Epigenetic silencing of long non-coding RNA BM742401 in multiple myeloma: impact on prognosis and myeloma dissemination

Background Long non-coding RNA (lncRNA) BM742401 is a tumor suppressor in gastric cancer and chronic lymphocytic leukemia. As the promoter and coding region of BM742401 are fully embedded in a CpG island, we hypothesized that BM742401 is a tumor suppressor lncRNA epigenetically silenced by promoter DNA methylation in multiple myeloma. Methods Methylation-specific PCR and quantitative bisulfite pyrosequencing were performed to detect the methylation of BM742401 in normal plasma cells, myeloma cell lines and primary myeloma samples. The expression of BM742401 was measured by qRT-PCR. The function of BM742401 in multiple myeloma cells was analyzed by lentivirus transduction followed by migration assay. Results BM742401 methylation was detected in 10 (66.7%) myeloma cell lines but not normal plasma cells, and inversely correlated with expression of BM742401. In primary samples, BM742401 methylation was detected in 3 (12.5%) monoclonal gammopathy of undetermined significance, 9 (15.8%) myeloma at diagnosis and 8 (17.0%) myeloma at relapse/progression. Moreover, BM742401 methylation at diagnosis was associated with inferior overall survival (median OS: 25 vs. 39 months; P = 0.0496). In myeloma cell line JJN-3, stable overexpression of BM742401 by lentivirus transduction resulted in reduced cell migration (P = 0.0001) but not impacting cell death or proliferation. Conclusions This is the first report of tumor-specific methylation-mediated silencing of BM742401 in myeloma, which is likely an early event in myelomagenesis with adverse impact on overall survival. Moreover, BM742401 is a tumor suppressor lncRNA by inhibiting myeloma cell migration, hence implicated in myeloma plasma cell homing, metastasis and disease progression.

the rate of 1% per year [3]. Genetically, multiple myeloma is a heterogeneous disease with about half of the patients carrying non-hyperdiploid karyotypes (such as recurrent translocations involving immunoglobulin gene located at 14q32), whereas the other half carrying hyperdiploid karyotype (such as trisomies of odd number chromosomes) [4]. Despite major advances, multiple myeloma remains an incurable disease [5,6].
Long non-coding RNA (lncRNA) is a novel class of RNA molecules of > 200 nucleotides in length without protein-coding capacity [7,8]. Functionally, lncRNAs may regulate gene expression at both transcriptional and post-transcriptional levels, and hence are involved in multiple biological processes including development, differentiation or carcinogenesis [9,10]. In particular, lncRNAs have been shown to be associated with the pathogenesis of multiple myeloma [11,12]. For instance, lncRNA CRNDE (colorectal neoplasia differentially expressed) was found to be upregulated in primary myeloma samples and cell lines as compared with healthy controls, and associated with poor OS, and knockdown of CRNDE inhibited myeloma cell proliferation and colony formation and increased apoptosis and cell cycle arrest in G0/G1 phase [13], suggesting an oncogenic role for CRNDE in myeloma. On the other hand, knockdown of lncRNA OIP5-AS1 has been shown to promote myeloma cell proliferation, cell cycle progression and inhibit apoptosis, suggesting OIP5-AS1 is a tumor suppressor in myeloma [14].
DNA methylation is an epigenetic mechanism for gene regulation without alteration of the DNA sequence [15], which refers to the addition of a methyl (-CH 3 ) group to carbon five position of the cytosine ring in a CpG dinucleotide catalyzed by DNA methyltransferases [16]. DNA regions enriched with CpG dinucleotides are called CpG islands [17,18]. In the mammalian genome, promoter-associated CpG islands are localized to or in close proximity to the promoter region of more than half of the human genes [19], and involved in the regulation of gene expression by DNA methylation [20]. Aberrant promoter DNA methylation contributes to carcinogenesis including blood cancers [21]. In normal cells, majority of promoter-associated CpG islands are unmethylated, associated with a euchromatin configuration, and hence transcriptionally ready or active for gene expression [22]. In contrast, cancer cells are characterized by global DNA hypomethylation, and locus-specific hypermethylation of promoter-associated CpG islands of tumor-suppressor genes, resulting in downregulation, and hence loss of tumor suppressor functions [23][24][25]. For instance, long non-coding RNA KIAA0495 has been shown to be silenced by promoter DNA methylation in myeloma [26].
By RNA-seq, BM742401, localized to 18q11.2, was found to be downregulated in gastric cancer cells compared with normal tissues, which was associated with poor survival in patients with gastric cancer, and hence a potential tumor suppressor. Moreover, ectopic overexpression of BM742401 inhibited gastric cancer metastasis through regulation of cell migration and invasion [27]. Recently, in chronic lymphocytic leukemia (CLL), BM742401 was also found to be a tumor suppressor lncRNA, which was frequently methylated in primary samples of CLL [28]. As the promoter and coding region of BM742401 are fully embedded in a CpG island, we hypothesized that BM742401 may also be a tumor suppressor lncRNA epigenetically silenced by promoter DNA methylation in multiple myeloma. To verify this hypothesis, we studied the methylation status of BM742401 promoter in healthy controls, myeloma cell lines and myeloma primary samples, and investigated its tumor suppressor function.

