let-7 miRNAs Regulate let-60/RAS in C. elegans During Development
The epidermis in C. elegans is made of three main cell types: seam, syncytial, and P cells. Temporal up-regulation of let-7 miRNA in the seam cells  is required for terminal differentiation at the adult stage , which is when these cells exit the cell cycle, fuse together, and excrete cuticular alae (the longitudinal external ridges on the sides of each animal) . To understand how the let-7 gene family affects cell differentiation, Dr. Frank Slack and his colleagues used a computational approach to identify eleven development genes that contained at least one let-7 complementary sequence (LCS) [1, 5]. Among the identified genes, let-60/RAS stood out because it has eight LCSs and ten nonconforming sites in the 3' untranslated region (3'UTR), as well as three LCSs in the coding sequence. The role of let-60/RAS in C. elegans vulval development is well characterized , and expression studies showed that temporal regulation of let-60/RAS by let-7 miRNA at least partially accounts for the role of let-7 in seam and vulval cell differentiation . Using genetic approaches and reporter constructs with the let-60/RAS 3'UTR, Johnson et al. also showed that the expression of let-60/RAS is regulated by two let-7 miRNA family members (let-7 and miR-84) via the 3'UTR of the let-60/RAS gene.
Human let-7 and RAS
let-60/RAS is the C. elegans orthologue of the human RAS genes, HRAS, KRAS, and NRAS. All three human RAS 3'UTRs contain multiple putative let-7 complementary sites with features of validated C. elegans LCSs. Many of the LCSs are conserved in rodents, amphibians, and fish, suggesting functional relevance. The presence of putative LCSs in human RAS 3'UTRs suggests that mammalian let-7 family members may regulate human RAS in a manner similar to the way let-7 and miR-84 regulate let-60/RAS in C. elegans.
let-7 regulates RAS expression in human cells
To confirm that let-7 regulates the expression of the human RAS genes, cultured cells were transfected with let-7 miRNA precursors and inhibitors, and RAS protein levels were monitored. HepG2 cells, which do not express let-7 at levels detectable by microarray analysis, were transfected with a let-7a Pre-miR miRNA Precursor Molecule using siPORT NeoFX Transfection Agent. The let-7a Pre-miR miRNA Precursor enters the miRNA pathway and modifies the expression of genes that are regulated by the endogenous miRNA. In three separate experiments, increasing let-7 miRNA levels decreased RAS expression by ~70% compared to cells transfected with a negative control (Pre-miR Negative Control #1 miRNA) (Figures 1AB), while GAPDH and p21 protein levels did not change (data not shown).
Figure 1. let-7 Influences the Expression of RAS in Human Cells. HepG2 cells were transfected in 24 well plates using siPORT NeoFX (Ambion) with either 30 nM let-7a Pre-miR miRNA Precursor Molecule (Ambion) or a negative control miRNA inhibitor (AB). HeLa cells were transfected in 24 well plates using Lipofectamine 2000 (Invitrogen) with either 30 nM let-7a Anti-miR miRNA Inhibitors (Ambion) or a negative control miRNA inhibitor (CD). Three days post-transfectio n, RAS expression was monitored by immunofluorescence (A, C) and quantified using MetaMorph software (Universal Imaging corporation) (B, D; triplicate assays). Modified with permission from Cell 120: 635647.
In reciprocal experiments, HeLa cells, which express high levels of let-7a, were transfected with a synthetic, antisense RNA to let-7a (let-7a Anti-miR miRNA Inhibitor), and the prediction that inhibiting let-7 activity may de-repress RAS expression held true (Figures 1CD). RAS protein levels were ~70% higher in cells transfected with the let-7a inhibitor compared to cells transfected with a negative control. Together, these results suggest that let-7 negatively regulates the expression of RAS in human cells.
let-7 regulates the Human RAS genes via their 3'UTRs
Portions of the NRAS 3'UTR (3.5 kb containing nine LCSs) and the KRAS 3'UTR (1 kb containing seven LCSs) were individually subcloned downstream of a luciferase reporter. When transfected into HeLa cells, the reporter constructs with the RAS 3'UTRs expressed 4- to 8-fold less luciferase activity than the control reporter construct with no insert (Figure 2A).
Figure 2. let-7 Regulates NRAS and KRAS Through Their 3' UTRs. The NRAS 3'UTR (3.5 kb containing nine LCSs) and the KRAS 3'UTR (1 kb containing seven LCSs) were subcloned downstream of t he luciferase reporter gene. The control plasmid did not contain any insert. HeLa cells were transfected in 12 well plates using Lipofectamine 2000 (Invitrogen) with the plasmids described. 24 h post-transfection, cells were harvested and assayed using the Dual-Luciferase assay (Promega). (A) Fold induction of reporter gene activity was expressed relative to the negative control vector. (B) To show specificity of the interaction between let-7 and RAS 3' UTRs, similar experiments were performed with an additional cotransfection of let-7 miRNA inhibitors (let-7a Anti-miR miRNA Inhibitors [Ambion] or a negative control miRNA inhibitor [Ambion]). Modified with permission from Cell 120: 635647.
