Overexpression of B-cell lymphoma 6 alters gene expression profile in a myeloma cell line and is associated with decreased DNA damage response.

B-cell lymphoma 6 (BCL6) attenuates DNA damage response (DDR) through gene repression and facilitates tolerance to genomic instability during immunoglobulin affinity maturation in germinal center (GC) B cells. Although BCL6 expression is repressed through normal differentiation of GC B cells into plasma cells, a recent study showed the ectopic expression of BCL6 in primary multiple myeloma (MM) cells. However, the functional roles of BCL6 in MM cells are largely unknown. Here, we report that overexpression of BCL6 in a MM cell line, KMS12PE, induced transcriptional repression of ataxia telangiectasia mutated (ATM), a DDR signaling kinase, which was associated with a reduction in γH2AX formation after DNA damage. In contrast, transcription of known targets of BCL6 in GC B cells was not affected, suggesting a cell type-specific function of BCL6. To further investigate the effects of BCL6 overexpression on the MM cell line, we undertook mRNA sequence analysis and found an upregulation in the genomic mutator activation-induced cytidine deaminase (AID) with alteration in the gene expression profile, which is suggestive of de-differentiation from plasma cells. Moreover, interleukin-6 exposure to KMS12PE led to upregulation of BCL6 and AID, downregulation of ATM, and attenuation of DDR, which were consistent with the effects of BCL6 overexpression in this MM cell line. Taken together, these results indicated that overexpression of BCL6 alters gene expression profile and confers decreased DDR in MM cells. This phenotypic change could be reproduced by interleukin-6 stimulation, suggesting an important role of external stimuli in inducing genomic instability, which is a hallmark of MM cells.

Multiple myeloma is a plasma cell neoplasm characterized by abnormal clonal plasma cell infiltration in the bone marrow. Genomic instability is one of the hallmarks of MM and affects the onset and progression of the disease.1, 2 However, the underlying mechanisms causing the instability of the genome in MM are largely unknown.3, 4, 5 Recent whole genome sequencing studies have revealed infrequent mutations of DNA repair pathway genes in MM.6, 7 Thus, the question is still to be answered.

Abnormalities in DNA repair pathways have been extensively studied in B‐cell neoplasms such as DLBCL. B‐cell lymphoma 6 is a transcriptional repressor originally identified by its involvement in 3q27 chromosomal translocations associated with DLBCL.8, 9 It is expressed in GC B cells and facilitates simultaneously rapid proliferation and tolerance to the genomic instability, which occurs during the process of immunoglobulin affinity maturation mediated by AID.10, 11, 12 B‐cell lymphoma 6 directly suppresses the genes involved in DNA damage sensing and cell cycle arrest, such as ATR, p53, and p21,13, 14, 15 and thereby attenuates the DDR.

Although BCL6 is generally downregulated as B cells differentiate into plasma cells, a recent study reported that BCL6 is upregulated in MM cells in the bone marrow microenvironment.16 However, the functional role of BCL6 in MM has not yet been well established. Therefore, we investigated the effects of BCL6 expression in MM cells, focusing in particular on DDR and genomic instability.

In this study, we established an MM cell line (KMS12PE‐BCL6) stably overexpressing BCL6, and examined the effects of BCL6 overexpression on the expression of DDR genes and γH2AX formation on DNA damage. Additionally, mRNA sequence analysis was carried out to detect differentially expressed genes in KMS12PE‐BCL6 cells. We found that BCL6 overexpression to KMS12PE led to upregulation of AID, downregulation of ATM, and attenuated DDR. Gene expression profiles in KMS12PE‐BCL6 cells are suggestive of de‐differentiation of plasma cells to GC B cells. These results imply that BCL6 overexpression is involved in genomic instability in MM cells.

Materials and Methods

Isolation of primary tumor cells

Bone marrow samples obtained from 28 newly diagnosed MM patients were subjected to the study after written informed consent. This study was approved by institutional review boards of Gunma University Hospital (Gunma, Japan) and Nishigunma National Hospital (Gunma, Japan). CD138 positive bone marrow plasma cells were isolated as a purity of >95% using the EasySep human PE‐positive selection kit (Stemcell Technologies, Vancouver, BC, Canada). These cells were used for the RNA isolation.

