Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Gastroenterology. Author manuscript; available in PMC 2010 June 18.
Published in final edited form as:
Gastroenterology. 2009 May; 136(5): 1689–1700.
PMCID: PMC2887605

The miR-106b-25 polycistron, activated by genomic amplification, functions as an oncogene by suppressing p21 and Bim


Background & Aims

Barrett’s esophagus (BE) is a highly premalignant disease that predisposes to the development of esophageal adenocarcinoma (EAC); however, the involvement of microRNAs (miRs) in BE-EAC carcinogenic progression is not known.


Esophageal cultured cells (HEEpiC, QhTRT, ChTRT, GihTRT and OE-33) and esophageal tissues (22 normal epithelia, 24 BE and 22 EAC) were studied. MiR microarrays and quantitative RT-PCR were employed to explore and verify differentially expressed miRs. Quantitative genomic PCR was performed to study genomic copy number variation at the miR-106b-25 polycistron and MCM7 gene locus on chromosome 7q22.1. In vitro cell proliferation, cell cycle, and apoptosis assays and in vivo tumorigenesis experiments were performed to elucidate biological effects of the miR-106b-25 polycistron. Western blotting and luciferase assays were performed to confirm direct mRNA targeting by miR-106b-25 polycistron.


The miR-106b-25 polycistron exerted potential proliferative, anti-apoptotic, cell cycle-promoting effects in vitro and tumorigenic activity in vivo. MiRs -93 and -106b targeted and inhibited p21, while miR-25 targeted and inhibited Bim. This polycistron was upregulated progressively at successive stages of neoplasia, in association with genomic amplification and overexpression of MCM7. In addition, miRs -93 and -106b decreased p21 mRNA, while miR-25 did not alter Bim mRNA, suggesting the following discrete miR effector mechanisms: 1) for p21, mRNA degradation; 2) for Bim, translational inhibition.


The miR-106b-25 polycistron is activated by genomic amplification and is potentially involved in esophageal neoplastic progression and proliferation via suppression of two target genes, p21 and Bim.


Barrett’s esophagus (BE) is a highly premalignant disease that predisposes to the development of esophageal adenocarcinoma (EAC), one of the most rapidly increasing cancers in developed nations 1. The prevalence of BE in United States endoscopic series ranges from 0.9% (of 51,311 patients) 2 to 4.5% (of 833 patients) 3. A recent study from Sweden reported a prevalence of BE of 1.6% (among 21,610 patients) 4. These findings imply that more than 3 million individuals may harbor BE in the United States alone 5. Among these patients, EAC susceptibility is a life-threatening long-term sequela. The incidence of EAC in BE patients is estimated 1.8 per 1,000 person-years, 30-fold higher than in the general population6.

The molecular genetics of BE and its evolution to EAC have been studied710. Nonetheless, molecular mechanisms underlying the BE-EAC carcinogenic sequence remain unclear, and a more thorough understanding of the molecular basis of this process would yield several benefits. These benefits include: 1) clues to biological pathways underlying Barrett’s-associated neoplastic transformation; 2) useful biomarkers of early cancer detection, disease progression, or ultimate prognosis; and 3) therapeutic targets for intervention in both the prevention and the treatment of this disease.

MicroRNAs (miRs) are a class of abundant, approximately 22 nucleotide non-coding RNAs that mediate post-transcriptional regulation of target mRNAs. Most known functions of miRs relate to negative gene regulation: miRs silence gene expression, usually by interfering with mRNA stability or protein translation. Cancer-specific miR fingerprints have been identified in every tumor type examined. Emerging evidence reveals that a number of miRs are involved in tumorigenesis and/or tumor progression11, 12. MiR regulatory mechanisms involved in cancer progression may be viewed as previously missing pieces of a puzzle that is now essential to understanding the complete molecular genetic landscape of cancer progression. However, thus far, there has been only one report that describes miR expression in BE and EAC13.

Herein, we describe novel findings showing that: (1) the miR 25-93-106b polycistron is activated in BE-associated neoplastic transformation; (2) DNA copy number variation is closely associated with activation of the miR 25-93-106b polycistron; and (3) the miR 25-93-106b polycistron exerts cell-proliferative and oncogenic effects via key downstream regulatory targets.

Materials and Methods

See also supplementary documents for full description.

Cell Culture

A human esophageal adenocarcinoma cell line, OE-33, three metaplastic BE-derived cell lines, QhTRT, ChTRT and GihTRT 14, human normal non-immortalized primary esophageal epithelial cells, HEEpiC, and human normal lung fibroblasts, WI-38 were used.

