Establishment and characterization of a novel vincristine‐resistant diffuse large B‐cell lymphoma cell line containing the 8q24 homogeneously staining region

Chromosome band 8q24 is the most frequently amplified locus in various types of cancers. MYC has been identified as the primary oncogene at the 8q24 locus, whereas a long noncoding gene, PVT1, which lies adjacent to MYC, has recently emerged as another potential oncogenic regulator at this position. In this study, we established and characterized a novel cell line, AMU‐ML2, from a patient with diffuse large B‐cell lymphoma (DLBCL), displaying homogeneously staining regions at the 8q24 locus. Fluorescence in situ hybridization clearly detected an elevation in MYC copy numbers corresponding to the homogenously staining region. In addition, a comparative genomic hybridization analysis using high‐resolution arrays revealed that the 8q24 amplicon size was 1.4 Mb, containing the entire MYC and PVT1 regions. We also demonstrated a loss of heterozygosity for TP53 at 17p13 in conjunction with a TP53 frameshift mutation. Notably, AMU‐ML2 cells exhibited resistance to vincristine, and cell proliferation was markedly inhibited by MYC‐shRNA‐mediated knockdown. Furthermore, genes involved in cyclin D, mTOR, and Ras signaling were downregulated following MYC knockdown, suggesting that MYC expression was closely associated with tumor cell growth. In conclusion, AMU‐ML2 cells are uniquely characterized by homogenously staining regions at the 8q24 locus, thus providing useful insights into the pathogenesis of DLBCL with 8q24 abnormalities.

Gene amplification, observed in the form of doubleminute chromosomes or homogeneously staining regions (HSRs), is recurrent and plays an important role in cancer [1]. HSR is rarely seen in hematopoietic neoplasms compared with solid tumors and is observed at a lower frequency in lymphoid neoplasms than in myeloid neoplasms [2].
Chromosome 8q24 is the most frequently amplified locus in many cancers, with MYC being the most likely oncogene at this locus. The MYC gene encodes a transcription factor that regulates the expression of many target genes that control cell proliferation. The deregulation of MYC, resulting from t (8;14) and gene amplification, leads to the constitutive overexpression of MYC in numerous cancers and plays a pathogenetic role in oncogenesis [3,4].
Plasmacytoma variant translocation 1 (PVT1) is also located at 8q24, 57 kb downstream of MYC, and extends over 200 kb in the direction of the telomeres. PVT1 is a non-protein-coding gene and a homologue of mouse Pvt1 [5]. The Pvt1 locus is a site of recurrent translocation in mouse plasmacytomas and a common integration site for the murine leukemia virus, which is capable of inducing T-cell lymphomas in mice. In contrast to the typical Burkitt lymphoma (BL), in which the t(8;14) translocation contains a breakpoint within MYC, the t(2;8) or t (8;22) variant translocations in BL contain breakpoints in PVT1 [6].
PVT1 produces a variety of noncoding RNAs, including several microRNAs [7]. The precise functions of the PVT1 region and its noncoding RNAs remain unclear, although the long noncoding PVT1 RNA has a documented role in stabilization of the MYC protein [8]. Moreover, several groups have reported that the amplification and subsequent overexpression of PVT1 have an oncogenic function in ovarian and breast cancers [9]. A genomewide screen using array comparative genomic hybridization (array-CGH) and gene expression profiling identified PVT1 as a candidate oncogene in breast and ovarian cancers, acute myeloid leukemia, and Hodgkin lymphoma [10,11]. Furthermore, we have previously reported two novel chimeric genes, PVT1-NBEA and PVT1-WWOX, in multiple myeloma cell lines [12].
In addition, a circular RNA obtained from exon 3 of PVT1 (circPVT1) has been shown to function as a promoter of cell proliferation in fibroblasts [13], an important prognostic factor in gastric cancer [14], and a diagnostic marker in osteosarcoma [15], and to have an oncogenic role in head and neck carcinoma and myeloid leukemia [16,17]. Hu et al. [18] have also described circPVT1 overexpression in B-cell acute lymphoblastic leukemia and performed functional studies indicating its role in B-cell proliferation. Taken together, these reports highlight the potential importance of the MYC/PVT1 locus in the pathophysiology of many cancers.
Diffuse large B-cell lymphoma (DLBCL) is the most common type of non-Hodgkin lymphoma and is known as a biologically heterogeneous tumor. Although approximately 70% of patients with DLBCL survive longer than five years when treated with immunochemotherapy involving rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone (R-CHOP), the remainder succumb to the disease [19]. A high international prognostic index, extranodal lesions, a non-germinal center B-cell phenotype, BCL2 expression, the deletion of CDKN2A, and MYC rearrangements have been recognized as high-risk features in DLBCL treated using R-CHOP [20]. The prognostic and biological significance of MYC rearrangements in DLBCL has been thoroughly investigated. However, the role of PVT1, which may be co-amplified with MYC, remains unclear.
We herein report the AMU-ML2 novel DLBCL cell line, obtained from a primary refractory patient. This cell line is uniquely characterized by a HSR at the 8q24 locus, containing MYC and PVT1 and displaying a high expression of PVT1 long noncoding RNAs. Here, we report the genetic and biological characteristics of this cell line in comparison with other B-cell lymphoma cell lines.

