MicroRNA mimicry blocks pulmonary fibrosis

Abstract

Over the last decade, great enthusiasm has evolved for microRNA (miRNA) therapeutics. Part of the excitement stems from the fact that a miRNA often regulates numerous related mRNAs. As such, modulation of a single miRNA allows for parallel regulation of multiple genes involved in a particular disease. While many studies have shown therapeutic efficacy using miRNA inhibitors, efforts to restore or increase the function of a miRNA have been lagging behind. The miR-29 family has gained a lot of attention for its clear function in tissue fibrosis. This fibroblast-enriched miRNA family is downregulated in fibrotic diseases which induces a coordinate increase of many extracellular matrix genes. Here, we show that intravenous injection of synthetic RNA duplexes can increase miR-29 levels in vivo for several days. Moreover, therapeutic delivery of these miR-29 mimics during bleomycin-induced pulmonary fibrosis restores endogenous miR-29 function whereby decreasing collagen expression and blocking and reversing pulmonary fibrosis. Our data support the feasibility of using miRNA mimics to therapeutically increase miRNAs and indicate miR-29 to be a potent therapeutic miRNA for treating pulmonary fibrosis.

Keywords: miR‐29; microRNA; mimic; pulmonary fibrosis; therapeutics.

Figure 1. Pharmacokinetic properties of miR-29b mimic

1. The double-stranded miR-29 mimics design contains a “guide strand” or “antisense strand” that is identical to the miR-29b, with a UU overhang on the 3′ end, modified to increase stability, and chemically phosphorylated on the 5′ end and a “passenger strand” or “sense strand” that contains 2′-O-Me modifications to prevent loading into RNA-induced silencing complex (RISC) as well as increase stability and is linked to cholesterol for enhanced cellular uptake. Several mismatches are introduced in the sense strand to prevent this strand from functioning as an antimiR.

2. Transfection experiments in NIH 3T3 show a dose-dependent decrease in Col1a1 with increasing amount of miR-29b mimic compared to either untreated or mock-treated cells. An siRNA directly targeting Col1a1 was taken along as a positive control. *P < 0.05 versus mock, ^#P < 0.05 versus untreated.

3. Northern blot analysis for miR-29b in different tissues 4 days after intravenous injection with 10, 50, 100, or 125 mpk miR-29b mimic indicates delivery to all tissues at the highest dose, with the most effective delivery taking place to the lungs and spleen compared to saline-injected mice. U6 is used as a loading control.

4. Real-time quantification of miR-29b mimicry indicates an increased level of miR-29b at the higher dose levels with the most efficient delivery to the lungs and spleen (n = 4 per group). *P < 0.05 versus saline-injected animals.

5. Northern blot analysis for miR-29b in different tissues 1, 2, 4, and 7 days after intravenous injection with 125 mpk of mimic indicates the presence of miR-29b mimic in all tissues examined, with a longer detection in lung and spleen. U6 is used as a loading control.

6. Real-time quantification of miR-29b mimicry indicates an increased level of miR-29b in all tissues measured which is maintained the longest in lungs and spleen (n = 4 per group). *P < 0.05 versus saline-injected animals.

Figure 2. miR-29b mimic blunts pathological signs of bleomycin-induced pulmonary fibrosis

A Real-time PCR analysis indicates a reduction in all miR-29 family members in response to bleomycin, while miR-29 mimic treatment resulted in the increased detection of miR-29b levels compared to either control- or saline-injected animals. *P < 0.05 versus Saline/Saline. B Real-time PCR analysis indicated a comparable decline in miR-29 levels in pulmonary biopsies of patients with idiopathic pulmonary fibrosis (IPF) compared to normal controls. *P < 0.05 versus Normal. C Histological examination by trichrome staining showing pronounced fibrotic and inflammatory response in response to bleomycin, which was blunted by miR-29b mimic treatment. Scale bar indicates 100 μm. D Hydroxyproline measurements to assay for total collagen content showed a significant increase following bleomycin treatment in both saline- and control-treated groups, while there was no statistical difference in the miR-29 mimic-treated group between saline- and bleomycin-treated mice. E–G Cytokine measurements on bronchoalveolar lavage (BAL) fluids indicated a significantly higher concentrations of IL-12 (E), IL-4 (F), and G-CSF (G) were detectable in BAL fluids from lungs from bleomycin-treated mice, which was reduced with miR-29b mimic (n = 4). *P < 0.05. H Bleomycin treatment increases the detection of immune cells in BAL fluids which was significantly reduced in the presence of miR-29b mimic, while the control mimic had no effect (n = 4)., *P < 0.05 versus Saline/Bleo, ^{^}P < 0.05 versus Control/Bleo.

Figure 3. In vivo mimicry of miR-29b represses the induction of miR-29 target genes during pulmonary fibrosis

A, B Bleomycin treatment increases the expression of Col1a1 (A) and Col3a1 (B), and the presence of miR-29b mimic inhibits Col1a1 and Col3a1 as measured by real-time PCR. MiR-29b mimicry has no effect on target repression under baseline conditions. (n = 6–8), *P < 0.05. C IGF1 levels in BAL fluids increase following bleomycin treatment which were significantly blunted in the presence of miR-29 mimic compared to both saline- and control mimic-treated mice (n = 4). *P < 0.05. D Immunohistochemistry demonstrated robust detection of IGF1 after bleomycin treatment, which was reduced in the miR-29b mimic-treated group compared to saline- or control mimic-treated mice. Scale bar indicates 50 μm.

Figure 4. Therapeutic mimicry of miR-29 attenuates bleomycin-induced fibrosis

A Hydroxyproline assessment showed a significant increase following bleomycin treatment in both saline- and control-treated groups; however, there was no statistical difference in the miR-29 mimic-treated group between saline- and bleomycin-treated mice. *P < 0.05 (n = 8). B, C Real-time PCR analysis showed a significant increase of Col1a1 (B) and Col3a1 (C) after bleomycin treatment. miR-29b mimic treatment normalized both Col1a1 and Col3a1 to vehicle-treated expression levels. *P < 0.05 (n = 8). D Histological examination by trichrome staining showing robust fibrosis in response to bleomycin, which was blunted by miR-29b mimic treatment. Scale bar indicates 50 μm. E, F Primary pulmonary fibroblasts from patients with IPF were treated with vehicle or TGF-β and transfected with control mimic or miR-29b mimic. Real-time PCR was performed for Col1a1 (E) and Col3a1 (F). miR-29b mimic treatment showed a dose-dependent reduction in both collagens. G, H A549 cells were treated with vehicle or TGF-β and transfected with control mimic or miR-29b mimic. Real-time PCR was performed for Col1a1 (G) and Col3a1 (H). miR-29b mimic treatment showed a dose-dependent reduction in expression of both Col1a1 and Col3a1.