Tissue Specimens

All patients provided written informed consent under a protocol approved by the Institutional Review Boards at the University of Maryland and Baltimore Veterans Affairs Medical Centers. Clinicopathological information is displayed in Supplementary Table S1.

MicroRNA Microarrays

MiRNA Labeling Reagent and Hybridization Kits and Human miR Microarray Kits (Agilent) were used. Data was collected and normalized to non-functional small RNA internal controls.

Quantitative RT-PCR (qRT-PCR) for miR Expression

TaqMan MicroRNA Assays, Human (Applied Biosystems) were used. All qRT-PCRs were performed in triplicate. RNU6B was used as an internal control.

Quantitative RT-PCR for mRNA Expression and Quantitative Genomic DNA PCR

iQ5 Multicolor Real-Time PCR Detection System and iQ SYBR Green Supermix (BIO-RAD) were used for both quantitative RT-PCRs of mRNA expression and quantitative genomic DNA PCRs. Primer sequences are available in Supplementary Table S2. All PCRs were performed in triplicate. Beta actin was used to normalize mRNA expression levels. An Alu sequence15 was used to normalize quantitative genomic DNA PCR.

Transfection of miR Mimics and Inhibitors

Synthesized RNA duplexes of miR mimics, miR inhibitors and small inteferring RNA duplex against p21 were purchased from Dharmacon. Cells were transfected with 60nM of each miR mimic, inhibitor or siRNA.

Cell Proliferation Assay

Cells were plated onto 96-well plates at day 0. At every other day until day 5, Cell Proliferation Reagent WST-1 (Roche) was added to each well, incubated at 37°C for 1 hour and an optical density was measured.

Ki-67 staining

OE-33 cells were transfected with miR-25-INH, miR-93-INH, miR-106b-INH, or NSC-INH. 48 hours after transfection, cells were fixed in 100% acetone. Ki-67 positive cells were detected by using anti-Ki-67 antibody (Invitrogen) and an UltraVision Detection System (Thermo Scientific Inc.).

In Vivo Tumorigenesis

Suspensions of 2 × 106 OE-33 cells transfected with miR inhibitors were mixed with BD Matrigel Basement Membrane Matrix (BD). Cells were injected subcutaneously into both flanks of athymic nude mice (Harlan) at day 0. Tumor growth was estimated by averaging the volumes of tumors at eight sites. Tumor volume was calculated according to the following formula: L2/2 × W, where L is the length and W is the width of the tumor. All animal procedures were performed in accordance with institutional guidelines under a protocol approved by the Johns Hopkins Animal Care and Use Committee.

Western Blotting

For CDKN1A (p21Cip1/WAF1), rabbit anti-p21Cip1/WAF1 from Invitrogen was used, while for BCL2L11 (Bim), rabbit-anti-Bim from Cell Signaling Technology was used. For an internal control, mouse anti-beta actin monoclonal antibody was used.

Luciferase Reporter Assay

Full-length p21 3′-untranslated region (3′-UTR) and truncated Bim 3′UTR, common to both Bim EL and Bim L, flanking miR responsive sites were amplified. We obtained plasmid clones pGL4.13 (Promega) containing correctly oriented inserts (pGL4.13-P21UTR and pGL4.13-BimUTR). Plasmid clones that contained reversely oriented inserts (pGL4.13-revP21UTR and pGL4.13-revBimUTR) were used as universal negative controls for the assay. We also constructed plasmids containing p21-3′UTR with mutated seed regions for the predicted miR-93/miR-106b binding sites (pGL4.13-mutP21UTR), along with plasmids containing the Bim-3′UTR with mutated seed regions for the predicted miR-25 binding sites (pGL4.13-mutBimUTR). Primer sequences are available in Supplementary Table S2. The constructed pGL4.13 vector and an internal control pRL-CMV (Renilla luciferase) vector were co-transfected 24 h after miR mimic or inhibitor transfection. 24 hours after plasmid vector transfection, the luciferase reporter assay was performed using a Dual-Glo luciferase assay kit (Promega). The luminescence intensity of Firefly luciferase was normalized to that of Renilla luciferase.

Statistical Analysis

Results of experiments were displayed as mean ± standard deviation in the figures. To evaluate statistical significances, Student’s unpaired t test was used, otherwise noted.


