control groups were inoculated into 96-well plates at a density of 3,000 cells/well/100 µl. The plates were precultured in the incubator at 37°C with 5% CO₂. Cell Counting Kit-8 (10 µl) reagent (Cofitt Life Science Company; www.cofitt.com) was subsequently added to each well and incubated for an additional 4 h. The absorbance was measured at a wavelength of 450 nm, using a microplate reader (BioTek Instruments, lnc.). The OD value of each well was measured for 3 consecutive days.

Statistical analysis was performed using SPSS 23.0 software (IBM Corp.). Edge R 4.1.0 software (https://www.r-project.org) was used to establish a linear model of negative binomial distribution according to various factors in the experimental design, and the dispersion coefficient of each factor was calculated (Fig. 1). The Quasi-Likelihood F test (P<0.05 and fold change ≥2.0) was used to screen the differentially expressed circRNAs. The volcano map was constructed to compare differences between the two groups (P<0.05 and fold change ≥2.0). The Heatmap. 2 function of R 4.1.0 software (https://www.r-project.org) was used to cluster the differentially expressed circRNAs. Unpaired independent sample t-test was used to compare the expression of circRNAs in the different cell lines and samples. P<0.05 was considered to indicate a statistically significant difference.

To identify the differentially expressed circRNAs in Ball-1 and HMy2.CIR cells, the circRNA expression profiles were analyzed via high-throughput circRNA microarray analysis. A total of 263 differentially expressed circRNAs were identified [log₂ (fold change) >2.0 (or ≥2.0) and P<0.05], of which 76 were upregulated and 187 were downregulated in Ball-1 cells compared with HMy2.CIR cells (Fig. 1A-C).

GO enrichment and KEGG pathway analyses of the differentially expressed circRNAs were performed using DAVID. GO enrichment analysis identified three terms, biological process (BP), cellular component (CC) and molecular function (MF).

For BP, upregulated circRNAs were mainly enriched in ‘macromolecule modification’ (Fig. 2A). For CC, upregulated circRNAs were mainly enriched in ‘intracellular parts’ (Fig. 2B). For MF, upregulated circRNAs were mainly enriched in ‘protein binding’ (Fig. 2C). KEGG pathway analysis demonstrated that upregulated circRNAs were mainly enriched in ‘Proteoglycans in cancer’, ‘MAPK signaling pathway’ and ‘protein processing in endoplasmic reticulum’ (Fig. 2D).

For BP, downregulated circRNAs were mainly enriched in ‘negative regulation of RNA biosynthetic processes’ and ‘negative regulation of nucleic acid templated transcription’ (Fig. 3A). For CC, downregulated circRNAs were mainly enriched in ‘intracellular organelles’ (Fig. 3B). For MF, downregulated circRNAs were mainly enriched in ‘transcription regulator activity and binding’ (Fig. 3C). KEGG pathway analysis demonstrated that downregulated circRNAs were mainly enriched in ‘natural killer cell-mediated cytotoxicity’ and ‘viral carcinogenesis’ (Fig. 3D).

To validate the microarray results, seven circRNAs downregulated in Ball-1 cells were selected and their expression was validated via RT-qPCR analysis. The seven differentially expressed circRNAs were selected according to the following criteria: i) High fold-change value, ii) significant P-value and iii) not reported in previous studies. The results are presented in Fig. 4. A total of six circRNAs (hsa_circ_0000745, chr15:87949594-87966067-, chr15:87898780-87966067-, hsa_circ_0006473, hsa_circ_0071106 and hsa_circ_0097878) were significantly downregulated in Ball-1 cells compared with HMy2.CIR cells (P=0.003; P=0.0041; P=0.0053; P=0.024; P=0.0032; P=0.0029), while has_circ_0104812 expression was slightly but significantly upregulated in Ball-1 cells compared with HMy2.CIR cells (P=0.042).

To validate the results from the cell lines, hsa_circ_0000745 and chr15:87949594-87966067- were selected to validate bone marrow samples from patients with B-ALL. The results demonstrated that these two circRNAs were significantly downregulated in B-ALL bone marrow samples compared with non-B-ALL bone marrow samples (P<0.01; Fig. 5). Therefore, these two circRNAs may serve as biomarkers for B-ALL. Detailed characteristics of the patients and controls are presented in Table SII.

hsa_circ_0000745 expression was overexpressed in HMy2.CIR cells, and the results demonstrated that high hsa_circ_0000745 expression significantly inhibited the proliferation of HMy2.CIR cells (P=0.034; Fig. S1). Transfection efficiency is presented in Fig. S2.