Case history
A 64-year-old man was referred to the clinic presenting pancytopenia, a 2-month history of anorexia, and general fatigue in November 2011. The laboratory and chest X-ray examinations revealed a white blood cell count of 1000 lL À1 , hemoglobin of 9.6 gÁdL À1 , a platelet count of 5000 lL À1 , LDH of 2397 UÁL À1 , and bilateral pleural effusion (Fig. S1A,B). Abnormal lymphocytes with Burkitt-like morphology were observed in the pleural effusion and bone marrow (Fig. S1C,D). The patient was diagnosed with DLBCL and commenced treatment with R-CHOP. Gbanding of cells revealed a complex karyotype; however, fluorescence in situ hybridization (FISH) using a MYC/ IGH probe set revealed a significant increase in MYC copy number, in the absence of a fusion signal. Subsequently, the patient underwent a more intensive regimen involving rituximab plus hyperfractionated cyclophosphamide, doxorubicin, vincristine, and dexamethasone, alternating with high-dose methotrexate and cytarabine (R-Hyper-CVAD/ MA) [21] after one cycle of R-CHOP. The patient responded to the treatment; however, he developed cytomegalovirus pneumonia after four cycles of R-Hyper-CVAD/MA and died of Trichosporon asahii sepsis at 6 months postdiagnosis.

Establishment of the AMU-ML2 cell line
The patient provided written informed consent for his cells from the pleural effusion to be used in a procedure approved by the Institutional Review Board of Aichi Medical University. The study methodologies conformed to the standards set by the Declaration of Helsinki. The cells were collected at the time of initial diagnosis, prior to chemotherapy. The cells were cultured in RPMI 1640 medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% heatinactivated fetal bovine serum (Thermo Electron, Melbourne, Vic., Australia) and 1% penicillin/streptomycin (GIBCO-BRL, Grand Island, NY, USA). Cultures were maintained at 37°C in 5% CO 2 , and the medium was partially exchanged every 5-7 days. After 2 months in culture, cell proliferation became continuous. The cell line was designated as AMU-ML2 after confirmation that cells had started growing again after the conventional freeze-thaw procedure.

Chromosomal analyses and spectral karyotyping
Chromosome preparations for G-banding, spectral karyotyping, and FISH were performed according to standard procedures (SRL Inc., Tokyo, Japan) [23,24]. A total of 20 metaphase spreads were analyzed by G-banding, and the karyotype was defined according to the International System for Human Cytogenetic Nomenclature [25]. Spectral karyotyping analyses were performed in metaphase spreads according to standard procedures (SRL Inc.) [26].

