This study was performed with the approval of the Institutional Review Board of Sichuan Provincial People's Hospital of China. Informed written consent was collected from each eligible patient, and the whole study was performed in accordance with the Declaration of Helsinki.

Between January 2017 and December 2018, 66 patients with MM (age range, 38–88 years; 37 men; 29 women) and ten healthy controls (age range, 31–83 years; six men; four women), who were treated at the Sichuan Academy of Medical Sciences and Sichuan Provincial People's Hospital were enrolled. The diagnostic criteria for active MM were based on the International Myeloma Working Group guidelines (18). The criteria for classification were according to the subtype of abnormal proliferative immunoglobulin. The criteria for staging were assessed using the R-ISS (3). The healthy control group consisted of transplant donors undergoing BM transplantation, who were recruited to provide BM (~1 ml) and peripheral blood samples (~5 ml).

All patients were treated with bortezomib (1.3 mg/m² subcutaneously on days 1, 4, 8 and 11 of every 21-day cycle or 1.3 mg/m² subcutaneously on days 1, 8, 15 and 22 of every 35-day cycle; three manufacturers permitted by medical insurance: QILU Pharmaceutical; Chia Tai Tianqing Pharmaceutical; Hansoh Pharmaceutical) in combination with dexamethasone (Tianjin Jinyao Pharmaceutical Co., Ltd.; 10 or 20 mg on the day of, and the day after the administration of each dose of bortezomib). In high-risk patients, either cyclophosphamide (300 mg/m² on days 1 and 8; Baxter Oncology GmbH) or lenalidomide (10 or 25 mg based on renal function measured from days 1–21; Beijing Shuanglu Pharmaceutical Co., Ltd.) was added to the combination. A total of 8–12 cycles were planned as induction therapy for all patients, and after induction therapy, patients entered maintenance therapy with either bortezomib (1.3 mg/m² once every 2 weeks for 2 years) or lenalidomide (25 mg once a day on days 1–21 and every 28 days until relapse).

Samples were initially collected from patients at the first diagnosis and at the required follow-up time points (once every 3 months); 5 ml blood was collected in serum collection tubes. Whole blood was allowed to stand for ~1 h at room temperature, before centrifugation at 1,200 × g for 10 min at 4°C, followed by the separation of serum. Successive centrifugation was performed for 10 min at 10,000 × g at 4°C to remove the cellular debris. The resulting serum was aliquoted into Eppendorf tubes and stored at −80°C to determine IL-6 and circulating miR-451a levels.

BM samples collected from both patients and controls were anticoagulated with EDTA-K₂ (BD Biosciences) at room temperature. Some BM samples were diluted 1:1 with RPMI-1640 basic medium (Gibco; Thermo Fisher Scientific, Inc.). Mononuclear cells were isolated with Ficoll cell isolation medium at room temperature (density 1.077 g/ml; Stemcell Technologies, Inc.). Then, plasma cells were isolated using CD138 magnetic beads at room temperature (Miltenyi Biotec, Inc.) for RNA extraction to analyze miR-451a expression and proteins were extracted to measure IL-6R levels. The remaining BM was used for MFC.

To concentrate the circulating miRNA, the extraction protocol was conducted by adding 10% PEG 8,000 solution (Beijing Solarbio Science & Technology Co., Ltd.). Total RNA was extracted from 500 µl serum using a mirVana miRNA Isolation kit (cat. no. AM1561; Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions. Then, miR-451a was reverse transcribed with a Bulge-LoopTM miRNA RT-qPCR Starter kit (Guangzhou RiboBio Co., Ltd.). The reaction conditions were: 42°C for 60 min and 70°C for 10 min. Relative quantification was performed via qPCR using the SYBR Green PCR kit (Guangzhou RiboBio Co., Ltd.). External references (cel-miR-39; Guangzhou RiboBio Co., Ltd.) were used to compare miRNA levels among different groups. The fold change was calculated based on the normalized mean differences (2^−ΔΔCq) (19). The PCR cycling conditions were as follows: Initial denaturation at 95°C for 10 min, followed by 40 cycles at 95°C for 2 sec, 60°C for 20 sec and 70°C for 10 sec. The following primers for miR-451a were used: Sense, 5′-ACCGTTACCATTACT-3′ and antisense, 5′-CTCACACGACTCACGA-3′. All reactions, including no-template controls, were performed in triplicate. Amplification and data analysis were performed using an ABI Prism 7,000 thermocycler (Applied Biosystems; Thermo Fisher Scientific, Inc.).

