ARK5 is transcriptionally regulated by the Large-MAF family ... - Nature

Oncogene (2005) 24, 6936–6944

& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00 www.nature.com/onc

ARK5 is transcriptionally regulated by the Large-MAF family and mediates IGF-1-induced cell invasion in multiple myeloma: ARK5 as a new molecular determinant of malignant multiple myeloma Atsushi Suzuki1,5, Shinsuke Iida2, Miyuki Kato-Uranishi2, Emi Tajima2, Fenghuang Zhan3, Ichiro Hanamura3, Yongsheng Huang3, Tsutomu Ogura1, Satoru Takahashi4, Ryuzo Ueda2, Bart Barlogie3, John Shaughnessy Jr3 and Hiroyasu Esumi*,1 1

Cancer Physiology Project, National Cancer Center Research Institute East, 6-5-1 Kashiwanoha, Kashiwa, Chiba 277-8577, Japan; 2Department of Internal Medicine and Molecular Science, Nagoya City University Graduate School of Medical Sciences, 1 Kawasumi, Mizuho-chou, Nagoya 467-8601, Japan; 3Donna D and Donald M Lambert Laboratory of Myeloma Genetics at the Myeloma Institute for Research and Therapy, University of Arkansas for Medical Sciences, Little Rock 72205, AR, USA; 4 Institute of Basic Medical Sciences, University of Tsukuba, 1-1-1, Tennodai, Tsukuba 305-8575, Japan

ARK5, AMP-activated protein kinase (AMPK)-related protein kinase mediating Akt signals, is closely involved in tumor progression, and its stage-associated expression was observed in colorectal cancer. In this study, we found ARK5 expression in multiple myeloma cell lines expressing c-MAF and MAFB. In addition, gene expression profiling of 351 clinical specimens revealed ARK5 expression in primary myelomas expressing c-MAF and MAFB, suggesting that ARK5 may be a transcriptional target of the Large-MAF family. Sequence analysis of the ARK5 gene promoter revealed that it contains two putative MAF-recognition element (MARE) sequences. In support of this hypothesis, ARK5 was induced when an MAFB or c-MAF expression vector was introduced into non-ARK5-expressing colon cancer cells. Furthermore, ARK5 promoter activity was dramatically decreased by mutation or deletion of MARE sequences. Chromatin immunoprecipitation assays revealed an interaction between the Large-MAF family proteins and MARE sequences in the ARK5 promoter. Moreover, in ARK5 mRNA-expressing multiple myeloma lines, but not in ARK5-negative lines, insulin-like growth factor (IGF)-1 increased invasion activity. IGF-1-induced invasion was reproduced when ARK5 was overexpressed in Burkitt’s lymphoma and plasmacytoma lines. Based on results, we conclude that ARK5 is a transcriptional target of the Large-MAF family through MARE sequence and that ARK5 may in part mediate the aggressive phenotype associated with c-MAF- and MAFB-expressing myelomas. Oncogene (2005) 24, 6936–6944. doi:10.1038/sj.onc.1208844; published online 25 July 2005 Keywords: ARK5; multiple family; invasion; IGF-1

myeloma;

Large-MAF

*Correspondence: H Esumi; E-mail: [email protected] 5 Current address: National Institute for Physiological Sciences, Aichi 444-8585, Japan Received 5 October 2004; revised 5 May 2005; accepted 7 May 2005; published online 25 July 2005

Introduction Multiple myeloma is an incurable neoplasm caused by the malignant transformation of plasma cells, usually in the bone marrow (Hideshima and Anderson, 2002), and its growth and survival are regulated by growth factors and cytokines, including insulin-like growth factor (IGF)-1 and interleukin (IL)-6 (Di Raimondo et al., 2000; Ferlin et al., 2000; Ishikawa et al., 2003; Klein et al., 2003). The same as in solid tumors, cell growth and survival in multiple myeloma are supported by the nutrient and oxygen through blood vessels, and thus antiangiogenesis therapy, for example, with thalidomide and vascular endothelial growth factor receptor inhibitors, has been proposed as a method of treatment (Drevs, 2003; Ria et al., 2003). In addition, it has recently been proposed that IGF-1 plays an important role in multiple myeloma malignancy by promoting invasion and migration (Asosingh, 2003; Tai et al., 2003; Qiang et al., 2004). Approximately 80% of multiple myeloma cases have chromosomal translocations involving the immunoglobulin heavy chain (IgH) locus (Bergsagel and Kuehl, 2001) and most of these translocations result in the dysregulation of cyclin D1, cyclin D3, c-Maf, MafB, or FGFR3 and MMSET (Hanamura et al., 2001; Seidl et al., 2003), and are likely primary oncogenic events in myeloma transformation. Through global gene expression profiling studies of a large population of newly diagnosed myeloma, we have shown that dysregulation of one of the three D-type cyclins is seen in virtually all myelomas and thus alteration of early G1>S phase of the cell cycle appears to be a unifying event in myeloma transformation (Bergsagel et al., 2005). The t(14;16) and t(14;20) translocations result in ectopic overexpression of the c-MAF and MAFB genes, respectively (Hanamura et al., 2001; Seidl et al., 2003), and t(14;16) seems to be highly correlated with plasma cell leukemia (Hayman and Fonseca, 2001). c-MAF and MAFB are members of the Large-MAF family of

