Role for Hyaluronan Synthase 3 in the Response to Vascular Injury
More than 1.5 million percutaneous coronary interventions are performed annually. The most important long-term complication is restenosis caused by inflammation, neointimal vascular smooth muscle cell (VSMC) hyperplasia, and excess accumulation of extracellular matrix. Delineating mechanisms that contribute to inflammation, neointimal hyperplasia, and constrictive vascular remodeling will, therefore, improve and expand the options for prophylactic and therapeutic percutaneous coronary interventions for the growing population of patients with arterial disease.
See accompanying article on page e9
Hyaluronan is a pericellular and extracellular matrix component of normal vasculature located throughout vessels from the luminal surface of the endothelium, to most prominently in the adventitia. The marked upregulation of hyaluronan production and accumulation in atherosclerotic lesions and in the neointima of restenotic vessels has generated great interest as to its role in vascular disease. Endothelial pericellular hyaluronan promotes leukocyte adhesion and transmigration, whereas hyaluronan binding to CD44 on leukocytes regulates inflammatory gene expression. Platelet-mediated hyaluronan cleavage generates bioactive hyaluronan fragments that stimulate leukocyte production of cytokines and chemokines. Moreover, hyaluronan is involved in the processes underlying the tissue response after vascular injury, in particular, VSMC proliferation and migration.
In most cases, the implication of hyaluronan in vascular disease has been based on the circumstantial evidence of elevated hyaluronan levels in vascular lesions or deduced from deleting or blocking hyaluronan receptors (such as CD44 and receptor for HA-mediated motility). Designing approaches to more directly define the role of hyaluronan in vascular disease has been difficult in view of the multitude of synthases (HAS1, HAS2, and HAS3) and multiple hyaluronidases involved in its metabolism. Each HAS isoform produces structurally identical hyaluronan. One might, therefore, expect that hyaluronan function is independent of the HAS by which it is synthesized. However, the expression patterns of each HAS differs, and the isoforms are differentially regulated in homeostasis and in pathological settings. Furthermore, hyaluronan is an essential component of the pericellular matrix, or alternatively, it can be released in a soluble form and be released and incorporated as part of the extracellular matrix. The composition and architecture of the matrix affect hyaluronan-dependent biochemical signaling, as well as the biophysical and biomechanical properties of tissues. The temporal and spatial relationship of hyaluronan with cells that express hyaluronidases that modify the molecular weight of hyaluronan is another determinant of hyaluronan function. In addition, one cannot rule out the possibility that hyaluronan synthases may affect vascular disease independent of hyaluronan. Taken together, evidence suggests that there is potential for HAS isoform–specific functions in tissue homeostasis and disease.
In the absence of any isoform-specific inhibitors of HAS activity and in spite of the potential for alternative HAS isoforms to compensate for the loss of a specific HAS, genetic deletion is the most direct approach to address isoform-specific HAS function. However, this has proven challenging because of embryonic lethality of genetic deletion of HAS2. This suggests that some HAS2-specific functions are not compensated for by HAS1 or HAS3. In contrast, genetic deletion of HAS1 or HAS3 has no effect on viability or homeostasis. Nonetheless, the article by Kiene et al1 in this issue of ATVB has broken new ground by demonstrating a vascular injury phenotype in HAS3-deficient mice, thus establishing an isoform-specific role for HAS3 or its hyaluronan product in vascular disease (Figure).
Using a ligation-induced carotid artery injury model, the authors observed attenuated neointimal hyperplasia in HAS3-null animals compared with wild-type control C57BL/6J mice. No changes were observed in medial and neointimal cell density, proliferation, or apoptosis. However, consistent with a lack of compensatory upregulation of HAS1 or HAS2, HAS3 deletion was associated with a reduction in vascular hyaluronan content, most dramatically in the media rather than the neointima. Readouts for endothelial function, blood pressure, and constrictive vascular remodeling post ligation were comparable between the 2 genotypes. Instead, transcriptome analysis of injured vessels from wild-type and HAS3-null mice revealed differential activation of pathways associated with a migratory VSMC phenotype. Further evidence from in vitro studies revealed that HAS3 overexpression in VSMCs supported a migratory phenotype in response to platelet-derived growth factor BB (PDGF-BB), whereas knockdown of HAS3 resulted in reduced PDGF-BB–induced migration. Interestingly, HAS3 knockdown also lead to a decrease in PDGF-B mRNA levels suggesting a potential autocrine loop involving HAS3, PDGF-B expression, and PDGF-BB–induced migration. Combined with the fact that the effect on hyaluronan content was most prominent in the media, the authors conclude that this axis promotes activation of a migratory phenotype in medial VSMCs and their relocation to the intima resulting in neointimal hyperplasia. On the basis of known differential effect on VSMC function, analysis of the effect of HAS3 deletion on the distribution of the various molecular weight forms of hyaluronan that accumulate in response to injury will be of interest in future studies.
Going forward, it will be interesting to see whether the injury phenotype described by the authors is directly attributable to the absence of HAS3-synthesized hyaluronan and whether its mechanism of action is via mechanotransduction versus receptor-mediated signaling. These studies also provide the foundation to extend the analysis of the effect of genetic deletion of HAS3 to even more complex pathophysiologic settings. Of particular interest would be injury in genetically prone mouse models of atherosclerosis, thus giving greater relevance to the status of patients undergoing percutaneous coronary intervention, not to mention the effect on atherogenesis itself. Nevertheless, the gene expression analysis presented by Kiene et al1 revealed upregulation of multiple inflammatory genes after injury. Intriguingly, differential induction of interleukin-13 (a profibrotic cytokine) in the absence of HAS3 suggests that a similar comparison between wild-type and HAS3-deficient atherosclerosis-prone mice could provide exciting new insights into the role of HAS3 and hyaluronan in such settings.
Drug-eluting stents that deliver immunosuppressants and antiproliferative agents have markedly reduced the onset of restenosis but possibly at the expense of increasing the risk of thrombosis. Targeting different activators of neointimal hyperplasia could, therefore, provide alternative targets for effective therapy. In this regard, it is worth mentioning that sirolimus-eluting stents have been posited to act, in part, by blocking accumulation of hyaluronan by VSMCs and reducing monocyte adhesion. This further suggests that targeting hyaluronan synthesis more directly may be worth exploring for potential translation to the clinic.
Sources of Funding
The authors’ work related to this subject has been supported by PHS grants PO1HL067663 and RO1AG0473.
- © 2016 American Heart Association, Inc.