• Open Access

Synergistic action of the microRNA-17 polycistron and Myc in aggressive cancer development

Authors

Errata

This article is corrected by:

  1. Errata: Correction Volume 99, Issue 1, 186, Article first published online: 7 December 2007

To whom correspondence should be addressed. E-mail: htagawa0279jp@yahoo.co.jp

Abstract

The c13orf25/miR-17 cluster, which is responsible for 13q31-q32 amplification in malignant lymphoma, contains the microRNA-17-18-19-20-92 polycistron. A previous study demonstrated that this polycistron could modulate tumor formation following transplantation of microRNA 17-19b into Eu-myc mice. Another study reported that Myc can upregulate the miR-17 cluster by binding directly upstream of the miR-17 locus. These findings suggest that Myc and the miR-17 cluster synergistically contribute to cancer development. In the study presented here, we observed recurrent 13q31-32 amplification in MYC-rearranged lymphomas (11 of 47 cases). Quantitative real-time polymerase chain reaction analysis of c13orf25 for MYC-rearranged lymphomas demonstrated that cases with 13q31-32 amplification showed significantly higher expression of c13orf25 than cases without such amplification, although cases without 13q31-32 amplification still showed slight upregulation of c13orf25. To investigate the relationship between Myc and the miR-17 polycistron in tumorigenesis, we engineered rat fibroblasts (Rat-1) that constitutively express the miR-17 polycistron (miR), Myc, or both miR and Myc. The highest level of miR expression was detected in Rat-1 transfected with both miR and Myc, whereas Myc transfectant cells alone also showed slight upregulation of miR. Furthermore, we demonstrated that nude mice injected with Rat-1 transfected with both miR and Myc presented more accelerated tumor growth than those injected with Myc transfectant cells. These results suggest that miR is stably upregulated in the presence of constitutive expression of Myc, and that the deregulation of miR and Myc synergistically contribute to aggressive cancer development, probably by repressing tumor suppressor genes. (Cancer Sci 2007; 98: 1482–1490)

MicroRNA (miRNA) are small RNA (20–24 nucleotides) and non-coding RNA that play important roles in gene regulation by pairing with the messages of protein-coding genes to specify mRNA cleavage or the repression of productive translation.(1–3) MiRNA display dynamic temporal and spatial expression patterns, and a disruption of these patterns may be associated with tumorigenesis. In normal somatic cells, the expression of miRNA is carefully controlled during differentiation to prevent progression to cancer. In the course of tumor development, however, some miRNA may be aberrantly upregulated, and indeed, recently some miRNA have been proven to be associated with tumorigenesis.(4–6) We previously demonstrated that the target gene of 13q31-32 genomic amplification in malignant lymphoma is c13orf25, which comprises two variants resulting from alternative splicing. The miRNA-17-18-19a-20-19b-92 cluster (miR-17 cluster) is contained within c13orf25 variant 2, indicating that c13orf25 is a pri-miRNA of the mature miRNA-17-92 (Fig. 1a).(5,7,8) Recently, overexpression of the miR-17 polycistron has also been detected in lung cancers.(9)

Figure 1.

13q31-32 amplification in MYC-rearranged malignant lymphomas. (a) Frequencies of genomic gains of 13q in a series of samples of 26 sporadic Burkitt's lymphomas (sBL) and 21 diffuse large B-cell lymphomas with MYC rearrangement. Bold bars indicate bacterial artificial chromosome (BAC) clones that showed recurrent genomic amplifications (n ≥ 2; log2 ratio ≥ +1.0). The arrow indicates the location of BAC RP11-121J17, which includes a miR-17 cluster. (b) Individual genome profiles of three cases of sBL with 13q amplification. Arrows indicate the location of BAC RP11-121J17. Genomic gain: log2 ratio > 0.2; genomic loss: log2 ratio < –0.2.

