Introduction
Multiple myeloma (MM) is a clonal B-cell malignancy that accounts for approximately 10% of all hematologic malignancies and is characterized by abnormal plasma cell proliferation and abnormal globulin secretion in bone marrow [
1]. Although survival outcomes in MM patients have significantly improved over the past 20 years, the disease remains incurable [
2,
3]. Additionally, chemotherapy resistance develops often in MM patients during treatment, stemming from a variety of factors including genetic abnormalities [
4]. Therefore, new therapeutic targets are urgently needed. Transcriptomic studies have provided important information on disease-related pathways and critical genes in addition to revealing new targets for the treatment of diseases [
5‐
7].
N6-methyladenosine (m6A) methylation is the most common post-transcriptional modification found on eukaryotic mRNA so far and also widely exists in a variety of bacteria and RNA viruses [
8]. m6A modification participates in the regulation of mRNA translation, splicing processing, as well as nuclear transport and degradation, which determines the entire life process of mRNA [
9]. Over the past few decades, the biological functions regulated by m6A have been shown to be involved in several types of tumor progression [
10]. However, the critical role of m6A modification in MM is still elusive, and the regulatory mechanisms of MM in general are not fully understood.
Heterogeneous nuclear ribonucleoprotein A2B1 (HNRNPA2B1) is one of the main m6A readers [
11,
12]. The reader is a methylated reading protein that recognizes RNA methylation modification and performs its functions by specifically binding to the m6A-modified region or altering the RNA secondary structure to facilitate binding to RNA [
13]. Recently, abnormal HNRNPA2B1 expression has been discovered in various disease states and linked closely to tumor progression. HNRNPA2B1 may play key roles in cancer development due to its ability to accelerate pre-mRNA processing through the function of RNA binding [
14].
In the present study, we discovered the hypermethylated status of MM and the high expression of HNRNPA2B1 in MM. Through transcriptome and m6A sequencing, we identified Toll-like receptor 4 (TLR4) as a target of HNRNPA2B1-mediated m6A modification. HNRNPA2B1 knockdown decreased the TLR4 mRNA level and simultaneously abolished TLR4 mRNA m6A modification. Through analysis of gene expression omnibus (GEO) datasets, HNRNPA2B1 and TLR4 were found to be adverse prognostic factors for survival among MM patients. Our results reveal the biological role of HNRNPA2B1 in mediating m6A modification in MM, indicating that HNRNPA2B1 plays an important role in MM pathogenesis via epigenetic regulation a critical target TLR4. From these findings, we propose that HNRNPA2B1 may be a novel therapeutic target for evaluating MM progression under m6A-based post-transcriptional regulation through TLR4 pathway.
Materials and methods
MM specimens and cell lines
Bone marrow specimens were obtained from 15 newly diagnosed MM patients and 15 normal controls (NC) at Harbin Medical University Cancer Hospital, China, between January 2019 and December 2020. The National Comprehensive Cancer Network clinical practice guidelines were used to diagnose MM.
The MM cell line, RPMI 8226 was purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The cells were maintained in RPMI 1640 medium (Gibco, Grand Island, NY, USA) supplemented with 10% (for RPMI 8226) fetal bovine serum (FBS, Hyclone, Logan, UT, USA) and 1% penicillin–streptomycin at 37 ℃ in a humidified incubator containing 5% CO2.
Total RNA was isolated using Trizol Reagent (Invitrogen, USA). cDNA was synthesized from 1 µg of total RNA using the Prime Script™ RT reagent kit (Takara Bio, Inc., Otsu, Japan). Relative gene expression was assessed by real-time RT-PCR using the SYBR Premix Ex Taq TM kit (Takara Bio, Inc., Otsu, Japan) on the ABI 7900HT Real-Time PCR system (Applied Biosystems Life Technologies, Foster City, CA, USA) according to the manufacturers’ instructions and suggested protocol. The primers used for qRT-PCR are listed in Additional file
1: Table S1. Each analysis was carried out in triplicate.
Cell proliferation and apoptosis assays
Aliquots of 3 × 104 cells were suspended in 100 µl RPMI 1640 medium and then seeded into 96-well plates. The CCK8 (Dojindo Molecular Technologies, Japan) assay was conducted according to the manufacturer’s instructions. Briefly, after cell culture, 10 µl CCK8 solution was added into each 96-well, and after incubation for 2 h, the absorbance at 450 nm wavelength was measured. Each analysis was carried out in triplicate.