FISH analysis
Detection of the 8q24 aberration and deletion of 17p in both metaphase and interphase nuclei was performed using the Vysis LSI IGH/MYC/CEP8 Tri-Color Dual Fusion Probes (Fig. S1A) and the Vysis LSI TP53/CEP17 Dual Fusion Probes (Abbot Molecular, Des Plaines, IL, USA). Bacterial artificial chromosome (BAC) and P1-derived artificial chromosome (PAC) clones RP1-160A22, RP1-193B12, RP1-109F14, and RP11-55J15 were obtained from Invitrogen (Carlsbad, CA, USA). FISH probes for the analysis of chromosome 6p and 8q breakpoints were prepared with DNA extracted from BAC clones and Poseidon Sub-Telomeric Probes for chromosome 8 (KREATECH, Amsterdam, the Netherlands) (Fig. S1A,B). DNA extraction was performed using the NucleoBond Xtra Midi kit (Macherey-Nagel, D€ uren, Germany). Labeling and hybridization of DNA were performed as previously described [27]. BAC/ PAC information was obtained from the National Center for Biotechnology Information website (https://www.ncbi.nlm. nih.gov/genome/gdv/), and the probes were confirmed to map to the precise chromosomal bands using metaphase spreads from peripheral blood lymphocytes of healthy donors.

Genome copy number analyses
Array-CGH was performed using the SurePrint G3 Human CGH 2 9 400K, Oligo Microarray Kit (G4448A; Agilent Technologies, Santa Clara, CA, USA), which contains approximately 400 000 60-mer oligonucleotides covering the entire human genome. Genomic DNA was extracted from AMU-ML2 cells using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). DNA labeling, hybridization, and washing were performed according to the manufacturer's protocols. Scanning analyses were performed using the Agilent Microarray Scanner (Agilent Technologies). The array data were analyzed to determine chromosomal copy numbers using the FEATURE EXTRACTIONS version 11.0 software program (Agilent Technologies) and the analytical software program DNA ANALYTICS, version 4.0 (Agilent Technologies). The ADM-1 algorithm (Threshold 6.0) was adopted to detect genomic aberrations [28]. The relationship between the log 2 ratio (X), the copy number in the reference sample (A), the average copy number in tumor cells (B), and the ratio of tumor cells (C) was determined as X = Log 2 .

DNA sequence analyses
Total RNA was isolated from AMU-ML2 cells using the NucleoSpin RNA kit (TaKaRa Bio, Inc., Tokyo, Japan). Complementary DNA was synthesized from 2 lg of total RNA by using the High-Capacity cDNA Reverse Transcription Kit (Invitrogen). PCR amplification of the open reading frame of TP53 was performed with a gene-specific primer set as follows: forward primer, 5 0 -ATGGAGGAGCCGCAGT CAGA; reverse primer, 5 0 -TCAGTCTGAGTCAGGCCCTT. Sequence analysis was performed using the BigDye Terminator v3.1 Cycle Sequencing Kit and an Applied Biosystems 3130 Genetic Analyzer (Foster City, CA, USA).

Western blot analyses
Western blot analyses were performed as described previously [29]. Briefly, proteins in the cell lysate were separated on 10% polyacrylamide gels, followed by transfer onto a polyvinylidene fluoride membrane (Merck Millipore, Darmstadt, Germany). The membrane was hybridized with a mouse monoclonal anti-c-Myc antibody (9E10; Wako) and a rabbit anti-b-actin antibody (13E5; Cell Signaling Technologies, Danvers, MA, USA). The membrane was visualized using ImmunoStar LD (Wako).

Real-time qRT-PCR
The expression levels of the MYC and PVT1 transcripts were quantified by real-time qRT-PCR using TaqMan Gene Expression Assays (MYC, Hs00153408_m1; PVT1, Hs00413039_m1; b-actin, Hs99999903_m1; Applied Biosystems) and the StepOnePlus TM Real-Time PCR System (Applied Biosystems). In brief, total RNA was extracted using the RNeasy Mini Kit (Qiagen) and cDNA was synthesized from total RNA by using the SuperScript III First-Strand Synthesis System (Invitrogen). The expression level of circPVT1 was quantified by real-time qRT-PCR using SYBR Green I (TaKaRa Bio, Inc.) with the StepOnePlus TM-Real-Time PCR System, as previously described [13,30,31]. cDNA was amplified using KOD FX Neo polymerase (Toyobo, Tokyo, Japan) with SYBR Green I. Primers for circPVT1 were follows: forward: 5 0 -GGTTCCACCAGCGT TATTC; reverse: 5 0 -CAACTTCCTTTGGGTCTCC [32].
GAPDH was used as an internal standard for the SYBR Green I-based analysis. Relative expression was determined using the 2 ÀDDCT method. The relative gene expression was determined using the gene expression in peripheral blood leukocytes (PBLs) from healthy donors as a control.

