Download What Determines Blood Vessel Structure? Genetic Prespecification

REVIEWS
PHYSIOLOGY 21: 388–395, 2006; 10.1152/physiol.00020.2006
What Determines Blood Vessel Structure?
Genetic Prespecification vs.
Hemodynamics
Vascular network remodeling, angiogenesis, and arteriogenesis play an important
role in the pathophysiology of ischemic cardiovascular diseases and cancer. Based on
recent studies of vascular network development in the embryo, several novel aspects
Elizabeth A. V. Jones,1, 2
Ferdinand le Noble,3
and Anne Eichmann2
1INSERM U36, Paris, France;
de France, Paris, France; and
3Laboratory for Angiogenesis and Cardiovascular Pathology,
Max Delbrück Centrum für Molekulare Medizin,
Berlin-Buch, Germany
[email protected]
2Collège
The online version of this article
contains supplementary data.
to angiogenesis have been identified as crucial to generate a functional vascular network. These aspects include specification of arterial and venous identity in vessels and
network patterning. In early embryogenesis, vessel identity and positioning are genetWe demonstrated that, during later stages of embryogenesis, blood flow plays a crucial role in regulating vessel identity and network remodeling. The flow-evoked
remodeling process is dynamic and involves a high degree of vessel plasticity. The
open question in the field is how genetically predetermined processes in vessel identity and patterning balance with the contribution of blood flow in shaping a functional vascular architecture. Although blood flow is essential, it remains unclear to what
extent flow is able to act on the developing cardiovascular system. There is significant
evidence that mechanical forces created by flowing blood are biologically active within the embryo and that the level of mechanical forces and the type of flow patterns
present in the embryo are able to affect gene expression. Here, we highlight the pivotal role for blood flow and physical forces in shaping the cardiovascular system.
Formation of the vascular tree is the result of a complex interplay between genetic factors and epigenetic
factors including hemodynamics and tissue oxygenation. The process of cardiovascular development
occurs in two steps, vasculogenesis and then vascular
remodeling. During vasculogenesis, cells are specified
as angioblasts, or endothelial cell precursors, that coalesce to form lumenized tubes. These lumenized tubes
form an initial network for blood flow. During remodeling, this network becomes a hierarchical vasculature
tree composed of arteries, veins, and capillaries. The
initial specification of angioblasts is believed to be
controlled by genetics, since blood flow is not present
at this stage, but the role of environmental inputs such
as mechanical forces and hypoxia in the remodeling
phase remains unclear (48). More explicitly, the specification of arteries and veins was initially believed to
be a result of flow; however, this theory has recently
been put into question since studies in zebrafish and
chick embryo indicate that angioblasts can express
arterial and venous markers before the formation of
tubes and the onset of flow (50). In fact, proper attribution of arterial-venous identity in early embryogenesis is essential for vascular remodeling. Without
388
specification, expansion into a branched vascular network is impeded. There is therefore now reason to
believe that genetically predetermined components
must play a role in cardiovascular remodeling and not
just the initial specification of angioblasts (for review,
see Ref. 24). The question from a biophysical point of
view, however, is whether the optimization of the vascular tree to be both hemodynamically efficient (to
reduce the load on the heart) and functionally efficient
(to deliver nutrients and oxygen) can be preprogrammed and, hence, controlled by genetically driven
processes.
Environmental Cues in Remodeling
Two independent locations of vasculogenesis occur in
the embryo: the blood islands in the extra-embryonic
membranes and the major vessels in the embryo
proper. The blood islands consist of hematopoietic
cells surrounded by endothelial cells and form in the
distal part of the yolk sac (FIGURE 1A). These
endothelial cells of the blood islands expand from a
thin band of endothelial and hematopoietic cells to
cover the entire yolk sac, forming a vascular network
1548-9213/05 8.00 ©2006 Int. Union Physiol. Sci./Am. Physiol. Soc.
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.1 on May 5, 2017
ically hardwired and involve neural guidance genes expressed in the vascular system.
REVIEWS
morphogenesis. Angioblast specification by ephrins
and other arterial and venous genes is needed to allow
arterial-venous differentiation, without which remodeling cannot occur. However, during later stages of
development, it is readily observed that arterial
endothelial cells are reused in growth of veins (chick
embryo or zebrafish). This leaves open the question:
To what extent is genetic prespecification before start
of flow needed, and how does it affect the final architecture?
The problem with viewing these genetics aspects of
remodeling in isolation, without the role of epigenetic
influences, is that the presence of flow is known to be
essential for remodeling to occur. Nearly a hundred
years ago, Chapman showed by surgically removing
the heart of chick embryos before the commencement
of circulation that the peripheral vasculature formed
but failed to remodel without blood flow and pressure
(6). Manner et al. surgically removed the heart of
young chicken embryos and incubated the embryos in
high levels of oxygen to remove the effects of hypoxia
and found that, although there were fewer deformities
in the embryo proper, the remodeling of the vasculature still did not occur (31). In the Ncx1 knockout
mouse, the heart is formed but does not beat. In these
embryos, the plexus forms in the yolk sac but is never
remodeled into vessels, even though other aspects of
development, such as limb and organ development,
are normal (49). At present, it is widely accepted that
there must be blood flow in an embryo for remodeling
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.1 on May 5, 2017
known as the capillary plexus (FIGURE 1B). Within
the embryo proper, the major vessels are formed
through the differentiation and migration of
angioblasts. These two locations eventually interconnect, allowing for a functional loop to be present in
which blood can flow. As the heart begins to beat,
blood flow is introduced into this network, and the
vessels remodel into a more hierarchical vascular tree
(FIGURE 1C). Remodeling involves changes in vessel
diameter, branching morphology, and recruitment of
peripheral cell types (pericytes, vascular smooth muscle) that stabilize the nascent vessels.