It has been reported that Eu-myc mice, transplanted with miR-17-19b and demonstrating overexpression of miRNA, tended to develop tumors significantly earlier than those without such overexpression.(5) Another study reported that Myc can upregulate the miR-17 polycistron by binding directly upstream of the miR-17 locus.(10) These findings suggest that constitutive expression of Myc, and deregulation of the miR-17 polycistron and Myc synergistically contribute to aggressive cancer development. Although it has been reported that there are two mechanisms for aberrant upregulation of the miR-17 cluster (one is genomic amplification and the other is the effect of Myc binding), it is still unclear which mechanism is more effective for overexpression of the miR-17 cluster, or which mechanism is more important for tumorigenesis.

The aim of the present study is to investigate the relationship between Myc and the miR-17 polycistron in tumorigenesis. To examine whether MYC-rearranged lymphoma could possess 13q31-32 genomic amplification, we carried out array comparative genomic hybridization (CGH) for a total of 47 cases of MYC-rearranged lymphomas. We further investigated the relationship between the miR-17 cluster and Myc by using rat fibroblast (Rat-1) transfectants that stably express the miR-17 polycistron or Myc, or both the miR-17 polycistron and Myc.

Materials and Methods

MYC-rearranged primary lymphoma samples.  All cases with sporadic Burkitt's lymphoma (sBL: n = 26) and diffuse large B-cell lymphoma (DLBCL) with MYC rearrangement (n = 21) were reviewed by expert pathologists, and diagnoses were established according to the criteria of the World Health Organization classification.

sBL  All cases of sBL showed typical pathological features such as a ‘starry sky’ pattern, and expressed BCL6 and a Ki-67 index over 95%. Histological analysis showed that 16 cases were classical Burkitt's lymphoma and 10 cases were atypical. All cases were CD10 positive.

DLBCL  Two cases of DLBCL presented diffuse, mixed small-large-cell lymphoma, three cases were large-cell immunoblastic lymphoma, two cases were large-cell centroblastic lymphoma and the remaining 14 DLBCL belonged to the diffuse large-cell category. Nine of the 21 cases were previously reported by Akasaka et al.(11) Seven cases were positive for CD10, and 14 cases were negative.

MYC rearrangement was detected in all 26 of the sBL and 21 of the DLBCL using fluorescence in situ hybridization (FISH), long-distance polymerase chain reaction (LD-PCR) and Southern blot analyses. The LSI MYC Dual Color Break-apart Rearrangement Probe-Hybridization set (Vysis, Downers Grove, IL, USA) was used for FISH. LD-PCR was carried out for all the cases as described previously.(11)

Cell lines.   Karpas 1718, Jeko-1, REC-1, G519, SP49 and JVM2 are cell lines derived from B cell lymphoma. Karpas 1718, Jeko-1 and REC-1 have been proven to possess 13q amplification and overexpression of the miR-17 cluster.(7,8) The HeLa cell line, confirmed not to possess any copy number alteration at 13q31-32, was used as a non-lymphoma control.(8)

Northern blot analysis.  Northern blotting for mature miRNA was carried out as described elsewhere.(8) For northern blot analysis of mature miRNA, 10 µg of RNA was separated on a 15% denaturing polyacrylamide gel. Total RNA from the HeLa cell line and sBL samples was extracted using the acid–phenol precipitation method. Northern blotting for MYC mRNA and transforming growth factor-β type II receptor (TβRII) (1101 bp derived from rat) was also carried out as described elsewhere.(8)

Western blot analysis.  Western blot analysis was carried out according to the manufacturer's protocol (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Antibody to TβRII was purchased from Santa Cruz Biotechnology.

Quantitative real-time reverse transcription–polymerase chain reaction.  Expression levels of c13orf25 (pri-miRNA-17 polycistron) mRNA were measured by means of real-time fluorescence detection. Briefly, the primers for c13orf25 variant 2 were: sense, 5′-tctactgcagtgaaggcact-3′; and antisense, 5′-tgccagaaggagcacttagg-3′. The polymerase chain reaction (PCR) fragment (170 bp) included mature miR-17-5p, miR-17-3p and miR-18 sequences. The real-time PCR using SYBR Green (Bio-Rad, CA, USA) and the primers was carried out with the Smart Cycler System (Takara Bio, Tokyo, Japan) according to the manufacturer's protocol. G6PDH served as an endogenous control; the expression levels of c13orf25 mRNA in each of the samples were normalized on the basis of the corresponding G6PDH content and recorded as relative expression levels.