MiR Microarrays Define Novel EAC-specific Candidate Tumor-suppressive miRs and Oncogenic miRs

As a first approach to study the involvement of miRs in BE-associated neoplasms, we performed miR microarray investigations of non-immortalized primary normal esophageal (NE) epithelial cells (HEEpiC), BE-derived cell lines (QhTRT, ChTRT and GihTRT), and an EAC-derived cell line (OE-33). Representative results of these experiments are displayed in Table 1. Several miRs were aberrantly regulated in BE or EAC cells vs. NE cells. In particular, we found that miRs -25, -93 and -106b, all of which form a polycistron on chromosome 7q22.1 (the miR-106b-25 polycistron), were significantly upregulated in OE-33 cells relative to BE and HEEpiC cells, suggesting possible involvement of this polycistron as an oncogene in BE-associated neoplastic transformation.

Table 1
MiRs Up- or Downregulated in EAC Cells

Quantitative Reverse-Transcriptase PCR (qRT-PCR) Verifies miR Microarray Results

Next, we validated miR microarray results for 7 miRs that were differentially up- or downregulated in BE or EAC cells (miRs -25, -93, and -106b in Figures 1A–1C;miRs -19b, -100, -125b and -205 in Supplementary Figure S1 and bold in Table 1). Using real-time qRT-PCR, we analyzed the same EAC and BE, and NE cells that were assessed on miR microarrays. Results of these qRT-PCRs verified those obtained by miR microarray assays, as shown.

Figure 1
Quantitative RT-PCR of miR-106b-25 polycistron in esophageal cultured cells and tissues

Clinical Significance of Aberrant miR Expression

We also performed qRT-PCRs of these differentially expressed miRs (except miR-19b, Figures 1D–1F and Supplementary Figure S2) in primary human esophageal tissue specimens to address the clinical significance of their aberrant expression. In all cases examined, tissue miR expression results correlated closely with results obtained in cell lines. In particular, all three members of the miR-106b-25 polycistron were not significantly upregulated in BE relative to NE tissues (p = 0.14, 0.26 and 0.35 respectively), however, notably, these miRs were markedly upregulated in EAC relative to NE and BE. These results implied progressive involvement of the miR-106b-25 polycistron in BE-EAC neoplastic evolution.

Regulatory Mechanisms Underlying miR-25, -93 and -106b Activation

Since the miR-106b-25 polycistron is located within intron 13 of the minichromosome maintenance protein 7 (MCM7) gene on chromosome 7q22.1, we hypothesized that activation of these miRs should correlate with MCM7 “mother gene” activation. To investigate this hypothesis, we first examined MCM7 mRNA levels in the same cell lines using expression microarrays and qRT-PCR (Figure 2A). Indeed, we found that MCM7 mRNA expression was upregulated in BE-derived vs. HEEpiC NE cells; and notably, MCM7 was more markedly upregulated in OE-33 EAC cells than in BE-derived cells, reflecting the same stage-related activation pattern seen with the miR-106b-25 polycistron. Analogous MCM7 mRNA expression results were obtained in tissue specimens (Figure 2B). Secondly, we performed quantitative genomic DNA PCR of the MCM7 mother gene to evaluate whether DNA copy number variation (CNV) constituted a mechanism underlying its activation. Interestingly, this experiment revealed a progressive increase in DNA copy number at the MCM7 locus during neoplastic evolution, proceeding from NE-derived, to BE-derived, and finally to EAC-derived cells (Figure 2C). Similarly, we observed significant DNA copy number increases at the MCM7 locus in primary esophageal tissue specimens (NE vs. BE: p=0.0022; NE vs. EAC: p=0.00013, Figure 2D). We also discovered significant de novo CNV of the MCM7 locus in paired NE and EAC tissue specimens from 11 patients (p=0.0037 by paired t test in Figure 2E). Figure 3 showed a close relationship between miRs -25, -93 and -106b expressions and CNV at the MCM7 locus in tissue specimens (Pearson’s correlation coefficient test; p=0.0014, 0.011 and 0.019 respectively).

Figure 2
MCM7 mRNA expression and CNV in esophageal cultured cells and tissues
Figure 3
Expression levels of miRs -25, -93 and -106b correlated significantly with CNV levels in primary esophageal tissues (Pearson’s correlation coefficient test, p = 0.0014, 0.011 and 0.019, respectively). The average value of NE was set at 1 for both ...