Primary plasma cells were purified from BM aspirates obtained from patients with MM and healthy volunteers using CD138 magnetic beads (Miltenyi Biotec, Inc.). Total RNA of cells was extracted using Tripure isolation reagent (Roche Diagnostics, Inc.) and reverse transcribed into cDNAs with a first stand cDNA synthesis kit using random primers (Roche Diagnostics). The duration of RT was as following: 25°C for 10 min, 55°C for 30 min, 85°C for 5 min, 10°C for 15 min and end at 4°C. qPCR was performed to analyze IL-6R expression using SYBR Premix Ex Taq (Takara Bio, Inc.). The qPCR consisted of an initial denaturation step at 95°C for 5 min, followed by 39 cycles at 95°C for 10 sec, 60°C for 30 sec, and a final elongation step at 95°C for 15 sec, 60°C for 1 min and 95°C for 15 sec. The following primers for IL-6R were used: Sense, 5′-CCTCTGCATTGCCATTGTTC-3′ and antisense, 5′-GAGATGAGAGGAACAAGCAC-3′.

The serum level of IL-6 in serum was measured using a human IL-6 ELISA kit (Abcam; cat. no. ab46027), according to the manufacturer's instructions. After color development was stopped, the absorbance was measured using a microtiter plate-computerized reader set to a wavelength at 450 nm. The sensitivity for IL-6 was 2 pg/ml.

Western blotting was used to analyze the levels of IL-6R in plasmocytes. Proteins were extracted from cells with a lysis buffer containing PMSF (Beijing Solarbio Science & Technology Co., Ltd.) and concentrations were measured using a BCA protein assay kit (Tiangen Biotech Co., Ltd.). A total of 20–30 µg protein extracts were subjected to electrophoresis on 10% SDS-PAGE gels and transferred to nitrocellulose membranes. The membranes were blocked with 10% skimmed milk for 2 h at room temperature and then incubated with the following primary antibodies overnight at 4°C: Rabbit anti-IL-6R (1:400; Abcam; cat. no. ab27321) and rabbit anti-GAPDH (1:1,000; Abcam; cat. no. ab8245). After incubation with the secondary antibody conjugated with horseradish peroxidase (1:2,000; Abcam; cat. no. ab6789) for 2 h at room temperature, the proteins were visualized with the Molecular Imager Gel Doc XR System (Bio-Rad Laboratories, Inc.) using Metal Enhanced DAB Substrate kit (Beijing Solarbio Science & Technology Co., Ltd.).

In follow-up experiments, MRD was assessed using MFC. A fixative-free erythrocyte lysis buffer (Beckman Coulter, Inc.) was used for the phenotypic characterization of the majority of participants. Intraprep permeabilization reagent (cat. no. A07803; Beckman Coulter, Inc.) was used to analyze intracellular immunoglobulin levels at room temperature. The number and viability of cells obtained were determined using staining cells with trypan blue for 10 min at room temperature and counted under a microscope. If the viability of recovered cells was ≥90%, 20 million cells were stained with two independent 6-color panels (10 million each): One tube contained CD19-FITC, CD20-PE, CD56-ECD, CD38-PECY5.5, CD138-APC and CD45-PECy7, and the other tube contained CD19-FITC, CD117-PE, CD56-ECD, CD38-PECY5.5, CD138-APC and CD45-PECy7. The intracytoplasmic panel consisted of the markers Cyκ-APC, Cyλ-APC750, CD19-PE, CD38-FITC and CD138-APC. In total, 10 million cells suspended at a final volume of 200 µl/tube were labeled in each tube. A minimum of 5 million events were recorded per tube in the mode of middle speed in a flow cytometer (Navios' Beckman Coulter, Inc.). The software used for analysis was Kaluza (version no. 2.1; Beckman Coulter, Inc.). Cell populations are considered abnormal if they have an atypical differentiation pattern, an increased or decreased expression level of normal antigens, an asynchronous maturational pattern or express aberrant antigens (20).

For miR-451a mimic transfections, the miR-451a mimic (5′-AAACCGUUACCAUUACUGAGUU-3′; 50 nM; Guangzhou RiboBio Co., Ltd.) and the corresponding negative control (50 nM; Guangzhou RiboBio Co., Ltd.) were separately transfected into 293T cells (purchased from China Center for Type Culture Collection) were performed separately in the absence of any other treatments. The transfections were performed using Lipofectamine 3000 (Thermo Fisher Scientific, Inc.). Then, 48 h after transfection, the expression of miR-451a was measured using RT-qPCR to confirm successful transfections, and the subsequent effect was only determined in cells expressing miR-451a.

