- Research focus
- Team members
- Read more
The Ehlers-Danlos syndrome (EDS) constitutes a clinically and genetically heterogenous group of heritable connective tissue disorders. Its hallmarks are joint hypermobility and skin hyperextensibility, but the clinical spectrum also includes severe physical disability and lifethreatening vascular and visceral complications.
EDS has fascinated people throughout the ages. The first report of this disorder dates back to the fourth century BC, when Hippocrates recorded that the Scythians were ‘unable to draw their bow because their shoulder joints were too lax’. For many centuries, affected individuals earned their livings as circus artists, and amazed their audiences in fairgrounds and circus sideshows by exhibiting contortionist tricks and a remarkable ability to stretch their skin. It eventually took until the second half of the 20th century to document the first biochemical and molecular defects in the biosynthesis of fibrillar collagens were identified and documented. More recently, the comprehensive clinical and molecular delineation of several novel types of EDS has not only revealed the impressivegenetic heterogeneity of this group of conditions, but also greatly expanded our insights into its pleiotropic clinical manifestations by including more diverse signs of connective tissue fragility.
Nevertheless, we still are faced with numerous challenges related to diagnostic and prognostic uncertainty, the mechanisms behind EDS, and the limited therapeutic possibilities. We aim to address all of these through a translational approach that includes both clinical, preclinical and fundamental research.
As documented in our previous studies, many EDS subtypes are caused by defects in fibrillar collagens or their modifying enzymes. The pathogenic mechanisms through which mutations in these genes cause the various connective tissue phenotypes remain poorly understood. Mechanical weakness, due to diminished collagen production and/or incorporation of structurally defective collagen molecules into the ECM, has long been thought to be the main driver of connective tissue fragility. However, as shown in other HCTD, this ‘mechanistic’ viewpoint is most likely oversimplified, and we expect that a complex interplay of cellular signaling pathways, cell-matrix and supramolecular interactions underlies the EDS phenotypes. In the second term of this Methusalem grant, we aim to further explore the contribution of several intra- and extracellular processes and pathways to the EDS-pathogenesis by means of a combination of molecular, biochemical, morphological, and ICC/IHC studies in humans, as well as in cellular and animal models.
First, we aim to explore the involvement of an unfolded protein response (UPR) and ER stress in the pathogenesis of EDS, as these processes have been shown to play an important role in the pathophysiology of other collagenopathies, such as OI. To this purpose we will select cultured fibroblasts (FB) from healthy controls and EDS patients with well-characterized defects in the helical, the N- and C-propeptide domains of collagen type I, III and V, and in molecules involved in collagen modification (e.g. ADAMTS2, PLOD1). We will study the rate of chain association, folding, secretion, pericellular processing and deposition of collagen in the ECM of the fibrillar collagens, in order to define which defects affect collagen protein expression and stability and whether the mutated proteins tend to accumulate in the ER. To this purpose, we will use pulse-chase and SDS-PAGE analyses and co-ICC staining with antibodies against fibrillar collagens, Golgi and ER marker proteins and molecular chaperones (e.g. HSP47 and PDI). Signs of ER stress will be traced by TEM, WB, and RT-qPCR for a large set of typical ER stress related molecules and genes (e.g. CHOP, ATF4, ATF6, BiP, CALR, PDI). To monitor autophagy, we will evaluate conversion of the endogenous autophagic protein light chain 3 (LC3) from its inactive LC3-I to its active LC3-II form. Since prolonged ER stress induces apoptosis, we will evaluate cleavage and consequently activation of caspases by means of WB and ICC. We will also evaluate ER stress responses in our previously generated transgenic mouse model overexpressing a col3a1 p.(Gly183Ser) substitution. TEM images on dermal FB isolated from these mice show, besides gross abnormalities in collagen fibril architecture in the ECM, severe dilatation of the ER with presence of autophagosomes and intracellular accumulation of enlarged vesicles. This suggests that mutant type III collagen protein is retained within the cell and causes an UPR. To further evaluate the overall collagen fibril organization and the integrity of the ECM, picrosirius red staining, ICC, IHC and (immuno) EM studies using antibodies against different ECM components (e.g. types I, III and V collagen, elastin) are planned on different tissues (e.g. skin, aortic wall). Pulse-chase studies and ER stress studies will be performed on dermal FB and other cell types (e.g. VSMC) as outlined above. In addition, we will cross the transgenic Col3a1 mice with ER stress reporter mice (ER stress-activated indicator (ERAI) transgenic mice), in order to monitor ER stress in vivo, which will allow us to accurately investigate the micro- anatomical distribution of the ER stress. Finally, we will treat the transgenic Col3a1 mice with taurine- conjugated ursodeoxycholic acid (TUDCA) or 4-phenyl bituryc acid (PBA), two chemical chaperones that alleviate ER stress. Because collagen overexpression by the transgenic Col3a1 mouse modeI could contribute to the observed phenotype and UPR signature, we will control for this by comparing the data with a transgenic mouse model overexpressing wt Col3a1.