Biological Impact of miR-25, -93 and -106b

In Vitro Effects

To elucidate cancer-relevant biological effects of miRs -25, -93 and -106b, we first performed cell number assays using WST-1 reagent. Inhibition of miRs -25, -93 and -106b induced significantly decreased numbers of OE-33 EAC cells (Figure 4A). We also performed miR-25-, -93- and -106b-mimic transfection of a BE-derived cell line (ChTRT), as well as of WI-38 normal lung fibroblasts. WI-38 cells were used as representative non-neoplastic cells because HEEpiC cells were not adequate for these experiments, due to their resistance to nucleic acid transfection and slow growth (data not shown). We also chose WI-38 cells because their expression levels of miRs -25, -93, and -106b were as low as those in HEEpiC cells (data not shown). Interestingly, miR mimic transfection induced significantly increased numbers of both BE and WI-38 cells, although the three miR mimics exerted this effect to differing degrees (Figures 4B and and3C).3C). Ki-67 staining showed that the number of Ki-67 positive cells was diminished in OE-33 cells transfected with miR-25-INH, miR-93-INH or miR-106b-INH relative to the number in cells transfected with NSC-INH (Supplementary Figure S7). Next, we performed experiments to assess the effects of these miRs on cell cycle progression. Inhibitors (INHs) of miRs -93 and -106b increased slightly but significantly the proportion of cells in G1 phase (nonspecific control inhibitor, NSC-INH: 51.46%, miR-93-INH: 53.38%, p=0.0028, and miR-106b-INH: 55.14%, p=0.032) in OE-33 cells. Similar effects of miRs-93 and -106b were observed with miR-93-MM and miR-106b-MM in ChTRT, GihTRT and QhTRT cells (data not shown). Next, to gauge the effects of these miRs on programmed cell death, we performed apoptosis assays using miR inhibitors. In these assays, miR-25-INH slightly but significantly augmented the proportion of cells in early apoptotic phase (NSC-INH: 5.00%; miR-25-INH: 5.76%, p=0.011).

Figure 4Figure 4
Cell Proliferation Assessed by WST-1 and in vivo Tumorigenesis Assays

In Vivo Tumorigenesis

Our discovery that the miR-106b-25 polycistron induced increased cell numbers in vitro prompted us to evaluate its effects in vivo. As shown in Figure 4D, inhibitors of miRs -25, -93 and -106b reduced in vivo tumorigenesis in nude mice. For all three miR inhibitors, the most marked inhibitory effects were observed at 5 days post-transfection (Day 4 in Figure 4D).

The miR-106b-25 Polycistron Directly Regulates p21 and Bim

We next focused our attention on mRNA targets of miRs -25, -93 and -106b. To identify these targets, we employed two principal strategies: 1) focusing candidate mRNA searches on results of our phenotypic experiments, e.g., cell cycle- and apoptosis-related genes; 2) performing database searches in miR target prediction engines, such as miRBase (, TargetScan ( and PicTar ( In our cell cycle experiments, miRs -93 and -106b exerted internally consistent effects on the G1/S checkpoint: Inhibitors of miRs -93 and -106b both increased the proportion of cells in G1 phase (i.e., induced G1 arrest) in OE-33 EAC cells. This finding prompted us to search for G1/S-regulatory proteins, and in combination with 3′UTR homology searches, we ultimately identified a CDK inhibitor and known tumor suppressor gene, CDKN1A (p21Cip1/WAF1)16. Western blotting revealed that p21 protein expression was sharply reduced by miR-93 and miR-106b mimics, and correspondingly, that p21 protein expression was sharply increased by miR-93 and miR-106b inhibitors (Figures 5A and and5B).5B). To evaluate direct binding to and silencing of p21 by miRs -25, -93, and -106b, we then performed luciferase assays of the p21 3′UTR. Luciferase reporter activity was significantly reduced by miRs -93 and -106b (p= 0.0015 and 0.0049 for miR-93 and p= 0.00078 and 0.00087 for miR-106b in Figures 5C and 5D, respectively) but not by miR-25 (Supplementary Fugures S6A and S6B), suggesting that miRs -93 and -106b bind directly to the p21 3′UTR and inhibit p21 protein expression.

Figure 5
p21 Western Blotting and Luciferase Assays

Similarly, in our apoptosis experiments, miR inhibitors increased the population of apoptotic cells, thus, in combination with 3′UTR homology searches, we identified the pro-apoptotic tumor suppressor gene, BCL2L11 (BCL2-like 11 or Bim)17. Bim protein exists in two major isoforms: Bim EL and Bim L, and both of them have the same 3′UTR sequence. Western blotting revealed that the expression of both Bim isoforms was reduced by a miR-25 mimic and increased by a miR-25 inhibitor (Figures 6A and 6B). Luciferase reporter activity was significantly reduced by miR-25 (p= 0.0012 and 0.0096 in Figures 6C and 6D, respectively) but not by miRs -93 and -106b (Supplementary Fugures S6C and S6D).