ARK5 in multiple myeloma A Suzuki et al

6937

basic/leucine zipper transcription factors, and are protooncogene products that act as transcription activators through interaction with other transcription factors such as c-Myb and c-Fos (Kataoka et al., 1994a, b; Hedge et al., 1998). Recently, the Staudt group has shown that c-MAF increases myeloma proliferation and adhesion to bone marrow through activation of the CCDN2 and integrinb7 genes (Hurt et al., 2004). We have recently shown that overexpression of c-MAF and MAFB is associated with a poor prognosis (Zhan et al., 2004). Thus, overexpression of the Maf family genes, as a result of chromosome 14 translocations, is suspected of being important to the malignant potential of multiple myeloma. ARK5 is the fifth member of the AMP-activated protein kinase (AMPK) catalytic subunit family established as a cellular stress-response factor (Hardie and Carling, 1997; Kemp et al., 1999, 2003). ARK5 is unique in that its activation is directly regulated by Akt and because activated ARK5 promotes cell survival during glucose starvation and death receptor stimulation (Suzuki et al., 2003a, b, 2004a). We recently demonstrated that ARK5 mRNA expression in colon cancer is stage associated and that liver metastatic foci of colon cancer express very high levels of ARK5 mRNA (Kusakai et al., 2004a, b). Moreover, both in vitro and

in vivo experiments have revealed an accelerated rate of invasion and metastasis by ARK5-expressing tumor cells (Suzuki et al., 2004b). Taken together, these data suggest that ARK5 may be a novel tumor progressionassociated factor. In the current study, we investigated the relationship between ARK5 and the Large-MAF family, especially c-MAF and MAFB, in multiple myeloma, and demonstrate that ARK5 plays a major role in the promotion of cell invasion in multiple myeloma as a result of the transcriptional regulation by the Large-MAF family.

Results Expression of c-MAF, MAFB, and ARK5 mRNAs in 19 multiple myeloma cell lines Expression of c-MAF, MAFB, and ARK5 mRNA was examined in 19 multiple myeloma cell lines. As shown in Figure 1, MAF (a), MAFB (b), and ARK5 (c) were detectable in five of 19, three of 19, and 11 of 19 lines, respectively. As shown in Figure 1d, these results were also confirmed with RT–PCR analysis; ARK5 expression was detected in six (Sachi-1, KM5, SK-MM1, JJN3, NCU-MM1, and KMS11) of eight cell lines

Figure 1 Expression of c-MAF, MAFB, and ARK5 mRNAs in 19 multiple myeloma cell lines. RNAs were extracted from 19 multiple myeloma cell lines, and real-time PCR (a–c) was performed using primer pairs for c-MAF (a), MAFB (b), and ARK5 (c), and RT–PCR (d) was performed. The absolute amount of target gene expression was calculated from the standard curve. The mRNA of interest in each sample was quantitatively assessed by dividing its expression level by that of b-actin as an internal control. Finally, the mRNA level of each gene is shown as a copy-number ratio (b-actin: 1). The real-time PCR assay of each sample was conducted in triplicate, and the mean value was recorded as the mRNA level Oncogene


6938

expressing c-MAF or MAFB mRNA (Sachi-1, KM5, SK-MM1, JJN3, NCU-MM1, KM4, ILKM2, and KMS11), but other multiple myeloma cell lines and reactive plasma cells derived from Castleman’s disease (reactive PC in Figure 1d) did not show mRNA expression of either MAF, MAFB, or ARK5. These results strongly suggest a close association between expression of ARK5, and MAF and MAFB expression. ARK5 expression is highly correlated with c-MAF or MAFB in newly diagnosed multiple myeloma To confirm our observation in multiple myeloma cell lines, we examined whether ARK5 expression correlated with c-MAF and/or MAFB in primary myeloma. We have previously demonstrated that myeloma can be separated into seven distinct molecular subgroups based on global gene expression patterns (Zhan et al., 2004). Spiked expression of c-MAF and/or MAFB mRNA, indicative of an IGH translocation into these loci, defined one of the seven subgroups. Myelomas in this subgroup tended to have IgA isotype (P ¼ 0.009), significantly elevated b2-micorglobulin levels (mean 9.09 mg/l) (P ¼ 0.003), increased incidence of chromosome 13 deletion as detected by interphase FISH (Po0.001) and shorter event-free (Po0.0001) and overall (P ¼ 0.0012) survival. In a search for genes that might be regulated by the c-MAF/MAFB transcription factor, we performed a correlation analysis for genes that were uniquely upregulated in myelomas with c-MAF and MAFB activation. This search revealed that ARK5 expression was uniquely elevated in c-MAF/ MAFB translocation-positive myelomas (Figure 2). c-MAF and MAFB induce ARK5 gene transcription through MAF-recognition element (MARE) sequences Both c-MAF and MAFB being known transcription factors (Kataoka et al., 1994a, b; Hedge et al., 1998) and our finding of a strong correlation between ARK5 and