Cell culture, plasmids, transfection, and retrovirus infection

Human MM cell lines, KMS11, KMS12PE, KMS12BM, KMS18, KMS26, ARH77, and RPMI8226 were provided by Dr. Otsuki (Kawasaki Medical School, Okayama, Japan), and the human DLBCL cell line, CTB‐1, was provided by the RIKEN BioResource Center through the National BioResource Project of the Ministry of Education, Culture, Sports, Science and Technology, Japan. These cell lines were cultured in RPMI‐1640 medium (Life Technologies, Delhi, India) with 10% FBS (Sigma‐Aldrich, St. Louis, MO, USA). Overexpression of BCL6 in KMS12PE cells was achieved by infection of the cells with a retroviral vector encoding BCL6 (pMY‐BCL6‐IRES‐EGFP). The virus was obtained and infected as previously described.17 The EGFP‐positive infected cells were isolated using a BD FACSAria II cell sorter (BD Biosciences, Franklin Lakes, NJ, USA).

Real‐time PCR

RNA was extracted using RNeasy Mini Kit (Qiagen, Venlo, the Netherlands) and cDNA was prepared using high‐capacity RNA‐to‐cDNA kit (Life Technologies). Real‐time PCR was carried out using a standard TaqMan PCR kit protocol on a 7300 Real‐time PCR System (Applied Biosystems, Waltham, MA, USA). Gene expression was normalized to that of GAPDH. The data is shown as the mean ± SD (n = 3 per bar). The statistical significance was determined using Student's t‐test. P < 0.01 was considered significant.

Quantitative ChIP

Chromatin immunoprecipitation was carried out using ChIP reagents (Nippon Gene, Tokyo, Japan), according to the manufacturer's instructions. Anti‐BCL6 (N‐3) (sc‐858; Santa Cruz Biotechnology, Santa Cruz, CA, USA) or anti‐IgG control (ab46540; Abcam, Cambridge, UK) and Dynabeads Protein G (Life Technologies) were incubated at 4°C overnight. Primers for the each promoter region of ATR and p53 genes and enhancer for p21 have been reported.13, 14, 18

Flow cytometry

For BCL6 staining, we used Foxp3 Staining Buffer Set (eBioscience, San Diego, CA, USA) and Alexa Fluor 647‐anti‐BCL6 antibody (clone K112‐91) (BD Biosciences) at 1:250 dilution. For γH2AX, the cells were stained as previously described,19 with FITC‐anti‐γH2AX (Merck Millipore, Darmstadt, Germany). After 3 h of incubation on ice, γH2AX was measured. For the CD138 assay, PE‐conjugated anti‐CD138 antibody (Beckman Coulter, Brea, CA, USA) was used. All measurements were carried out on a FACSCanto II Flow Cytometer (BD Biosciences) and analyzed with FlowJo software (TreeStar, San Carlos, CA, USA). The statistical significance was determined using the χ2‐test by the population comparison platform of FlowJo. P < 0.01 was considered significant.

Treatment with BCL6 BTB domain inhibitor 79‐6 and IL‐6

The BCL6 inhibitor 79‐6 (Merck Millipore) was dissolved in DMSO. The BCL6‐overexpressed KMS12PE (KMS12PE‐BCL6) cells (5 × 105/mL) were exposed to 50 μM 79‐6 or DMSO control for 8 h for RNA quantification. KMS12PE cells (2.5 × 105/mL) were treated with 100 or 122 ng/mL recombinant human IL‐6 (R&D Systems, Minneapolis, MN, USA) or PBS supplemented with 0.1% BSA as the control for 16 h, then harvested for RNA extraction.

