Figure 1. miRNA Expression in Lung and Colon Cancer Patients. The miRNA expression profiles of tumor vs normal adjacent tissues were compared for lung (A) and colon (B) cancer patients. The miRNAs are provided in rows; the patients are presented in columns. Green in the heat map shows miRNAs that are down-regulated in the tumor sample relative to the normal adjacent tissue sample, and red shows miRNAs that are up-regulated in the tumor sample relative to the normal adjacent tissue sample.
miRNA Expression in Cancer Samples
miRNAs are small, siRNA-like molecules encoded in the genomes of animals and plants that regulate gene expression through sequence-specific interactions with target mRNAs. More than 200 miRNAs have been identified in humans, and many of the miRNA genes are located at chromosomal fragile sites associated with cancer . Studies of miRNA expression profiles in cancer samples have identified a handful of miRNAs that are differentially regulated in tumors, suggesting a possible link between miRNAs and oncogenesis [2-4]. We used the mirVana miRNA microarray platform (see, mirVana Microarray Platform for Sensitive miRNA Array Analysis, sidebar below) to compare the miRNA expression profiles of tumor and normal adjacent tissues from more than sixty patients with lung, colon, breast, bladder, pancreatic, prostate, or thymus cancer.
miRNA Expression in Lung Cancer
Twenty-two tumor and normal adjacent tissue (NAT) samples from lung cancer patients were analyzed using mirVana miRNA array platform. The arrays were analyzed, and the relative expression of each miRNA was compared between the tumor and normal adjacent tissues from each patient. The various miRNAs were clustered based on their relative expression in tumors across different patients (Figure 1A). Six miRNAs (miR-126, -30a, -143, -145, -188, and -331) were expressed at significantly lower levels in the tumors of more than 70% of the patients. Three miRNAs (miR-21, -189, and -200b) were expressed at significantly higher levels in the tumors of more than 70% of the patients. The differential expression of a number of these miRNAs was verified by Northern analysis (Figure 2).
Figure 2. Validation of miRNA Array Expression Results in Lung Cancer Patients. Total RNA samples from two lung cancer patients were analyzed for expression of miR-16, miR-21, miR-143, miR-145, and let-7 using Northern analysis. The graphs show the relative abundance of each miRNA (ratio of tumor:NAT) from the array analysis and Northern phosphoimager analysis.
miRNA Expression in Colon Cancer
Twenty-five tumor and NAT samples from colon cancer patients were analyzed using the miRNA array platform. Like the lung cancer comparisons, the various miRNAs were clustered based on their relative expression in tumors across the different colon cancer patients (Figure 1B). Five miRNAs (miR-143, -145, -195, -130a, and -331) were expressed at significantly lower levels in the tumors of more than 70% of the patients. Four miRNAs (miR-223, -21, -17, and -106) were expressed at significantly higher levels in the tumors of more than 70% of the patients.
miRNAs as Cancer Markers
It is interesting that eight different miRNAs were differentially expressed between the tumor and normal adjacent samples for most of the lung and colon patient samples that we analyzed (Figure 3). These same miRNAs were also found to be differentially expressed in the breast, thymus, bladder, pancreatic, and prostate cancer patients that we analyzed, suggesting that these miRNAs might control cellular processes that when altered lead to cancer.
Figure 3. Differentially Expressed miRNAs in Multiple Cancer Types. miRNA array analysis comparing tumor and normal adjacent tissues from patients with various types of cancer was used to identify miRNAs that are differentially expressed in cancer. The percentage of patients exhibiting up- or down-regulation of a given miRNA was calculated for each cancer type. The eight that were most often differentially expressed across sample types are presented.
miRNAs as Regulators of Oncogene Expression
To address whether specific miRNAs might be participating in cancer through the mis-regulation of oncogenes, we scanned the 3' untranslated regions (UTRs) of 150 well-known oncogenes for sequences with significant homology to the miRNAs identified in our microarray analysis. We selected potential target sites based on two criteria:
(1) Perfect complementarity between positions 2-9 of the miRNA and the oncogene. This miRNA core sequence has been identified as critical to the activities of miRNAs and the known miRNA target sites have essentially 100% complementarity at this site .
(2) Overall Tm of the miRNA/mRNA interaction. In addition to the core sequence, overall binding stability between miRNAs and mRNAs has been shown to be an important indicator of miRNA activity .
