Determinants of Vessel Wall Development and Homeostasis
Our laboratory is interested in the development and maintenance of the human cardiovascular system, particularly the structural homeostasis of the arterial wall. One goal is to understand the genetic factors that predispose to aortic aneurysm, a condition accounting for 1–2 percent of deaths in industrialized countries. While our initial approach has been to study Marfan syndrome (MFS), it is anticipated that a comprehensive understanding of this relatively rare disorder will promote a greater understanding of vascular wall biology and identify pathways that may be relevant to the more common forms of aortic aneurysm. MFS is an autosomal-dominant connective tissue disorder that affects about 1 in 5,000 individuals. Manifestations include long-bone overgrowth, skeletal muscle myopathy, lens dislocation, emphysema, mitral valve thickening and dysfunction, and aortic aneurysm with a predisposition for early vascular rupture and sudden death. In 1991, we showed that primary mutations in fibrillin-1, the major component of extracellular microfibrils, cause MFS.
Doctrine posited that microfibrils are needed for elastic fiber assembly during embryogenesis. The inference is that individuals with MFS are born with an obligate structural predisposition for the development of aortic aneurysm. Remarkably, our study of mice with targeted mutations in both copies of the fibrillin-1 gene (Fbn1) revealed normal elastic fiber formation but abnormal elastic fiber maintenance during postnatal life. Fragmentation of elastic fibers was preceded by a predictable sequence that included an intense fibroproliferative response of vascular smooth muscle cells and recruitment and activation of matrix-degrading enzymes.
Enid Neptune has shown that fibrillin-1–deficient mice show primary failure of alveolar septation in early postnatal life. This abnormality correlates with increased transforming growth factor-β (TGFβ) activation and signaling. Administration of TGFβ-neutralizing antibody can rescue alveolar septation, documenting a direct link between lung disease and cytokine dysregulation. Neptune has now focused on identifying molecules downstream of TGFβ that influence lung maturation and is exploring the relevance of TGFβ to other etiologies for developmental emphysema.
In related work, Ronald Cohn has now extended this model to the skeletal muscle hypoplasia that can be seen in MFS. Fibrillin-1–deficient mice show small muscle fibers that are also reduced in number, abnormally formed (split fibers), and surrounded by excessive extracellular matrix. This associates with increased TGFβ signaling and pronounced failure of muscle regeneration after induced injury. Cohn has shown that both steady-state muscle architecture and muscle regeneration can be rescued by TGFβ antagonism in vivo. He has also shown that excess TGFβ signaling attends failed muscle regeneration in other myopathic states, including Duchenne muscular dystrophy, and that TGFβ antagonists achieve similar protection in a validated mouse model of this disorder. Cohn and Christel van Erp are studying the relevance of this mechanism and treatment strategy to other etiologies of myopathy.
Jennifer Habashi and Daniel Judge have focused on the most critical problem in MFS, i.e., aortic aneurysm. Here, increased TGFβ activity correlates with postnatal deterioration of elastic fiber architecture, aortic wall thickening due to excess matrix accumulation, and progressive aortic root enlargement. Administration of TGFβ-neutralizing antibody greatly attenuates phenotypic severity. Habashi and Judge asked whether there is an FDA-approved drug that could mimic this effect. Attention focused on losartan, an angiotensin II type 1 (AT1) receptor blocker that lowers blood pressure (a desired effect for people with aortic aneurysm) and also leads to a clinically significant reduction in TGFβ signaling. Losartan completely prevented aortic pathology in a mouse model of MFS. Indeed, treated Marfan mice could not be distinguished from their wild-type littermates by any parameter tested. Furthermore, AT1 blockade rescued features of MFS outside of the cardiovascular system, including lung septation and skeletal muscle myopathy. On the basis of this work, the Pediatric Heart Network of the National Heart, Lung, and Blood Institute launched a large multicenter clinical trial of losartan for MFS in the fall of 2007. Members of the laboratory are attempting to expand and extend therapeutic options. Peter Matt and Florian Schoenhoff have demonstrated increased circulating TGFβ in mice and people with MFS. The level of circulating TGFβ is decreased with losartan in a dose-dependent manner and correlates with aortic size in mouse models, suggesting that circulating TGFβ may serve as both a prognostic and therapeutic marker, allowing the tailoring of therapeutic protocols to the individual. Matt and Schoenhoff are also using a mouse model of acute aortic dissection for biomarker development.