Figure 6
Bim Western Blotting and Luciferase Assays

Regulatory Effects of the miR-106b-25 Polycistron on p21 and Bim

Next, to investigate whether suppression of p21 and Bim protein expression was due to mRNA degradation, we performed qRT-PCRs of p21 and Bim mRNA after transfecting mimics of miRs -25, -93, and -106b. An siRNA directed against p21 served as a positive control. qRT-PCRs showed that p21 mRNA expression was diminished approximately 30% by the miR-93 mimic and 40% by the miR-106b mimic, vs. 94% by the control p21-siRNA (Figure 7A). Bim mRNA expression did not change following transfection of a miR-25 mimic (Figure 7B).

Figure 7
p21 and Bim mRNA Expression Following miR Mimic Transfection


Although the first microRNA, lin-4, was discovered in 199318 and the second miR, let-7, in 200019 in Caenorhabditis elegans, only recently has the study of this class of small regulatory RNAs in humans become more widespread20. Experimental evidence that has accumulated in recent years has led oncologists to speculate that unrevealed molecular actors, particularly non-coding RNAs previously been classified as “junk,” play important roles in carcinogenesis. In fact, approximately 70% of the genome has been found to be transcribed in mammals21. Thus, it is a natural extension of this logic to explore novel noncoding RNA-based mechanisms underlying cancer-related gene dysregulation, even mechanisms unanticipated prior to the discovery of lin-4 and let-7. MicroRNAs (miRs) are now intensively studied in cancer research, and these molecules are predicted to control as much as 30% of all gene expression22.

To our knowledge, the current findings constitute the first report documenting (1) activation of the miR-106b-25 polycistron by DNA copy number variation (CNV); (2) miR biologic effects in esophageal cells; and (3) in vivo tumorigenic properties of the miR-106b-25 polycistron.

Several of the miRs listed in Table 1 have been previously implicated in carcinogenesis. Among miRs upregulated in the current study, miR-200a in ovarian cancer23 and miR-15b in gastric cancer24 have been reported as potential oncomiRs. Among miRs that were downregulated in the current study, members of the let-7 family have been implicated in lung cancer25, as have members of the miR-34 family in colorectal cancer26 and of the miR-125 family in prostate cancer27. The miR-17-5p–92 and 106a–92 polycistrons have also been documented to function as oncomiRs and are highly homologous to each other, as well as to the miR-106b-25 polycistron12. Surprisingly, members of the miR-17-5p–92 and 106a–92 polycistrons (viz.,miRs-17-5p, -19b, -20a, -20b, -92 and -106a in Table 1) were downregulated in EACs in the current study, suggesting the possibility that the miR-106b-25 polycistron, with its overexpression and high homology to these other polycistrons, may actually compensate for a “perceived” deficit in oncogenic activity of the (downregulated) miR-17-5p–92 and 106a–92 polycistrons.

Among BE tissue specimens in the current study, there were 12 BE metaplasias, 4 low-grade dysplasias (LGDs), and 8 high-grade dysplasias (HGDs; Supplementary Table S1). Expression levels of miRs -25, -93 and -106b did not differ significantly between NE and any histologic grade of BE. Moreover, all clinical characteristics, including prognosis (not shown), did not correlate with expression levels of miRs -25, -93, or -106b.

We found that expression of miRs -25, -93, -106b, and MCM7 mRNA in NE, BE and EAC correlated closely in both cell lines and tissue specimens (Figures 1, ,2A2A and and2B);2B); however, relative levels of these three miRs differed from those of MCM7 mRNA, particularly in EAC cells and tissues. Specifically,in EAC, the fold-change of MCM7 mRNA expression was higher than that of the miR-106b-25 polycistron (Figures 1A–1C vs.vs.2A,2A, and Figures 1D–1Fvs. 2B2B ). This contrasting result suggests that unique regulatory mechanisms may also underlie miR-106b-25 polycistron vs.MCM7 mRNA expression in EAC.