MYC knockdown
The pRetrosuper Myc shRNA and pMKO.1-puro GFP shRNA were gifts from Addgene (plasmids #15662 and #10675, respectively; Cambridge, MA, USA) [33,34]. To obtain cells stably expressing decreased levels of MYC, the MYC shRNA vector (AMU-ML2/MYCsh) or the control GFP shRNA vector (AMU-ML2/GFPsh) was introduced into AMU-ML2 cells. The retroviral plasmids were packaged into 293T cells using the pCL10A vector. Viral supernatants were harvested 96 h after transfection and filtered before infection. The cells were infected with retroviruses in the presence of 8 lgÁmL À1 Polybrene (Sigma-Aldrich). Antibiotic selection (puromycin; 0.3 lgÁmL À1 ; Wako) was begun 48 h after infection and continued for at least 3 days. Following infection and antibiotic selection, the cells were examined for MYC protein levels using western blotting.

Microarray gene expression analyses
The experimental procedure for the cDNA microarray analysis was based on the manufacturer's protocol (Agilent Technologies). In brief, cDNA synthesis and cRNA labeling with the cyanine 3 (Cy3) dye were performed using the Agilent Low Input Quick Amp Labeling Kit. The Cy3labeled cRNA was purified, fragmented, and hybridized on a Whole Human Genome 4 9 44k Oligo Microarray Chip containing 43 377 oligonucleotide probes, using a Gene Expression Hybridization kit. The microarray slide was washed and scanned using an Agilent DNA microarray scanner (Agilent Technologies). The scanned data were quantified using the FEATURE EXTRACTION software program, version 11.0.1.1 (Agilent Technologies). The signal intensities were then normalized as described elsewhere [35]. The background signals were also normalized, and the microarray expression data were rank-ordered based on the differential expression in AMU-ML2/MYCsh cells versus AMU-ML2/GFPsh cells as follows: Up-and downregulated genes were called when exhibiting a > 0.27 increase (fold change > 2.0) or < À0.27 decrease (fold change < 0.5), respectively, in AMU-ML2/MYCsh cells versus AMU-ML2/ GFPsh cells.

Karyotyping and FISH analysis of 8q24 abnormalities in AMU-ML2 cells
The karyotyping of AMU-ML2 cells by G-banding and spectral karyotyping revealed the following karyotype: 46,XY,del(6)(p21p23),der(8)(8pter?8q24::hsr::6p21? 6pter),add(9)(p13),del(17)(p?) [21] (Fig. 1A,B). The FISH analysis of AMU-ML2 cells using a MYC/IGH probe set revealed no fusion signal for IGH and MYC; however, a significant increase in the MYC copy number was observed, corresponding to the HSR on chromosome 8q24 (Fig. 1C). The copy number of the amplicon at 8q24.21 was approximately 26 per tumor cell. A FISH analysis using RP11-55J15, containing a segment of PVT1 but not the MYC gene (Fig. S2A), also showed a significant increase in the copy number at the PVT1 gene locus (Fig. 1D). The FISH analysis of AMU-ML2 cells using an RP1-160A22/RP11-55J15 probe set revealed that chromosome 6p21.2-22.2 translocated on chromosome 8q (white arrow), and a significant increase in the PVT1 copy number was observed (white triangle), corresponding to the HSR on chromosome 8q24 (Fig. 1D). RP11-55J15 contained a segment of PVT1 but not the MYC gene (Fig. S2A), and also showed a significant increase in the copy number at the PVT1 gene locus. The FISH analysis of AMU-ML2 cells using a RP1-160A22/RP1-193B12 probe set revealed fusion signal for both probes on chromosome 6p (white triangle); however, the other RP1-193B12 was absent on chromosome 6p, and the other RP1-160A22 translocated on chromosome 8q (white arrow) (Fig. 1E). The FISH analysis of AMU-ML2 cells using a RP1-160A22/Sub-Telomere 8qter probe set revealed a fusion signal for both probes on chromosome 8q telomere (Fig. 1F). The FISH analysis of AMU-ML2 cells using a TP53/ CEP17 set revealed that one of TP53 disappeared on chromosome 17p (Fig. 1G). The copy number of the amplicon at 8q24.21 was approximately 26 per tumor cell.
MYC, PVT1, and circPVT1 mRNA levels in AMU-ML2 and other B-cell lymphoma cell lines To investigate the influence of the 8q24.21 amplification on MYC, PVT1, and circPVT1 expression, we performed quantitative RT-PCR (qRT-PCR) analysis in AMU-ML2 cells and other B-cell lymphoma cell lines, for which the chromosomal status at 8q24 is summarized in Fig. 3A. qRT-PCR analysis showed that the expression of MYC, PVT1, and circPVT1 was significantly higher in AMU-ML2 cells than that in PBLs from healthy donors. In addition, the expression of PVT1 and circPVT1 in AMU-ML2 cells was the highest among the cell lines used in this study (Fig. 3B,C).