The evidence that remodeling of the vasculature,
and therefore the final vascular architecture, is genetically predetermined derives from the observation that
certain genes are selectively expressed in either arterial or venous vascular compartments. Before the onset
of blood flow, expression of proteins from the ephrin
or neuropilin family (51) demarcate arterial and
venous angioblasts. Mutant mice for ephrinB2 and
ephB4 die because of failure to form a proper arterialvenous network (17, 50). Many of these arterial and
venous genes, such as Unc5B, are also expressed in
endothelial tip cells that are involved in branching
morphogenesis decisions in the early vasculature, and
exposure of endothelial cell filopodia to ligands of
these genes cause retraction of the filopodia (30). It is
therefore postulated that arterial and venous genes
expressed in the vascular system may control two
crucial events: vessel identity and branching
FIGURE 1. Vascular development in the embryo and extraembryonic membranes
The vascular system forms in two specific sites in the embryo: one in the blood islands of the yolk sac, the other in the embryo proper. Formation of
extraembryonic blood vessels consists of 1) endothelial precursor specification and 2) expansion and interconnection of these precursor cells to form
a honeycomb-shaped network called the capillary plexus. The expansion of the endothelial cells (blue) from the blood islands occurs much earlier
than the expansion of the erythroblasts (red). The vessels in the embryo proper are specified and then migrate to form two lumenized vessels, called
the dorsal aorta and cardinal vein. After the initiation of blood flow, the vessels remodel to a more hierarchical tree.
PHYSIOLOGY • Volume 21 • December 2006 • www.physiologyonline.org
389
REVIEWS
transport by the blood, early stages of remodeling
occur normally. Hypoxia increases the number of
blood vessels by inducing sprouting. The heart can
increase perfusion volume and pressure by increasing
stoke volume or heart frequency; however, both
parameters are limited in their degree of adaptation,
and there needs to be a method to prune vessels, not
only create new vessels. The final vascular structure
must be an optimization between regulation of
peripheral oxygen diffusion and cardiac activity. Such
optimization principle has been put forward several
decades ago as Murray’s law and seems to hold in
many flow transportation systems (37). For a functional vascular system to be established, a mechanism to
reduce vascular volume and prune inefficient vessels
must therefore also be present for remodeling to
occur.
The initiation of blood flow creates mechanical
forces on the developing endothelial cells. As a fluid
flows through a tube, it exerts a force tangential to the
tube, called shear stress, and another that is perpendicular to the tube wall and is caused by the pressure in
the vessel, called circumferential stretch (FIGURE 2).
Mature vessels can react to both these forces. Shear
stress from blood flow has been shown to cause
changes in morphology, cytoskeleton organization,
ion channel activation, and gene expression within
endothelial cells in vitro (7). Observation of endothelial cells in flowless embryos, such as the Titin–/–, show
that endothelial cells are more globular and do not
flatten, reminiscent of cell shapes associated with
migrating angioblasts rather than differentiated
endothelial cells (32). Many genes upregulated by
abnormal flow patterns in adult circulatory systems
are important to cardiovascular development, such as
PDGF-␤ (41), connexin43 (14), and Flk1 (14). Our
work in the mouse embryo has shown that the levels of
shear stress present during the remodeling process are
within the range known to activate these pathways
(21). Pressure also activates certain genetic pathways
(reviewed in Ref. 28) and leads to changes in smooth
muscle cell coverage in adult blood vessels (5). In
these mature systems, chronic elevations in shear
stress lead to increased vessel diameter, whereas
chronic increases in pressure lead to decreased vessel
diameter (13, 36). This work has led to a pressureshear hypothesis, whereby vessels adapt to shear
stress as a function of local pressure (40). Thus the two
blood flow-induced forces, pressure and shear stress,
are associated with long-term rearrangement of the
FIGURE 2. Forces created by flowing blood
cardiovascular
system in adults. The question
Fluid flow within a tube creates two types of forces: shear
remains,
however,
whether embryonic vasculature has
stress, which is a force tangential to the vessel wall, and
pressure, which creates a circumferential stretch perpendi- the same mechanisms as adult vasculature to alter in
cular to the vessel wall. The velocity profile (shown by the
arrows) within the vessel and the viscosity of the fluid deter- response to these forces. It is known that postnatal
arterial development is affected by shear stress (9, 52),
mine the amount of shear stress to which the endothelial
cells are exposed. The pressure at any given point in the
although the stage of development at which shear senvasculature is related to pressure at the outlet of the heart
sitivity is initiated is not known. The response to presand the pressure losses that occur given the flow path to
sure in adults is largely driven by smooth muscle cells.
that point.
