Trimannose-coupled antimiR-21 for macrophage-targeted inhalation treatment of acute inflammatory lung damage

Abstract

Recent studies of severe acute inflammatory lung disease including COVID-19 identify macrophages to drive pulmonary hyperinflammation and long-term damage such as fibrosis. Here, we report on the development of a first-in-class, carbohydrate-coupled inhibitor of microRNA-21 (RCS-21), as a therapeutic means against pulmonary hyperinflammation and fibrosis. MicroRNA-21 is among the strongest upregulated microRNAs in human COVID-19 and in mice with acute inflammatory lung damage, and it is the strongest expressed microRNA in pulmonary macrophages. Chemical linkage of a microRNA-21 inhibitor to trimannose achieves rapid and specific delivery to macrophages upon inhalation in mice. RCS-21 reverses pathological activation of macrophages and prevents pulmonary dysfunction and fibrosis after acute lung damage in mice. In human lung tissue infected with SARS-CoV-2 ex vivo, RCS-21 effectively prevents the exaggerated inflammatory response. Our data imply trimannose-coupling for effective and selective delivery of inhaled oligonucleotides to pulmonary macrophages and report on a first mannose-coupled candidate therapeutic for COVID-19.

Conflict of interest statement

Technical University of Munich has filed an intellectual property right on the therapeutic use of mannose-coupled antimiR-21 with D.R. and S.E. named as inventors. S.E. and T.F. are founders of RNATICS GmbH, a biotech company focussed on macrophage RNA therapeutics. After the completion of this study, D.R. joined RNATICS GmbH. Other authors do not have any conflicts.

© 2023. The Author(s).

Figures

Fig. 1. Identification of miR-21 in macrophages as therapeutic target for acute inflammatory lung disease.

a Overview of the study design. b Scatter plot of differentially expressed microRNAs in bleomycin lungs compared to control lungs. Each dot represents one microRNA. Differentially expressed microRNAs are highlighted in red (FDR < 0.05, −0.75 < log2FC > 0.75, mean normalised counts > 1000). Control n = 3, Bleomycin n = 3. c Scatter plot of differentially expressed microRNAs in COVID-19 lungs compared to control lungs. Each dot represents one microRNA. Differentially expressed microRNAs are highlighted in red (FDR < 0.05, −0.75 < log2FC > 0.75, mean normalised counts > 1000); Control n = 12, COVID-19 n = 14. d Lollipop graph represents the top 10 expressed microRNAs in (left) mouse and (right) human lung macrophages (n = 3-4). e Representative staining for macrophage marker CD68 in bleomycin-induced mouse lungs (n(PBS) = 3, n(bleo) = 3) and COVID-19 patients (n(COVID-19) = 3, n(Control) = 3). Scale bar represents 50 μm. f Experimental strategy. Wild-type and macrophage-specific miR-21 deficient (miR-21 cKO) mice were administered with PBS or 2 U/kg bleomycin (bleo) into the lungs using a micro sprayer. Two weeks later, lung function was assessed, and lungs were harvested for morphometry and isolation of cells. g Mean tracings of pressure-volume curves. Gray shaded area behind the curves indicates standard error mean. PBS and bleomycin treatment are represented by dashed and solid lines, respectively. h Lung function as indicated by quasi-static elastance and quasi-static compliance. i Sirius red/fast green staining of representative lung sections and quantification. Scale bar represents 100 μm. f–i WT PBS n = 3, miR-21 cKO PBS n = 3, WT bleo n = 3, miR-21 cKO bleo n = 3. Data are mean ± SEM and individual values and were analysed using 2-way ANOVA with Tukey’s post-test (two-sided). Source data are provided as a Source Data file.

Fig. 2. Development of macrophage-targeted, carbohydrate-coupled antimiR-21 (RCS-21).

a Screening strategy to identify surface receptor genes that are enriched in macrophage populations compared to other cell populations in the lung (mRNA log2 fold change > 3 and FDR < 0.05) and are abundant in both interstitial and alveolar macrophages (TPM > 100 in MP-A and MP-I and TPM < 10 in other cells). n = 3 per cell type. b Scatter plot showing surface receptor genes that are enriched in macrophage populations compared to other cell populations in the lung. c Comparative analysis identified mannose receptor 1 (MRC1) as the surface receptor enriched and abundantly expressed in both alveolar and interstitial lung macrophages. d Feature plot of Mrc1 in mouse lungs after bleomycin injury (data from GSE141259). e Feature plot of MRC1 in human lungs after COVID-19 (data from DUOS-000130). f Representative immunofluorescent staining for MRC1 in PBS- and bleomycin-treated mouse lungs (staining was carried out for three mice of each group). Nuclei were stained with Sytox Green. Scale bar represents 100 μm. g Chemical structure of trimannose-conjugated LNA-antimiR-21 (RCS-21). h Schematic illustration of uptake of RCS-21 by MRC1 in macrophages. MP-A alveolar macrophages, MP-I interstitial macrophages, EC endothelial cells, EpC epithelial cells. Source data are provided as a Source Data file.