DNA damage induction

For induction of DNA damage, cells were exposed to 0, 3, 5, and 10 Gy IR using the RX‐650 cabinet X‐ray system (Faxitron X‐ray, Tucson, AZ, USA) and then incubated at 37°C for 1 h before analysis. Cells were also treated with different concentrations of etoposide (0, 1, 5, 10, 50, and 100 μM) for 30 min, washed with fresh media, and incubated at 37°C for 1 h before analysis. Formation of γH2AX was assessed by flow cytometry and immunofluorescence staining. For real‐time PCR and immunoblot analysis of DDR genes, cells were incubated at 37°C and collected 30 min after irradiation.

Immunofluorescence staining

After irradiation and incubation for 1 h at 37°C, cells were permeabilized with 0.5% Triton X and blocked and stained with anti‐γH2AX antibody (ab22551; Abcam) at 1:800 dilution. Cy3 conjugated donkey anti mouse IgG (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) was used as a secondary antibody, at 1:500 dilution for 1 h, and mounted with Prolong Gold with DAPI (Life Technologies). All the images were captured by a Leica DMLB fluorescent microscope (Leica Microsystems, Wetzlar, Germany). The mean density of γH2AX expression per nuclei were measured using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Immunoblot analysis

Cells were harvested and lysed in RIPA lysis buffer (Santa Cruz Biotechnology), then frozen and thawed twice, centrifuged at 20 600 for 10 min. The supernatant was collected as whole cell lysates. The protein (80 μg) was used for the immunoblot, described previously.17 Band densities were quantified with ImageJ software, and the relative protein amount was determined by comparison of the protein/β‐actin ratios. The following antibodies were used for immunoblot analysis: ATM (2C1[1A1], ab78), phospho‐ATM (Ser1981) (10H11.E12, ab36810), phospho‐ATR (Ser428) (EPR2184, ab178407), β‐actin (ab8227; all Abcam), BCL6 (N‐3, sc‐858), ATR (N‐19, sc‐1887), p53 (DO‐1, sc‐126), p21 (C‐19, sc‐397; all Santa Cruz Biotechnology), phospho‐p53 (Ser15) (#9284; Cell Signaling Technology, Beverly, MA, USA), rabbit IgG‐HRP, mouse IgG‐HRP (both R&D Systems), and goat IgG‐PO (Jackson ImmunoResearch Laboratories).

Cycloheximide chase assay

Cycloheximide (Wako Pure Chemicals, Tokyo, Japan) was dissolved in DMSO. KMS12PE‐BCL6 cells were exposed to cycloheximide (80 μg/mL) for 1 h at 37°C, and irradiated with 0 or 10 Gy. At 0, 0.5, 1, 2, and 4 h after IR, cells were harvested and lysed in RIPA lysis buffer, and supernatant was collected. Twenty‐five micrograms of protein was separated by SDS‐PAGE and used for immunoblot analysis as described above.

Next generation mRNA sequencing

The RNA concentration and purity were determined using a NanoPhotometer (Implen, Munich, Germany) and the integrity was verified using an Agilent RNA 6000 Pico Kit and Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA). The RNA integrity number for all samples was at least 9. Libraries were generated using a NEBNext Ultra Directional RNA Library Prep Kit for Illumina (New England BioLabs, Ipswich, MA, USA). The libraries were quantified with EnSpire (PerkinElmer, Waltham, MA, USA) and validated using an Agilent 2100 Bioanalyzer for size, and then sequenced using NextSeq 500 (Illumina, San Diego, CA, USA). The count file was generated by RNA‐express software version 1.0 (Illumina), and was moved onto EdgeR to identify differentially expressed genes. Genes with a false discovery rate <0.05 were defined as significantly different. The mRNA sequence data were submitted to the DDBJ/EMBL/GenBank databases under accession number DRA004263.