390
PHYSIOLOGY • Volume 21 • December 2006 • www.physiologyonline.org
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.1 on May 5, 2017
to occur. The essential signal that flow imparts, however, is not known. Blood flow carries nutrients, oxygen,
and signaling molecules to the vessels as well as creating physical forces on the endothelial cells of the vessel
wall. Therefore, the initiation of blood flow brings
many different signals to the embryo.
Reductions in tissue oxygen availability and hypoxia
have been linked to changes in network configuration.
Hypoxia, an environmental and therefore nongenetic
signal, is known to be able to trigger vessel sprouting
into avascular regions through induction of VEGF production (33, 39). VEGF is one of the most important
proteins during vascular development and can induce
migration, proliferation, and the survival of endothelial cells (12). Hypoxia-induced genes are regulated by
a transcription factor known as hypoxia inducible factor-1, or HIF-1. Embryos that lack HIF-1␣ fail to
undergo the remodeling process (45). Paradoxically,
the oxygen-carrying ability of the blood does not seem
to be required for remodeling during the first few days
in which the heart is beating. In the yolk sac, the mesenchymal tissue and the blood vessels of the yolk sac
have equal access to oxygen since the yolk sac is the
link between the maternal and embryonic environment before the formation of the placenta. When
zebrafish (38), Xenopus (48a), chick (8a), or mouse
embryos (E8.5–E9.5; unpublished results) are cultured
in carbon monoxide, effectively ablating oxygen
REVIEWS
the cytoskeleton (42). How a cell attaches to surrounding cells and the composition of the extracellular
matrix, something that is genetically determined,
therefore, mediates how cells react to flow. It has
recently been shown that cotransfection of only three
proteins (VE-cadherin, PECAM-1, and VEGFR2) can
endow nonendothelial cells with the ability to align
with the direction of flow in vitro (48a). The flow that
endothelial cells are exposed to is also genetically
determined by the development of the heart. Since
oscillatory flow is more biologically active, the formation of heart valves, which stops retrograde or backward flow, induces an important change in the hemodynamics of the early embryo (21). Thus, even when
an external nongenetic signal such as flow is considered, it is important to keep in mind that it is the genetics of cell fate determination that set up the initial conditions to which the cells are reacting and that determine the type and location of these forces.
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.1 on May 5, 2017
These cells are not present initially on embryonic
blood vessels, and how pressure would affect a bare
vessel segment is not known. Therefore, adult vasculature and embryonic vasculature may not react similarly to these forces, but it seems unlikely that hemodynamic forces would have such profound effects on the
geometry of adult vasculature and be completely inert
within the embryo.
Understanding the role of shear stress and forces in
biology is complicated by the fact that different types
of flow seem to induce different patterns of gene
expression (for review, see Ref. 42). It is therefore not
sufficient to analyze the level of shear stress present;
one must also look at whether the flow is laminar or
turbulent and at the role of oscillatory shear stress. The
presence of laminar flow, which is a type of fluid flow
where the streamlines of the fluid motion are parallel,
has been found to have atheroprotective effects on
mature blood vessels (4). Laminar shear stress reduces
levels of apoptosis (10) and induces many anti-apoptotic genes, such as Bcl-XL (3). Although laminar flow
prevents apoptosis, it also keeps cells from proliferating by inhibiting DNA synthesis (2, 29). Through
microarray analysis, it appears that physiological levels of laminar or turbulent shear stress downregulate
more genes than it upregulates and that gene expression profiles are significantly different depending on
the type of flow endothelial monolayers are exposed to
(16). Disturbed flow, a term that includes both turbulent flow and eddies caused by laminar flow separation,
is a much more biologically active flow. It is rarely seen
in the mammalian cardiovascular system and is generally indicative of disease. Vessel bifurcations are
prone to flow separation and eddy formation, and it
has been observed that atherosclerotic plaques
formed preferentially at these locations (9).
Microarray analysis has identified genes, such as
KLF2, that are differentially expressed when endothelial cells are exposed to flow typical of arteroprotected
regions of the vascular tree compared with flow profiles similar to arteroprone regions (37b). KLF2 is
essential for the upregulation of many genes associated with endothelial response to disturbed flow, such
as eNOS. In zebrafish embryos, KLF2 is upregulated
after the onset of flow but absent in zebrafish mutants
that lack blood flow (37b), and lack of this protein
results in embryonic lethality in mice (50a). Disturbed
flow is not, however, indicative of disease within the
embryo and has been observed in normal development (21). Disturbed flow is important during the formation of cardiac valves in zebrafish embryos (18a).
Only by localized assessment of gene expression
changes in regions where disturbed flow is observed
can a possible role for flow patterns in embryonic
development be assessed.
The role of hemodynamics does not negate a role for
genetics in the remodeling process. The signals that
are created by flow are sensed through integrins and
“Recent evidence has shown that proper
establishment of vessel identity, whether
arterial, venous, or capillary, is intrinsically
linked to the remodeling process.”
