A new conserved player in lymphatic morphogenesis

The lymphatic system comprises a network of interconnected and blind-ended vessels whose central function is to maintain fluid homeostasis¹. Similar to the vessels of the circulatory system, the lymphatic system consists of a hierarchical network of small and large caliber vessels, which function together to allow fluid drainage. Lymphatic capillaries are lined by endothelial cells that are highly discontinuous, making them permeable to surrounding fluid to facilitate resorption. By contrast, larger lymphatic collecting ducts possess a thin smooth muscle layer and valves to promote unidirectional transport of fluid towards connection points with the circulatory system at the subclavian veins. In addition to fluid homeostasis, the lymphatic system and the constituent lymph nodes are a portal for lymphocytes, as well as antigen-presenting cells, while also functioning in lipid absorption in the digestive system¹. Given their wide-ranging importance in these normal physiological processes, it is not surprising that the lymphatic system is implicated in a number of pathological conditions. Indeed, inflammation is intimately associated with lymphatic function and is a central cause of lymphedema in a wide range of acquired and congenital diseases¹. Notably, congenital forms of lymphedema are associated with mutations in a number of genes important for lymphatic vessel formation, including those encoding the receptor tyrosine kinase FLT4 and its ligand, VEGFC².

The lymphatic system is also essential for cardiac homeostasis and has been increasingly implicated in related diseases. Importantly, lymphatic vessels can play a central role to facilitate repair of cardiac tissue following myocardial infarction (MI)³. Indeed, injection of VEGFC following MI can lead to sustained improvement in cardiac function, including increased ejection fraction, and reduction in both end systolic volume and hypertrophy. In these cases, the presence of additional lymphatic vessels induced by VEGFC is thought to better resolve post-MI edema and inflammatory responses, both of which contribute to fibrosis³. In addition to a possible direct therapeutic role for lymphatic vessels in the heart, they may also be beneficial in the context of atherosclerosis. Several studies have found that improving lymphatic function is generally associated with increased clearance of cholesterol from tissue, which in turn could promote reverse cholesterol transport to reduce circulating levels³. Importantly, improved lymphatic function could facilitate cholesterol removal from the arterial wall, the major site of atherosclerotic plaque formation in the circulatory system. Thus, therapeutic growth of new lymphatic vessels could have multiple beneficial effects on the progression and resolution of heart disease³. Given their potential in this regard, a better understanding of the basic mechanisms that govern lymphatic vessel growth is essential.

Fortunately, the lymphatic system is conserved in vertebrates, making it feasible to utilize model organisms to gain a better understanding of how lymphatic vessels develop and function ⁴. In particular, major insights into lymphatic morphogenesis have been revealed from developmental studies in both mouse and zebrafish models. In mouse, lymphatic specification is initiated at embryonic day 9 with the induction of the transcription factor Sox18 in selected endothelial cells lining the cardinal vein ⁵. Sox18 induces the expression of the Prox1 transcription factor in these lymphatic progenitors, while Vegfc in the surrounding mesenchyme subsequently induces their migration out of the vein to form primitive lymph sacs ⁶. A similar process occurs in the zebrafish trunk, although Sox18 appears to be dispensable in this case ^{7, 8}. Much like the developing circulatory system, primitive lymphatic vessels subsequently undergo extensive remodeling and maturation to form a functional network, including development of mature collecting ducts and formation of valves. Similar to blood vessel maturation, this process is driven largely by signaling through the Angiopoietin/Tie and Ephrin/Eph receptor tyrosine kinase pathways⁴. Subsequently, valve formation is initiated by the coordinated induction of a transcriptional regulatory network including Gata2, Foxc2, and Prox1 ⁴. In turn, these transcription factors induce the expression of extracellular matrix components, such as integrin α9, that are essential for valve leaflet morphogenesis⁹ and subsequent lymphatic function. Accordingly, human mutations in GATA2 or FOXC2 can lead to congenital defects in lymphatic function and lymphedema ².

In this issue of Circulation Research, two groups have taken advantage of both zebrafish and mouse models to identify a new molecule that is essential for proper formation of the lymphatic system^{10, 11}. In the course of a forward genetic screen, Karpanen et al. identified a zebrafish mutant that failed to form a thoracic duct, an early lymphatic vessel formed during development in this model¹¹. This mutant also exhibited edema, suggesting a deficiency in lymphatic function. Subsequent positional cloning of the mutation revealed a premature stop codon in the svep1 gene, which encodes Polydom, a large multi-domain protein that plays a role in cell adhesion to the extracellular matrix¹². Rescue experiments with bacterial artificial chromosome (BAC) transgenes, as well as mRNA, confirmed that lymphatic defects were due to loss of svep1 function. In parallel, Morooka et al. applied genome editing to generate a targeted deletion in zebrafish svep1, which led to the same defects in thoracic duct formation¹⁰, further confirming the requirement for this gene in lymphatic development. More careful assessment of the mutant phenotype in zebrafish revealed that early lymphatic specification was unaffected, with Prox1 expression apparent in progenitors in the cardinal vein, although sprouting of these lymphatic progenitors was deficient. This phenotype was much milder that that associated with loss of Vegfc or its receptor, Flt4, suggesting that Svep1 acts independently of this pathway. Indeed, activation of the serine/threonine kinase ERK, an important downstream effector of Vegfc, was not changed in the lymphatic progenitors of svep1 mutants¹¹.