c-MAF/MAFB expression, suggested that ARK5 may be regulated by c-MAF and MAFB. We tested this hypothesis by examining the effect of forced overexpression of c-MAF and MAFB on ARK5 expression. Multiple myeloma cell lines are not useful for gene-transfection assays because of their inherently low transfectability (lipofection: 0.1–0.3%; adenovirus infection: 0.1–0.5%). Thus to test this hypothesis, we used two human colorectal cancer cell lines HCT-116 and DLD-1 that do not express c-MAF, MAFB, or ARK5. Endogenous ARK5 mRNA expression was clearly induced in both cell lines after transfection of Myc-tagged mouse MAFB or FLAG-tagged mouse c-MAF (Figure 3a). c-MAF and MAFB induce the gene transcription through the interaction with MARE on the promoter region of target gene, and two sequences (TGCTGACT CAGCA and TGCTGACGTCAGCA) were identified as the consensus sequence of MARE (Kataoka et al., 1994a, b). In addition, Motohashi et al. (1997) reported that TGCTGAC is working as the core sequence of MARE and that 50 -side thymidine is important for the binding between c-MAF/MAFB and MARE. In the current study, sequence analysis revealed that the ARK5 promoter contains two MARE-core sequences (Figure 3b). A deletion analysis was then performed to identify whether the MARE elements in the ARK5 gene promoter are responsible for c-MAF and MAFB responsiveness. Strong reporter gene activity was observed with ARK5 promoter-luciferase constructs containing full-length or D-I mutant transfected into the HCT-116 or DLD-1 cell lines overexpressing MycMAFB and FLAG-c-MAF (Figure 3c). An approximately 0.5-fold decrease was observed with the D-II mutant promoter, but only marginal activity was observed with the D-III, -IV, and -V mutants (Figure 3c). In addition, we also examined whether c-MAF and MAFB physically interacted with the MARE site on the promoter region of ARK5, using a chromatin immunoprecipitation (ChIP) assay. As shown in Figure 4a, PCR analysis showed that a genomic fragment including

Figure 2 Correlative expression of ARK5 mRNA with c-MAF or MAFB mRNA expression in clinical specimens of multiple myeloma. Box plots of log 2-transformed gene expression signal for c-MAF (a), MAFB (b), and ARK5 (c) in 351 newly diagnosed myelomas distributed into seven subgroups as described (Zhan et al., 2004). Samples and sample sizes are indicated on the x-axis. The top, bottom, and middle lines of each box correspond to the 75th percentile (top quartile), 25th percentile (bottom quartile), and 50th percentile (median) of the log base 2-transformed Affymetrix Signal for each gene. One-way ANOVA test for differences in expression for each gene across the myeloma molecular subgroups was applied to the expression of c-MAF, MAFB, and ARK5. Note the overexpression of c-MAF, MAFB, and ARK5 in group V, which is characterized by the presence of c-MAF or MAFB translocations Oncogene


6939

Figure 3 Large-MAF family promotes ARK5 gene transcription. (a) The HCT-116 (H) and DLD-1 (D) cell lines were transfected with expression vector containing (I) or not containing (E) Myc-MAFB (MafB) or FLAG-c-MAF. Expression of Myc-MAFB, FLAGc-MAF, and actin was detected by Western blotting with antibodies against Myc-tag (MafB), and FLAG-tag (c-Maf), and actin (Actin). ARK5 mRNA expression was examined by RT–PCR at 48 h after transfection. (b) The ARK5 promoter is represented schematically (Full: full-length; D-I–D-V: deletion mutants). Each construct was prepared by using pGL2 plasmid containing luciferase (luc). Open boxes represent the position of MARE-core sequence. Consensus and core sequence of MARE, and putative MARE sequence detected in the promoter region of the ARK5 gene (MARE-1: M1; MARE-2: M2) were represented. (c) HCT-116 cells (closed) and DLD-1 cells (open) were transfected with expression vector containing or not containing (Empty þ ) Myc-MAFB ( þ MafB) or FLAG-c-MAF ( þ c-Maf), ARK5 reporter gene assay constructs (Full and D-I–D-V), and b-galactosidase expression vector. Cell extracts were collected 48 h after transfection, and luciferase activity was measured. The results were normalized to bgalactosidase activity and are means of the triplicate measurements. The bars represent the s.e. values

each MARE site was contained in immunoprecipitates with antibodies against Myc- or FLAG-tag, when Myc-MAFB or FLAG-c-MAF expression plasmids were introduced into HCT-116 or DLD-1 cells. Furthermore, the interaction of endogenous c-MAF or MAFB to each MARE site was confirmed by ChIP assay using two multiple myeloma cell lines (Sachi-1: MAFB ( þ ); JJN3: c-MAF ( þ ) as shown in Figure 1d; Figure 4b). These results suggest that c-MAF and MAFB bind to the MARE sites in the promoter region of the ARK5 gene. To further elucidate whether ARK5 gene transcription-promoted MAF/MAFB is mediated by MAREs, the six constructs shown in Figure 5a were prepared. Deletion mutants that lack the core sequence of MARE and two-point mutants that lack Large-MAF family-binding activity (Kataoka et al., 1994a, b) were constructed in this study (Figure 5a). As seen in Figure 5b, reporter gene activity induced by overexpression of c-MAF and MAFB was significantly decreased when either MARE sequence was mutated, indicating that ARK5 gene transcription is regulated by c-MAF and MAFB through both MAREs in the ARK5 promoter.

ARK5 mediates tumor invasion activity induced by IGF-1 downstream of Akt Since we previously reported that ARK5 mediates tumor invasion in colon cancer cells (Kusakai et al., 2004a; Suzuki et al., 2004b), we used a Matrigel invasion chamber assay to determine whether ARK5 also induced multiple myeloma invasion. As shown in Figure 6a, expression of ARK5 mRNA was observed in four of 11 multiple myeloma cell lines, that is, JJN3, KM5, SK-MM1, and Sachi-1, but not in seven other cell lines. Using the Matrigel-coated invasion chamber (Kusakai et al., 2004b; Suzuki et al., 2004b), we were able to show that ARK5 promotes tumor invasion downstream of Akt (Suzuki et al., 2004b), and that Akt activation upon the stimulation with IGF-1 has been well known (Alessi et al., 1996; Datta et al., 1999). In addition, IGF-1 has been reported to be important to the malignant potential of multiple myeloma, especially with regard to invasion migration (Asosingh, 2003; Tai et al., 2003; Qiang et al., 2004). Therefore, we examined the effects of IGF-1 and/or Matrigel on Akt phosphorylation in multiple myeloma cell lines. As shown in Figure 6b, all Oncogene