Plasticity in the Embryonic
Vasculature
Recent evidence has shown that proper establishment
of vessel identity, whether arterial, venous, or capillary,
is intrinsically linked to the remodeling process. Vessel
identity can be defined by a set of markers, such as
ephrinB2 (50) and Dll4 (11, 15, 46) for arteries and
EphB4 (50) and neuropilin-2 (18) for veins. These genes
are believed to be important in guidance, boundary formation between arteries and veins, and other functions.
Abolition of these genes interferes with vascular remodeling in the yolk sac vessels (1, 11, 23, 50). The importance of these genes is not clear, however, since changes
in cardiac activity also interfere with yolk sac vascular
remodeling, even if arterial and venous markers are
properly expressed. This indicates that arterial/venous
identity alone is not sufficient to induce proper branching morphogenesis. Instead, branching requires a physical cue generated by hemodynamics and cardiac output regulation. Recent evidence has revealed a high
level of plasticity within the networks, such that flow
dynamics are able to change identity (27, 35). The idea
of flow-driven plasticity is essential, because plasticity
could in fact be the vital link between the genetic theories and epigenetic theories of vascular formation.
In avian models, endothelial plasticity was demonstrated in transplantation studies with isolated arterial
and venous vessels of quail embryos grafted into a
chick embryo host (FIGURE 3A). The cell fate of the
PHYSIOLOGY • Volume 21 • December 2006 • www.physiologyonline.org
391
REVIEWS
FIGURE 3. Experimental manipulation of arterial-venous fate
Quail arterial cells are grafted into a chick host between day 2 and day 14 of
gestation. Before day 7 of gestation, these cells, which can be identified by antibody staining for quail-specific proteins, can be found in the aorta or the veins of
the chick host. Similarly, if venous cells are grafted, they migrate to either the
artery or vein of the host before day 7. These cells change expression of arterial
and venous markers to match the new environment in which they are found (A).
Arterial and venous markers can also be changed by rerouting flow (B). In chick
embryos, if the major artery (red) is ligated, flow reroutes such that venous flow
passes through these vessels. In response, these endothelial cells begin to
express venous marker (black) and downregulate arterial markers (red).
392
PHYSIOLOGY • Volume 21 • December 2006 • www.physiologyonline.org
cells expressing arterial or venous markers are localized in a posterior-arterial and anterior-venous pole.
After the onset of cardiac activity and the start of perfusion, the vitelline artery forms in the posterior arterial pole by flow-driven fusion of preexisting capillaries branching from the aorta at the level of somite 21.
During the subsequent stages, the arterial network
expands and some small capillary side branches are
selectively disconnected from the arterial network.
These segments do not regress or apoptose; instead,
the disconnected vessels reconnect to the venous
plexus. The disconnected segments lose their arterial
identity and start to express venous markers.
Detailed high-magnification intra-vital imaging
shows that the disconnected segments, which are
filled with blood and pressurized due to connections
with more distal parts of the arterial network, grow
and extend over the arteries before reconnecting to
the veins of the primary network (26, 27). The disconnected segment lacks tip cells or filipodia and is
instead driven by its luminal pressure and guided by
the strain fields generated by the surrounding large
caliber vessels (37a). The relatively high pressure in
the arteries repels the expanding disconnected segments, which avoid the arteries and can only reconnect to lower pressure veins. Such avoidance of the
arterial segments is also observed in the zebrafish
parachordal vessel, which starts at the venous level,
crosses the intersegmental artery without fusing to it,
and reconnects to the cardinal vein in the trunk (20).
Arterial and venous identity can also be altered by
changing the flow environment of the endothelial cells
(FIGURE 3B). Rerouting flow by artificially obstructing
the vitelline artery results in perfusion of the arterial
tree with blood of venous origin, which transforms the
arteries into veins, both morphologically and with
respect to gene expression. Perfusing veins with arterial blood can likewise transform them into arteries
(27). It is undetermined whether flow-regulated vessel
identity is controlled by the physical properties of the
flow signal itself (related to the pressure profile) or by
chemical differences in arterial and venous flow such
as oxygen tension. Remarkable differences in pressure
profiles between arteries and veins exist, but any
molecular signaling cascade able to recognize such
differences is unknown, and the levels present in the
embryo are significantly lower, around 1–2 mmHg,
compared with the adult (18b). Therefore, any genetic
cascade directly activated by arterial or venous profiles in the adult may not activate similarly in the
embryonic situation. Even if pressure profiles cannot
be directly “sensed,” plastic deformations of endothelial cells and the cytoskeleton, the concept of tensegrity (19), may provide a rational framework to explain
changes evoked by arterial or venous flow, even in the
relatively low-pressure environment of the embryo.
In vivo studies of the developing trunk vasculature
in zebrafish have shown that endothelial cell plasticity
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.1 on May 5, 2017
quail cells could be assessed using the monoclonal
antibody QH1, which selectively recognizes quail
endothelial cells. In early stages of development, quail
endothelial cells of either arterial or venous origin
could colonize both chick arteries and veins (35). The
expression of arterial- and venous-specific genes
changed according to the vessel identity where the
grafted endothelial cells integrated. This plasticity
was, however, progressively lost at later stages of
development. After embryonic day 11, grafts of arteries
yielded endothelial cells that colonized only arteries,
whereas venous grafts colonized mainly host veins.