Apoptosis was evaluated by Annexin V-APC/propidium iodide (PI) staining. Briefly, cells were resuspended in binding buffer, and then Annexin V-APC and PI were added to each sample for 15 min and incubation at room temperature in darkness. Flow cytometry CantoII (BD Biosciences, San Jose, CA, USA) was used to evaluate the numbers of apoptotic cells. Each analysis was carried out in triplicate.
Adenovirus vector plasmid for transfection of RPMI 8226 cells
Adenovirus expressing HNRNPA2B1-shRNAs was purchased from Hanheng Bioscience Co., Ltd. (Shanghai, China). The HNRNPA2B1 shRNA sequences were: sh1,5′-TCGAGGCCATGGCTGCAAGACCTCATTCAATTTCAAGAGAATGAATGAGGTCTTGCAGCCATGGTTTTTTA-3′; sh2, 5′-AGCTTAAAAAACCATGGCTGCAAGACCTCATTCAATTCTCTTGAAATTGAATGAGGTCTTGCAGCCATGGCC-3′; and sh3, 5′-TCGAGGAGGAACAGTTCCGTAAGCTCTTTATTTCAAGAGAATAAAGAGCTTACGGAACTGTTCCTTTTTTTA-3′. Aliquots of 5 × 106 cells were plated onto 25-cm2 plates for 48 h before transfection with Ad-eGFP or Ad-HNRNPA2B1 with RPMI 1640 medium for 4 h at 37 °C. The cells were then washed twice with RPMI 1640 medium and once with RPMI 1640 medium containing 10% FBS, and then incubated for 2 days at 37 °C. After the formation of a stable cell line, RT-qPCR and western blot analyses were conducted to screen the RPMI 8226 with HNRNPA2B1 knockdown.
RNA m6A quantification
After extraction, the quality and quantity of the total RNA were measured by the NanoDrop (Thermo Fisher Scientific, Waltham, MA, USA). The EpiQuik m6A RNA Methylation Quantification Kit was then used to examine the m6A modification level according to the manufacturer’s instructions. Briefly, RNA was coated onto the assay wells, which were then washed and incubated with capture antibody. After antibody addition, the m6A level was detected by measurement of the absorbance at 450 nm (optical density [OD]450). Each analysis was carried out in triplicate.
RNA-sequencing
To assess the RNA degradation and contamination, 1% agarose gel electrophoresis was conducted, and the NanoPhotometer® spectrophotometer (IMPLEN, CA, USA) was used to test the RNA purity. RNA integrity was assessed by using the Bioanalyzer 2100 System (Agilent Technologies, CA, USA).
For RNA-sequencing of 13 paired samples, mRNA Capture Beads with Oligo(dT) (VAHTS, Nanjing, China) were first used to capture mRNA. mRNA was then purified with binding and washing buffer. mRNA was then randomly fragmented and reverse transcribed into cDNA in the corresponding buffer. Random primers were used to synthesize the first cDNA strand, and dNTPs/DNA polymerase I (ABclonal, MA, USA) was used to synthesize the second cDNA strand. After dA—tailing of 3' DNA fragments, UMI (Novogene, Beijing, China) and sequence adapter were used to ligate DNA. After PCR, the product was purified and collected using DNA Clean Beads (Beckman Coulter, CA, USA) and nuclease-free H2O. Each analysis was carried out in triplicate. After quality assessment, the prepared library was sequenced on the Illumina platform (Illumina Novaseq, Inc., USA).
M6A-RNA immunoprecipitation (MeRIP) assay and m6A sequencing
The mRNA m6A of 15 paired samples was sequenced by MeRIP-seq at Novogene (Beijing, China). The degradation and contamination of the extracted RNA were tested as described above. For immunoprecipitation, fragmented mRNA (~ 100 nt) was incubated with anti-m6A antibody at 4 ℃ for 2 h (Synaptic Systems). m6A modified-mRNA was then enriched and detected through qRT-PCR or next generation sequencing using Illumina HiSeq 2000 (Illumina Inc.). Each analysis was carried out in triplicate. The NEBNext ultraRNA library prepare kit for Illumina (Illumina, Inc., USA) was used for library construction. Finally, the library preparation was sequenced by an Illumina Novaseq platform.
Dot blot for measuring m6A
Nitrocellulose membrane was fixed in plates. Then 10 µg RNA per well was added to the dot bolt apparatus. Then RNA was crosslinked to the membrane using an ultraviolet (UV) crosslinker. The membrane was then washed with washing buffer and blocked with blocking buffer at room temperature for 30 min. After that, the membrane was incubated with anti-m6A antibody at 4 °C. Then the membrane was washed, incubated with western blotting reagent, and exposed to autoradiography.