Patient information
Bone marrow samples were obtained from patients with MGUS (n = 24), newly diagnosed myeloma (n = 57) and myeloma relapse/progression (n = 47). Diagnosis of myeloma was based on standard criteria of the International Myeloma Working Group (IMWG) [29]. Complete staging work-up consisted of bone marrow examination, skeletal imaging, serum and urine protein electrophoresis, and/or serum free light chain levels. Of the 57 patients with newly diagnosed myeloma, there were 24 females and 33 males, with a median age of 71 (35-88) years. Apart from 11 patients lacking International Staging System (ISS) data [30], there were 10 stage I, 22 stage II, and 14 stage III cases. There were 12 IgA, 40 IgG, 4 light chain, and 1 nonsecretary myelomas. According to the IMWG criteria, "relapse" was defined as the reappearance of the same paraprotein detected by serum/urine protein electrophoresis, appearance of new bone lesion or extramedullary plasmacytoma, or unexplained hypercalcemia after prior complete remission; while "progression" as increase of M-protein by 25% from lowest confirmed response value with an absolute rise of serum M-protein of ≥ 0.5 g/dL [31]. The study has been approved by the Institutional Review Board of Queen Mary Hospital (UW 05-269 T/932), and written informed consent was obtained from patient for publication of this article and any accompanying data or images. DNA of patient samples are extracted from bone marrow buffy coat, whereby malignant plasma cells are enriched by ficoll gradient centrifugation.  [32] and MMKKF (unpublished) were established from the myelomatous pleural effusion of myeloma patients. Cell lines were cultured in RPMI-1640 medium (IMDM for LP-1, DMEM + IMDM for MMLAL), supplemented with 10% or 20% fetal bovine serum, 50 U/mL of penicillin and 50 μg/mL streptomycin, in a humidified atmosphere of 5% CO 2 at 37 °C. All culture reagents were purchased from Invitrogen (Carlsbad, CA, USA).

Methylation-specific polymerase chain reaction (MSP)
Detailed procedures of MSP have been previously described [33,34]. Primer sequences and conditions are in Table 1.

Quantitative reverse transcription polymerase chain reaction (qRT-PCR)
Total RNA was isolated using mirVana ™ miRNA Isolation Kit (Ambion, Austin, TX, USA). Reverse transcription was performed using QuantiTect Reverse Transcription Kit (Qiagen). BM742401 was quantified using SYBR Green Master Mix (Applied Biosystems, Waltham, MA, USA) with GAPDH as endogenous control. Primer sequences of qRT-PCR for BM742401 and GAPDH were listed in Table 1. Expression of BM742401 was calculated by ∆CT method.