Array-based CGH.  Array CGH was carried out for DLBCL cases using previously described methods with a glass slide of Aichi Cancer Centre (ACC)-array-slide version 4.0 (Aichi, Japan). The array consisted of 2304 bacterial artificial chromosome and P-1-derived artificial chromosome clones, covering the whole human genome with roughly 1.3-Mb of resolution. Bacterial artificial chromosome clone RP11-121J17, which contains the c13orf25 gene, was included in the platform. DNA preparation, labeling, array fabrication and hybridization were carried out as described previously.(7) The thresholds for the log2ratio of gains and losses were set at the log2ratio of +0.2 and –0.2, respectively.

Construction of plasmids expressing the miR-17 cluster and retrovirus infection.  We constructed Rat-1 cells stably expressing mouse Myc or both miR-17-19b (or miR-17-92) and mouse Myc. The entire coding region of the miR-17-92 (miR17-19b) polycistron was amplified by means of reverse transcription-PCR using the gene-specific primers 5′-tgtcagaataatgtcaaagtgct-3′ and 5′-cactaccacagtcagttttgcat-3′ for miR-17-19b, and 5′-ccaaactcaacaggccgggacaag-3′ and 5′-cactaccacagtcagttttgcat-3′ for miR-17-92. The PCR products were cloned into an appropriate cloning site of the PMXpuro or PMXneo vectors.(12) The construct of PMX vector (1.5 µg), Myc (1.5 µg), miR-17-19b (or miR-17-92) and miR17-19b-1 (or miR-17-92 + Myc) (1.5 µg) was stably transfected into Rat-1 cells by means of retroviral transfection using FuGene 6 (Promega, Madison, WI, USA). The transfection was accomplished via Plat-E cells, a potent retrovirus packaging cell line that was generated based on the 293T cell line.(12) At 48 h after infection, the cells were drug selected with puromycine (3 µg/mL; Gibco BRL, Gaithersburg, MD, USA) or G418 (1.5 g/L; Gibco BRL). The Myc, miR-17-19b and miR17-19b (or miR-17-92 + Myc) expression constructs were transfected into Rat-1 fibroblasts using the polybrene (1.5 µg/mL; Invitrogen, San Diego, CA, USA) reagent according to the manufacturer's instructions. MiR-17-19b infection was also conducted for Rat-1 cells stably expressing Mycpuro through contact with PMX-neo-miR17-19b (miRneo). The culture medium was iscove + 10% fetal calf serum (Rat-1).

Transformation assay.

Growth assay.  After transfection of the vector, Myc and miR + Myc to Rat-1 cells, 3 × 105 cells were plated in triplicate in 6-cm dishes, and the number of cells was counted twice in 1 week.

Colony-formation assay.  Drug-selected cells were trypsinized and split to a low melting point soft agarose gel (Cambrex Bio Science, Baltimore, MD, USA) at 1 × 103 cells/dish and plated to six wells in triplicate. After 14 days, the dishes were stained with Giemsa and the number of colonies was counted.

Nude mice assay.  A total of 2 × 106 Rat-1-transfected cells were injected subcutaneously into the right or left side of the body of 6–8-week-old female CD-1 athymic nude mice (Charles River Co., Yokohama, Japan). Tumor growth was monitored twice a week.

Luciferase reporter assay.  The firefly luciferase vector was modified from the pGL3 Control Vector (Promega), such that the TβRII 3′ untranslated region (UTR) was inserted into the XbaI site immediately downstream of the stop codon. Similarly, miR-17-5p (pGL-17), miR-20 (pGL3-20) and their complementary sequences (miR-17c, miR-20c) were inserted immediately downstream at the stop codon of the luciferase gene (XbaI site). Rat-1 cells expressing miR + Myc and vector (mock) were cultured to 80–90% confluence in 12-well plates. Cells were transfected with 0.8 µg of the firefly luciferase reporter vector and 0.16 µg of the control vector containing Renilla luciferase, pRL-TK (Promega), in a final volume of 1.0 mL using Lipofectamine 2000 (Invitrogen). Firefly and Renilla luciferase activities were measured consecutively using the Dual-luciferase assays (Promega) 30 h after transfection (two independent experiments, each with three culture replicates). The following sequences were inserted into the PGL3 control vector: human TβRII 3′ UTR wild type, 5′-tgttagcacttcctcaggaaatgagattgatttttacaatagccaataacatttgcactttattaatgcct-3′; and TβRII 3′ UTR mutated, 5′-tgttagcacttcctcaggaaatgagattgatttttacaatagccaataacatttgAaAtAtattaatgccta-3′.