6940

Figure 4 ChIP assay. (a) DLD-1 and HCT-116 cells were transfected with expression vector containing or not containing (Empty) Myc-MAFB ( þ MafB) or FLAG-c-MAF ( þ c-MAF), and then immunoprecipitation after crosslinking between chromatin and each overexpressed transcriptional factor, was performed using antibody against Myc (IP-Ab: aMyc) or FLAG (IP-Ab: aFLAG). Each of the immunoprecipitates was used for PCR (after purification of chromatin: ChIP) or Western blotting (IP/WB) to detect MARE-1 (M-1) and MARE-2 (M-2) sequence, and expressed Myc-MAFB (by anti-Myc antibody: MafB) and FLAG-c-MAF (by anti-FLAG antibody: c-Maf). (b) Sachi-1 and JJN3 cells were used for ChIP assay. After crosslinking treatment, each cell extract was reacted with or without (No Ab) antibody against MAFB (aMafB) or c-MAF (ac-Maf), and then each immunoprecipitate was used for PCR (after purification of chromatin: ChIP) or Western blotting (IP/WB) to detect MARE1 (M-1) and MARE-2 (M-2) sequence, and MAFB (by anti-MAFB antibody: MafB) and c-MAF (by anti-c-MAF antibody: c-Maf)

cell lines, with the exception of cell line ODA, showed only a slight increase in Akt activation on Matrigel, and Akt was dramatically phosphorylated by IGF-1. In addition, an accelerated phosphorylation of Akt was observed in KMS-12PE, XG7, SK-MM1, and Sachi-1 cell lines when grown in the presence IGF-1 and Matrigel (Figure 6b). As expected, ARK5 mRNAexpressing multiple myeloma cell lines showed high activity in the Matrigel-invasion assay (Figure 6c), whereas only marginal activity was detected in other cell lines (Figure 6c). ARK5 mRNA-expressing cell lines, but not the nonexpressing cell lines, showed increased invasion activity in the presence of IGF-1 (Figure 6c), leading us to hypothesize that ARK5 mediates tumor invasion activity in multiple myeloma cell lines. We showed IGF-1-induced invasion activity of ARK5 Oncogene

Figure 5 Promotion of ARK5 gene transcription by the LargeMAF family through MAREs. (a) Schematic representations of the reporter gene assay constructs are shown. Open box: wild-type core sequence of MARE (M1-WT and M2-WT); closed box: LargeMAF family-binding mutant (two-point mutants) (M1-MT and M2-MT); cut box: core-sequence-deletion mutant (M1-D and M2D). (b) HCT-116 cells (closed) and DLD-1 cells (open) were transfected with expression vector containing Myc-MAFB ( þ MafB) or FLAG-c-MAF ( þ c-Maf), ARK5 reporter gene assay constructs (Full, D-I, D-II, M1-WT, M1-MT, M1-D, M2WT, M2-MT, and M2-D), and b-galactosidase expression vector. Cell extracts were collected from the cells 48 h after transfection, and luciferase activity was measured. The results were normalized to b-galactosidase activity and are means of triplicate measurements. The bars represent the s.e. values

mRNA-expressing multiple myeloma cell lines in this study. Recently, Hurt et al. (2004) reported upregulated expression of integrinb7 mRNA in Large-Maf familyoverexpressing multiple myeloma. Akt activation induced by integrin through integrin-linked kinase has been well known (Hannigan et al., 1996; Delcommenne et al., 1998; Persad et al., 2000; Persad and Dedhar, 2003). Therefore, ARK5 may play an essential role downstream of integrin signaling in that case.


6941

green fluorescence protein was very low (o0.1%: data not shown). Therefore, we used human Burkitt’s lymphoma P3HR1 and mouse plasmacytoma NSO cell lines, instead of human multiple myeloma cell lines, in the current study. Since P3HR1 cell line is classified as B-cell line, the biological and genetic character is thought to be similar to plasma cells. In addition, the NSO cell line was derived from the mouse, not the human, but it is classified as a plasmacytoma line, so the biological and genetic character of the NSO cell line is very similar to the human multiple myeloma cell line. Transfections of ARK5, dominant-negative (DN)-Akt1, and constitutive active (CA)-Akt1 were successful in NSO cells (data not shown) and P3HR1 cells (Figure 6d). ARK5-transfected cells P3HR1 and NSO cell lines showed increased invasion activity and this ARK5-promoted invasion was markedly suppressed by the introduction of DN-Akt1 (Figure 6e). IGF-1 treatment further increased the invasion activity and DN-Akt1 suppressed the promotion of the invasion activity (Figure 6e). We previously reported that ARK5induced tumor invasion activity was dramatically increased by the introduction of CA-Akt1 into human pancreas cancer cell lines (Suzuki et al., 2004b); however, the effect of CA-Akt1 introduction was very weak in the current study (Figure 6e). Thus, IGF-1 increases ARK5-induced tumor invasion through Akt activation, suggesting that the Akt/ARK5 system mediates the tumor invasion induced by IGF-1 in multiple myeloma cell lines and in the P3HR1 and NSO cell lines. Figure 6 ARK5-promoted invasion activity. (a) Total RNAs were extracted from KM4, KM7, KMS-12PE, ODA, U266, XG7, NCUMM1, JJN3, KM5, SK-MM1, and Sachi-1 cells, and RT–PCR was performed to detect mRNA expression of ARK5 and GAPDH. (b) In all, 11 multiple myeloma cell lines were subjected to serumstarved medium for 16 h, and then cells were seeded onto normal tissue culture plate (1 and 3) or Matrigel-coated culture plate (2 and 4) and incubated for 2 h in the presence (3 and 4) or absence (1 and 2) of 100 ng/ml IGF-1. After incubation, cell extracts were collected and Western blotting was performed to detect phosphorylated (pAkt) or total (tAkt) Akt. (c) The invasion activity of 10 multiple myeloma cell lines in the presence (closed) or absence (open) of 100 ng/ml IGF-1 for 24 h was assessed. The numbers of invading cells are means of data from three experiments, and the bars represent the s.e. values. (d) ARK5, DN-Akt1, and/or CA-Akt1 expression vectors (Insert: þ ) were introduced into P3HR1 cells, and expression of each factor was detected by Western blotting with antibodies against ARK5 and Myc-tag (for DN-Akt1 and CA-Akt1). Empty vector (Insert: ) was also introduced into cells as a control. (e) The invasion activity of the P3HR1 and NSO cells transiently transfected with 1 mg ARK5, DN-Akt1, DN-ARK5, or their empty vector (Vector), in the presence ( þ ) or absence () of 100 ng/ml IGF-1 for 24 h was assessed. The numbers of invading cells are the means of data from three experiments, and the bars represent the s.e. values