Lentivirus transduction
The full-length cDNA of BM742401 was amplified and cloned into the XbaI and EcoRI sites of a pCDHCMV-MCS-EF1-copGFP lentivector (System Biosciences, Palo Alto, CA, USA; named empty vector) as described before [28], and the reconstructed vector was named BM742401 vector. BM742401 vector and empty vector were then co-transfected with pPACK packaging plasmid mix respectively into 293TN cells, followed by collection of supernatants at 48 h after transfection and concentration of pseudoviral particles by PEG-it ™ Virus Precipitation Solution (System Biosciences). After pseudoviral titer estimation using 293TN cells, JJN-3 cells were transduced for 48 h by the pseudoviral particles with multiplicity of infection at 4. GFP-positive JJN-3 cells were

Migration assay
To test the effect of BM742401 overexpression on myeloma cell migration, we used bone marrow stromal cells (BMSCs) as a source for secreting chemoattractant for myeloma cells.
In the migration assay, a pilot transwell experiment was conducted to find out the optimal experimental conditions. At 24 h before migration assay, JJN-3 cells transduced with empty vector were starved by washing with PBS and resuspending in RPMI-1640 medium without FBS. The next day, in each of the transwell permeable support (8.0-μm polycarbonate membrane, 6.5-mm insert, and 24-well plate; Corning Costar, Tewksbury, MA, USA), 1 × 10 5 starved JJN-3 cells were seeded in 200 μl RPMI-1640 medium. In the lower chamber, one of the following three conditions was used: (1) 500 μl RPMI-1640 medium with 20% FBS; (2) 500 μl BMSC conditioned medium (described below); or (3) 2 × 10 5 BMSCs in 500 μl DMEM medium with 10% FBS that had been seeded on the day before. BMSCs were cultured from normal bone marrow donors as previously described [35]. "BMSC conditioned medium" was generated by mixing the filtered culture medium of BMSCs (at 37 °C in 5% CO 2 for 24 h) with 20% FBS in RPMI-1640 medium at the ratio of 1:1. After 72 h of incubation at 37 °C, the GFP-positive cells that had migrated to the lower chambers were counted using fluorescence microscope (Axiovert 135, ZEISS microscopy, Germany). The rationale of the use of BMSCs in the lower chamber stemmed from the concept that myeloma cell homing is mediated by SDF-1 produced by BMSCs that bind to the CXCR4 receptor on myeloma cells, hence myeloma cells would migrate to the bone marrow niches due to this concentration gradient [36]. As the highest myeloma cell migration occurred with BMSCs laid at the bottom of the lower chamber (Additional file 1: Figure S1), which hence was adopted for subsequent transwell experiments to compare migration efficiency between JJN-3 cells transduced with empty vector and BM742401 vector. Triplicate experiments were performed for each group, and the means and standard deviations were calculated.

Trypan blue exclusion assay
Cell death was analyzed by trypan blue (Sigma-Aldrich) at day 3 after seeding cells. Cells in five random microscopic fields were counted for each group under microscope. Dead cell (stained in blue) percentage = average number of dead cells per microscopic field/average number of total cells per microscopic field.

MTS assay
The number of viable cells in proliferation was measure by CellTiter 96 ® AQ ueous One Solution Cell Proliferation Assay Kit (Promega, Madison, WI, USA) following the manufacturers' instructions. Relative proliferation percentage of BM742401 overexpressed cells compared with control cells was calculated at day 5 after seeding cells.

Statistical analysis
Overall survival (OS) was measured from the date of diagnosis to the date of last follow-up or death. OS of patients with and without BM742401 methylation were compared. Survival was plotted by the Kaplan-Meier method, and compared by the log-rank test. The difference between JJN-3 cells transduced with BM742401 vectors and empty vectors in migration assay was studied by Student's t test. All P values were two-sided and P < 0.05 was defined as significant difference.

Methylation of BM742401 in healthy controls and human myeloma cell lines (HMCLs)
MSP was carried out to examine methylation of BM742401 in the bisulfite-converted DNA of healthy controls [peripheral blood (n = 10) and CD138-sorted bone marrow plasma cell (n = 7)] and HMCLs (n = 15). Direct sequencing of the M-MSP products from positive control with methylated DNA confirmed complete bisulfite conversion and MSP specificity, as indicated by conversion of all unmethylated cytosines into thymidines after PCR, whereas all methylated cytosines remained unchanged (Fig. 1a). None of the healthy controls showed methylation of BM742401 (Fig. 1b). By contrast, in HMCLs, BM742401 was completely meth-  (Fig. 1c). Moreover, these MSP methylation statuses (MM, MU, and UU) were verified using quantitative bisulfite pyrosequencing, which showed that completely methylated HMCLs were associated with a higher methylation level between 63.4% to 85.4%, partially methylated HMCLs carried an intermediate methylation level of 36.9% to 49.6%, and completely unmethylated HMCLs were associated with a lower methylation level from 15.1% to 23.6% (Additional file 2: Figure S2). These data suggested that BM742401 was methylated in a tumorspecific manner in myeloma.