Statistical analysis.  Fisher's exact test was used for comparison between two groups of frequencies of gain or loss for each single clone, and the Mann–Whitney U-test was used for detecting significant differences in the expression levels of c13orf25 between groups with or without 13q31–32 genomic amplifications. All of the statistical analyses were conducted using the STATA version 8 statistical package (StataCorp, College Station, TX, USA).

Results and Discussion

Overexpression of the miR-17 polycistron induced by 13q31-32 amplification in MYC-rearranged aggressive lymphomas.  For the detection of primary tumor cases that showed both MYC and miR-17 polycistron deregulation, we carried out array CGH for 47 cases of B-cell lymphoma with MYC rearrangement. These lymphomas included 26 cases of sBL and 21 cases of DLBCL. We found that the two groups both showed gains of 1q21-32 (sBL, 10 cases; DLBCL, 11 cases), 12q12-13 (sBL, five cases; DLBCL, seven cases) and 13q31-32 (sBL, six cases; DLBCL, five cases). Among these gains, recurrent genomic amplification was found only at 13q31-32 (three cases per group; Fig. 1a,b), which harbors the miR-17 polycistron.(7,8)

To examine expression of the c13orf25/miR-17 cluster, we carried out quantitative real-time PCR and northern blotting for the miR-17 cluster of primary cases with or without 13q amplification. Quantitative real-time PCR analysis of the sBL samples (Fig. 2a) demonstrated that sBL with amplification of 13q31-32 showed significantly higher expression of the miR-17 cluster than the cases without this amplification (Mann–Whitney, P < 0.001).

Figure 2.

Overexpression of the miR-17 cluster in lymphoma with or without MYC rearrangement. (a) Real-time quantitative polymerase chain reaction of the miR-17 cluster in lymphomas with (n = 14) and without (n = 27) MYC rearrangement. Sporadic Burkitt's lymphoma (sBL) = 14 cases, diffuse large B-cell lymphoma (DLBCL) = 12 cases, and adult T cell lymphoma (ATLL) = 15 cases. (b) Northern blot analysis of miR-19a and -20 in sBL samples with (n = 5) and without (n = 5) 13q amplification. Average c13orf25 expression: sBL with 13q amplification = 21.94, sBL without 13q amplification = 2.23, DLBCL with 13q amplification = 6.99, DLBCL without 13q amplification = 0.74, ATLL = 0.27.

Quantitative real-time PCR of the c13orf25/miR-17 cluster was also examined for malignant lymphomas without MYC rearrangement. These lymphomas included 12 DLBCL, comprising six cases with 13q amplification and six cases without, as well as 15 adult T-cell lymphoma cases. Although we found overexpression of the miR-17 cluster in DLBCL with 13q amplification (average c13orf25 expression, 6.99), the average expression level was lower than that of sBL with 13q amplification (average c13orf25 expression, 21.94). Furthermore, DLBCL without 13q gain or amplification showed miR-17 cluster expression that was lower than that of sBL without 13q amplification (average, 2.23). The expression of c13orf25 of adult T-cell lymphoma (average, 0.27) was far lower than that of sBL and DLBCL without 13q amplification (average, 0.74). These results suggest that MYC deregulation also could enhance expression of the miR-17 cluster, whereas 13q amplification alone could induce sufficiently high upregulation of the c13orf25/miR-17 cluster. In primary lymphomas, 13q31-32 amplification thus plays a more important role in overexpression of the miR-17 cluster than does endogenous upregulation of the cluster resulting from Myc overexpression. Upregulation of mature miR-17, -19 and -20 was also confirmed by northern blot analysis of sBL samples. As shown in Fig. 2b, expression of miR-19a and 20 was higher in cases with 13q amplification than in those without. Overexpression of miR-17-5p was also detected, but the expression of miR-17-3p and miR-18 was very low in these primary samples (data not shown). Upregulation of the miR-17 cluster itself might be most important for its tumorigenicity. 13q31-32 amplification plays essential roles in stabilizing and upregulating the miR-17 cluster in malignant lymphomas. MYC deregulation has a lesser effect on miR-17 cluster upregulation than on 13q amplification.