Signals in the vessel wall may at least partially regulate
endothelial identity. When arterial endothelial cells
from later stage embryos were dissociated from the
vessel wall, these cells colonized both host arteries
and veins. The responsible paracrine signaling pathways remain to be determined but could involve
nerves since functional innervation of the arterial wall
starts around that stage (43, 44).
Using in vivo time-lapse imaging of developing
chick embryos, we recently demonstrated that vessel
plasticity is controlled by hemodynamics (27). We
observed that prior to the onset of flow, endothelial
REVIEWS
for the formation of a functional vascular loop.
Taken together, arterial-venous differentiation and
branching morphogenesis in the yolk sac plexus is
flow driven and requires a high degree of endothelial
cell plasticity. In perfused networks, the expression of
arterial- and venous-specific genes is the consequence of local hemodynamic cues that may regulate
vessel wall remodeling.
Remodeling as a Force Equilibrium
One remaining factor that must be considered with
regard to vascular remodeling is the principle of force
equilibrium and changing tissue material properties.
Flowing blood creates a force on surrounding mesenchyme. The mesenchyme itself exerts a force
through tissue growth onto the vessels. As the physics
of the situation change, the conformation must be
energetically feasible or else the vessel location would
require an active and energy-consuming effort by the
tissues for the entire lifetime of the organism.
Mesenchymal tissue growth within the yolk sac is
rarely considered when remodeling is studied. The
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.1 on May 5, 2017
is not only present in the yolk sac but also within the
vascular network of the embryo proper. In early stage
embryos, primary vessel sprouts originating from the
aorta grow dorsally between the somites to form the
intersomitic vessel, which subsequently branches laterally into the caudal and cranial regions to form the
dorsal lateral anastomosing vessel (DLAV) (25). At this
stage, the intersomitic vessel is connected to the aorta
and is therefore of arterial origin. During subsequent
stages, secondary sprouts form from the cardinal vein
and grow parallel and in close proximity to the intersomitic artery. Occasionally, these secondary sprouts
fuse with the preexisting intersomitic artery, which
progressively loses its connection with the aorta as the
intersegmental vessel turns into a vein. As a consequence, the flow direction in the intersomitic arteries
is reversed (20). The exact nature of this process is not
understood, but it is possible that the fusion of the secondary sprouts with the intersomitic artery branching
from the aorta is stochastic, needed to balance local
perfusion and probably related to local pressure profiles (20). Although these vessels are initially arterial,
plasticity during the remodeling process is essential
FIGURE 4. Time-lapse microscopy of murine yolk sac vasculature during early remodeling
Endothelial cells of the yolk sac were followed through time-lapse microscopy (A–D; movie 1, online) using a transgenic mouse that expresses GFP (green) in its endothelial cells (34), using a technique previously described (22). The
embryo is initially at approximately the 8 somite stage. Extensive expansion of mesenchymal tissue is present, and
some vessels of the yolk sac regress during the process (arrows). Expansion of the mesenchymal tissue could, through
this mechanism, alter flow trajectories in the early yolk sac (E). Images were taken every 10 min at ⫻10 magnification
on a Zeiss LSM5 PASCAL for a total duration of 11.7 h.
PHYSIOLOGY • Volume 21 • December 2006 • www.physiologyonline.org
393
REVIEWS
Conclusions
The remodeling and maturation of the embryonic cardiovascular system is a complicated process requiring
both genetic and environmental inputs. The initial
specification of vasculature is genetic predetermination to enable guidance and formation of proper connection between vessels. Remodeling, however, is both
genetic and environmentally controlled. Ablation of
certain genes results in a failure of remodeling, but lack
of physical cues, whether oxygen tension or hemodynamics, also causes failures in vascular remodeling. It
is very difficult at present to separate genetics factors
from the epigenetic factors, since most genes
expressed in the vasculature are also expressed in the
heart. If the abnormal cardiac expression causes even
small changes in the flow conditions, this could result
in significant vascular abnormalities. The flow in
embryos where arterial or venous genes have been
ablated has not been analyzed, making the interpretation of gene ablation studies difficult. It will therefore
be essential to develop techniques to expose endothelial cells in which important genetic signals are ablated
394
PHYSIOLOGY • Volume 21 • December 2006 • www.physiologyonline.org
to normal flow or to expose wild-type cells to abnormal
flow. This may be possible through studies using heartspecific or vascular-specific conditional ablation of
genes or through the use of embryonic stem cells and
other in vitro models. In this way, the relative importance of genetics vs. epigenetics can be investigated.
References
1.
Adams RH, Wilkinson GA, Weiss C, Diella F, Gale NW,
Deutsch U, Risau W, and Klein R. Roles of ephrinB ligands and
EphB receptors in cardiovascular development: demarcation
of arterial/venous doains, vascular morphogenesis, and
sprouting angiogenesis. Genes Dev 13: 295–306, 1999.