Resistance of AMU-ML2 cells to vincristine
To clarify the effect of anticancer drugs used in the chemotherapy of DLBCL on AMU-ML2 cells, we performed an MTT assay using AMU-ML2 and other B-cell lymphoma cell lines following treatment with vincristine, doxorubicin, prednisolone, and cyclophosphamide. The MTT assay showed that AMU-ML2 cells exhibited resistance to vincristine (at 100, 500, and 1000 nM), whereas cell survival in other B-cell lymphoma cell lines was almost completely suppressed in a dose-dependent manner (Fig. 3D). The MTT assay also showed that doxorubicin dose-dependently decreased the cell survival rate in all cell lines used in this study (Fig. 3E). Prednisolone partially suppressed cell proliferation in AMU-ML2 and B-cell lymphoma cell lines, whereas cyclophosphamide did not exhibit any tumor-suppressive effects (Fig. 3F,G).

The role of MYC in proliferation and gene expression in AMU-ML2 cells
As our data showed that MYC expression was higher in AMU-ML2 cells than that in PBLs from normal healthy volunteers, we next investigated the effects of MYC expression on cell proliferation using RNA interference. Our western blot analysis showed robust MYC expression in AMU-ML2 cells expressing the control GFPsh vector; the expression of this protein decreased in cells expressing MYCsh (Fig. 4A). Therefore, we examined the effect of MYC knockdown on cell proliferation using an MTT assay. The MTT assay showed that the optical densities reflecting the cell numbers were significantly higher at both days 1 and 3 in cells expressing GFPsh than in those expressing MYCsh, strongly suggesting that MYC expression is closely associated with cell proliferation in AMU-ML2 cells (Fig. 4B).
To further investigate the role of MYC in tumorigenesis in AMU-ML2 cells, we performed a comprehensive gene expression analysis using Agilent cDNA microarrays. A heatmap analysis revealed different gene expression patterns between cells expressing GFPsh and those expressing MYCsh (Fig. 4C). We also observed that MYC knockdown downregulated the expression of 172 genes by < 0.5-fold and upregulated the expression of 214 genes by > 2.0-fold compared with that in cells expressing GFPsh (Tables S1  and S2). Moreover, we found that MYC knockdown significantly decreased the expression of cyclin D1 (CCND1), BCL2, and STAT1 but not that of GAPDH in AMU-ML2 cells (Fig. 4F). By using a PANTHER classification analysis, we showed that 42.3% of the downregulated genes encoded binding proteins, including CCND2, BCL2, IRF4, SLAMF1, and SLAMF7 ( Fig. 4D,E). Notably, this analysis also showed that 8.0% of the downregulated genes encoded either inflammation-or apoptosis-related signaling molecules (Fig. S3). Therefore, we performed a gene set enrichment analysis (GSEA) to investigate whether the expression of a specific set of oncogenesis-associated genes significantly differed between the MYC knockdown and control cells. Genes with oncogenic signatures showed a significant inactivation of cyclin D signaling-related genes (NKG7, IFITM1, and PDE2), of genes induced by mTOR signaling (ITPR1, ATF3, and BATF), and of Ras signaling-associated genes (NR4A3, DOCK4, and SATB1) (Fig. 4E). Furthermore, a GSEA using the Kyoto Encyclopedia of Genes and Genomes database showed a significant inactivation of genes associated with peroxisome proliferator-activated receptor-, hematopoietic-, and cytokine-cytokine receptor-related signaling (Fig. S4). Collectively, these results suggest that MYC expression is closely associated with tumor cell growth in AMU-ML2 cells.