In vitro experiment for overexpression of the miR-17 polycistron in the presence of constitutive Myc expression.  In primary lymphoma cases, upregulation of the miR-17 polycistron is induced via the mechanism of DNA copy number amplification. However, 13q amplification and 8q24 translocation induce deregulation of the miR-17 polycistron and MYC, respectively. These genetic alterations may then jointly lead to the aggressive development of lymphomas. Because it is known that N-myc-overexpressed Rat-1 is capable of transformation,(13) we speculated that additional expression of the miR-17 polycistron could enhance the tumorigenicity of this fibroblast. To validate this speculation, we used the retroviral transfection system to engineer Rat-1 that constitutively expressed either the miR-17 polycistron or Myc or both.

We inserted 674-bp (miR-17-18-19a-20-19b: miR-17-19b) or 755-bp (miR-17-18-19a-20-19b-92: miR-17-92) fragments spanning a given miRNA genomic region into a modified PMX vector (Fig. 3a), so that they were under the control of a left terminal repeat (LTR) promoter instead of the genomic copy amplification. As a result, we succeeded in establishing Rat-1 cells with stable expression of the miR-17 polycistron (miR-17-19b) (Fig. 3b). Because expression of mature miRNA in the miR-17-92 transfectant was lower than that of the miR-17-19b transfectant, we used the latter for the subsequent tumorigenesis assays.

Figure 3.

Figure 3.

Overexpression of the miR-17 cluster in the presence of constitutive expression of Myc. (a) Schematic illustrations of c13orf25 and the miR-17 polycistron, and of the PMX vector constructs PMX-Myc (puromycin resistant), PMX-miR (neomycin resistant) and PMX-miR-Myc (puromycin resistant). (b) The expression of miR-20 of rat fibroblast (Rat-1) cells transfected with Mycpuro, miR-17-19bneo, miR-17-19bneo + Mycpuro, miR-17-92neo, miR-17-92neo + Mycpuro, and vector control (mock). The quantitative expression for miR-20 was monitored using a densitometer. (c) Expression of MYC mRNA of Rat-1 transfectants. (d) Expression of miR-19a, miR17-5p, miR-17-3p and miR-18 in Rat-1 cells transfected with miRpuro + Mycpuro, Mycpuro and mock. Exposure times are indicated below the boxes. (e) Expression of mature miR-17-5p, miR-19a, miR-17-3p and miR-18 of indicated B cell lymphoma cell lines. (f) Comparison of mature miR-20 expression in Rat-1 transfectants, B cell lymphoma cell lines with 13q amplification (Karpas 1718 and REC1),(7,8) and the HeLa cell line, which was derived from an epithelial cancer cell line without 13q amplification. The figure shows standard deviations for three independent experiments. (g) Rat-1 expressing Mycpuro were subsequently transduced with miR-17-19bneo. ‘Mycpuro + miRneo’ represents a Rat-1 cell transduced with both Mycpuro and miR-17-19bneo by dual drug selection as shown in the schematic illustration next to the photograph.

Figure 3.

Figure 3.