2.
Akimoto S, Mitsumata M, Sasaguri T, and Yoshida Y. Laminar
shear stress inhibits vascular endothelial cell proliferation by
inducing
cyclin-dependent
kinase
inhibitor
p21(Sdi1/Cip1/Waf1). Circ Res 86: 185–190, 2000.
3.
Bartling B, Tostlebe H, Darmer D, Holtz J, Silber RE, and
Morawietz H. Shear stress-dependent expression of apoptosis-regulating genes in endothelial cells. Biochem Biophys
Res Commun 278: 740–746, 2000.
4.
Berk BC, Abe JI, Min W, Surapisitchat J, and Yan C.
Endothelial atheroprotective and anti-inflammatory mechanisms. Ann NY Acad Sci 947: 93-109; discussion 109–111,
2001.
5.
Birukov KG, Bardy N, Lehoux S, Merval R, Shirinsky VP, and
Tedgui A. Intraluminal pressure is essential for the maintenance of smooth muscle caldesmon and filamin content in
aortic organ culture. Arterioscler Thromb Vasc Biol 18:
922–927, 1998.
6.
Chapman WB. The effect of the heart-beat upon the development of the vascular system in the chick. Am J Anat 23:
175–203, 1918.
7.
Chiu JJ, Wang DL, Chien S, Skalak R, and Usami S. Effects of
disturbed flow on endothelial cells. J Biomech Eng 120: 2–8,
1998.
8.
Cirotto C and Arangi I. Chick embryo survival under acute
carbon monoxide challenges. Comp Biochem Physiol A 94:
117–123, 1989.
8a. Cirotto C and Arangi I. Chick embryo survival under acute
carbon monoxide challenges. Comp Biochem Physiol A 94:
117–123, 1989.
9.
Cornhill JF and Roach MR. A quantitative study of the localization of atherosclerotic lesions in the rabbit aorta.
Atherosclerosis 23: 489–501, 1976.
10. Dimmeler S, Haendeler J, Rippmann V, Nehls M, and Zeiher
AM. Shear stress inhibits apoptosis of human endothelial
cells. FEBS Lett 399: 71–74, 1996.
11. Duarte A, Hirashima M, Benedito R, Trindade A, Diniz P,
Bekman E, Costa L, Henrique D, and Rossant J. Dosage-sensitive requirement for mouse Dll4 in artery development.
Genes Dev 18: 2474–2478, 2004.
12. Ferrara N and Davis-Smyth T. The biology of vascular
endothelial growth factor. Endocr Rev 18: 4–25, 1997.
13. Folkow B. Structure and function of the arteries in hypertension. Am Heart J 114: 938–948, 1987.
14. Gabriels JE and Paul DL. Connexin43 is highly localized to
sites of disturbed flow in rat aortic endothelium but connexin37 and connexin40 are more uniformly distributed. Circ Res
83: 636–643, 1998.
15. Gale NW, Dominguez MG, Noguera I, Pan L, Hughes V,
Valenzuela DM, Murphy AJ, Adams NC, Lin HC, Holash J,
Thurston G, and Yancopoulos GD. Haploinsufficiency of ␦-like
4 ligand results in embryonic lethality due to major defects in
arterial and vascular development. Proc Natl Acad Sci USA
101: 15949–15954, 2004.
16. Garcia-Cardena G, Comander J, Anderson KR, Blackman BR,
and Gimbrone MA Jr. Biomechanical activation of vascular
endothelium as a determinant of its functional phenotype.
Proc Natl Acad Sci USA 98: 4478–4485, 2001.
17. Gerety SS and Anderson DJ. Cardiovascular ephrinB2 function is essential for embryonic angiogenesis. Development
129: 1397–1410, 2002.
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.1 on May 5, 2017
yolk sac is, however, expanding throughout the remodeling process, and the growth of these tissues pushes
against the immature vessels. In chick embryos, if outgrowth of the yolk sac is prevented by physical insertion of a metallic ring around the yolk sac, vascular
remodeling does not occur normally (unpublished
results). In vivo time-lapse imaging of mice embryos
that express GFP in their endothelial cells reveals that
the expansion of mesenchymal tissues seems to pinch
off or disconnect certain vessels within the yolk sac
(movie 1, online, FIGURE 4). Thus growth of the mesenchymal tissue may be able to affect branching geometry within the yolk sac (FIGURE 4B). Similar longitudinal forces affect remodeling in adults. If an artery is
exposed to longitudinal strain, it remodels to adapt to
this strain. If strain on an artery is released, vessels take
on a more tortuous trajectory (20a). A classic theory is
that the vessels that carry the largest flow enlarge and
vessels with the smallest flow regress (48). If true, the
mechanism for vessel-diameter adaptation during
remodeling could arise from a balance between the
outward push created by blood flow and the inward
force created by the growth of mesenchymal tissue.
Even if vascular remodeling is seen to have a component of force-driven deformation, this does not
negate a role for genetics in the remodeling process.
This is because the deformation of a tissue by a force is
defined by its material properties, and those are all
genetically encoded. The more stiff a tissue is, the less
it can be deformed when exposed to flowing blood,
and the stiffness of a tissue is given by its attachment to
the extracellular matrix, the expression of adhesion
proteins, and the nature of the cytoskeleton.