Second, we will use our transgenic Col3a1 mouse model to study the contribution of dysregulation of the TGFβ- and other signaling pathways (mitochondrial dysfunction, VSMC differentiation) to the arterial fragility observed in vascular EDS (vEDS). Interestingly, a recent study in humans demonstrated increased TGFβ2 secretion from dermal fibroblasts (FB) of vEDS patients, although downstream signaling of the TGFβ pathway was normal. This study also revealed a chronic inflammatory component to vEDS. Biomechanical experiments on skin and aorta of our male and female transgenic Col3a1 mice with the highest copy number of the transgene provide evidence for increased biomechanical weakness of these tissues and TEM studies show an altered structure of the aortic wall, making this a valuable model to study the pathophysiological basis of vEDS. We will investigate the different pathways involved in vascular disease, including the TGFβ, BMP and Notch signaling pathways, together with cell-matrix interactions, VSMC dysfunction, oxidative stress, and inflammation at different time points, similarly as outlined for the MFS mouse models. Although the phenotype of our transgenic col3a1 mouse model mimics the human vEDS phenotype quite faithfully, we are aware of the disadvantages of an overexpression system. Therefore, if needed, we will generate a knock-in mouse model, by means of homologous recombination, with the same mutation to validate the physiologic relevance of our findings.
Third, we aim to explore the mechanisms behind the phenotypic overlap between proteoglycan (PG) deficient types of EDS and the kyphoscoliotic type of EDS (EDS type VIA) using ZF models. PGs consist of one or more variable glycosaminoglycan (GAG) chains, which are linear polysaccharides consisting of repeating disaccharide blocks, attached to a core protein. Depending on the composition of these blocks, the PG superfamily is subdivided into two major groups: chondroitin sulfate (CS)/dermatan sulfate (DS) PGs and heparan sulfate (HS) PGs. Biosynthesis of CS/DS and HS chains starts with the formation of a tetrasaccharide linker region, a process catalyzed by the stepwise action of specific glycosyltransferases (encoded respectively by XYLT1/2, B4GALT7, B3GALT6 and B3GAT3). Bi-allelic mutations in B4GALT7 and B3GALT6 cause progeroid EDS-like phenotypes, associated with a (usually) mild skeletal dysplasia, but, intriguingly, some B3GALT6 mutations cause a severe skeletal dysplasia without any EDS features. Mutations in XYLT1 and B3GAT3 on the other hand have recently been identified in patients with skeletal dysplasia and Larsen syndrome with cardiac defects respectively, but EDS features were absent in both conditions. It remains currently unresolved why deficiency of these enzymes, all involved in the biosynthesis of the linker region shared among both CS/DS and HS chains, results in different clinical phenotypes. We hypothesize that enzyme- (or even mutation-) specific alterations in GAG chain composition could explain the phenotypic differences between the different conditions. Based on the phenotypic overlap between the B4GALT7 and B3GALT6 linkeropathies and musculocontractural EDS, caused by deficient activity of two enzymes involved in DS synthesis (D4ST-1 and DSE), we hypothesize that the EDS-like features in these disorders areprimarily caused by deficient biosynthesis of DS PG. TEM studies on skin specimens from patients with B3GALT6, CHST14 en DSE mutations show abnormal collagen fibrils, which could be explained by the observed alterations in GAG constitution of decorin, a small leucine-rich proteoglycan that regulates collagen fibril organization. The clinical overlap between these conditions and EDS VIA, caused by deficiency of lysylhydroxylase 1 (PLOD1), required for collagen cross-linking and the attachment of either galactose or glucosyl-galactose residues to the helix, suggests interacting pathogenic pathways for these disorders. To further explore this we will generate homozygous knockout (KO) ZF models for b3galt6, b3gat3, dse and chst14 using the TALEN and/or CRISPR-Cas system, and purchase plod1a, xylt1 and b4galt7 KO through the Zebrafish Mutation Project. In vitro wound healing assays will be performed. Mutants will be bred into different transgenic backgrounds to study cartilage morphogenesis (Tg(Col2a1aBAC:mcherry)), heart development (Tg(fli1a:EGFP), Tg(tie2:GFP)) and vasculature (Tg(fli1a:EGFP)), complimented by TEM of these tissues. We will perform qualitative and quantitative studies of GAGs, by means of whole-mount IHC for HS, CS and DS, a 1,9-dimethylmethylene blue assay to quantify the amount of sulfated GAG chains and strong anion exchange high performance liquid chromatography (SAX-HPLC) for detailed GAG chain analysis. The changes in GAG chain composition in mutants compared to wt will be correlated with ultrastructural changes in collagen structure and ECM organization. To this end we will perform TEM on mutant zebrafish embryos, picrosirius red and alcian blue staining on whole mount embryos, and IHC for ECM components on whole mount and sections of embryos.
Last updated: 17 July 2018 - 17:02
Copyright 2018 Center for Medical Genetics, Gent.