NCBI GEO microarray dataset analysis
GEO datasets GSE116294, GSE80608 and GSE141260 were used to compare the expression of HNRNPA2B1 at different stage of monoclonal gammopathy. GSE116294 consisted of 69 clinical samples, including 4 normal samples, 50 MM samples at diagnosis and 15 plasma cell leukemia (PCL) samples at diagnosis. GSE80608 contained 10 control, 10 monoclonal gammopathy of undetermined significance (MGUS) and 10 MM samples. GSE141260 contained 10 control and 10 MM samples. Additionally, GSE4202, GSE24080 and GSE136400 were used to estimate the relationship between the expression of HNRNPA2B1/TLR4 and the disease prognosis of MM.
Based on the Affymetrix-GPL570 platform, the data for gene expression and survival were obtained from the GEO datasets GSE4204 (newly diagnosed MM, n = 538) and GSE24080 (newly diagnosed MM, n = 559). In GSE4204, the samples were divided into two subgroups based on the median gene expression values, the survival curves were drawn to analyze overall survival (OS) between different expression subgroups. In GSE24080, the clinical endpoints were OS and event-free survival (EFS) at 24 months, and we compared the expression levels of hnRNPA2B1 and TLR4 between two subgroups with different survival outcomes.
In addition, GSE136400 dataset was used to analyze the relationship between TLR4 expression and survival of MM patients. The dataset consisted of 1424 samples from different times, including before treatment, post-treatment but pre-transplantation, post-transplantation, pre-consolidation, etc., and we selected all samples before treatment (n = 354) for further analysis. The gene expression data were analyzed using the GPL27143 platform Affymetrix Human Genome U133 Plus 2.0 Array. X-tile software (v3.6.1, Yale University, New Haven, CT, USA) was used to confirm the cut-off value of TLR4 expression, and the patients were divided into two subgroups: TLR4high and TLR4low. Survival curves were drawn to analyze OS and progression-free survival (PFS) between the two subgroups.
Western blot analysis
After collection of cells and tissues, radioimmunoprecipitation assay buffer (Beyotime Institute of Biotechnology, Shanghai, China) was added into the pellet to extract proteins. After extraction, concentration of total protein was measured using a BCA protein assay kit (Beyotime Institute of Biotechnology). Agarose gel electrophoresis was performed after the addition of 2X sodium dodecyl sulfate loading buffer and incubated at 100 °C for 5 min. After transfer onto polyvinylidene difluoride membranes and blocking using 5% milk, the membranes were incubated with primary antibodies against TLR4 (1:500; Affinity Biosciences, Jiangsu, China), HNRNPA2B1 (1:500; Affinity Biosciences) and GAPDH (1:500; Affinity Biosciences) overnight at 4 °C. Then the secondary antibodies were applied to the membranes. After a series of washing steps, the membranes were visualized using chemiluminescence (Beyotime Institute of Biotechnology). Each analysis was carried out in triplicate.
Statistical analysis
Statistical analyses were conducted by GraphPad Prism 8 software (GraphPad Software, Inc., San Diego, CA, USA). Survival curves were drawn by the Kaplan–Meier approach, and the log-rank test was used for comparisons between groups. The differences between two groups were analyzed by unpaired t tests, and differences between multiple groups were analyzed by one-way analysis of variance (ANOVA). In all statistical analyses, p-values < 0.05 were considered statistically significant.
Discussion
M6A is the most common RNA chemical modification in human mRNA. Recently, with the advancement of high-throughput sequencing and antibody-based technology, researchers have been able to accurately map the site of m6A and further study its biological functions [
17]. Although studies have shown potential functions of m6A in regulating mRNA decay, translation, and processing, the pathologic significance of m6A in MM has remained unknown. In the present study, we found that m6A levels were significantly elevated in MM due to upregulation of the methyl reader HNRNPA2B1. Our observations in MM cells with HNRNPA2B1 knockdown indicated that HNRNPA2B1 plays a critical role in promoting MM proliferation and inhibiting MM apoptosis. In addition, we found that TLR4 is a downstream target of m6A modification mediated by HNRNPA2B1. In summary, our results suggest that HNRNPA2B1 and its associated m6A modification promote MM progression by regulating the expression of the key gene TLR4 at the post-transcriptional level. Therefore, our findings reveal a new epigenetic regulatory mechanism contributing to the progression of MM.
