Mechanisms of lymphatic endothelial cell junction transformations

The lymphatic vascular network encompasses a series of blind-ended, unidirectional vessels that routes fluid, macromolecules, and cells from the interstitial compartments of virtually every organ for return into the systemic circulation¹. Lymphatic vessels of the gut have specialized functions that include absorption of dietary lipids and fat-soluble vitamins, cholesterol transport, and gut immunosurveillance. This process begins with uptake of interstitial fluid by the lymphatic capillaries which coalesce and drain into collecting vessels equipped with valves that serve to propel lymph towards the thoracic duct and back into the systemic circulation. Lymphatic capillaries absorb interstitial fluid through discontinuous intercellular contacts between endothelial cells called button-like junctions, while collecting vessels, lined by less permeablezipper-like junctions facilitate the transport of lymph and prevent fluid backleak². Although several studies have revealed junctional plasticity in initial lymphatics whereby button junctions transform into zippers or vice versa the mechanisms underlying these transformations are not well understood.

In this issue of Circulation Research, Zarkada et al elegantly investigated the effects of chylomicrons on lacteal permeability and lipid absorption in the intestine³. By comparing postnatal day 0.5 (P0.5) mice fed with milk to prenatal embryonic day 20.5 (E20.5) mice that were not fed, the authors discovered that lacteals from fed mice had predominantly button-type junctions while non-fed mice had more zipper junctions. The authors hypothesized that that the presence of dietary lipids promoted junction opening, thus enabling transport of lipids from the intestinal epithelial cells into the lymphatic vascular system. These findings were confirmed in cultured LECs treated with chylomicrons or arachidonic acid, in which zipper junctions were transformed into buttons and junction-anchored stress fibers appeared in conjunction with increased phosphorylation of ROCK. Conditional knockdown of ROCK1 and 2 resulted in mutant mice with increased zipper junctions and less chyle in their mesenteric lymphatic vessels, suggesting that ROCK activity is necessary for formation of lacteal button junctions and dietary fat uptake.

This study went on to show that VEGFR2 phosphorylation was significantly reduced in lacteals from P0.5 mice compared to E20.5 mice. Loss of lymphatic Vegfr2 did not affect morphology of the lacteal junctions, but injecting exogenous VEGF-A increased the ratio of zipper-like junctions in control mice, whereas mice with conditional knockdown of Vegfr2 in LECs mostly maintained buttoned junctions. In contrast, inducible overexpression of Vegfr2 in LECs was sufficient to increase lacteal zipper junctions, an effect that was independent of VEGFR2 Y949 phosphorylation. Although VEGFR3 does not directly bind to VEGF-A, conditional knockdown of Vegfr3 in LECs resulted in more lacteal buttoned junctions after VEGF-A injection compared to the control mice. Silencing VEGFR3 did not affect VEGFR2 expression levels, however, PI3K/AKT signaling was altered suggesting the formation of VEGFR2/-3 heterodimers. VEGF-A-induced zippering of lacteal junctions was decreased by PI3K inhibitor Wortmannin and in Akt1 knockout mice, while lymphatic-specific knockdown of Plcγ1 or Erk1/2 showed no effect. These studies demonstrated that PI3K/AKT1 signaling, but not PLCγ1/ERK1/2 signaling, mediates the zippering of LEC junctions in response to VEGF-A. In vitro studies showed that VEGF-A/VEGFR2/PI3K signaling in LECs reduces stress fiber formation by inhibiting RhoA/ROCK/MLC via activation of RAC1.

Taken together, this study demonstrates chylomicrons derived from ingested lipids activate ROCK, which phosphorylates MLC leading to assembly of actomyosin stress fibers and opening of junctions, thereby facilitating lacteal uptake of chylomicrons. Lymphatic VEGFR2 is inhibited by chylomicrons, while VEGF-A signals through VEGFR2/VEGFR3 heterodimers to counteract ROCK activation and junction opening. VEGFR2/VEGFR3 activates PI3K/AKT1 and RAC1, thereby reducing RhoA and ROCK-dependent MLC phosphorylation. This sequence of events results in the relaxation of actin stress fibers, formation of cortical actin, and zippering of junctions, which prevent the uptake of chylomicrons by lacteals (Figure 1).

Figure1. Factors that promote transformation of lymphatic endothelial cell junctions.