As seen in Figure 4, potential target sites in the 3' UTRs of known oncogenes were identified for all of the miRNAs that were observed to be routinely differentially expressed in tumor samples. Interestingly, KRAS2, MYCL1, and CBL have multiple predicted miRNA binding sites which could provide the cooperative miRNA binding that has been implicated as an important factor in miRNA regulation [6, 7]. Many of the genes listed in Figure 4 become oncogenic when they are over-expressed, thus it is conceivable that reduced expression of a miRNA could lead to up-regulation of one or more oncogenes and subsequently lead to oncogenesis.
Figure 4. Cancer-related miRNAs and Their Putative Oncogene Targets.
Measuring the Effect of miRNAs on Oncogene Expression
Confirming miRNA target site predictions can be done in a variety of ways. In Drosophila and C. elegans, genetic approaches have been applied wherein mutations in the miRNA and the putative miRNA target site(s) are made and shown to result in similar phenotypes [8, 9]. In mammalian cells, where genetic approaches are far more difficult, reporter constructs have been used to show that the 3' UTRs of putative target genes are regulated in cells at levels that are disproportionate to reporter vector controls that contain mutations in the putative miRNA binding sites . In addition, vectors and oligonucleotides have been used to introduce or inhibit miRNAs in cells to determine the effects on endogenous levels of putative target genes [10, 11]. We have taken the latter approach to validate our miRNA target site predictions.
Ambion recently introduced a series of precursor miRNAs (Pre-miR miRNA Precursor Molecules, Cat# 17100) and miRNA inhibitors (Anti-miR miRNA Inhibitors, Cat# 17000) that can be transfected into mammalian cells to either introduce miRNAs into cells or inhibit the activity of miRNAs in cells, respectively . We used a Pre-miR miRNA Precursor Molecule and Anti-miR miRNA Inhibitor to one of the miRNAs that were found to be down-regulated in tumor samples to determine if our target site predictions were correct. In these experiments, cultured cells that express undetectable levels of the miRNA were transfected with the Pre-miR Precursor Molecule of the miRNA using siPORT NeoFX Transfection Agent (Ambion). Immunofluorescence assays were used to measure the two putative oncogene targets of the miRNA in the transfected cells. The proteins from both oncogenes were expressed at almost three-fold lower levels in cells transfected with the Pre-miR miRNA than in cells transfected with Pre-miR Negative Control #1 (Ambion) (Figure 5A). In a reciprocal experiment, cells that naturally express high levels of the miRNA were transfected with the Anti-miR miRNA Inhibitor for the miRNA. As expected, the proteins from both oncogenes were higher in cells transfected with the miRNA-specific Anti-miR than in cells transfected with the Anti-miR miRNA Negative Control #1 (Ambion) (Figure 5B). These results are consistent with the model that the miRNA regulates the expression of the two oncogenes.
Figure 5. Oncogene Levels Modulated in Cells Treated with Cancer-related miRNA Precursor and Inhibitor Molecules.A schematic of each experiment is shown at the left of each panel. (A) Tissue culture cells were transfected with a cancer-related or negative control Pre-miR miRNA Precursor Molecule. The protein expression of two oncogenes that were predicted to be regulated by the cancer-related miRNA was measured by immunofluorescence. Transfection of the cancer-related Pre-miR miRNA Precursor Molecule resulted in a greater than 2-fold decrease in the expression of the proteins from both oncogenes. (B) An Anti-miR miRNA Inhibitor targeting the same cancer-related miRNA was introduced into cells that express high levels of the miRNA. Immunofluorescence 48 hours after transfection revealed an approximately two-fold increase in the expression of both oncogenes.
miRNA expression studies in cancer patient samples suggest that at least a few miRNAs are key regulators of cellular processes that control unencumbered cell proliferation in humans. Based on target site prediction and validation of two targets of these miRNAs, it appears that misregulation of these key miRNAs could participate in cancer progression by failing to regulate the expression of one or more oncogenes.
David Brown, Jaclyn Shingara, Kerri Keiger, Jeffrey Shelton, Kathy Lew, Brian Cannon, Sean Banks, Steve Wowk, Mike Byrom, Angie Cheng, Xiaowei Wang, Emmanuel Labourier Ambion, Inc.
Anti-miR miRNA Inhibitor
Anti-miR miRNA Inhibitor
20 nmol (4 x 5 nmol)
Anti-miR miRNA Inhibitor
Pre-miR miRNA Precursor
Pre-miR miRNA Precursor
20 nmol (4 x 5 nmol)
Pre-miR miRNA Precursor Molecule