Results

Overexpression of BCL6 to KMS12PE cell line did not repress expression of DDR genes

To determine the role of BCL6 in MM cells, we first examined the expression levels of BCL6 in tumor cells in the bone marrow of MM patients. We quantified the RNA expression levels by real‐time PCR. Expression levels of BCL6 in CD138‐positive cells from 28 newly diagnosed patients distributed in a wide range (Fig. 1a). Next we tested seven MM cell lines and a BCL6‐positive DLBCL cell line, CTB‐1, as the positive control. Unlike the tumor cells from the patients, all MM cell lines examined expressed very low levels of BCL6 (Fig. 1b). To investigate the function of BCL6 in MM cells, we overexpressed BCL6 in KMS12PE cells, which expressed BCL6 at the highest level among the tested cell lines. We used a retrovirus system to obtain the cells overexpressing BCL6 (KMS12PE‐BCL6) and the mock‐infected cells (KMS12PE‐Empty). The expression level of BCL6 after retroviral infection was comparable to that of CTB‐1 and the median of that in these patients (Fig. 1c). Increased protein levels of BCL6 were also confirmed by immunoblotting and flow cytometry analysis (Fig. 1d,e, respectively).

Figure 1

B‐cell lymphoma 6 (BCL6) overexpression in the KMS12PE multiple myeloma (MM) cell line did not repress the expression of DNA damage response genes. (a) Expression of BCL6 RNA in CD138‐positive cells from bone marrow samples of newly diagnosed MM patients was measured by real‐time PCR and is presented as the fold change relative to the median (dotted line). (b) Real‐time PCR was used to measure the expression levels of RNA in MM cell lines KMS11, KMS12PE, KMS12BM, KMS18, KMS26, ARH77, and RPMI8226; diffuse large B‐cell lymphoma cell line CTB‐1 was used as the positive control. The expression levels are presented as the fold change relative to that of KMS12PE. (c) Expression of RNA in CTB‐1, KMS12PE‐BCL6, and KMS12PE‐Empty is presented as the fold change relative to that of the median of the MM patients. BCL6 protein expression levels of KMS12PE‐BCL6 and ‐Empty cells were determined by immunoblot analysis (d) and flow cytometry (e). β‐Actin was used as a loading control. (f) Real‐time PCR results for , p53, and p21 in KMS12PE‐BCL6 cells are presented as the fold change relative to each expression level of KMS12PE‐Empty cells. ns, not significant.

As BCL6 is known to directly repress the transcription of DDR genes, such as ATR, p53, and p21, by binding to the respective promoter/enhancer regions and recruiting transcriptional co‐repressors in BCL6‐positive B‐cell lymphoma cell lines, we evaluated the expression levels of these genes in this cell line.13, 14, 15, 18 Unlike in DLBCL, none of these genes were repressed by BCL6 overexpression in KMS12PE (Fig. 1f). There were two possibilities: (i) BCL6 does not bind to these target genes; or (ii) BCL6 does bind to those regions, but cannot work as a transcriptional repressor. To answer this question, we investigated the binding of BCL6 to these genes.

B‐cell lymphoma 6 binds to DDR genes in KMS12PE‐BCL6 cells

To examine whether BCL6 binds to the reported promoter or enhancer regions in KMS12PE‐BCL6 cells, ChIP assay was undertaken for the promoters of ATR, p53, and the enhancer of p21, which encompasses the BCL6 binding sites in GC B cells.13, 14, 18 Binding of BCL6 to each of these regions was clearly observed in KMS12PE‐BCL6 cells (Fig. 2a). One of the major repressor functions of BCL6 is mediated by its dimerization and formation of the BTB domain interface, which serves as a docking site for co‐repressors.20, 21 To further explore the functional effect of BCL6 expression against these DDR genes in KMS12PE‐BCL6 cells, the cells were exposed to BCL6 inhibitor, 79‐6, a small inhibitor molecule of BCL6, which was designed to bind to the groove of BCL6 BTB domain and disrupt the ability of BCL6 to recruit co‐repressors SMRT and NCOR, and reactivate the expression of BCL6 target genes.22 As a positive control, we used CTB‐1, a BCL6‐positive DLBCL cell line. Exposure to 79‐6 did not upregulate the expression levels of ATR, p53, or p21 in KMS12PE‐BCL6, whereas the inhibitor upregulated these genes in the positive control cell line, CTB‐1 (Fig. 2b). These results suggest that BCL6 binds to the reported DDR genes in MM cells, but unlike in B cells, it does not repress their transcription, which suggests the cell type‐specific function of BCL6.