REVIEWS
18. Herzog Y, Kalcheim C, Kahane N, Reshef R, and
Neufeld G. Differential expression of neuropilin-1
and neuropilin-2 in arteries and veins. Mech Dev
109: 115–119, 2001.
18a. Hove JR, Koster RW, Forouhar AS, AcevedoBolton G, Fraser SE, and Gharib M. Intracardiac
fluid forces are an essential epigenetic factor for
embryonic cardiogenesis. Nature 421: 172–177,
2003.
18b. Hu N and Clark EB. Hemodynamics of the stage
12 to stage 29 chick embryo. Circ Res 65:
1665–1670, 1989.
19. Ingber DE. Tensegrity: the architectural basis of
cellular mechanotransduction. Annu Rev Physiol
59: 575–599, 1997.
20. Isogai S, Lawson ND, Torrealday S, Horiguchi M,
and Weinstein BM. Angiogenic network formation
in the developing vertebrate trunk. Development
130: 5281–5290, 2003.
21. Jones EA, Baron MH, Fraser SE, and Dickinson
ME. Measuring hemodynamic changes during
mammalian development. Am J Physiol Heart Circ
Physiol 287: H1561–H1569, 2004.
22. Jones EA, Crotty D, Kulesa PM, Waters CW, Baron
MH, Fraser SE, and Dickinson ME. Dynamic in vivo
imaging of postimplantation mammalian embryos
using whole embryo culture. Genesis 34:
228–235, 2002.
23. Krebs LT, Xue Y, Norton CR, Shutter JR, Maguire
M, Sundberg JP, Gallahan D, Closson V, Kitajewski
J, Callahan R, Smith GH, Stark KL, and Gridley T.
Notch signaling is essential for vascular morphogenesis in mice. Genes Dev 14: 1343–1352, 2000.
24. Lawson ND and Weinstein BM. Arteries and veins:
making a difference with zebrafish. Nat Rev Genet
3: 674–682, 2002.
25. Lawson ND and Weinstein BM. In vivo imaging of
embryonic vascular development using transgenic
zebrafish. Dev Biol 248: 307–318, 2002.
26. le Noble F, Fleury V, Pries A, Corvol P, Eichmann A,
and Reneman RS. Control of arterial branching
morphogenesis in embryogenesis: go with the
flow. Cardiovasc Res 65: 619–628, 2005.
27. le Noble F, Moyon D, Pardanaud L, Yuan L, Djonov
V, Matthijsen R, Breant C, Fleury V, and Eichmann
A. Flow regulates arterial-venous differentiation in
the chick embryo yolk sac. Development 131:
361–375, 2004.
28. Lehoux S and Tedgui A. Signal transduction of
mechanical stresses in the vascular wall.
Hypertension 32: 338–345, 1998.
29. Levesque MJ, Nerem RM, and Sprague EA.
Vascular endothelial cell proliferation in culture
and the influence of flow. Biomaterials 11:
702–707, 1990.
31. Manner J, Seidl W, and Steding G. Formation of
the cervical flexure: an experimental study on
chick embryos. Acta Anat (Basel) 152: 1–10,
1995.
32. May SR, Stewart NJ, Chang W, and Peterson AS.
A Titin mutation defines roles for circulation in
endothelial morphogenesis. Dev Biol 270: 31–46,
2004.
33. Mazure NM, Chen EY, Yeh P, Laderoute KR, and
Giaccia AJ. Oncogenic transformation and hypoxia synergistically act to modulate vascular
endothelial growth factor expression. Cancer Res
56: 3436–3440, 1996.
34. Motoike T, Loughna S, Perens E, Roman BL, Liao
W, Chau TC, Richardson CD, Kawate T, Kuno J,
Weinstein BM, Stainier DY, and Sato TN. Universal
GFP reporter for the study of vascular development. Genesis 28: 75–81, 2000.
35. Moyon D, Pardanaud L, Yuan L, Breant C, and
Eichmann A. Plasticity of endothelial cells during
arterial-venous differentiation in the avian
embryo. Development 128: 3359–3370, 2001.
36. Mulvany MJ. Small artery remodeling in hypertension. Curr Hypertens Rep 4: 49–55, 2002.
37. Murray CD. The physiological principle of minimum work. I. The vascular system and the cost of
blood volume. Proc Natl Acad Sci USA 12:
207–214, 1926.
37a. Nguyen TH, Eichmann A, Le Noble F, and Fleury
V. Dynamics of vascular branching morphogenesis: the effect of blood and tissue flow. Phys Rev E
Stat Nonlin Soft Matter Phys 73: 061907, 2006.
37b. Parmar KM, Larman HB, Dai G, Zhang Y, Wang ET,
Moorthy SN, Kratz JR, Lin Z, Jain MK, Gimbrone
MA jr, and Garcia-Cardena G. Integration of flowdependent endothelial phenotypes by Kruppellike factor 2. J Clin Invest 116: 49–58, 2006.