One novel finding was that CNV of MCM7 locus correlated closely with miR-106b-25 polycistron expression, not only in NE, BE and EAC cell lines (Figures 1A–1C vs. 2C),2C), but also in tissues (Figures 1D–1F vs. 2D2D and and3).3). CNV of MCM7 in the three metaplastic BE-derived cell lines (QhTRT, ChTRT and GihTRT) averaged 1.4-fold higher than in HEEpiC NE cells, while CNV of MCM7 in OE-33 EAC-derived cells was much higher (5.8-fold; Figure 2D). These findings suggest that cumulative CNV occurs during BE neoplastic transformation, possibly at the transition between HGD and EAC. We also observed a close correlation between miR-106b-25 polycistron expression and CNV in matched RNA-DNA pairs from the same individuals (p=0.0014, 0.011 and 0.019, respectively; Figure 3). These findings also support the conclusion that the miR-106b-25 polycistron is activated by genomic copy number variation at least in a proportion of cases. To our knowledge, this is the first direct evidence of a relationship between miR-106b-25 polycistron activation and CNV during neoplastic evolution of any human cancer. There have been several articles describing concomitant de novo genomic amplification, a type of CNV, and miR overexpression28, 29. Recent reports have noted that although most CNV is inherited as a polymorphic trait30, de novo CNV can also occur31. Figure 2E shows that among our paired normal-tumor DNA samples, most manifested de novo CNV, i.e., somatic DNA amplification, at the MCM7 locus. Although there were only two paired BE-EAC samples in the current study, we observed the same trend in de novo CNV for MCM7 in these two paired BEs and EACs (Relative amplification: 1.293 to 1.504 and 1.159 to 1.177). Further studies are needed to ascertain whether inherited polymorphic CNV occurs in normal subjects, since only normal tissues from EAC patients were included in the current study. Taken together, these results strongly imply that genomic DNA copy number variation at the MCM7 locus was at least partially responsible for coordinate activation of both the miR-106b-25 polycistron and of the MCM7 gene, and that this CNV increased progressively during BE-associated neoplastic evolution.

We also studied cancer-related biologic effects of the miR-106b-25 polycistron. All three members of this polycistron induced increased cell numbers in normal esophageal, BE-derived and EAC-derived cell lines. These three miRs exhibited unique degrees of cell type-specific targeting in their cell number-increasing effects (Figure 4): i.e., miRs-93 and -106b induced greater increases in cell numbers in normal fibroblasts than did miR-25, while miR-25 induced greater increases in cell numbers in BE-derived cells, and miR-93 was the most but miR-25 the least influential in EAC-derived cells. Thus, which member of the miR-106b-25 polycistron most markedly affected cell numbers depended on cell type. The miR-25 mimic effect rapidly decreased from days 3 to 5 only in WI-38, but not in ChTRT, QhTRT and GihTRT cells (QhTRT and GihTRT not shown in Figure 4). In contrast, the effects of miRs -93 and -106b persisted at least until day 5 in all cell lines examined. We were unable to determine exactly what caused this unique difference in effects; however, we speculate that WI-38 may have degraded or inactivated miR-25 more rapidly than did the other cell lines we examined. One notable feature of oncomiR polycistrons, including miR-106b-25, is that coordinated activation involving the entire polycistron may enhance its overall oncogenic effect. Moreover, MCM7 itself is oncogenic in prostate cancer32. Thus, DNA copy number variation occurring at the MCM7 locus may enhance miR-derived oncogenesis by activating MCM7 simultaneously with the miR-106b-25 polycistron.

OE-33 EAC cells grow slowly in athymic nude mice and are mildly rejected during the course of several weeks (Supplementary Figure S3) with peak tumor size usually occurring within 7 days and maximal tumor volume reaching only 40 mm3 on average, a size at which experimental effects are difficult to discern. We overcame this difficulty by introducing Matrigel reagent, which accelerated cell growth and appeared to protect these cells from their potentially hostile in vivo environment. According to qRT-PCR measurement of miRs -25, -93 and -106b following miR mimic transfections, levels of these miRs remained quite high (24.0-, 23.6-, and 40.0-fold change, respectively) at 6 days post-transfection (Supplementary Figure S4), indicating that these transfected small RNAs continued to be biologically effective for at least 6 days post-transfection. As shown in Figure 4D, antagonists of miRs -25, -93 and -106b exerted inhibitory effects on in vivo tumorigenesis in nude mice, further supporting the in vitro cell number-increasing effects of these miRs.

