Figure 1.
MiRNA binding sites in Nf1 3′-UTR.
(A) Relative position of conserved miR-103, miR-137, miR-27, miR-128, miR-182 and miR-153 sites in Nf1 3′-UTR. (B) Sequence conservation of miR-103, miR-137 and miR-128 binding sites. (C) Alignment of mature miR-103, miR-137 and miR-128 sequences with Nf1 transcript. (D) Predicted hybridization of miRNAs and Nf1 transcript using the RNAhybrid algorithm [36]. The minimum free energy required for the hybridization is indicated.
Figure 2.
MiR-103, miR-128 and miR-137 directly reduce NF1 reporter protein expression.
(A) Schematic of miRNA plasmids and their expression levels in HEK293 cells. Representative gel showing the products of RT-PCR reactions amplified with primers specific for mature miR-103, miR-128, and miR-137 in HEK293 cells transfected with scramble or miR-128 vectors. The amount of starting template for each condition was equilibrated relative to U6 RNA. Cycles were falling within the linear range of amplification for each primer pair. (B) Schematic of Nf1 3′-UTR reporter construct and miRNAs tested for Nf1 3′-UTR regulation of expression. HEK293 cells were co-transfected with both the reporter gene (0.2 µg/reaction) and miRNA expression vectors (pri-mir-103, pri-mir-128, pri-mir-137, pri-mir-27, pri-mir-182 and pri-mir-153) (0.8 µg/reaction) and luciferase activity was measured 48 hours later. The fold change from control values as well as the P-value for each miRNA condition is presented in the table. These assays demonstrated that only miR-103, miR-128 and miR-137 significantly reduce NF1 reporter protein levels. Subsequently, mutagenesis of the predicted miR-103, miR-128 and miR-137 binding sites alleviated this inhibitory effect indicating that miR-103, miR-128 and miR-137 directly suppress Nf1 expression by targeting the identified seed regions in the Nf1 3′-UTR. The average value of three single (scramble 1, scramble 2, miR-218) and two double (miR-101/181, miR-218/377) miRNA constructs expressing miRNA predicted not to bind Nf1 3′-UTR were used as controls for the single or double expression constructs, respectively. Data show the mean ± s.e.m from 6 independent transfections (*, P<0.05; **, P<0.01).
Figure 3.
MiR-128 and miR-137 synergize to reduce NF1 reporter protein expression.
HEK293 cells were co-transfected with both the reporter gene (0.2 µg/reaction) and variable concentrations of pri-mir-128 and pri-mir-137 expression vectors. Luciferase activity was measured 48 hours later. (A) pri-mir-128 plasmid concentration was kept at 0.4 µg/reaction while pri-mir-137 concentration varied from 0.1–0.4 µg/reaction. (B) pri-mir-137 plasmid concentration was kept at 0.4 µg/reaction while pri-mir-128 concentration varied from 0.1–0.4 µg/reaction. Scramble 2 plasmid was supplemented in some of the reactions to achieve final miRNA plasmid concentration of 0.8 µg/reaction. These assays demonstrated that miR-128 and miR-137 synergistically reduce NF1 reporter expression when levels of the less abundant miRNA in the reaction are 50% or higher of the more abundant miRNA. The average value of two single miRNA expression plasmids (combination of scramble 2 plus pri-mir-9 or pri-mir-181 or pri-mir-218) predicted not to bind Nf1 3′-UTR were used as controls. Data show the mean ± s.e.m from 4 independent transfections (*, P<0.05; **, P<0.01, ***, P<0.001).
Figure 4.
MiR-103, miR-128 and miR-137 reduce endogenous NF1 protein but not mRNA expression.
(A) Representative gel of the RT-PCR amplification of Nf1 mRNA in HEK293 cells transfected with scramble or miR-128 vectors for 48 hours. The amount of starting template for each condition was equilibrated relative to U6 RNA. Cycles were falling within the linear range of amplification for each primer pair. (B) Representative Western Blot analysis demonstrating that miR-128, miR-128/103, and miR-128/137 reduce endogenous NF1 protein levels in human cells. HEK293 cells were transfected with miR-128 constructs for 48 hours. 15 µg of whole-cell lysate was then loaded in each lane. B-tubulin was used as an internal control for loading. Data show the mean ± s.e.m from 4 (for RT-PCR) and 8 (for Western Blot) independent experiments (***, P<0.001). B
Figure 5.
MiR-103, miR-128, miR-137 and Nf1 mRNA are co-expressed in the nervous system.
Representative gels of the RT-PCR amplification products of miR-103, miR-128, miR-137, and NF1 mRNA levels in: (A) Different murine tissues of embryonic day 18 animals; (B) Hippocampus of different ages; (C) Cortex of different ages; and (D) Different neural cell types. The amount of starting template for each condition was normalized to U6 RNA. The number of PCR cycles for each miRNA RT-PCR assay is different between gels. Cycles were falling within the linear range of amplification for each primer pair. SCG, superior cervical ganglion; TG, trigeminal ganglion; E, embryonic day; P, postnatal day.
Figure 6.
MiR-128 alone or together with miR-103 or miR-137 reduces endogenous NF1 protein levels in neurons.
(A) Representative images of transfected hippocampal neurons (green) stained with NF1 antibody (red). Merged images are shown in the third column of the panel. E16 murine hippocampal neurons were transfected with scramble or pri-mir-128 plasmids immediately after plating by using Lipofectamine 2000. Immunocytochemistry was carried out 40 hours after transfection. (B) Average decrease in NF1 protein levels. 80 neurons were analyzed per experiment. Data show the mean ± s.e.m from 6 independent experiments (*, P<0.05; **, P<0.01). (C) Freshly dissociated hippocampal neurons were transfected with Nf1 reporter vector plus antisense 2′-O-methyl inhibitors (miR-128 alone or together with miR-103 or miR-137) and assayed 48 hours later. Data show the mean ± s.e.m from 4 independent experiments (*, P<0.05).