38. Pelster B and Burggren WW. Disruption of hemoglobin oxygen transport does not impact oxygendependent physiological processes in developing
embryos of zebra fish (Danio rerio). Circ Res 79:
358–362, 1996.
39. Plate KH, Breier G, Weich HA, and Risau W.
Vascular endothelial growth factor is a potential
tumour angiogenesis factor in human gliomas in
vivo. Nature 359: 845–848, 1992.
40. Pries AR, Secomb TW, and Gaehtgens P. Design
principles of vascular beds. Circ Res 77:
1017–1023, 1995.
41. Resnick N, Collins T, Atkinson W, Bonthron DT,
Dewey CF Jr, and Gimbrone MA Jr. Plateletderived growth factor B chain promoter contains
a cis-acting fluid shear-stress-responsive element.
Proc Natl Acad Sci USA 90: 4591–4595, 1993.
42. Resnick N, Yahav H, Shay-Salit A, Shushy M,
Schubert S, Zilberman LCM, and Wofovitz E. Fluid
shear stress and the vascular endothelium: for
better and for worse. Prog Biophys Mol Biol 81:
177–199, 2002.
43. Rouwet EV, Tintu AN, Schellings MW, van Bilsen
M, Lutgens E, Hofstra L, Slaaf DW, Ramsay G, and
Le Noble FA. Hypoxia induces aortic hypertrophic
growth, left ventricular dysfunction, and sympathetic hyperinnervation of peripheral arteries in the
chick embryo. Circulation 105: 2791–2796, 2002.
44. Ruijtenbeek K, le Noble FA, Janssen GM, Kessels
CG, Fazzi GE, Blanco CE, and De Mey JG. Chronic
hypoxia stimulates periarterial sympathetic nerve
development in chicken embryo. Circulation 102:
2892–2897, 2000.
45. Ryan HE, Lo J, and Johnson RS. HIF-1 alpha is
required for solid tumor formation and embryonic
vascularization. EMBO J 17: 3005–3015, 1998.
46. Shutter JR, Scully S, Fan W, Richards WG,
Kitajewski J, Deblandre GA, Kintner CR, and Stark
KL. Dll4, a novel Notch ligand expressed in arterial endothelium. Genes Dev 14: 1313–1318, 2000.
47. Territo PR and Burggren WW. Cardio-respiratory
ontogeny during chronic carbon monoxide exposure in the clawed frog Xenopus laevis. J Exp Biol
201: 1461–1472, 1998.
48. Thoma R. Untersuchungen über die Histogenese
und Histomechanikdes Gefässsytems. Stuttgart,
Germany: Ferdinand Enke, 1893.
48a. Tzima E, Irani-Tehrani M, Kiosses WB, Dejana E,
Schultz DA, Engelhardt B, Cao G, DeLisser H, and
Schwartz MA. A mechanosensory complex that
mediates the endothelial cell response to fluid
shear stress. Nature 437: 426–431, 2005.
49. Wakimoto K, Kobayashi K, Kuro OM, Yao A,
Iwamoto T, Yanaka N, Kita S, Nishida A, Azuma S,
Toyoda Y, Omori K, Imahie H, Oka T, Kudoh S,
Kohmoto O, Yazaki Y, Shigekawa M, Imai Y,
Nabeshima Y, and Komuro I. Targeted disruption
of Na+/Ca2+ exchanger gene leads to cardiomyocyte apoptosis and defects in heartbeat. J Biol
Chem 275: 36991–36998., 2000.
50. Wang HU, Chen ZF, and Anderson DJ. Molecular
distinction and angiogenic interaction between
embryonic arteries and veins revealed by ephrinB2 and its receptor Eph-B4. Cell 93: 741–753,
1998.
50a. Wani MA, Means RT Jr, and Lingrel JB. Loss of
LKLF function results in embryonic lethality in
mice. Transgenic Res 7: 229–238, 1998.
51. Wilting J, Birkenhager R, Eichmann A, Kurz H,
Martiny-Baron G, Marme D, McCarthy JE, Christ
B, and Weich HA. VEGF121 induces proliferation
of vascular endothelial cells and expression of flk1 without affecting lymphatic vessels of chorioallantoic membrane. Dev Biol 176: 76–85, 1996.
52. Wong LC and Lagille BL. Developmental remodeling of the internal elastic lamina of rabbit arteries:
effect of blood flow. Circ Res 78: 799–805, 1996.
PHYSIOLOGY • Volume 21 • December 2006 • www.physiologyonline.org
395
Downloaded from http://physiologyonline.physiology.org/ by 10.220.33.1 on May 5, 2017
20a. Jackson ZS, Gotlieb AI, and Langille BL. Wall tissue remodeling regulates longitudinal tension in
arteries. Circ Res 90: 918–925, 2002.
30. Lu X, Le Noble F, Yuan L, Jiang Q, De Lafarge B,
Sugiyama D, Breant C, Claes F, De Smet F,
Thomas JL, Autiero M, Carmeliet P, TessierLavigne M, and Eichmann A. The netrin receptor
UNC5B mediates guidance events controlling
morphogenesis of the vascular system. Nature
432: 179–186, 2004.