Figure 2

B‐cell lymphoma 6 (BCL6) binds to DNA damage response genes in KMS12PE‐BCL6 multiple myeloma cells. (a) KMS12PE‐BCL6 and ‐Empty cells were used for ChIP assay with antibodies against BCL6 and control IgG. Regions of the promoters of and p53, and enhancer of p21 encompassing the BCL6 binding sites were amplified by real‐time PCR. (b) CTB‐1 and KMS12PE‐BCL6 cells were exposed to 50 μM 79‐6 or DMSO for 8 h and subjected to RNA quantification. RNA expression levels of , p53, and p21 are presented as the fold change relative to KMS12PE‐BCL6 cells exposed to DMSO. **P < 0.01. ns, not significant.

Formation of γH2AX after DNA damage was repressed by BCL6 overexpression in KMS12PE cells

To further analyze the role of BCL6 under DNA damaged conditions, we exposed the cells to IR or etoposide, both of which induce DNA double‐strand breaks. Phosphorylation of Ser139 in γH2AX by ATM is a primary event after DNA double‐strand breaks.23 Formation of γH2AX is required for the assembly of DNA repair proteins at the sites of damage as well as for the activation of checkpoint proteins, which lead to cell cycle arrest. To examine whether BCL6 plays a role in DNA damage response, formation of γH2AX was analyzed in KMS12PE‐BCL6 cells. Quantitative analysis of γH2AX foci formation by immunofluorescence staining indicated that the dose‐dependent increase in the mean density of the γH2AX signal following IR exposure was significantly suppressed by BCL6 expression (Fig. 3a,b). Analysis of γH2AX by flow cytometry also showed similar results, in which the induction of γH2AX was significantly lower in KMS12PE‐BCL6 cells than KMS12PE‐Empty cells in response to both IR and etoposide exposure (Fig. 3c,d, respectively). These results suggest that BCL6 attenuates the initial step of the DDR signaling pathway through repression of γH2AX formation in KMS12PE cells.

Figure 3

Histone H2AX (γH2AX) formation after DNA damage was repressed by B‐cell lymphoma 6 (BCL6) overexpression in KMS12PE multiple myeloma cells. (a) KMS12PE‐BCL6 and ‐Empty cells were irradiated by 3, 5, or 10 Gy and subjected to immunofluorescence staining for γH2AX. Original magnification, ×1000. Only the 3 Gy results are shown. (b) Mean densities of γH2AX expression per nucleus in 100 cells of each group were measured by densitometry. The graph shows the data as mean ± SD (n = 100 per bar). **P < 0.01. ns, not significant. γH2AX formation in KMS12PE‐BCL6 and ‐Empty cells after several doses of ionized radiation (c) or etoposide (d) was measured by flow cytometry. The results are shown as the mean fluorescence intensity of γH2AX. *P < 0.01.

Expression level of ATM repressed by BCL6 overexpression in KMS12PE cells

Ionized radiation‐induced γH2AX foci formation is known to depend on ATM, but not on ATR.23 We have already shown that overexpression of BCL6 failed to suppress ATR expression (Fig. 1e). To further examine the possibility that the repression of γH2AX formation in KMS12PE‐BCL6 is mediated by ATM, the total protein expression and phosphorylation form of ATM was assessed by immunoblotting before and after IR exposure. The reproducible results showed that overexpression of BCL6 caused a reduction in both the total ATM proteins and its phosphorylated forms after DNA damage induced by IR (approximately 39% and 28% reduction for total protein and phosphorylated forms, respectively). Phosphorylation of p53 was also reduced by BCL6 overexpression after IR exposure (approximately 30% reduction) (Fig. 4a,b), which is a reasonable result as ATM is responsible for Ser15 phosphorylation of p53.