This model outlines a mechanism by which lipids (chylomicrons) promote button junction formation in lacteals via upregulation of ROCK-mediated phosphorylation of myosin light chain (MLC) and subsequent cytoskeletal contraction. In contrast, zipper junctions are maintained via VEGF-A activation of VEGFR2/VEGFR3 heterodimers, which promotes PI3K/Akt-mediated Rac1 inhibition of cytoskeletal contraction. Junction plasticity in other lymphatic capillary beds can be influenced by several factors (infection, oxidative stress, glucocorticoids, biomechanical forces) and may play important roles in normal physiology and provide new therapeutic targets for conditions including fatty liver disease and respiratory infections.

The authors persuasively show that chylomicrons can modulate lacteal junctions and thus regulate their own absorption through ROCK-dependent cytoskeletal contractions that subsequently open lymphatic endothelial junctions. Although the focus of this work is the effects of dietary lipoproteins on intestinal lymphatic absorption, the lymphatic vascular network is also responsible for uptake and transport of lipids extravasated into interstitial compartments throughout the body. The intestinally generated chylomicrons do not have a role in lipid-lymphatic interactions in peripheral tissue, however, the principles laid out in this study may be applicable to otherreclaimed lipids that are exquisitely position to modulate the permeability of peripheral lymphatic vessels. Indeed, LDL cholesterol has been shown to affect lymphatic permeability. For instance, cultured LECs exposed to oxLDL became significantly less permeable and showed a skewed ratio of Vegfr2/Vegfr3⁴. Mice with fatty liver disease and elevated oxLDL treated with rVEGFC (cys156ser) targeting VEGFR-3 corrected the lymphatic vessel dysfunction and lessened liver inflammation and progressive hepatic fibrosis. Thus, in addition to being the primary conduit for absorption and transport of lipoproteins throughout the body, dietary lipoproteins and possibly lipoproteins within the peripheral interstitium may modulate architecture and function of the lymphatic vessels.

Beyond modulation of lacteal junctions, the authors extended their study to investigate transformation of lymphatic junctions in the skin. While lipids were not the stimulus used in these experiments, the downstream ROCK signaling pathway that was shown to control lacteal LEC button to zipper-like junctions was also found to play a role in lymphatic junction morphology in dermal initial lymphatics. Functional studies showed reduced uptake of Evans Blue dye the popliteal lymphatic network of ROCK1/2iLKO mice exposed to VEGF-A, results that paralleled decreased absorption in intestinal lacteals. These observations underscore the likely generalizability of their findings to lymphatic junction transformation across different lymphatic capillary beds. Other studies have observed that buttons in initial lymphatics are replaced by zippers during Mycoplasma pulmonis infection in mouse airways thus limiting dissemination of infection^5,6. Conversion of buttons to zippers after infection can be reversed by administration of dexamethasone ⁶. (Figure 1) Whether the conversion of zipper to button junctions in lymphatic vessels has a role in tissues clearance and anti-inflammation in the many disease conditions treated with steroids is an important clinical question that remains open.

There are few therapies that specifically target lymphatics. Much of the current focus is on pharmacotherapeutic agents that modify lymphatic growth, e.g., VEGF-C activators/inhibitors of lymphangiogenesis. Additional therapies aimed at contractile dynamics and lymphatic valve competence in the collecting lymphatic vessels are being developed. Zarkada et al highlight the potential that future therapies will target the lymphatic absorptive capacity which could lessen the impaired tissue clearance present in numerous disease conditions. Indeed, previous work from this groups has shown that converting lacteal LEC button to zipper-like junctions prevented chylomicron entry into lacteals reducing lipid absorption that protects against diet-induced obesity⁷. They now show that chylomicron-derived lipids trigger lacteal junction opening via ROCK dependent contraction of junction-anchored stress fibers. Since RhoA/ROCK is a ubiquitous pathway with many upstream stimuli that controls a wide array of downstream effects this underscores the possibility that different pharmacotherapeutic agents can be developed to target LEC junctional remodeling. The study used a variety of genetically engineered mouse models and in vitro approaches that convincingly support their conclusions. The limitations of the study include the use of neonatal animals which will require confirmation of applicability of these observations in adults. Further studies will need to determine the impact of disease conditions that affect lipids or connective tissue matrix surrounding the stress fibers pulling open the endothelial junctions.