Habashi and Jef Doyle are comparing the relative benefits of selective AT1 blockade with losartan to the benefits of limiting production of angiotensin II with an angiotensin-converting enzyme inhibitor (ACEi) that diminishes signaling through both the AT1 and AT2 receptor. This addresses the current controversy regarding whether AT2 signaling is protective or detrimental in MFS. Results in mouse models of MFS document that targeting of the AT2 receptor results in larger aortic dimensions, that losartan provides significantly better protection than ACEi, and that AT2 signaling is required for the full protective effect of losartan. We are now exploring whether an AT2 agonist will be synergistic with losartan and are elucidating the mechanism of protection by AT2 signaling.
Mark Lindsay is exploring potential events downstream of TGFβ that may contribute to vascular disease. He and David Loch have found that medial cells isolated from the aortic root of patients and mice with MFS express abnormally high levels of selected vascular smooth muscle cell (VSMC) markers but fail to express markers of terminal VSMC differentiation. Lindsay is using lineage-tracing studies to determine whether this represents dedifferentiation of VSMCs derived from normal embryonic origins, or perhaps population of the aortic media with myofibroblasts derived from pathologic endothelial-to-mesenchymal transition (EnMT), a process known to be driven by TGFβ. Preliminary results support the EnMT hypothesis and suggest that agents that suppress EnMT (e.g., statins and fasudil) may be protective in MFS; this is currently being tested in mouse models. Kathleen Kent and Juan Calderon are mapping genetic modifiers of MFS, and Tammy Holm and Doyle are using mouse models to explore the relevance of noncanonical TGFβ-signaling cascades in aortic disease.
During the course of patient care, Bart Loeys and I recognized a subset of patients that had some features of MFS but showed particularly aggressive vascular disease, with rupture at aortic sizes not associated with risk in MFS and/or death in early childhood. Other distinguishing features included the common presence of widely spaced eyes, a split uvula, and generalized arterial tortuosity with a predisposition for aneurysm and dissection throughout the arterial tree. At the most severe end of the spectrum, patients showed premature fusion of the skull bones, cleft palate, foot deformities, and various forms of structural congenital heart disease. We determined that this new syndrome (now called Loeys-Dietz syndrome, or LDS) is caused by mutations in either of the two genes that encode the TGFβ receptor. Analysis of patient-derived tissues showed that the net effect is too much TGFβ activity. David Loch and Elena Gallo are using patient cells and knock-in mouse models of LDS to define the mechanisms by which heterozygous loss-of-function mutations in TGFβ receptor genes lead to paradoxically enhanced signaling and to explore treatment strategies. We have recently shown that patients with LDS have a greatly increased predisposition for gastrointestinal disease, including food allergy, eosinophilic esophagitis, and Crohn's disease. Pamela Frischmeyer-Guerrerio and Anthony Guerrerio are investigating the molecular basis for perturbations of immunologic tolerance in LDS.
Upon recognition of the significant overlap between LDS and the vascular form of Ehlers-Danlos syndrome (EDS IV), we identified TGFβ receptor mutations in a large group of patients with EDS IV who lack the type III collagen abnormalities characteristic of this disorder and fail to show the craniofacial features of LDS (now designated LDS-II). To reconcile why excess TGFβ signaling (and the attendant increase in collagen production) phenocopies a collagen-deficient state, Timothy Cooper is exploring the hypothesis that, like fibrillin-1, type III collagen negatively regulates TGFβ superfamily signaling.
We have contributed to the identification of the genes for other aortic aneurysm syndromes, including arterial tortuosity syndrome and recessive cutis laxa. The emerging view is that too much TGFβ activity is a final common pathway to arterial aneurysms, and that therapeutic strategies for MFS (e.g., losartan) hold promise for the treatment of a wide array of vascular diseases. David Kim and Gallo are defining the molecular basis of additional syndromic and nonsyndromic presentations of aortic aneurysm.
Loeys and Elizabeth Bell have recently demonstrated that domain-specific mutations in fibrillin-1 underlie stiff skin syndrome (SSS), an autosomal-dominant presentation of congenital scleroderma (progressive fibrosis and hardening of the skin). These mutations impair the physical interaction between cells and fibrillin-1 that are normally mediated by integrins. Loss of integrin signaling correlates with increased fibrillin-1 deposition in the dermis and consequent concentration of TGFβ with increased TGFβ signaling. Similar events are seen in a more common acquired presentation of scleroderma called systemic sclerosis (SSc). Bell has now made mouse models of SSS that recapitulate progressive skin fibrosis and confirm the central role of impaired integrin binding to fibrillin-1 in disease pathogenesis. These mice are currently being used to explore treatment strategies for SSS and SSc.
This work was supported in part by grants from the National Institutes of Health, the National Marfan Foundation, the Smilow Foundation, and the Scleroderma Research Foundation.
As of April 20, 2010