Negative regulation of miR target mRNAs may be due to either protein translation inhibition or to mRNA degradation. Figures 7A show that miRs -93 and -106b caused at least some degree of p21 mRNA instability (though not as much as that caused by p21 siRNA), while miR-25 did not affect Bim mRNA levels (Figure 7B). These findings were further corroborated by Ivanovska et al.33, who found that miR-106b reduced p21 mRNA by approximately 40% in human mammary epithelial cells. However, our data only partially agrees with a recent study of gastric cancers, which showed no apparent effects of miRs -25, -93 and -106b on RNA stability34. Taken together, our findings suggest that these two mRNA targets are regulated by discrete miR effector mechanisms: 1) for p21, mRNA degradation; 2) for Bim, translational inhibition.

Identifying miR target mRNAs poses a challenge to investigators. Target prediction engines often yield hundreds of potential candidate mRNAs, compelling investigators to struggle with the choice of which targets to focus on. One promising strategy to overcome this challenge is to compare global miR and mRNA expression profiles in the same starting materials35. Judging from our own results (Figures 7A and 7B), p21 would have been detected by such a comparison, while Bim would not. Such global comparisons are useful only when miRs effect a decrease in their target mRNA expression levels; moreover, this strategy may also unearth mRNAs whose stability or expression are not directly targeted by particular miRs. Thus, more efficient and innovative approaches to identify mRNA targets of miRs are still needed.

In conclusion, the miR-106b-25 polycistron exerted potential proliferative, anti-apoptotic, cell cycle-promoting effects in vitro, tumorigenic properties in vivo and was progressively activated progressively at successive stages of neoplasia from NE to BE and finally to EAC, potentially due to de novo genomic DNA copy number variation of the MCM7 locus at chromosome 7q22.1. The miR-106b-25 polycistron thus represents a potential oncogene in BE-EAC carcinogenic evolution, as well as in other human cancers, and may act via regulation of two specific target genes, p21 and Bim.

Supplementary Material


Grant Support: This work was supported partially by NIH awards CA85069 and CA01808.


Barrett’s esophagus
copy number variation
esophageal adenocarcinoma
normal epithelium


Financial Disclosures: The authors declare that they have no competing financial interests.