The online software TargetScan (TargetScan Human 7.0; http://www.targetscan.org/vert_70/) was used to investigate the relationship between miR-451a and IL-6R. The predicted binding site for miR-451a to human IL-6R 3′UTR was 5′-UUGCCAAA-3′ (17,21). The human IL-6R 3′UTR with a mutation in the miR-451a seed sequence (mutant; mut) and 3′UTR (wild-type; WT) were amplified and inserted between the restriction sites XhoI and NotI of the firefly and Renilla luciferase reporter vector pmiR-RB-REPORT (Guangzhou RiboBio Co., Ltd.), respectively. Cells were plated on 24-well plates at 1×10⁵ cells/well and transfected with luciferase reporter vectors (30 ng) using Lipofectamine 3,000 (Thermo Fisher Scientific, Inc.). At 24 h post-transfection, firefly and Renilla luciferase activities were measured using a Dual-Glo Luciferase assay system (Promega Corporation), according to the manufacturer's instructions. The Renilla luciferase signal was normalized to that of the firefly luciferase signal for each individual analysis.

Data are presented as the medians and interquartile ranges. The experiments were repeated three times. The Kolmogorov-Smirnov test was used to analyze whether the data were normally distributed. The Kruskal-Wallis rank sum test was used to analyze differences in the data among the four groups. Dunn's test was used as the post hoc test after Kruskal Wallis. Comparisons between two groups were performed using the Mann-Whitney U test. After the Mann-Whitney U test, a Bonferroni was conducted to correct the P-value. Correlations were determined by calculating Spearman's correlation coefficients. Statistical analyses were performed using SPSS version 19.0 (IBM Corp.). P<0.05 was considered to indicate a statistically significant difference.

In the present study, the median age of the patients was 63.0 years (age range, 38–88 years), and 56.1% of patients were male. The patients received immunomodulatory drugs or proteasome inhibitors as induction therapy without autologous stem cell transplantation. In total, 15 of these patients (22.7%) were R-ISS stage I, 17 patients (25.8%) were R-ISS stage II and 34 patients (51.5%) were R-ISS stage III (Table I).

Low circulating miR-451a levels were detected in patients with MM, and were only 0.39 times the value of the control group (U=4.00; P<0.001; Fig. 1A). Among the 66 patients with MM, the median expression of miR-451a was 0.73 times the value of the control group in R-ISS stage I MM; and that in R-ISS stage II MM was 0.41 times the value of the control group (Fig. 1B). Moreover, patients with R-ISS stage III MM presented the lowest level, at 0.24 times the value of the control group (Fig. 1B).

The expression of miR-451a was compared between cells in the BM and circulation to evaluate the consistency. The expression of miR-451a in the circulation exhibited a strong positive correlation with miR-451a expression in BM (R^s=0.98; P<0.001; Fig. 1C). The concentration of IL-6 was strongly, negatively correlated with miR-451a expression (R^s=−0.97; P<0.0001; Fig. 1D). The plasma cells from patients with MM demonstrated an upregulated expression of IL-6R compared with the control subjects via western blotting. Furthermore, the expression levels increased gradually from stage I to stage III (Fig. 1E). IL-6R levels were strongly and negatively correlated with miR-451a expression (R^s=−0.95; P<0.001; Fig. 1F).

After two courses of consolidation chemotherapy, 19 patients achieved complete remission (CR). In the 2-year follow-up period, MRD was observed using MFC and changes in circulating miR-451a expression were evaluated. For 10 patients, the expression of circulating miR-451a remained steady in the follow-up period (seven patients with R-ISS stage I, two with R-ISS stage II and one with R-ISS stage III). The change in miR-451a expression was <50%. The 10 patients were in continuous CR (Fig. 2A). However, the other nine patients (two patients with R-ISS stage I, three with R-ISS stage II and four with R-ISS stage III) demonstrated an abrupt decrease in circulating miR-451a expression during the follow-up period after CR (Fig. 2B-J). The value at the observation point was >50% lower compared with the value measured 3 months prior (50%; Fig. 2G-J). The turning points in the trend appeared 4–8 weeks before abnormal plasma cells were obtained via MFC (Fig. 3A-I). Thus, when MFC results were still negative, significant changes in circulating miR-451a expression had occurred. Subsequently, clinical relapse occurred in these nine patients.

Significantly higher miR-451a expression was observed in the mimic group compared with the mimic control cells and in non-transfected cells (both P<0.001; Fig. 4A). The dual-luciferase reporter assay results revealed that co-transfection of the miR-451a mimic and IL-6R-wild-type (WT) resulted in decreased the luciferase activity compared with co-transfection of the mimic control and IL-6R-WT (P<0.001; Fig. 4B). However, there was no effect of co-transfection of the miR-451a mimic and IL-6R-mut had no effect on the luciferase activity (P>0.05; Fig. 4B). Furthermore, the overexpression of miR-451a significantly downregulated IL-6R expression both at the mRNA (P<0.001; Fig. 4C) and protein (Fig. 4D) levels. Collectively, these results suggested that IL-6R was a target gene of miR-451a and was negatively regulated by miR-451a (Fig. 4E).