Figure 4

Expression level of was repressed by B‐cell lymphoma 6 (BCL6) overexpression in KMS12PE multiple myeloma cells. (a, b) KMS12PE‐BCL6 and ‐Empty cells were irradiated by 10 Gy and analyzed by immunoblotting. (c) and p53 (d) RNA in KMS12PE‐BCL6 and ‐Empty cells with or without ionized radiation (IR) are shown as the fold change relative to control cells without IR. (e) Expression levels of of KMS12PE‐BCL6 cells after exposure to 50 μM 79‐6 (BCL6 inhibitor) are shown. (f, g) KMS12PE‐BCL6 cells were treated with cycloheximide (80 μg/mL) to block protein synthesis with or without 10 Gy IR. At 0, 0.5, 1, 2, and 4 h after IR, cell samples were collected and analyzed by immunoblotting. Relative band intensities were calculated by the protein/β‐actin ratios and normalized to the control cells without IR in each protein (b), and at 0 h after IR (g). **P < 0.01. ns, not significant.

Moreover, real‐time PCR showed that the BCL6 overexpression repressed ATM transcripts both before and after IR at same level (Fig. 4c), whereas the p53 gene was not affected (Fig. 4d). To investigate whether the repression of ATM by BCL6 was mediated through its BTB domain, the cells were exposed to 79‐6. Exposure to 79‐6 failed to affect the expression of ATM mRNA in KMS12PE‐BCL6 cells, suggesting that the BTB domain is not involved in this process in MM cells (Fig. 4e).

To further investigate the reason for the discrepancy between transcriptional repression and protein level change before and after IR, we speculated that the reduction in ATM transcription may not have affected the protein levels in a steady state because of slow degradation. We looked at the protein degradation time course of ATM with and without IR using cycloheximide. As shown in Figure 4(f,g), the ATM protein level was steady without IR but rapidly reduced after IR, suggesting faster degradation following DNA damage.

Overexpression of BCL6 led to upregulation of AID and downregulation of CD138 in KMS12PE cells

To investigate the effects of BCL6 overexpression on KMS12PE cells, mRNA sequence analysis was carried out. The total number of differentially expressed genes by KMS12PE with BCL6 overexpression was 2274, including 1265 upregulated genes and 1009 downregulated genes (Fig. 5a). Pathway analysis by iPathwayGuide software (Advaita, http://www.advaitabio.com/ipathwayguide) revealed significantly impacted pathways, such as those associated with cell adhesion and cytokine–cytokine receptor interaction (both P < 0.001). Neither the DNA repair pathway nor cell cycle checkpoints were extracted as a source of significantly different genes by BCL6 overexpression. Among the differentially expressed genes, we found an unexpected upregulation of the AID gene (Fig. 5a,b), which was verified by real‐time PCR (Fig. 5c). As the expression of AID and BCL6 is primarily restricted to GC B cells,24, 25 we tried to seek some other genes, which were known to be preferentially expressed in GC B cells or differentiated plasma cells from mRNA sequence data. As a result, we identified SPIB and CIITA as upregulated genes and XBP1 and SDC1 as downregulated genes (Fig. 5a,b). SPIB is required for normal GC reactions and prevents plasma cell differentiation,26, 27 and CIITA is required for the expression of MHC class II proteins and characteristics of B cells and is extinguished in terminally differentiated plasma cells.28 XBP1 is one of the essential transcriptional regulators for plasma cell differentiation,29, 30 and SDC1 encodes terminally differentiated plasma cell marker CD138. These findings were verified by real‐time PCR (Fig. 5d). Decreased CD138 expression in BCL6‐transfected cells was confirmed by flow cytometric analysis (Fig. 5e). In addition, real‐time PCR also showed BCL6 overexpression repressed expression of PRDM1, a master regulator of plasma cell differentiation (Fig. 5f), which is consistent with previous reports,30 although the mRNA sequence analysis did not show the downregulation significantly. PRDM1 has been shown to directly or indirectly control CIITA and XBP1.31, 32 Taken together, the gene expression profile of BCL6‐transfected KMS12PE cells suggested the de‐differentiation from plasma cells to B cells.

