Therapeutic potential of synthetic microRNA mimics based on the miR-15/107 consensus sequence

Abstract

MicroRNA expression is frequently suppressed in cancer, and previously we demonstrated coordinate downregulation of multiple related microRNAs of the miR-15/107 group in malignant pleural mesothelioma (PM). From an alignment of the miR-15 family and the related miR-103/107, we derived a consensus sequence and used this to generate synthetic mimics. The synthetic mimics displayed tumour suppressor activity in PM cells in vitro, which was greater than that of a mimic based on the native miR-16 sequence. These mimics were also growth inhibitory in cells from non-small cell lung (NSCLC), prostate, breast and colorectal cancer, and sensitised all cell lines to the chemotherapeutic drug gemcitabine. The increased activity corresponded to enhanced inhibition of the expression of target genes and was associated with an increase in predicted binding to target sites, and proteomic analysis revealed a strong effect on proteins involved in RNA and DNA processes. Applying the novel consensus mimics to xenograft models of PM and NSCLC in vivo using EGFR-targeted nanocells loaded with mimic led to tumour growth inhibition. These results suggest that mimics based on the consensus sequence of the miR-15/107 group have therapeutic potential in a range of cancer types.

Fig. 1. Design of mimics based on the consensus sequence of the miR-15/16 family.

A The 5 members of the miR-15 family with detectable expression in PM samples and the related miR-103 and miR-107 were aligned. To generate a consensus sequence, positions at which one base was most frequently detected are shaded. Red denotes positions at which more than one base was equally prevalent or where less than half the aligned sequences contained that position. B A graphical representation of the consensus sequence logo generated with WebLogo [65]. C Sequences of four mimics based on the consensus sequence (bases differing between the four sequences are in red) and their individual percentage homologies to the endogenous microRNAs used to generate the consensus.

Fig. 2. Consensus mimics inhibit growth in PM cell lines.

A A panel of PM cell lines was transfected with microRNA mimics at a final concentration of 5 nM and proliferation was measured after 72 h. Data are mean ± SD (n = 3). B Cells transfected with the indicated microRNA mimics (5 nM) were transferred to 24-well plates and colony forming ability was measured after 7–10 days. A representative of three independent experiments is shown. C To determine the relative potency of the growth inhibitory activity of the mimics, cells were transfected with a microRNA mimics at a final concentration of 0.039 to 10 nM and proliferation was measured 96 h after transfection. Data are mean ± SD of triplicate measurements and are representative of three independent experiments.

Fig. 3. Activity of consensus mimics in other cancer cell types.

A C lines derived from a range of tumour types were transfected with microRNA mimics at a final concentration of 5 nM and proliferation was measured after 72 h. Data are mean ± SD (n = 3). B The relative potency of the growth inhibitory activity of the mimics was determined as in Fig. 2. Data are mean ± SD of triplicate measurements and are representative of three independent experiments.

Fig. 4. Effects of consensus mimics on gene expression.

A Protein expression was measured in MSTO cells 72 h after transfection with miR-16 mimic, the consensus mimics 107.2 or 107.4 or a control mimic. The number of downregulated proteins following each treatment is shown. B SWATH-MS-based proteomic analysis of predicted miR-16 target genes that were downregulated in mimic transfected cells compared to controls. C The effect on target gene mRNA expression was measured 72 h after transfection of H28 or VMC23 cells with miR-16, the consensus mimics 107.2 or 107.4, or control mimic (5 nM). Data are mean ± SD (n = 3). D The binding of miR-16 and two consensus mimics to sites in target gene 3′UTRs was measured using luciferase reporter genes in MSTO and H28 cells. Data are mean ± SD of triplicate measurements and representative of three independent experiments. E Predicted binding sites and free energy calculations for miR-16 and the consensus mimics for 2 sites in the CCND1 3′UTR.

Fig. 5. Effects of consensus mimics on sensitivity to gemcitabine.

Cell lines were transfected with microRNA mimics at a final concentration of 5 nM, and gemcitabine at the indicated concentrations was added one day post-transfection. Proliferation was measured after 72 h. Data are mean ± SD of technical replicates (n = 3) and representative of three independent experiments producing similar results.

Fig. 6. Consensus mimics control tumour growth in vivo.

Xenograft tumours were formed by implantation of 1 × 10⁶ PM (A) or A549 (B) cells in nude mice. Once tumours reached a size of 100 mm³, treatment began. Mice were treated via tail-vein injection 4× per week (arrows) with 1 × 10⁹ EDVs containing the consensus mimic 107.2, control mimic, or saline. Tumour volume was measured on the indicated days. *P < 0.05, **P < 0.01. C Delivery of the consensus mimics to the xenografts was confirmed by the detection of the novel mimic sequences in RNA-seq analysis.