Fig. 3. Delivery of inhaled RCS-21 to pulmonary macrophages in vivo.

a Wild-type mice were administered with either unconjugated, N-acetylgalactosamine- (GalNAc) or trimannose-conjugated (RCS-21) antimiR-21-FAM by inhalation. Two hours later, mice were sacrificed and cells isolated from bronchoalveolar lavage fluid (BALF) and lung tissue were assessed for FAM signals by flow cytometry. b Percentage of FAM-positive cells in different lung macrophage fractions. Unconjugated n = 3 per group, RCS-21 n = 3 per group, GalNAc-conjugated n = 2 per group. Data are mean ± SEM and asymmetric nonlinear regression analysis was used for curve fitting and compare the groups. c Top, median fluorescence intensity of FAM signals in macrophages. Bottom, representative histogram of macrophage subsets. 1.25 mg/kg Unconjugated n = 3 and 1.25 mg/kg RCS-21 n = 3. Data are mean and individual values, and were analysed using two-sided Student’s t-test. d Left, representative immunofluorescent staining of 5 μm mouse lung tissue cryosections for CD68 as a marker for macrophages. Nuclei were stained with DAPI. Scale bar represents 50 μm. White arrow indicates FAM-positive macrophages. Right, quantification of the same. PBS n = 3, 1.25 mg/kg Unconjugated n = 3 and 1.25 mg/kg RCS-21 n = 3. Data are mean and individual values and were analysed using two-sided one-way ANOVA with Sidak’s post-test. FAM fluorescein amidites, MP macrophages. Source data are provided as a Source Data file.

Fig. 4. Therapeutic efficacy of inhaled RCS-21 in a preclinical animal model of acute inflammatory lung damage.

a Experimental strategy. Wild-type mice were administered with PBS or 2 U/kg bleomycin (bleo) into the lungs using a micro sprayer. 2.5 mg/kg RCS-21 or control oligo (containing mismatch-miR-21 sequence) was applied by inhalation 4 days after injury. 10 days later, lung function was assessed, and lungs were harvested for morphometry and isolation of cells. b Mean tracings of pressure-volume curves. Gray shaded area behind the curves indicates standard error mean. PBS and bleomycin treatment is represented by dashed and solid lines, respectively. c Lung function as indicated by quasi-static elastance and quasi-static compliance. d Left, representative staining and analysis of lung sections using Sirius red/Fast green. Scale bar represents 100 μm. Right, quantification of fibrosis. e Left, representative staining of CD68. Right, quantification of CD68 positive cells. Scale bar represents 50 μm. b–e Control oligo: PBS n = 3, bleo n = 7. RCS-21: PBS n = 3, bleo n = 6. Data are mean ± SEM and individual values and were analysed using two-way ANOVA with Tukey’s post test (two-sided). Source data are provided as a Source Data file.

Fig. 5. RCS-21 inhibits inflammatory activation of human lung tissue in response to SARS-CoV-2 (Omicron).

a Overview of the study. b RNA-Seq of uninfected and infected hPCLS. BAM files were aligned onto the SARS-CoV-2 reference genome. c MicroRNA Red Scope staining for hsa-miR-21-5p in hPCLS (blue: nucleus). Scale bar represents 200 μm. Representative of n = 2 individuals per condition. d DESeq2 normalised reads of miR-21-5p in indicated hPCLS groups. Box plots represent the 25th and the 75th percentiles. The horizontal line represents the median, the red dots represent the mean for each group. Statistical analysis was performed using two-sided one-way ANOVA with Tukey’s post test. e Principal component analysis of the transcriptomes. f Cumulative distribution curves of mRNA transcriptomes to asssess miR-21-5p activity. Rightward shift of predicted miR-21 targets (red) indicates decreased miR-21 activity (RCS-21 vs. control). Statistical analysis was performed using two-sided Kolmogorov-Smirnov (KS) test. g Heatmap of conserved predicted targets of miR-21. h Volcano plots representing differentially expressed genes between infected and uninfected hPCLS (upper panel) and between infected RCS-21-treated and infected control hPCLS (lower panel). Significantly upregulated genes (log2FC > 1, FDR < 0.05) are highlighted in violet and significantly downregulated genes (log2FC < -1, FDR < 0.05) in green. i GO bubble plots represent the biological processes for upregulated (log2FC > 1) and downregulated genes (log2FC < −1) between SARS-CoV-2 infected and uninfected hPCLS (upper panel) and between infected RCS-21-treated and infected control hPCLS (lower panel), FDR significance cutoff < 0.01, upregulated pathways are highlighted in violet, downregulated pathways are highlighted in green. As a statistical test DAVID uses a modified Fisher’s Exact test; d–in(uninfected) = 5, n(infected) = 6, n(RCS-21) = 5. Source data are provided as a Source Data file.