Figure 5

B‐cell lymphoma 6 (BCL6) overexpression led to upregulation of and downregulation of CD138 in KMS12PE multiple myeloma cells. (a) Global differential expression map of mRNA sequencing of KMS12PE‐BCL6 versus ‐Empty cells. The Y‐axis represents the logarithm of the fold change (log2 FC) and the X‐axis represents the average logarithm of counts per million of reads (log2 CPM). Black dots represent differentially expressed genes (n = 2274); gray dots, not significantly differentially expressed genes (n = 9889); red dots, genes we focused on. (b) Log2 FC of genes which relate to GC B cell or plasma cell type are shown. Real‐time PCR was used to measure RNA expression level of (c), ,,, and (d), and (f) in KMS12PE‐BCL6 cells, presented as the fold change relative to that of KMS12PE‐Empty. The data are shown as the mean ± SD (n = 3 per bar). **P < 0.01. ns, not significant. (e) We determined the expression level of CD138 in KMS12PE‐BCL6 and ‐Empty cells and uninfected KMS12PE cells, referred to as virus (−), using flow cytometry.

Interleukin‐6 exposure upregulated BCL6 and AID, repressed ATM, and attenuated DDR in KMS12PE cells

In the bone marrow microenvironment, IL‐6 is reported to induce BCL6 expression in tumor cells of MM patients.16 We also confirmed that BCL6 was upregulated by treatment with recombinant IL‐6 in KMS12PE and other cell lines except for RPMI8226 (Fig. 6a). Furthermore, IL‐6 stimulation repressed ATM transcription in KMS12PE and KMS12BM (Fig. 6b), but not in other cell lines (data not shown). We also showed an attenuated γH2AX formation induced by IR‐dependent DNA damage in KMS12PE cells after IL‐6 treatment (Fig. 6c). Moreover, IL‐6 stimulation downregulated SDC1 (CD138) (Fig. 6d) and upregulated AID (Fig. 6e) in this cell line. These phenotypic and functional properties in IL‐6‐treated KMS12PE cells resemble those of BCL6‐transfected KMS12PE cells.

Figure 6

Interleukin‐6 (IL‐6) exposure upregulated and , downregulated , and attenuated DNA damage response in KMS12PE multiple myeloma cells. Real‐time PCR was used to measure RNA expression levels of (a) in multiple myeloma cell lines KMS11, KMS12PE, KMS12BM, KMS18, KMS26, ARH77, and RPMI8226, (b), (d), and (e) in KMS12PE cells treated with 122 ng/mL recombinant human IL‐6 for 16 h. The expression levels are presented as the fold change relative to each expression level of the cells without exposure to IL‐6. (c) Histone H2AX (γH2AX) formation in KMS12PE cells with or without exposure to 100 ng/mL IL‐6 after exposure to 0, 5, or 10 Gy ionized radiation, and incubation for 1 h was assessed by flow cytometry. The results are shown as the mean fluorescence of γH2AX. *P < 0.01.

Discussion

The expression levels of BCL6 in some of the patient samples were significantly higher than a B cell tumor cell line, which supports our experimental system using an MM cell line stably expressing BCL6. Using this model, the following results were observed. First, overexpression of BCL6 induced downregulation of ATM, a major DDR signaling kinase, but not other DDR genes. ATM repression was associated with decreased γH2AX formation following DNA damage. Second, decrease in the ATM protein levels became obvious in KMS12PE‐BCL6 cells after irradiation due to the rapid degradation of ATM protein. Third, mRNA sequence analysis identified AID, primarily expressed by GC B cells, as one of the upregulated genes in KMS12PE‐BCL6 cells. Upregulation in other GC B cell‐related genes, SPIB and CIITA, and downregulation of plasma cell differentiation genes, such as XBP1, SDC1, and PRDM1, were observed. Finally, IL‐6 exposure to KMS12PE cells resulted in phenotypic and functional changes similar to those observed in KMS12PE‐BCL6 cells: upregulation of BCL6 and AID, downregulation of ATM and SDC1, and attenuation of DDR. These results suggest that BCL6 overexpression in these MM cells leads to de‐differentiation of plasma cells to GC B cells and has a significant impact on genomic instability.