Methylation and expression of BM742401 in HMCLs
To study if methylation was correlated with repression of BM742401, qRT-PCR was employed to measure the expression levels of BM742401 in HMCLs. Results showed that HMCLs with methylation of BM742401 had significantly lower expression levels of BM742401 ( Fig. 2a; MM vs. UU, P = 0.041; MM + MU vs. UU, P = 0.047) than HMCLs that were completely unmethylated.
To further testify if promoter DNA methylation resulted in downregulation of BM742401, MOLP-8 cells, which were completely methylated for BM742401, were treated with 5-AzadC, a demethylation agent. Upon treatment with 5-AzadC, the promoter of BM742401 was demethylated as evidenced by the emergence of U-MSP signal on day 5 (Fig. 2b). Moreover, by qRT-PCR, BM742401 was simultaneously re-expressed by 5.2 to 9.9 folds with different concentrations of 5-AzadC (Fig. 2b). Therefore, in myeloma cells, methylation-mediated silencing of BM742401 was reversible.

Methylation of BM742401 in primary bone marrow samples
By MSP, methylation of BM742401 was detected in primary bone marrow samples of 3 (12.5%) MGUS, 9 (15.8%) myeloma at diagnosis, and 8 (17.0%) myeloma at relapse/progression (Fig. 3a). Methylation frequency of BM742401 was not significantly different among those consecutive clinical stages of myeloma (MGUS vs. myeloma at diagnosis: P = 1.000; myeloma at diagnosis vs. myeloma at relapse/progression: P = 1.000). In contrast to absence of methylation in normal, presence of methylation in MGUS with a frequency comparable to consecutive stages from MGUS to myeloma at diagnosis and relapse/progression indicated BM742401 methylation might be an early event in the pathogenesis of myeloma.

Tumor suppressive function of BM742401 in myeloma cells
As BM742401 was frequently methylated in HMCLs and primary samples, we postulated that it might act as a tumor suppressor. By lentivirus transduction, BM742401 was stably overexpressed by 9397.0 folds in JJN-3 cells compared with empty vector control ( Fig. 4a and Additional file 3: Figure S3; P = 0.0009). Moreover, overexpression of BM742401 resulted in reduced cell migration of JJN-3 cells by transwell migration assay (Fig. 4b and  c; P = 0.0001), but not affecting cell death ( Fig. 4d; P = 0.1009) or proliferation ( Fig. 4e; P = 0.2401) by trypan blue exclusion assay and MTS assay respectively. Therefore, BM742401 exhibits its tumor suppressor property in myeloma by inhibiting cell migration.

There are a number of interesting observations in this study
Firstly, methylation of BM742401 was tumor-specific as it was absent in normal controls, whereas frequently detected in HMCLs and primary myeloma samples, which is similar to the tumor-specific methylation of other tumor suppressive protein coding genes [37,38], miRNAs [39,40] and lncRNA [26] in myeloma. In contrast, methylation of some miRNAs, such as miR-9-2 and miR-373 [41,42], occurred in both cancer cells and their normal counterparts, and hence methylated in a tissue-specific manner, thereby unimportant in carcinogenesis.
Secondly, methylation-mediated silencing of BM742401 was shown to be reversed by treatment of demethylating agent, consistent with the reversible silencing of BM742401 shown in CLL [28], indicating that promoter DNA methylation is also a mechanism for repression of tumor suppressor lncRNAs in myeloma.
Thirdly, in primary samples, methylation of BM742401 appeared as early as MGUS, at a frequency comparable to that of active myeloma at diagnosis and relapse/progression. Therefore, it is likely that methylation of BM742401 is an early event in the pathogenesis of myeloma, similar to methylation of miR-203 [40] and miR-342 [34]. By contrast, miR-129-2 methylation was implicated in the progression from MGUS to symptomatic myeloma [43], and miR-34b/c methylation at relapse/progression of myeloma [39]. Moreover, methylation of BM742401 correlated with shorter OS in newly diagnosed myeloma, similar to CDKN2A [44,45] and DAPK1 [46] methylation, suggesting an