To test this hypothesis, the effect of ARK5 overexpression on invasion activity was examined in P3HR1 and NSO cell lines. Although we examined several technologies, such as lipofection, viral infection, and cell fusion, all trials failed and the efficacy of plasmid introduction measured using expression plasmid with

Discussion Two members of the Large-MAF family, c-MAF and MAFB, are targets of recurrent gene activating translocations in approximately 5% of newly diagnosed myeloma (Hanamura et al., 2001; Siegel and Massague, 2003). In the current study, we found a close association between the expression of ARK5 and Large-MAF family mRNA expression in multiple myeloma cell lines and a large series of newly diagnosed myelomas with clinical follow-up showing that patients with LargeMAF family overexpression have significantly inferior survival when treated with high-dose therapy and autotransplants. Previous data have shown that overexpressed Large-MAF in myeloma cell lines induce an invasive phenotype. The relationship between ARK5 and invasion has been established (Kusakai et al., 2004a; Suzuki et al., 2004b). We show that ARK5 was a direct target of Large-MAF and that this activation was mediated through MARE sequences. The data presented suggest that activation of Large-MAF results in increased expression of a potential target gene ARK5, and since ARK5 plays an important role in IGF-1/Aktinduced invasion, we suggest that ARK5 inactivation might be a novel therapeutic approach in a myeloma with a molecularly defined lesion and a poor prognosis. Oncogene


6942 Materials and methods Cell lines and transfection Human multiple myeloma cell lines ODA, Sachi-1, U266, ILKM2, ILKM3, ILKM8, SK-MM1, XG7, FR4, NOP1, KM4, KM5, KM7, NCU-MM1, NCU-MM2, JJN3, AMO1, KMS11, and KMS-12-PE (Iida et al., 1997, 2000; Yoshida et al., 1999), human Burkitt’s lymphoma cell line P3HR1, and mouse plasmacytoma cell line NSO were maintained in RPMI1640 medium supplemented with 10% heat-inactivated fetal calf serum (FCS). Human colorectal cancer cell lines HCT-116 and DLD-1 were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% FCS (Sigma Chemical Co. Ltd, St Louis, MO, USA). For transfection, cells were seeded onto a fibronectin-coated plate (Becton, Dickinson Company, Franklin Lakes, NJ, USA), and transfection was performed with TransFast Transfection Reagent (Promega Corp., Madison, WI, USA) with some modifications. The cells were exposed to the plasmid/TransFast Reagent mixture for 4 h. Antibodies and plasmids Antibodies against Akt (total and phosphorylated at Ser473) were purchased from Cell Signaling Technology Inc, (Beverly, MA, USA). Expression vectors of CA-Akt1 and DN-Akt1, and anti-Myc-tag antibody were purchased from Upstate Inc. (Lake Placid, NY, USA). Antibody against FLAG-tag was purchased from Sigma. Antibodies against c-MAF and MAFB were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). Expression vectors of human ARK5 (Suzuki et al., 2003a), mouse c-MAF (Kajihara et al., 2003), and mouse MAFB (Huang et al., 2000) were used in the experiment in the same manner as reported previously. Patients Plasma cells were isolated by CD138 immunomagnetic bead selection from bone marrow aspirates of 351 newly diagnosed patients with multiple myeloma who went on to be treated with high-dose chemotherapy and tandem peripheral blood stem cell transplantation as part of an NCI-sponsored clinical trial Total Therapy 2. The Institutional Review Board of the University of Arkansas for Medical Sciences approved the research studies and all subjects provided written, informed consent. Gene expression profiling Gene expression profiling was performed as described previously (Zhan et al., 2002, 2003; Tian et al., 2003). Briefly, RNA from CD138-enriched plasma cells (greater than 90% pure as determined by CD38/CD45 FACs analysis) derived from pretreatment bone marrow aspirates for 351 patients with multiple myeloma. RNA was converted to biotinylated cRNA and hybridized to U133Plus2.0 oligonucleotide microarrays capable of investigating the expression of approximately 20 000 genes. The expression value ‘Signal’ was derived from the MAS5.01 software program and was transformed by the log base 2 and each sample was normalized to give a mean of 0 and a variance of 1.