Overexpression of the miR-17 cluster in the presence of constitutive expression of Myc. (a) Schematic illustrations of c13orf25 and the miR-17 polycistron, and of the PMX vector constructs PMX-Myc (puromycin resistant), PMX-miR (neomycin resistant) and PMX-miR-Myc (puromycin resistant). (b) The expression of miR-20 of rat fibroblast (Rat-1) cells transfected with Mycpuro, miR-17-19bneo, miR-17-19bneo + Mycpuro, miR-17-92neo, miR-17-92neo + Mycpuro, and vector control (mock). The quantitative expression for miR-20 was monitored using a densitometer. (c) Expression of MYC mRNA of Rat-1 transfectants. (d) Expression of miR-19a, miR17-5p, miR-17-3p and miR-18 in Rat-1 cells transfected with miRpuro + Mycpuro, Mycpuro and mock. Exposure times are indicated below the boxes. (e) Expression of mature miR-17-5p, miR-19a, miR-17-3p and miR-18 of indicated B cell lymphoma cell lines. (f) Comparison of mature miR-20 expression in Rat-1 transfectants, B cell lymphoma cell lines with 13q amplification (Karpas 1718 and REC1),(7,8) and the HeLa cell line, which was derived from an epithelial cancer cell line without 13q amplification. The figure shows standard deviations for three independent experiments. (g) Rat-1 expressing Mycpuro were subsequently transduced with miR-17-19bneo. ‘Mycpuro + miRneo’ represents a Rat-1 cell transduced with both Mycpuro and miR-17-19bneo by dual drug selection as shown in the schematic illustration next to the photograph.

Myc overexpression was confirmed in Rat-1 transfected with Myc, as well as in the cells transfected with both miR-17-19b and Myc (miRpuro + Mycpuro/miR + Myc) (Fig. 3c) but not in Rat-1 transfected with miR-17-19b or the vector control. Northern blot analysis of miR-20 of Rat-1 transfectants showed that these cells transfected with miR + Myc expressed higher levels of mature miR-20 than the other transfectants. Rat-1 transfected with miR showed upregulation of miR-20 as well, although the expression was weaker than that of the cells transfected with miR + Myc (Fig. 3b). Rat-1 transfected with Myc also showed slightly higher expression of mature miR-17-5p, -19a, -19b and -20 than cells that were transfected with the PMX vector alone. Because O'Donnell et al. recently reported that Myc can lead to upregulation of endogenous miR-17 polycistron via direct binding upstream of the miR-17-5p sequence,(10) the slight elevation of mature miR-17-19b in Rat-1 transfected with Myc should be due to endogenous upregulation. In fact, slight upregulation of the miR-17 polycistron by MYC deregulation was also observed in primary lymphomas (Fig. 2a). The expression levels of mature miR-17-5p, -19 and -20 in Rat-1 transfected with miR + Myc were enhanced more than expected. For instance, as shown in Fig. 3b, the expression level of miR-20 in Rat-1 transfected with miR + Myc was 7.5, which is 1.6 times higher than expected, that is, 4.6 (miR + Myc = 3.3 + 1.3).

Expression levels of the other miRNA within the miR-17 polycistron were also examined, and upregulation of miR-17-5p and miR-19a was confirmed in miR + Myc transfectant cells, whereas miR-17-3p and miR-18 did not show any upregulation, as reported previously (Fig. 3d).(8) This suggests that miR-17-3p and miR-18 may not play an essential role in tumorigenesis.(8)

It is worth noting that the expression patterns of Rat-1 transfected with the miR-17 cluster were the same as those of lymphomas. For instance, the expression of miR17-3p and miR-18 were not upregulated in the lymphoma cell lines with 13q amplification (Karpas 1718, Jeko-1 and REC-1) (Fig. 3e,f).

We also established Rat-1 expressing both miR and Myc by using dual transfection (Fig. 3g). Rat-1 cells transfected with Myc (puromycine resistant, Mycpuro) were additionally transfected with the miR-17 polycistron (neomycin resistant, miRneo). The expression levels of mature miR-17-5p, -19 and -20 were also enhanced 1.5–1.6 times more than expected after the miR-17-19b transfection. Because miR-17-19b sequences inserted into the PMX vector did not contain a Myc-binding locus, it is unlikely that the Myc gene exerts an endogenous effect leading to miR-17 polycistron expression. Overexpression of these mature miRNA in miRpuro + Mycpuro (miR + Myc) or Mycpuro+ miRneo (Myc + miR) transfectant cells were not only due to the LTR promoter of the PMX vector but also to the ‘exogenous’ effect of Myc on the miR-17 polycistron. It is thus tempting to speculate that the Myc protein or mRNA may act as a ‘stabilizer’ for stable high-level expression of the mature miR-17-19b.

