Multi-level transcriptome analysis of LMNA related dilated cardiomyopathy identifies disease driving processes

Introduction: : Mutations in the LMNA gene, encoding the nuclear envelope proteins lamin A and C, are responsible for a plethora of human diseases known as laminopathies, often specifically affecting striated muscle. The most prevalent laminopathy is LMNA-related dilated cardiomyopathy (LMNA-DCM). The molecular pathogenic mechanisms remain largely unresolved, leaving patients with limited and ineffective treatment options. Cardiac tissues from individuals with LMNA-DCM and cardiac-specific laminopathy mouse models, with anecdotal evidence of nuclear defects, exhibit increased levels of DNA damage and activation of p53, however it is unclear whether these defects are causative, or downstream consequences of aberrant signaling. Several preclinical models have used pharmacological and/or genetic interventions to target these altered signaling mechanisms in laminopathies; however, none of the existing approaches have been curative, and most have shown at most modest effects on survival. This low phenotypic rescue could be attributed in part to the fact that, so far, studies of LMNA-DCM in human cardiac tissue and preclinical animal models have largely been devoid of temporal analyses, being limited to the late disease stage, when severe phenotypes arise. To more efficiently identify the underlying molecular mechanism that is altered by LMNA-DCM, we generated a mouse model of LMNA-DCM with inducible cardiomyocyte-specific deletion of lamin A/C (Lmna-cKO) and performed multi-level transcriptomic studies at early and late disease timepoints (Figure 1A). The Lmna-cKO mice provide a unique advantage to distinguish between early effects of lamin A/C deletion that likely represent disease drivers, and longer-term effects that may result from declining health and tissue function.

Materials and

Methods: : Lmna-cKO mice were attained by crossing mice possessing loxP sites flanking exon 2 of the Lmna gene (Lmna fl/fl) with mice expressing a tamoxifen-inducible Cre recombinase (MerCreMer, MCM) driven by the α-myosin heavy chain (αMHC) promoter, limiting expression to cardiomyocytes. Wild-type Lmna littermates expressing αMHC-MerCreMer without the floxed Lmna gene served as controls. Animals were injected with tamoxifen at 9-10 weeks of age. Left ventricle free wall tissue was isolated from mice at both 11- and 25-days post-injection (dpi). To identify early disease-driving genes, we combined bulk (high sensitivity and resolution) and single-nucleus (sn) RNA sequencing (cell type-specificity), and Slide-seq (spatial resolution). From the bulk RNA-seq, we cross-referenced differentially expressed genes (DEGs) between Lmna-cKO and Control at both early (11-dpi) and late (25-dpi) disease timepoints to retain only those that showed intensifying changes over disease progression (from 11-dpi to 25-dpi), allowing for highly sensitive identification of genes already showing changes early in disease termed emerging DEGs (eDEGs). snRNA-seq datasets for each experimental condition were integrated into a single dataset and annotated for cell types. DEG analyses were then performed between subclusters identified amongst each individual cell type-specific cluster. Spatial transcriptomics of cardiac tissue slices from Control and Lmna-cKO mice provided additional validation of our bulk and single nucleus results. To define the cell-cell signaling between all cell types at the tissue level, we computed the strength of cell-cell interactions in each experimental condition, using CellChat, and then compared cell-cell communication between Lmna-cKO and Control samples at each timepoint.

Results, Conclusions, and Discussions:: Lmna-cKO mice presented with increasing levels of fibrosis (Figure 1B), followed by significant enlargement of Lmna-cKO hearts at 25-dpi (Figure 1C), closely recapitulating LMNA-DCM progression. Using bulk RNA-seq, we captured eDEGs, genes whose disease-contribution intensified over the course of disease progression (Figure 1D). Comparative gene ontology (GO) analysis between our list of identified eDEGs and DEGs identified at 25-dpi showed a clear enrichment of eDEGs for GO terms associated with apoptotic processes mediated by DNA damage and heart function (Figure 1E). These results suggest that nuclear damage damage is present early on in disease, supporting our hypothesis that the fragile characteristic of Lmna-deficient nuclei is a driving feature of LMNA-DCM pathology. Our snRNA-seq analysis identified disease-specific cardiomyocyte subpopulations (CM3 and CM5) in the Lmna-cKO hearts (Figure 1F) that largely drove the expression of eDEGs involved in the downregulation of heart function and upregulation of apoptotic signaling pathways. The snRNA-seq data also indicated the contribution of several non-cardiomyocyte cell types (fibroblast and innate immune cells) to early disease pathogenesis, even though these cells express lamin A/C in the Lmna-cKO mice. Cell-cell communication of Lmna-cKO-specific CM3 with non-cardiomyocyte cells was particularly strong with fibroblast subpopulations Fib1 and Fib4, as well as myeloid subclusters Mye1 and Mye2 (Figure 1G), in Lmna-cKO versus Control samples. Spatial transcriptomics further validated close proximity between CM3, Fib1, and Mye1 (Figure 1H) subpopulations in the Lmna-cKOsamples at 11-dpi. The enrichment of Fib1 for pro-fibrotic genes, combined with the communication and spatial results, suggest that Lmna-cKO-specific CM3 cardiomyocytes drive the observed fibrosis and downstream inflammation observed in LMNA-DCM.
Collectively, our results indicate that LMNA-DCM pathology is driven by aberrant processes from multiple cell types. We hypothesize that mechanically induced damage in a subset of cardiomyocytes (CM3 and CM3) results in activation of pathogenic signaling pathways and inflammatory signaling, leading to aberrant ECM deposition from fibroblasts and recruitment of innate immune cells, ultimately causing severe heart dysfunction. Further work will focus on validating identified signaling pathways and determining if modulation of Lmna-deficient nuclei mechanics can help rescue LMNA-DCM phenotypes and increase mouse survival.

Acknowledgements and/or References (Optional)::