within each subgroup were created in such a way that the top, bottom, and middle lines of each box corresponded to the 75th percentile (top quartile), 25th percentile (bottom quartile), and 50th percentile (median) of the log base 2-transformed Affymetrix Signal for each gene. The one-way ANOVA test for differences in expression for each gene across the myeloma molecular subgroups was applied to the expression of c-MAF, MAFB, and ARK5. RNA extraction and first-strand cDNA synthesis Total RNA was extracted from 19 multiple myeloma cell lines with Trizol reagent (Invitrogen Life Technologies, Groningen, The Netherlands) and treated with Amplification Grade DNaseI (Invitrogen). A 1 mg sample of total RNA was reverse transcribed to obtain single-strand cDNA with random primers and Superscript II (Invitrogen) and scaled up to a final volume of 42 ml. This 42 ml of cDNA was diluted in ddH2O at 1 : 128. Contamination by genomic DNA was identified with the LightCycler Quick System (Roche Molecular Biochemicals, Mannheim, Germany) and b-actin primers. This assay allowed us to exclude poor quality RNA that was contaminated by genomic DNA. Real-time PCR The primer pairs were designed so that they were located at different exons in order to prevent amplification from the contaminating genomic DNA. Only the primer set was selected from the same exon for the MAFB gene, because the MAFB gene has no intronic sequences. PCR was carried out in a total volume of 20 ml of reaction mixture containing 2 ml of diluted cDNA, 2 ml of FastStart DNA master SYBR Green I (Roche), and 0.5 mM of each primers with a LightCycler Quick System 330 (Roche). For precise quantitative determination of the transcripts, we also assessed the expression levels of b-actin and 18S ribosomal RNA as qualitative and quantitative internal controls. The primer set for b-actin was purchased from Roche Molecular Biochemicals, and the PCR conditions were according to the manufacturer’s instructions. The standard curve for each gene was generated by evaluating serially diluted plasmid clones containing cDNA inserts of the gene as templates. The absolute amount of target gene expression was calculated from the standard curve. The quantitative assessment of the mRNA of interest in each sample was performed by dividing its expression level by that of either b-actin mRNA or 18S ribosomal RNA as an internal control. Finally, the mRNA level of each gene was recorded as the copy-number ratio. The real-time PCR assay of each sample was conducted in triplicate, and the mean value was used as the mRNA level. The PCR primer pairs used in this study are listed in Table 1. RT–PCR Total RNA extraction was performed with Isogen purchased from Nippon Gene Co. Ltd (Osaka, Japan). The concentration of extracted RNA was measured at A260, and 0.5 mg total RNA was reverse transcribed with AMV transcriptase (Takara Bio Co. Ltd, Kyoto, Japan). After reverse transcription, PCR was performed with an LA PCR Kit (Takara) and primers for human ARK5 and GAPDH. PCR was performed for 25 cycles.

Statistical analysis Affymetrix Signal was log 2 transformed in 351 newly diagnosed myelomas that are distributed into seven subgroups, as described (Kataoka et al., 1994a, b). Box plots of expression Oncogene

Preparation of ARK5 promoter and its deletion mutants Based on the results of a Genomic BLAST Search, primers with the NheI (upstream primer) or XhoI (downstream primer)


6943 Table 1

List of primer pairs of c-MAF, MAFB, and ARK5 mRNAs, and 18S rRNA, for real-time PCR

Gene name

Oligo

Sequence

c-MAF

Forward Reverse

50 -TGGCATCAGAACTGGCAATGA-30 50 -CTGCACGGCGTGCTCATG-30

MAFB

Forward Reverse

50 -TTCAACCTTGTTGGTGCTG-30 50 -AATTTGACCATAAGACAAGGCTGT AGT-30

ARK5

Forward Reverse

50 -GAGTCCACTCTATGCATC-30 50 -ATGTCCTCAATAGTGGCC-30

18S rRNA

Forward Reverse

50 -TGGTTGATCCTGCCAGTAG-30 50 -TCGGCATGTATTAGCTCTAG-30

GAPDH

Forward Reverse

50 -GAGTCAACGGATTTGGTCGT-30 50 -GACAAGCTTCCCGTTCTCAG-30

Table 2 List of primers for the construction of ARK5 promoters carrying MARE sequence mutants Promoter name

Sequence

M1-WT

50 -gccccacccctatctACGCGTccctttGCTGACtctctttttggac tcagc-30 50 -gccccacccctatctACGCGTccctttGagGACtctctttttggact cagc-30 50 -gccccacccctatctACGCGTcccttt******tctctttttggactca gc-30

M1-MT M1-D M2-WT M2-MT M2-D

Ltd). After cell treatment or transfection, cells were lysed with PBS containing 1% NP-40. Cell extracts were collected by centrifugation, and the luciferase activity in each cell extract was measured with a Luminescenser (ATTO Corp., Tokyo, Japan). Normalization was performed by cotransfection with b-galactosidase expression vector, and b-galactosidase activity was measured with the b-galactosidase Activity Assay System (Promega). ChIP assay The ChIP assay was performed using the kit purchased from Upstate according to the manufacturer’s instructions. Cells were crosslinked with 1% formaldehyde and the genomic DNA was sheared by sonication. The samples were then immunoprecipitated with antibody against FLAG-tag, Myc-tag, c-Maf, or MafB. The immunocomplex was heated at 651C for 4 h to revert the crosslinking between DNA and proteins. The purified DNA was then dissolved in 20 ml of Tris buffer (10 mM Tris, pH 8.5) and analysed by PCR (35 cycles) with primers: MARE-1 – 50 -TGGCTCTTCCTGGCTCAT CC-30 (upper) and 50 -GATAATTGTTTTCCCCCTCT-30 ; and MARE-2 – 50 -AAACAGTACCAGCCTATGGT-30 and 50 -GGAACCTTTAGGAGCC AGTG-30 . Matrigel-invasion assay

