Connective Tissue Mediated Vascular Disease

National Heart, Lung, And Blood Institute

Investigators

Linked publications, trials & patents

Abstract

Due to COVID, much of our in-laboratory effort slowed due to restrictions on time in lab. As such, we transitioned to several new computational projects that could be performed while teleworking. Defining domains and modifying elastin gene expression: Once present in the extracellular matrix, elastic fibers have a half-life of 70 years and little to no new elastin is made to replace or repair damaged fibers. As such, improved understanding of the mechanisms guiding elastin gene expression have the potential to impact developmental disorders of elastic tissues and aging related outcomes. Through our work with patients with elastic fiber disease, we have identified several individuals with overlapping deletions upstream of ELN, sparing the coding exons, who present with vascular features of elastin insufficiency, such as supravalvar aortic stenosis. Using computational methods, we have identified a minimal region of overlap for the deletions and have identified underlying areas containing likely enhancers of elastin expression. Investigations using patient derived fibroblasts will allow us to confirm differences in elastin gene expression and deposition in these patients and future work will include manipulation of these specific regions using CRISPR in cell or animal systems to test their unique impact on gene expression. Such findings should identify new genomic regions relevant to rare elastic fiber disease and that can potentially be manipulated to re-initiate elastin gene expression outside of the typical developmental window. An initial summary of the patient-identified region of interest is underway as part of an invited manuscript due this fall. Using induced pluripotent stem cells to investigate elastic fiber assembly We have also optimized a platform to generate induced pluripotent stem cell (iPS)-derived vascular smooth muscle cells (iVSMC) using chemically defined and serum free media. Single cell RNA-sequencing of differentiating cells from iPSCs confirms the transition of TBXT+ mesoderm cells to SM22a+ iVSMC from day 2 to day 3 of differentiation. Using published single cell RNA-seq study of human embryonic heart (Cui et al, Cell Rep 2019), we confirmed that our differentiation platform generates cells that have high similarity to developing smooth muscle cells/ fibroblasts lineage as compared to other cell types, including endothelial, cardiac and hematopoietic lineages. As in human and mouse tissues, elastin is expressed by our cell lines for only a brief period, with initial elastin mRNA expression noted on day 2 and peaking on day 3 of differentiation and deposition of elastin in the matrix over the following week. Using this optimized platform, we have modeled elastin-mediated vasculopathy in a dish using patient-derived iPSCs carrying different mutations in ELN (n=2) gene as well as CRISPRed iPSC lines targeting ELN (n=3). This year, we confirmed that the cell genomes had been minimally altered by reprogramming. Likewise, minimal off target genetic effects were detected in the CRISPR lines. We are now working to confirm the expression level of ELN mRNA and deposition of elastic fiber using quantitative real-time PCR and immunostaining, respectively in these ELN-insufficient iPS lines. We are also piloting new methods to assess quality-based differences in elastic fibers secreted by iVSMCs produced by donor lines with missense rather than nonsense variants. Toward this end, decellularized extracellular matrix from iVSMC will be tested using proteomic approaches, like mass spectrometry or higher resolution imaging like 2 photon or electron microscopy. In addition to elastin quantity and quality, we will evaluate for functional outcomes on the cell, such as changes to cell proliferation and mobility. Identification of novel genes important for elastic fiber assembly As noted above, elastin mRNA is readily detectable on days 2 and 3 of differentiation, but then disappears by day 5. Elastin protein, on the other hand, is detectable in cells by day 5 and appears in the matrix in robust quantities by day 9. Interestingly, deposition of elastin in the extracellular matrix does not occur until other elastic fiber assembly genes such as EFEMP2, FBLN5, and LOX are up-regulated. To identify additional genes that may play a role in elastic fiber production, we have developed new computational methods using single cell data drawn from the literature to identify genes that are consistently co-expressed in the same cell with elastin across tissue types and time periods. Current efforts have confirmed the co-expression of a set of previously implicated elastic fiber assembly genes but have also revealed new candidates likely to be key to this process. Together, these genes make up the elastogenic cell profile. This gene list is enriched with extracellular matrix genes, but additional novel pathways that may be relevant to the secretion and cell surface assembly process are also being discovered. Single cell additionally offers the opportunity to study genes thought to be important for elastic fiber assembly that are not made concurrently by the same cellssuggesting either the need for an individual cell to pause and hold its developing elastin globules intracellularly until the other assembly proteins are upregulated or possibly collaboration among subtypes to produce and secrete all of the proteins necessary for elastic fiber assembly. Once a final list of relevant genes is obtained, newer technologies such as RNA-FISH can be used to look at the relationship between these mRNAs and cell types in intact tissues. Mutations in the copper binding domain of lysyl oxidase produce aberrant elastic fibers Lysyl oxidase is a copper binding enzyme responsible for crosslinking tropoelastin monomers to one another to generate insoluble elastin in the extracellular space. The Loxb2b370.2Clo (c.G854T; p.C285F) mouse Lox variant falls within the copper binding domain of the molecule. In Lox+/ c.G854T aortas, mRNA levels are similar to WT, but enzymatic activity is reduced. Those aortas reveal a relative increase in pro-Lox protein at age E19, but relatively less detectable mature Lox throughout the developmental series. Cell culture studies, however, show normal levels of mature Lox protein in the conditioned media from fibroblasts generated from Loxc.G854T/ c.G854T mice, suggesting that it is appropriately secreted. Pressure-diameter testing reveals a narrower aorta in 3 month C57Bl/6 Lox+/- males relative to C57Bl/6 WT, but dilates with age faster than a WT. This phenotype shows both sex and genetic background effects, with males and animals in a congenic hypertensive background dilating faster. When studied by two photon imaging, Lox+/ c.G854T mutants reveal disorganized lamellae. Lox+/ c.G854T mice also show increase numbers of holes in their elastic lamellae relative to WT, a phenotype that also becomes more pronounced with age or increased blood pressure. Ultrastructural imaging of the vessel wall using Fib-SEM shows in detail the structural derangements that occur in the Lox mutants and immunostating reveals a preponderance proteoglycan in the extracellular matrix replacing the elastin. Likewise, aortas from Lox+/ c.G854T mice exposed to elastase dilate faster suggesting that the elastin and collagen produced in these mutants is structurally inferior, making it increasingly susceptible to proteolytic damage. Indeed, analysis of the endogenouse elastase-1 protein level in the aorta from 3 month old Lox+/ c.G854T mice showed that the mutants do actually have increased amount elastase in their tissues. Likewise, RNAseq experiments showed increased turnover of the extracellular matrix with expression differences noted in genes responsive to TGF-B and estrogen. This manuscript is being revised for resubmission.