Discussion
A HSR is occasionally observed in solid tumors, but rarely detected in DLBCL [36]. In this study, we established a novel DLBCL cell line, AMU-ML2, using patient cells, which was uniquely characterized by a    HSR at the 8q24 locus, containing MYC and PVT1 (Fig. 1). Our aCGH analysis clearly identified the 8q24 amplicon size as ranging from 128 611 to 130 073 kb in the HSR (Fig. 2, Table 1). Moreover, we found that the amplicon contained the entire MYC and PVT1 sequences and was amplified at more than 20 copies per cell. To our knowledge, this is the first to report of a DLBCL cell line showing an amplicon containing the entire MYC and PVT1 genes at 8q24. As the patientderived cells were collected before the initiation of chemotherapy and established within 2 weeks, the detected chromosomal aberrations reflect the actual pathogenesis for the onset of aggressive DLBCL and do not reflect chemotherapy and/or long-term cell culture. The expression levels of both MYC and PVT1 were significantly higher in AMU-ML2 and other B-lymphoma cells with 8q24 abnormalities than those in PBLs from healthy donors and Farage cells lacking 8q24 abnormalities (Fig. 3B). The expression of MYC and PVT1 appears similar to that observed in gene amplification and translocation events at immunoglobulin loci.
MYC, a candidate oncogene at the 8q24 amplification, has been reported to play an important role in the pathogenesis of lymphoma and leukemia [37]. In this study, we show that the MYC-shRNA-mediated knockdown significantly suppressed cell proliferation in AMU-ML2 cells (Fig. 4A,B). Furthermore, a cDNA microarray analysis revealed that the CCND2 cell-cycle-promoting gene, the IRF4 oncogenic transcription factor, and the BCL2 anti-apoptotic gene were all downregulated following MYC knockdown (Fig. 4C). Although MYC is assumed to repress the transcription of BCL2 directly or through p53, BCL2 expression was repressed by MYC inhibition in AMU-ML2 cells. This suggests the dysfunction of p53 in the AMU-ML2 background. Our GSEA also showed that MYC knockdown inactivated gene expression for oncogenic gene sets, including the Ras, mTOR, and cyclin D signaling pathways (Fig. 4E). These results strongly suggest that the survival and proliferation of AMU-ML2 cells strongly depend on the aberrant MYC expression. Recent reports have suggested that the deregulation of PVT1 consequent to gene amplification and chromosomal translocation may contribute to tumorigenesis and drug resistance [9,38]. Patients with DLBCL often suffer from resistance to chemotherapy, including R-CHOP. Resistance to cisplatin by PVT1 overexpression has been reported in gastric and ovarian cancers [39]. In addition, it has been reported that PVT1 promotes the development of multidrug resistance by mediating the mTOR/HIF-1a/P-glycoprotein pathway and/or the MRP1 signaling pathway [40]. Moreover, overexpression of circPVT1 has been recently shown to serve as a prognostic factor in gastric cancer [14]. In the present study, we found that AMU-ML2 cells displayed vincristine resistance, in contrast to other B-cell lymphoma cell lines (Fig. 3D). As the expression of PVT1 and circPVT1 in AMU-ML2 cells was the highest among the cell lines tested (Fig. 3B,C), the overexpression of PVT1 and circPVT1 may contribute to vincristine resistance in AMU-ML2 cells. Although the pathogenic role of PVT1 in lymphoma is not precisely known, the examination of whether the microRNAs and/or circPVT1 transcribed from the PVT1 locus are linked to tumorigenesis and drug sensitivity in AMU-ML2 cells is worthy of further study.
The co-amplification of MYC and PVT1 has been reportedly observed in several types of solid tumors and is associated with a shortened survival [9]. However, the biological differences underlying the deregulation of MYC alone, or that of both MYC and PVT1, are not well characterized in DLBCL. Future clinical studies utilizing RNA-seq, whole-genome sequencing, and functional experiments will help clarify the precise incidence, clinical implications, and biological significance of the co-amplification of MYC and PVT1 in DLBCL.
In addition to the HSR at 8q24, we have identified several other CNA loci (Table 1, Fig. 2A). TP53 is a well-known tumor suppressor gene at the 17p13.1 locus.
Loss of heterozygosity and inactivating mutations in the TP53 gene are frequently observed in many cancers and constitute a poor prognosis for DLBCL. In the present study, our aCGH analysis showed a monoallelic 7.8-Mb deletion, which contains the TP53 gene (Fig. 2D). Moreover, the identification of a TP53 frameshift mutation in AMU-ML2 cells suggested that p53 does not function in these cells. It has been reported that the disruption of the p14/ARF-MDM2-p53 pathway that accompanies MYC overexpression contributes to the pathogenesis of BL [41]. It may therefore be possible that the pathophysiology of DLBCL is partly associated with the HSR at 8q24 as well as with the dysregulation of p53 in AMU-ML2 cells. We also detected a chromosomal breakpoint and a single copy deletion in the HIST1 gene in AMU-ML2 cells ( Table 1, Fig. 2B). The HIST1 gene spans over 2 Mb and contains all the replication-dependent H1 histone genes and other core histone genes at the 6p22-p21 locus. Moreover, it encodes histone proteins, which associate with the double-stranded helical DNA molecule to form the chromatin, and play a role in gene regulation [42]. In AMU-ML2 cells, the deletion of a part of HIST1, after the t(6;8) translocation, may result in the haploinsufficiency of HIST1, which may influence chromatin remodeling and cellular gene expression.
In conclusion, the present study is the first to show a HSR containing both MYC and PVT1 at the 8q24 locus, in the AMU-ML2 novel DLBCL cell line. We also demonstrated a loss of heterozygosity for TP53 at 17p13 with a frameshift mutation of TP53, suggesting that the high expression of MYC and TP53 dysfunction may contribute to cell survival in DLBCL. The AMU-ML2 cell line is useful for investigating the roles of MYC and PVT1 and the interaction of the products of both genes in lymphomagenesis. Furthermore, it would be of interest to investigate the molecular mechanism through which AMU-ML2 cells induce vincristine resistance. Our finding that MYC expression is closely related to the expression of oncogenic genes, including those in the Ras, mTOR, and CCND signaling pathways, provides new insights that may aid in the development of novel molecular-targeted drugs for the treatment of patients with DLBCL. Further studies are required to clarify the role of PVT1 in the pathophysiology of DLBCL.