50 -ataaataaagtttaACGCGTatatttGCTGACactacttttccaa tgcaga-30 50 -ataaataaagtttaACGCGTatatttGagGACactacttttccaat gcaga-30 50 -ataaataaagtttaACGCGTatattt******actacttttccaatgc aga-30

Underlined bases: MluI site used for cloning; bold bases: core sequence of MARE; bold and small-letter bases: mutation in core sequence of MARE; bold asterisks: deleted bases

site were synthesized, and PCR was then performed with the primers for genomic DNA extracted from PANC-1 cells. The PCR fragment digested with NheI and XhoI was ligated into pGL2-basic. Deletion mutants were prepared by Exo III/MUNG Bean Nuclease Deletion Kit (Stratagene Cloning Systems, La Jolla, CA, USA), and five deletion mutants (D-I– D-V) were sequenced. MARE sequence mutants were also constructed in this study. Plasmid construction was performed by the PCR method as a template full-length ARK5 promoter in pGL2basic vector. Using the wild-type, two-point mutant (GCTGAC-GagGAC), which lacks the ability to bind with c-Maf and MafB, and the deletion mutant containing the primer, each construct was amplified and ligated into the MluI and XhoI site of pGL2-basic. After sequencing, each construct was used for a reporter gene assay. The primers used to construct the ARK5 promoter carrying the MARE sequence mutants are listed in Table 2. Reporter gene assay The reporter gene assays were performed by using the PicaGene Luciferase Assay System (Toyo Ink MFG Co.

Matrigel-invasion assays were performed with a MatrigelCoated Invasion Chamber (Becton-Dickinson Company). Cells (5 104) were seeded into each chamber with culture medium, and the medium in the inner and outer chamber was changed after 6 h. After 24 h, the number of cells that had invaded was counted under a phase-contrast microscope. Western blotting Proteins for Western blotting analysis were prepared by lysing cells for 30 min with PBS containing 1% NP-40. All procedures were carried out at 41C. Supernatants were collected by centrifugation at 15 000 r.p.m. for 15 min, and their protein concentrations were determined with a BCA protein assay kit (Pierce Biotechnology Inc., Rockford, IL, USA) with bovine serum albumin as the standard. Sample proteins separated by SDS–PAGE were transferred onto nitrocellulose membranes by a semidry blotting system. The membranes were blocked at room temperature for 1 h with PBS containing 5% (w/v) skim milk, washed with a mixture of PBS and 0.05% Tween-20 (Sigma, Tween-PBS), and then incubated overnight at room temperature with antibody diluted with PBS. After washing with Tween-PBS, the membranes were incubated with a 2000-fold diluted HRPconjugated anti-rabbit IgG antibody. The membranes were then washed with Tween-PBS and developed with the ECL system (Amersham Biosciences, Uppsala, Sweden). Acknowledgements This work was partly supported by Grants for the 2nd- and 3rd-Term Comprehensive 10-year Strategy of Cancer Control from the Ministry of Health, Welfare and Labour, Grant for the Medical Frontier Program from the Ministry of Health, Welfare and Labour, and Research Grant of the Princess Takamatsu Cancer Research Fund. AS is the recipient of a Research Resident Fellowship from the Foundation for Promotion of Cancer Research. This research was supported by The Fund to Cure Myeloma (JS) and by NIH Grants CA55819 (JS) and CA97513 (JS) from the National Cancer Institute, Bethesda, MD, USA. Oncogene


6944 References Alessi DR, Andjelkovic M, Caudwell B, Cron P, Morrice N, Cohen P and Hemmings BA. (1996). EMBO J., 15, 6541–6551. Asosingh K. (2003). Verh. K. Acad. Geneeskd. Belg., 65, 127–134. Bergsagel PL and Kuehl WM. (2001). Oncogene, 20, 5611–5622. Bergsagel PL, Kuehl WM, Zhan F, Sawyer J, Barlogie B and Shaughnessy Jr J. (2005). Blood, 106, 296–303 [E-pub ahead of print, March 8]. Datta SR, Brunet A and Greenberg ME. (1999). Genes Dev., 13, 2905–2927. Delcommenne M, Tan C, Gray V, Rue L, Woodgett J and Dedhar S. (1998). Proc. Natl. Acad. Sci. USA, 95, 11211–11216. Di Raimondo F, Azzaro MP, Palumbo G, Bagnato S, Giustolisi G, Floridia P, Sortino G and Giustolisi R. (2000). Haematologica, 85, 800–805. Drevs J. (2003). Anticancer Res., 23, 1159–1161. Ferlin M, Noraz N, Hertogh C, Brochier J, Taylor N and Klein B. (2000). Br. J. Haematol., 111, 626–634. Hanamura I, Iida S, Akano Y, Hayami Y, Kato M, Miura K, Harada S, Banno S, Wakita A, Kiyoi H, Naoe T, Shimizu S, Sonta SI, Nitta M, Taniwaki M and Ueda R. (2001). Jpn. J. Cancer Res., 92, 638–644. Hannigan GE, Leung-Hagesteijin C, Fitz-Gibbon L, Coppolino MG, Radeva G, Filmus J, Bell JC and Dedhar S. (1996). Nature, 379, 91–96. Hardie DG and Carling D. (1997). Eur. J. Biochem., 246, 259–273. Hayman SR and Fonseca R. (2001). Curr. Treat. Options Oncol., 2, 205–216. Hedge SP, Kumar A, Kurschner C and Shapiro LH. (1998). Mol. Cell. Biol., 18, 2729–2737. Hideshima T and Anderson KC. (2002). Nat. Rev. Cancer, 2, 927–935. Huang K, Serria MS, Nakabayashi H, Nishi S and Sakai M. (2000). Gene, 242, 419–426. Hurt EM, Wiestner A, Rosenwald A, Shaffer AL, Campo E, Grogan T, Bergsagel PL, Kuehl WM and Staudt LM. (2004). Cancer Cell, 5, 191–199. Iida S, Hanamura I, Suzuki T, Kamiyama T, Kato M, Hayami Y, Miura K, Harada S, Tsuboi K, wakita A, Akano Y, Taniwaki M, Nitta M and Ueda R. (2000). Int. J. Hematol., 72, 85–91. Iida S, Rao PH, Butler M, Corradini P, Boccadoro M, Klein B, Chaganti RS and Dalla-Favera R. (1997). Nat. Genet., 17, 226–230. Ishikawa H, Tsuyama N, Abroun S, Liu S, Otsuyama K, Zheng X and Kawano MM. (2003). Leukemia Lymphoma, 44, 1477–1481. Kajihara M, Sone H, Amemiya M, Katoh Y, Isogai M, Shimano H, Yamada N and Takahashi S. (2003). Biochem. Biophys. Res. Commun., 312, 831–842. Kataoka K, Fujiwara KT, Noda M and Nishizawa M. (1994a). Mol. Cell. Biol., 14, 7581–7591. Kataoka K, Noda M and Nishizawa M. (1994b). Mol. Cell. Biol., 14, 700–712.

