LUMENating Blood Vessels

Abstract

The acquisition of a lumen is an essential step in vascular morphogenesis. In this issue of Developmental Cell, Xu et al. (2011) show that the small GTPase Rasip is a critical regulator of cytoskeleton dynamics and cell adhesion, which together drive the emergence of vascular lumens.

The vascular system consists of a series of interconnecting ducts responsible for the transport of blood and the trafficking of immune cells. Understanding the processes that regulate the acquisition, maintenance, and remodeling of vascular lumens is thus essential to built concrete knowledge of vascular development and function. Unlike lumen formation in epithelial tubes, the identification of the molecular players that regulate tube morphogenesis by endothelial cells has lagged behind other advances in vascular biology. In the last decade, genetic models in zebrafish and mice, and ingenious in vitro systems that recapitulate vascular morphogenesis have been developed. The combination of these systems has paved the way towards initial characterization of the process by which vessels gain lumens (Iruela-Arispe and Davis, 2009). In this issue of Developmental Cell, Xu et al. (2011) report findings that add to the understanding of lumen formation.

In vitro studies using three-dimensional systems contributed significantly to the body of knowledge regarding vascular invasion and lumen formation. These studies brought to light three categories of molecules: regulators of cytoskeletal dynamics, polarity proteins, and adhesion receptors. In particular, it was found that lumen formation requires integrin signaling, activation of Cdc42, Rac1, pak2/4, Raf kinases, and the Par3/Par6/atypical PKC complex (Koh et al., 2008; Koh et al., 2009).

More recently, our knowledge of vascular lumen formation has been expanded through the serendipitous exploration of transgenic mutants. Deletion of beta1 integrin in the endothelial cell compartment resulted in disruption of endothelial cell polarity and arrest of lumen formation (Zovein et al., 2010). Because the deletion occurred by mid-gestation (E10.5–14.5), removal of beta1 integrin was found to manifest differently in newer vessels and in established endothelium. Loss of beta1 integrin in the established endothelium resulted in endothelial stratification and cuboidal cell shapes. In contrast, absence of lumen was evident in small arteries, which exhibited mislocalization of cell-cell adhesion molecules and apical endothelial adhesion with obliteration of lumens. Further mechanistic analysis identified the polarity protein Par3 (pard3) as a critical downstream regulator. In fact, overexpression of Par3 was able to partially rescue the polarity and lumen defects imparted by lack of beta1 integrin (Zovein et al., 2010). These results are consistent with findings from in vitro models. They also indicate that changes in adhesion and lumen defects are downstream of polarity cues and that these in turn are triggered by beta1 integrin.

Xu et al. (2011) now present work that uncovers additional molecular players and adds to the understanding of the contribution of the cytoskeleton in lumen formation. In particular, the group demonstrated that Rasip1, an endothelial-specific Ras interacting protein, is required for organization of lumens during development. Genetic inactivation of Rasip1 in mice blocks the emergence of patent lumens in all blood vessels leading to developmental arrest and lethality by E10.5. Two binding partners of Rasip1 – non-muscle myosin heavy chain II (NMMHCII) and Arhgap29, a RhoA specific GAP – were identified with obvious roles in the process. These findings clearly direct attention to the cytoskeletal apparatus. Indeed, the lack of Rasip1 upset the balance between Cdc42 / Rac1 and RhoA, with mutant cells showing high levels of actin stress fibers and decreased acetylated tubulin. Furthermore, Rasip1/Arhgap29 work together to suppress ROCK, Myosin Light Chain, and NMMHCII, keeping the actomyosin complex in a state of adequate contraction. Depletion of Rasip1 or Arghgpa29 blocked in vitro lumen formation by altering the cytoskeleton, increasing contraction, and reducing integrin-dependent adhesion. Indeed, the authors found a significant reduction of beta1 integrin activation (by 70–80%) but observed no change in integrin protein levels. These results suggest that Rasip1 contributes integrin signaling. In agreement with previous findings (Zovein et al, 2010), the authors also observed defective Par3 trafficking/localization and ectopic tight junction distribution. An emerging theme from both studies is that endothelial cell polarity is a critical first step in lumen formation. In this context, Rasip1 appears to function as a signaling node essential for coordinating cytoskeleton dynamics and cell adhesion that lead to cell polarity and lumen formation (Figure 1).

Figure 1. Acquisition of endothelial cell polarity and lumen formation.

Prior to lumen emergence, vessels consist of solid cords of cuboidal endothelial cells that express cell-cell adhesion molecules (ZO1, VE-Cadherin) in the apical, as well as lateral aspects of each cell. The formation of a lumen is triggered by activation of integrins, cytoskeletal dynamics, and rearrangement of cell-cell adhesion molecules from the apical to the lateral cell membrane with concurrent changes in cell shape. Rasip and Arhgap29 participate in the activation of the cytoskeleton by reducing levels of active Rho and increasing levels of Cdc42 and Rac. In addition, Rasip alters integrin activation status and distribution of Par3 to junctional complexes. The effect on Par3 is likely indirect (dotted arrow).

While Xu et al. (2011) has assembled another big piece of the “lumen puzzle”, multiple questions still remain. For example, how does endothelial cell shape and trafficking of membranes change during lumen formation? How is the movement of cell-cell adhesion molecules coordinated and correctly routed to the lateral aspects of the membrane in the process of lumen formation? What are the layers of regulatory cross talk between cell-matrix and cell-cell interactions? Do all vascular lumens utilize the same molecular pathways in their formation? On this last question, similarly to epithelial tube morphogenesis, it is unlikely that all vessels will form lumens in the same fashion. Data from in vitro systems and zebra fish, for example, indicate that some capillaries are likely to gain lumens via fusion of internal vesicles (Kamei et al., 2006). In contrast, the aorta, in a manner seemingly distinct from other vessels, forms by the coalescence and subsequent fusion of two strings of vascular cords. The transition from a cord to a tube is rapid and dependent on the processes described by Xu et al. (2011), but expansion of the lumen and the accumulation of fluid in the lumen appears to require the polarized expression of the sialomucins CD-34 and podoxalyxin (Strilic et al., 2009). Coordination of actin contraction and apical distribution of negatively charged sialomucins result in the repulsion of juxtaposed apical membranes through the accumulation of interstitial fluid in the emerging lumen (Strilic et al., 2009).

As additional studies advance our understanding of lumen formation, it is clear that these insights into the nuances of the regulatory process will continue to add complexity and clarify organ or vessel-specific differences. While these studies better our understanding of how vessels acquire lumens, an equally pressing question pertains to how the dynamic control of vessel diameter is sensed and regulated. Blood vessels are remarkable structures with the strength to absorb pressure and the flexibility to adapt to homeostatic needs. Unraveling the operative molecular web that coordinates alterations in vessel diameter will clearly be our next challenge in understanding vascular development.