In spite of the pivotal role of ATM in the genotoxic stress response, there are a few reports on its transcriptional regulation.33, 34, 35 Although the detailed mechanism of this repression is unknown, our study is the first to show that BCL6 leads to the transcriptional repression of ATM in myeloma cells. As the BCL6‐BTB domain inhibitor did not affect ATM expression in KMS12PE‐BCL6 cells, further studies are necessary to determine the mechanisms of BCL6‐dependent downregulation of ATM in MM cells.

In spite of the downregulation of ATM transcription in KMS12PE‐BCL6 cells, we found no difference in the ATM protein levels before DNA damage. We also showed the fast degradation of ATM protein in DNA damage response, which is in concordance with the previous report that ATM cleavage occurs after DNA damage.36

Additionally, we found that BCL6 overexpression in KMS12PE cells led to changes in the gene expression profile, implying de‐differentiation from plasma cells to GC B cells. Our finding is consistent with a recent study that showed that exogenous expression of BCL6 in a plasma cell line leads to repression of plasma cell‐specific transcripts and re‐expression of B‐cell surface markers.37 Furthermore, we found upregulation of AID in KMS12PE‐BCL6 cells. AID is a member of the APOBEC family, a well‐known DNA mutator, which targets not only immunoglobulin genes but also a number of others, such as oncogenes. AID is implicated in oncogenic mutations and chromosomal translocations, especially those involving immunoglobulin genes, which results in B‐cell and plasma cell tumorigenesis.38, 39, 40 It has been reported that the AID/APOBEC‐mediated mutational signature is associated with poor prognosis of myeloma patients.41 Taken together, BCL6‐dependent AID upregulation is a threat to the genomic integrity of myeloma cells and thus may affect disease progression.

We showed that IL‐6 treatment of KMS12PE cells resulted in phenotypic and functional changes consistent with those obtained by BCL6 overexpression. This report is the first to show that IL‐6, possibly through the upregulation of BCL6, can affect DDR or genomic instability in MM cells. There are several reports suggesting that some cytokines attenuate DNA repair by disruption of nuclear localization of DNA repair proteins or upregulation of AID.42, 43 In MM cells, there is a report that IL‐6 stimulation represses basal expression of excision repair genes.44 Koduru et al.5 showed that a tumor microenvironment can upregulate AID in MM cells isolated from a patient's bone marrow. Our results indicating that IL‐6 downregulated genomic guardian ATM and upregulated genomic mutator AID are in line with those of previous reports suggesting that genomic integrity could be threatened by extracellular factors. Thus, our results shed light on the mechanisms of the high frequency of somatic mutation in MM cells.45 Given whole genome sequencing studies that showed that mutations of DNA repair pathway genes occur infrequently in MM patients, mechanisms other than cell intrinsic factors may operate in the dysregulation of DDR.6, 7

There are several limitations to the present study. Our findings were obtained by examining a single cell line and the expression levels of ATM and AID in response to IL‐6 were different. What makes the difference between these cell lines are unknown. Further studies are necessary to solve these questions.

Disclosure Statement

The authors have no conflict of interest.

activation induced cytidine deaminase

apolipoprotein B mRNA‐editing enzyme, catalytic polypeptide‐like

ataxia telangiectasia mutated

ataxia telangiectasia and Rad3 related

B‐cell lymphoma 6

class II major histocompatibility complex transactivator

DNA damage response

diffuse large B‐cell lymphoma

germinal center

interleukin‐6

ionized radiation

multiple myeloma

phycoerythrin

PR/SET domain 1

syndecan 1

Spi‐B transcription factor

X‐box binding protein 1

phosphorylated H2AX