Synergistic relationship of the miR-17 polycistron and Myc in aggressive cancer development.  Because Rat-1 fibroblasts have been shown to be capable of transformation due to the constitutive expression of N-myc,(13) we speculated that deregulation of the miR-17 polycistron and Myc may collaborate for the augmentation of tumorigenic activity. The constitutive expression of Myc can induce tumorigenicity in Rat-1 cells. In fact, Rat-1 overexpressing both miR and Myc presented a full-blown neoplastic transformation phenotype accompanied by anchorage-independent growth and showed the most accelerated cell growth (Fig. 4a).

Figure 4.

Tumorigenic assays. (a) Growth assay. Rat fibroblast (Rat-1) cells containing the indicated vectors were drug selected and then subjected to a growth assay. Standard deviations for three independent transductions are shown. (b) Colony assay. Rat-1 cells containing Myc, the miR-17 polycistron (miR), miRpuro + Mycpuro and Mycpuro + miRneo were plated in soft agar and counted after 2 weeks. ‘miRpuro + Mycpuro’ represents a Rat-1 cell transfected with both miR-17-19b and Myc by simultaneous construct as shown in Fig. 3a. ‘Mycpuro + miRneo’ represents a Rat-1 cell transfected with both Myc and miR-17-19b by dual transfection as shown in Fig. 3g. Standard deviations for three independent experiments are shown. (c) Nude mice assay. A total of 2 × 106 Rat-1 cells expressing Myc (n = 8), miR-17-19b (n = 8), miR + Myc (n = 8) and mock (n = 6) were injected subcutaneously. The ‘miRpuro + Mycpuro’ transfectant was used in this experiment. Averages and standard deviations are shown. The tumor size was monitored twice a week until days 35–42.

A cell-growth assay demonstrated that Rat-1 cells transfected with miR + Myc showed more accelerated cell growth than other transfectants (Fig. 4a). It was interesting that, even without Myc activation, miR-17-19b conferred a growth advantage to the cells, although to a lesser extent than that observed in cells with Myc or miR + Myc (or Myc + miR). This suggests that overexpression of the miR-17-19b polycistron alone also causes tumorigenicity.

Next, we carried out a colony assay and found that there were more colonies of Rat-1 transfected with miRpuro + Mycpuro (miR + Myc) or transfected with Mycpuro + miRneo (Myc + miR) than of cells transfected with Myc or with miR alone (Fig. 4b). Although Rat-1 transfected with miR alone did generate colonies in soft agar, there were fewer colonies than of cells transfected with Myc and miR + Myc (or Myc + miR). In addition, a colony assay using Rat-1 transfected with miR-17-92 + Myc did not show a significantly higher number of colonies than cells transfected with Myc alone (data not shown).

The tumorigenic activity of miR-17-19b was further investigated with an athymic nude mouse assay. The cells containing miR + Myc, Myc and miR showed efficient tumor growth in nude mice, whereas the mice injected with miR + Myc transfectant showed more accelerated tumor growth than those injected with Myc transfectant (Fig. 4c). These results suggest that Myc overexpression could enhance the tumorigenic potential of the miR-17 polycistron, and that deregulation of the miR-17 polycistron and Myc jointly contribute to aggressive cancer development.

Our results for the expression pattern and tumorigenicity of the miR-17 polycistron were similar to the results of He et al.(5) It should be noted that, according to their report, miR-17-92 transfectant cells did not show tumorigenicity. This may be due to the lower expression level of mature miR-17, -19 and -20 compared with that of miR-17-19b + Myc transfectant cells, although the reason remains unclear. In addition to these previously reported findings, we demonstrated in our Rat-1 experiment that upregulation of the miR-17 polycistron alone shows tumorigenicity. The result of our experiment using Rat-1 cells indicates that the miR-17 polycistron and Myc possess tumorigenic potential not only in lymphomas but also in solid tumors. The miR-17 polycistron might thus repress essential target tumor suppressor genes regardless of tumor phenotypes.