HNRNPA2B1 participates in multiple biological processes in many diseases, especially cancers. One study demonstrated that down-regulation of HNRNPA2B1 can inhibit the cell proliferation, tumor invasion, and cell cycle progression through the PI3K/AKT signaling pathway of cervical cancer cells, thus triggering cell apoptosis [
18]. HNRNPA2B1 also was identified as an oncogene in head and neck cancer, which promotes epithelial to mesenchymal transition through the AKT/PKB signaling pathway [
19]. Additionally, the protein level of HNRNPA2B1 can be modulated by its ubiquitination status through miR503HG, thereby regulating HCC metastasis and migration in hepatocellular carcinoma (HCC) [
20]. Recently, the role of HNRNPA2B1 in MM was also explored by Jiang and colleagues, who found that HNRNPA2B1 expression is a negative prognostic factor in MM. Mechanistically, interleukin enhancer binding factor 3 (ILF3) was identified as an m6A target site of HNRNPA2B1 [
21]. In the present study, we identified TLR4 as the HNRNPA2B1 target site. Notably, we did not observe ILF3 among the core genes identified by RNA-sequencing and MeRIP sequencing. These inconsistent results may be due to different experiment conditions, as well as slight differences in the sequencing technologies employed.
TLR4, as a pattern recognition receptor, mainly mediates endogenous immunity and immune presentation, regulating the inflammatory response and participating in the expression of inflammatory factors. Recent studies have suggested that TLR4 expression correlates with cancer progression and TLR4 plays a key role in disrupting tumor cell apoptosis regulation, decreasing tumor cell apoptosis, and promoting tumor cell proliferation. For example, TLR4 activation releases immunosuppressive exosomes, promotes tumor progression, and accelerates the metastatic process [
22]. In MM, TLR4 expression was increased in the bone marrow cells of MM patients compared to healthy volunteers. Flow cytometric analyses also indicated upregulation of TLR4 in MM cells [
23]. Mechanistically, TLR4 suppresses ER stress-related apoptosis and promotes MM cell proliferation and survival through the PERK-CHOP pathway [
24]. A recent study highlighted the overexpression of TLR4 according to disease stage and implied that TLR4 inhibition reduced mesenchymal stromal cell activity and decreased MM cell growth [
25]. Consistent with these previous studies, our results also imply that a high level of TLR4 expression correlates with poor prognosis in MM patients.
m6A methylation and its regulators links epigenetic transcriptomics to tumor development. Studies have shown that m6A modification plays critical and diverse roles in the progression of various cancers [
26]. m6A methylation and its enzyme METTL3 are upregulated in glioblastoma stem cells and HCC, and the high expression of METTL3 induces SOCS2 degradation by m6A methylation, thus promoting tumor cell growth and survival [
27]. In another study, fat mass and obesity-associated protein (FTO), an m6A eraser, showed altered expression in intrahepatic cholangiocarcinoma, and low FTO expression predicted poor prognosis [
28]. These studies imply oncogenic functions of m6A methylation. On the other hand, several studies have shown that m6A may work as a tumor suppressor in some cancers. For example, one study found that METTL4 down-regulation led to m6A modification dysregulation, which was related to tumor metastasis and poor prognosis in HCC [
29]. In our study, we suggested that m6A modification has an oncogenic function in MM. Dysregulation of the m6A level seems to act as a “double-edged sword” in cancer progression, and further studies are needed to elucidate the key role of m6A.
The findings of this study must be considered in light of some limitations. According to the results in Fig.
1B, HNRNPA2B1 and eukaryotic translation initiation factor 3 (eIF3, B, D, E, K) were both overexpressed genes. eIF3 is the most complex eukaryotic translation initiation factor and is crucial for tumorigenesis [
30‐
32]. eIF3 regulates cap-dependent translation processes and also plays an important role in cap-dependent translation regulation by binding to putative internal ribosome entry sites [
33,
34]. The role of eIF3 in MM is not fully understood, and this can be a future direction of research. eIF3 may also represent a potential mechanism for the regulation of TLR4 in MM. However, our study did not include experiments to exclude it. Further exploration of the role of eIF3 in MM is needed in the future.
In summary, the results of the present study indicate that HNRNPA2B1 acts as oncogene in MM development. Our experiments suggested the HNRNPA2B1 promotes MM proliferation and inhibits apoptosis through the epigenetic regulation of TLR4 by m6A modification. Moreover, HNRNPA2B1 expression was significantly increased in MM and shown to correlate with poor prognosis in MM patients. Altogether, our results suggest HNRNPA2B1 as a potential therapeutic target for MM.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.