To verify that this result is related to let-7 expression in HeLa cells, the luciferase reporter vectors were co-transfected with the let-7 miRNA inhibitor or a negative control miRNA inhibitor. As seen in Figure 2B, decreasing let-7 activity in HeLa cells de-repressed expression of the reporter gene. In these assays, luciferase activity increased by 2- to 2.5-fold compared to cells transfected with the negative control inhibitor. Along with the identification of several potential LCSs in the RAS 3'UTR, the data presented in Figure 2 indicate that let-7 miRNA can mediate RAS expression through the RAS 3'UTR.
let-7 in Human Lung Cancer
In a related research project, Ambion's miRNA microarray technology (167 miRNA prob es, see Highly Sensitive microRNA Array Performance) was used to assess let-7 expression levels in tissue from 21 different cancer patients, including 12 lung cancer patients with stage IB or IIA squamous cell carcinoma. Interestingly, the lung tumor samples had more than 50% reduction in levels of let-7 miRNA relative to the normal adjacent tissues from the same patients (Figure 3A). Only sporadic reduction in let-7 was detected in breast and colon cancer samples. A similar finding was independently discovered by Takamizawa and colleagues . In addition, several human let-7 family members have been mapped to chromosomal intervals that are deleted in lung cancers , providing a possible explanation for the reduced let-7 expression that we observed.
let-7 miRNA, RAS mRNA, and RAS Protein in Lung Cancer
Misexpression or mutation of RAS (HRAS, KRAS, and NRAS) is associated with human cancer . The observations that (1) RAS is an oncogene, (2) let-7 is down-regulated in lung tumors, and (3) RAS expression is regulated by let-7 miRNA all suggest that reduced let-7 in lung tissue could lead to over-expression of RAS and increased cell proliferation. A prediction from this model is that let-7 miRNA and RAS protein expression should be inversely proportional in lung cancer samples.
To test this hypothesis, both RNA and protein were isolated from three pairs of lung squamous cell carcinoma and NAT using the mirVana PARIS Kit, so that RAS protein and RNA levels and let-7 miRNA levels could be measured in the same samples. In all cases, RAS protein levels in tumor cells were at least ten times higher than in the corresponding NAT (Figure 3C, top), and let-7 miRNA levels in the tumor were 4- to 8-fold lower than in the corresponding NAT (Figure 3C, middle). This result indicates that RAS protein expression correlates better with miRNA expression levels than with RAS mRNA in lung.
Figure 3. let-7 is Poorly Expressed in Human Lung Tumors. (A) The mirVana miRNA Isolation Kit (Ambion) was used to isolate miRNA from 21 breast, colon, and lung tumors, as well as from associated normal adjacent tissue (NAT). Microarray analysis using 167 miRNA probes (mirVana miRNA Probe Set, Ambion) was used to examine let-7 miRNA expression profiles (tumor relative to NAT) in these samples. After labeling tumor miRNA samples with Cy3 (Amersham) and NAT miRNA samples with Cy5 (mirVana miRNA Labeling Kit, Ambion), the microarray was hybridized for 14 h. Scanning was performed with the GenePix 4000B (Axon) (B) Northern analysis was performed using paired RNA preparations from two of the patients from Figure 3A and a radiolabeled probe specific for let-7c and 5S rRNA (control). (C) The mirVana PARIS Kit (Ambion) was used to isolate protein and RNA fr om three additional pairs of tissues from lung cancer patients. Western blots were performed using antibodies specific for RAS (US Biological) or GAPDH (Ambion) (top). Northern blot analysis was performed using radiolabeled probes for let-7 and U6 snRNA (middle). NRAS and -actin mRNA, as well as 18S rRNA (for normalization) were quantified by real-time RT-PCR (bottom). Modified with permission from Cell 120: 635647.
This study suggests that let-7 miRNA might function as a tumor suppressor in lung. Misregulation of let-7 might contribute to oncogenesis by triggering up-regulation of RAS protein expression. Analysis of miRNA expression profiles in clinical samples, combined with rapid and powerful loss and gain of function studies in cellular models, will undoubtedly reveal new critical biological functions for members of the miRNA family.
Jaclyn Shingara, Mike Byrom, Rich Jarvis, Angie Cheng, Emmanuel Labourier, Kerri Keiger, Brian Cannon, Vince Pallotta, Sean Banks, David Brown Ambion, Inc.