Supporting information
Additional supporting information may be found online in the Supporting Information section at the end of the article. The amplicon detected by aCGH analysis (green), the gene structure of MYC and PVT1, the PVT1-encoded microRNAs, BAC clone (RP11-55J15, green bar), and Vysis FISH probe (LSI/MYC, shown in red) are depicted. The FISH probe for MYC (red bar) covers an 821-kb region containing the entire MYC and PVT1 genes. The RP11-55J15 BAC clone partially covers the PVT1 region, but not the MYC region. The size of the 8q24 amplicon (green bar) detected by aCGH approximately spans 1462 kb, containing the entire MYC and PVT1 genes. PVT1 encodes at least six microRNAs (miR-1204, miR-1205, miR-1206, miR-1207-5p, miR-1207-3p, and miR-1208; blue bar).
The black horizontal bars indicate exons in each gene. (B) Genomic features at 6p22-p21. The deletion detected by aCGH (purple), gene structure including a HIST1 gene cluster, and PAC clones (RP1-97D16, black bar; RP1-160A22, red bar; RP1-193B12, green bar; RP3-408B20 and RP1-109F14, black bar) used for FISH analysis are depicted. The positional data for genes, microRNAs, and PAC/BAC clones were obtained from the NCBI website (https://www.ncbi. nlm.nih.gov/) and the DNA ANALYTICS software (Agilent Technologies). The positions (Mb) indicate the distance from the telomeric end on the short arm of each chromosome. Mb, mega base. Fig. S3. Results of Panther Classification Analysis. Gene ontology analyses using the Panther Classification System. The downregulated genes in cells expressing MYCsh were classified using PANTHER-Gene List Analysis (www.pantherdb.org). The percentages of genes classified into each pathway are shown as a pie chart. Fig. S4. GSEA with Kyoto Encyclopedia of Genes and Genomes (KEGG) gene sets. GSEA was conducted using GSEA v2.2.4 software and the Molecular Signatures Database (Broad Institute). All of the raw data were formatted and applied to the KEGG gene sets (C2).