Deregulation of the miR-17 polycistron can be assumed to contribute to aggressive cancer development by repressing tumor suppressor genes. Among the hundreds of predicted target genes of the miR-17 polycistron, the predicted tumor suppressor targets are p130 (RBR2), PTEN, MeCp2 and TβRII.(14,15) Screening for these candidate targets by western and northern blot analyses revealed that the expression levels of TβRII in Rat-1 transfected with miR + Myc were remarkably repressed compared to those of the other transfectants. The miR + Myc transfectant was found to produce a 2–3-fold decrease in TβRII RNA levels and a 6–8-fold decrease in protein levels (Fig. 5a,b). It was noteworthy that the cells expressing Myc also effected a slight reduction in the expression levels because of endogenous upregulation of the miR-17 polycistron. Target genes containing sequences that are completely complementary to the miRNA are generally degraded by an RNA interference mechanism, whereas targets with partially complementary sequences at their 3′ UTR are subjected to translation inhibition and, to a lesser extent, to mRNA degradation.(16–18) To determine whether TβRII was directly regulated by the miR-17 polycistron, we carried out a luciferase activity assay and, as seen in Fig. 5c, transfection experiments showed significant repression of luciferase activity in the wild-type construct compared to the mutant TβRII. This suggests that TβRII is the direct downstream target of the miR-17 polycistron. Recently, another study found that TβRII is also a target for miR-17-5p and -20.(19)

Figure 5.

Transforming growth factor-β receptor type II (TβRII) as a target of the miR-17 cluster. (a) Western blot analysis of TβRII for the indicated rat fibroblast (Rat-1) transfectants. α-Tublin was use as a control. (b) Northern blot analysis of TβRII (rat) for the indicated Rat-1 (rat) transfectants. Ethidium bromide (EtBr) was used as a control. (c) Luciferase assay. pGL3-TβRIImut, pGL3-inserted mutated TβRII 3′ untranslated region (UTR); pGL3-TβRII, pGL3-inserted wild-type TβRII 3′ UTR; PGL3-17 or -20, negative control containing a miR-20 sequence. PGL3-17c and pGL3-20c are positive controls containing their complementary sequence inserted into the pGL3 XbaI site. The firefly luciferase vector was modified from the pGL3 Control Vector (Promega), such that TβRII 3′ UTR was inserted into the XbaI site immediately downstream of the stop codon. Similarly, miR-17-5p (pGL-17), miR-20 (pGL3-20) and their complementary sequences (miR-17c, miR-20c) were inserted immediately downstream of the stop codon of the luciferase gene (XbaI site).

Although we could not provide evidence that repression of TβRII is crucial for lymphoma with 13q amplification, we believe that our findings represent important information about the tumorigenic function of the miR-17 polycistron, whose deregulation may render the pathological features of cancers more aggressive by repressing tumor suppressor genes.

It is tempting to speculate that overexpression of the miR-17 cluster may be capable of repressing Myc-inducible tumor suppressors and thus induce more aggressive development of tumors with miR and Myc overexpression than of tumors with Myc overexpression. Furthermore, we speculate that the overexpression of other families in the miR-17 polycistron, such as the miR-19 family (miR-19a and miR-19b), may also play a crucial role in tumorigenicity. Further investigation is needed to elucidate not only the exact role of TβRII downregulation in lymphomagenesis but also the possible participation of other targets of the miR-17 polycistron.

Acknowledgments

We express our appreciation to Ms Hiroko Suzuki and Ms Yumiko Kasugai for their outstanding technical assistance. PMX vector was kindly provided by Dr Toshio Kitamura. We also thank Dr Hitoshi Ohono (Kyoto University) and Dr Naoya Nakamura (Tokai University) for contributing tumor samples of diffuse large B cell lymphoma with MYC rearrangement.

This work was supported in part by Grants-in-Aid from Trade and Industry, from the Ministry of Health, Labor and Welfare, from the Ministry of Education, Culture, Sports, Science and Technology (M.S.), from the Grant-in-Aid B Japan Society for the Promotion of Science (M.S.), from the Foundation of Promotion of Cancer Research (M.S.), from the Princess Takamatsu Cancer Research Fund (M.S.), from the Project Research Foundation of Aichi Cancer Center (H.T.), from the Grant-in-Aid C from the Japan Society for the Promotion of Science (H.T.), from Mochida Memorial Foundation for Medical and Pharmaceutical Research (H.T.), and from the Charitable Trust Laboratory Medicine Foundation of Japan (H.T.).

Ancillary