Conflicts of interests: The authors declare that they have no potential investigator conflicts of interest.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Shaheen NJ. Advances in Barrett’s esophagus and esophageal adenocarcinoma. Gastroenterology. 2005;128:1554–66. [PubMed]
2. Cameron AJ, Lomboy CT. Barrett’s esophagus: age, prevalence, and extent of columnar epithelium. Gastroenterology. 1992;103:1241–5. [PubMed]
3. Hirota WK, Loughney TM, Lazas DJ, et al. Specialized intestinal metaplasia, dysplasia, and cancer of the esophagus and esophagogastric junction: prevalence and clinical data. Gastroenterology. 1999;116:277–85. [PubMed]
4. Ronkainen J, Aro P, Storskrubb T, et al. Prevalence of Barrett’s esophagus in the general population: an endoscopic study. Gastroenterology. 2005;129:1825–31. [PubMed]
5. Sampliner RE. A population prevalence of Barrett’s esophagus--finally. Gastroenterology. 2005;129:2101–3. [PubMed]
6. van Soest EM, Dieleman JP, Siersema PD, et al. Increasing incidence of Barrett’s oesophagus in the general population. Gut. 2005;54:1062–6. [PMC free article] [PubMed]
7. Jankowski JA, Wright NA, Meltzer SJ, et al. Molecular evolution of the metaplasia-dysplasia-adenocarcinoma sequence in the esophagus. Am J Pathol. 1999;154:965–73. [PubMed]
8. Selaru FM, Zou T, Xu Y, et al. Global gene expression profiling in Barrett’s esophagus and esophageal cancer: a comparative analysis using cDNA microarrays. Oncogene. 2002;21:475–8. [PubMed]
9. Kimos MC, Wang S, Borkowski A, et al. Esophagin and proliferating cell nuclear antigen (PCNA) are biomarkers of human esophageal neoplastic progression. Int J Cancer. 2004;111:415–7. [PubMed]
10. Schulmann K, Sterian A, Berki A, et al. Inactivation of p16, RUNX3, and HPP1 occurs early in Barrett’s-associated neoplastic progression and predicts progression risk. Oncogene. 2005;24:4138–48. [PubMed]
11. Akao Y, Nakagawa Y, Naoe T. let-7 microRNA functions as a potential growth suppressor in human colon cancer cells. Biol Pharm Bull. 2006;29:903–6. [PubMed]
12. O’Donnell KA, Wentzel EA, Zeller KI, et al. c-Myc-regulated microRNAs modulate E2F1 expression. Nature. 2005;435:839–43. [PubMed]
13. Feber A, Xi L, Luketich JD, et al. MicroRNA expression profiles of esophageal cancer. J Thorac Cardiovasc Surg. 2008;135:255–60. discussion 260. [PMC free article] [PubMed]
14. Palanca-Wessels MC, Klingelhutz A, Reid BJ, et al. Extended lifespan of Barrett’s esophagus epithelium transduced with the human telomerase catalytic subunit: a useful in vitro model. Carcinogenesis. 2003;24:1183–90. [PMC free article] [PubMed]
15. Schmid CW. Alu: structure, origin, evolution, significance and function of one-tenth of human DNA. Prog Nucleic Acid Res Mol Biol. 1996;53:283–319. [PubMed]
16. Gartel AL, Radhakrishnan SK. Lost in transcription: p21 repression, mechanisms, and consequences. Cancer Res. 2005;65:3980–5. [PubMed]
17. O’Connor L, Strasser A, O’Reilly LA, et al. Bim: a novel member of the Bcl-2 family that promotes apoptosis. Embo J. 1998;17:384–95. [PubMed]
18. Lee RC, Feinbaum RL, Ambros V. The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell. 1993;75:843–54. [PubMed]
19. Reinhart BJ, Slack FJ, Basson M, et al. The 21-nucleotide let-7 RNA regulates developmental timing in Caenorhabditis elegans. Nature. 2000;403:901–6. [PubMed]
20. Ruvkun G, Wightman B, Ha I. The 20 years it took to recognize the importance of tiny RNAs. Cell. 2004;116:S93–6. 2 p following S96. [PubMed]
21. Carninci P, Kasukawa T, Katayama S, et al. The transcriptional landscape of the mammalian genome. Science. 2005;309:1559–63. [PubMed]
22. Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell. 2005;120:15–20. [PubMed]
23. Iorio MV, Visone R, Di Leva G, et al. MicroRNA signatures in human ovarian cancer. Cancer Res. 2007;67:8699–707. [PubMed]
24. Xia L, Zhang D, Du R, et al. miR-15b and miR-16 modulate multidrug resistance by targeting BCL2 in human gastric cancer cells. Int J Cancer. 2008;123:372–9. [PubMed]
25. Lee YS, Dutta A. The tumor suppressor microRNA let-7 represses the HMGA2 oncogene. Genes Dev. 2007;21:1025–30. [PubMed]
26. Toyota M, Suzuki H, Sasaki Y, et al. Epigenetic silencing of microRNA-34b/c and B-cell translocation gene 4 is associated with CpG island methylation in colorectal cancer. Cancer Res. 2008;68:4123–32. [PubMed]
27. Ozen M, Creighton CJ, Ozdemir M, et al. Widespread deregulation of microRNA expression in human prostate cancer. Oncogene. 2008;27:1788–93. [PubMed]
28. Ota A, Tagawa H, Karnan S, et al. Identification and characterization of a novel gene, C13orf25, as a target for 13q31–q32 amplification in malignant lymphoma. Cancer Res. 2004;64:3087–95. [PubMed]
29. He L, Thomson JM, Hemann MT, et al. A microRNA polycistron as a potential human oncogene. Nature. 2005;435:828–33. [PubMed]
30. Redon R, Ishikawa S, Fitch KR, et al. Global variation in copy number in the human genome. Nature. 2006;444:444–54. [PMC free article] [PubMed]
31. Egan CM, Sridhar S, Wigler M, et al. Recurrent DNA copy number variation in the laboratory mouse. Nat Genet. 2007;39:1384–9. [PubMed]
32. Ren B, Yu G, Tseng GC, et al. MCM7 amplification and overexpression are associated with prostate cancer progression. Oncogene. 2006;25:1090–8. [PubMed]
33. Ivanovska I, Ball AS, Diaz RL, et al. MicroRNAs in the miR-106b family regulate p21/CDKN1A and promote cell cycle progression. Mol Cell Biol. 2008;28:2167–74. [PMC free article] [PubMed]
34. Petrocca F, Visone R, Onelli MR, et al. E2F1-regulated microRNAs impair TGFbeta-dependent cell-cycle arrest and apoptosis in gastric cancer. Cancer Cell. 2008;13:272–86. [PubMed]
35. Lim LP, Lau NC, Garrett-Engele P, et al. Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature. 2005;433:769–73. [PubMed]