Oncogene

Kemp BE, Mitchelhill KI, Stapleton D, Michell BJ, Chen ZP and Witters LA. (1999). Trends Biochem. Sci., 24, 22–25. Kemp BE, Stapleton D, Campbell DJ, Chen ZP, Murthy S, Walter M, Gupta A, Adams JJ, Katsis F, Van Denderen B, Jennings IG, Iseli T, Michell BJ and Witters LA. (2003). Biochem. Soc. Trans., 31, 162–168. Klein B, Tarte K, Jourdan M, Mathouk K, Moreaux J, Jourdan E, Legouffe E, De Vos J and Rossi JF. (2003). Int. J. Hematol., 78, 106–113. Kusakai G, Suzuki A, Ogura T, Kaminishi M and Esumi H. (2004a). J. Exp. Clin. Cancer Res., 23, 263–268. Kusakai G, Suzuki A, Ogura T, Miyamoto S, Ochiai A, Kaminishi M and Esumi H. (2004b). Am. J. Pathol., 164, 987–995. Motohashi H, Shavit JA, Igarashi K, Yamamoto M and Engel JD. (1997). Nucleic Acids Res., 25, 2953–2959. Persad S, Attwell S, Gray V, Delcommenne M, Troussard A, Sanghera J and Dedhar S. (2000). Proc. Natl. Acad. Sci. USA, 97, 3207–3212. Persad S and Dedhar S. (2003). Cancer Metast. Rev., 22, 375–384. Qiang YW, Yao L, Tosato G and Rudikoff S. (2004). Blood, 103, 301–308. Ria R, Roccaro AM, Merchionne F, Vacca A, Dammacco F and Ribatti D. (2003). Leukemia, 17, 1961–1966. Seidl S, Kaufmann H and Drach J. (2003). Lancet Oncol., 4, 557–564. Siegel PM and Massague J. (2003). Nat. Rev. Cancer, 3, 807–821. Suzuki A, Kusakai G, Kishimoto A, Lu J, Ogura T and Esumi H. (2003a). Oncogene, 22, 6177–6182. Suzuki A, Kusakai G, Kishimoto A, Lu J, Ogura T, Lavin MF and Esumi H. (2003b). J. Biol. Chem., 278, 48–53. Suzuki A, Kusakai G-I, Kishimoto A, Shimojo Y, Miyamoto S, Ogura T, Ochiai A and Esumi H. (2004a). Oncogene, 23, 7067–7075. Suzuki A, Lu J, Kusakai G, Kishimoto A, Ogura T and Esumi H. (2004b). Mol. Cell. Biol., 24, 3526–3535. Tai YT, Podar K, Catley L, Tseng YH, Akiyama M, Shringarpure R, Burger R, Hideshima T, Chauhan D, Mitsiades N, Richardson P, Munshi NC, Kahn CR, Mitsiades C and Anderson KC. (2003). Cancer Res., 63, 5850–5858. Tian E, Zhan F, Walker R, Rasmussan E, Ma Y, Barlogie B and Shaughnessy J. (2003). N. Engl. J. Med., 349, 2483–2494. Yoshida S, Nakazawa N, Iida S, Hayami Y, Sato S, Wakita A, Shimizu S, Taniwaki M and Ueda R. (1999). Leukemia, 13, 1812–1816. Zhan F, Hanamura I, Anaissie E, Burlington B, Huang Y, Ramussen E, Xu H, Tian E, Santra M, Sawyer J, Zangari M, Lee C, Fassas A, Van Rhee F, Talamo G, Thertulian R, Walker R, Crowley J, Tricot G, Barlogie B and Shaughnessy J. (2004). Blood, 104, 26. Zhan F, Hardin J, Bumm K, Zheng M, Tian E, Sanderson R, Tang Y, Wilson C, Zangari M, Anaissie E, Morris C, Muwalla F, Van Rhee F, Fassas A, Ticot G, Crowley J, Barlogie B and Shaughnessy J. (2002). Blood, 99, 1745–1757. Zhan F, Tian E, Bumm K, Smith R, Barlogie B and Shaughnessy J. (2003). Blood, 101, 1128–1140.