Similar phenotypes were observed by both groups in svep1 knockout mice as well. Previous work from the Sekiguchi Lab had identified Svep1 as a high-affinity ligand for the integrin α9β1¹², suggesting it may play a role in lymphatic valve formation, hence their interest in determining its functional role in lymphatic development. In mice lacking svep1, both groups noted severe edema and early postnatal death suggestive of lymphatic defects. Accordingly, svep1-deficient mouse embryos displayed defects in lymphatic patterning in the skin, although early specification of lymphatic progenitors from the cardinal vein appeared normal, similar to results from the zebrafish studies. Analysis of mesenteric lymphatics revealed a failure to form collecting ducts and absence of valves. Similar observations were made by both groups in a number of different anatomical locations, including the intestine and heart, suggesting a global requirement for svep1 in lymphatic maturation and valve morphogenesis. Together, these observations suggested that svep1 is dispensable for initial specification of lymphatic progenitors, but is required at later stages during lymphatic maturation and valve formation.

A striking finding made in both papers is that svep1 was not expressed by endothelial cells, but was rather found in a closely associated peri-endothelial cell. Using recombinant BAC-mediated transgenesis in zebrafish, Karpanen et al. observed that svep1-expressing cells were present in the vicinity of lymphatic progenitors as they arose from the cardinal vein. These cells remained in close association with lymphatic endothelial cells as lymphatic morphogenesis proceeded. Similar observations were made by Morooka et al. in both zebrafish and in mice, using a lacZ knock-in allele in the latter case, as well as immunostaining of endogenous Svep1 protein. Neither group was able to definitively determine the identity of these cells, although they did not appear to be neural or muscle cells. Interestingly, ccbe1 (collagen and calcium-binding EGF domain-containing protein 1), which was likewise identified through forward genetic screening for lymphatic mutants in zebrafish, is also known to be required in a non-endothelial cell autonomous manner and is expressed in cells adjacent to the developing lymphatic system¹³. Whether ccbe1 and svep1 are expressed in the same population is not clear from the present studies. However, these observations suggest the existence of an important peri-endothelial support cell that may be specifically required to pattern the early lymphatic vasculature.

How might Svep1 promote lymphatic maturation and valve morphogenesis? As noted above, Morooka et al. initially identified Svep1 as a binding partner for integrin α9β1, which itself is required for lymphatic valve morphogenesis⁹. However, svep1 mutant mice show much broader defects in the lymphatic system that those associated with integrin α9-deficiency¹⁰, while integrin α9 mutant zebrafish do not recapitulate the svep1 mutant phenotype¹¹. These results suggest that Svep1 acts independently of integrin α9 during lymphatic maturation, although a functional interaction during valve formation cannot be ruled out. Interestingly, Morooka et al. found that Svep1 can bind directly to Ang2, which is essential for lymphatic maturation and valve morphogenesis¹⁴. Moreover, expression of a number of genes important for valve formation, including Foxc2 and Tie1 were reduced in svep1-deficient mouse embryos. These observations suggest that Svep1 may play a role in signaling through the Ang/Tie pathway, although further biochemical confirmation is needed to support this possibility. An intriguing consideration in light of these studies is the possible role of Svep1 in transducing mechanosensitive forces to promote lymphatic maturation and valve formation. After early specification of lymphatic progenitors and formation of nascent lymphatic sacs, much of the expansion and maturation of the lymphatic system is driven by a variety of mechanosensory inputs¹⁵. Importantly, integrin β1, which can also mediate adhesion via Svep1¹², is essential for early steps in mechanosenory-induced lymphatic expansion and maturation. While purely speculative, it will be interesting to determine whether integrin β1 function in this context involves interaction with Svep1 during lymphatic development.

Our understanding of the signals and mechanisms that govern development of the lymphatic system still lags considerably behind what we known concerning blood vessels. However, the last decade has seen major advances in our knowledge of how this essential system is formed. Both the zebrafish and mouse models have greatly aided in these studies. In particular, the zebrafish has provided a powerful genetic platform to identify of novel lymphatic genes, such as ccbe1. At the same time, the mouse model enables detailed assessment of lymphatic function in the context of development and disease. Importantly, both models together have yielded crucial insights into causative genes responsible for congenital lymphedema in humans. As such, moving forward it will be important to investigate the possible involvement of new genes, such as Svep1, and their role in the pathology of congenital lymphedema in humans.