<<

. 37
( 66 .)



>>

et al., 2002). In fact, nodal™s asymmetric expression in the chicken em-
bryo appears to result from a juxtacrine Notch--Delta-based signaling
system, such as that described above for epithelial patterning (Raya
et al., 2003, 2004). A molecularly realistic dynamical model for the
generation of axis-associated gene expression patterns in the chicken
can be found in Raya et al. (2004).
7 PATTERN FORMATION: SEGMENTATION, AXES, AND ASYMMETRY 185




A




D
C
B




AP position
Mediolateral position

Fig. 7.12 Formation of the midline and enfolding of the anteroposterior (AP) axis
according to the reaction“diffusion model of Meinhardt. (A) Schematics of the
hypothesized processes involved in axis formation. Cells close to the blastopore (red
annulus) move towards the Spemann“Mangold organizer (blue). The spot-like organizer
redirects the cells in its vicinity: initially they are drawn to the organizer but then lose
their attraction to it, extending from the posterior towards the anterior, so that they
leave as a uni¬ed stripe-like band. Stem cells in the organizer region may contribute to
this band to form the most central (“medial” as opposed to “lateral”) element (green) of
the dorsal surface. Such highly coordinated cellular motion and strong positional
speci¬cation along the enfolding AP axis require the simultaneous action of several
feedback loops. (B“D) A simpli¬ed simulation of the scenario just described. A
reaction“diffusion system tuned to make stripes (green; compare with the zone
formation in Fig. 7.11) is triggered by the organizer (blue). The organizer itself is the
result of a self-enhancing system activated in a spot-like manner. (In B“D the blastopore
is shown in yellow.) Repulsion between the spot system (the organizer) and the stripe
system (the notochord) causes elongation of the latter. Saturation in self-enhancement
(due to the factor s A in Eq. 7.9a) ensures that the stripe system does not disintegrate
into individual patches and thus establishes the midline. This, in turn, acts as a sink for a
ubiquitously produced substance (pink), which could be the product of the BMP-4 gene
(Dosch et al., 1997); the local concentration of this substance (shades of pink) is a
measure of the distance from the midline. This simulation is simpli¬ed in that the actual
movement of cells toward the organizer is not considered; instead, the midline elongates
by the addition of lines of new cells next to the blastopore. (A redrawn and B“D
reprinted from Meinhardt, 2001, with permission from the University of the Basque
Country Press.)
186 BIOLOGICAL PHYSICS OF THE DEVELOPING EMBRYO


Embryonic axial asymmetry is re¬‚ected in left--right differences
in the expression pattern of certain genes, such as Sonic hedgehog and
nodal itself, and eventually in the left--right asymmetry of organs such
as the heart and intestines. Although total inversion of the symmetry
of internal organs (situs inversus totalis) usually has little adverse im-
pact on health, partial inversions, in which the heart, for example, is
predominantly on the right, are highly deleterious (Aylsworth, 2001),
and perhaps for this reason vertebrate embryos have the means to en-
sure that 99.99% of humans, for example, have the standard left--right
asymmetry.
If we entertain Meinhardt™s hypothesis that the axis-forming sys-
tem of vertebrates is governed by a reaction--diffusion mechanism
then we are confronted with the question how left--right asymmetry
is so reliably generated in these organisms. Supp and coworkers found
that a mouse genetic variant called iv, characterized by random posi-
tioning of the internal organs, carries a mutation in the microtubule-
associated protein left--right dynein (LRD) (Supp et al., 1997). Dyneins
form the part of the molecular motor that controls the rotation of
cilia, and LRD is expressed in ˜˜monocilia” (cilia are motile extensions
of the cell surface and in this case each cell has only one) in Hensen™s
node in the mouse. The mutation in LRD in the iv mutant has the
result that these monocilia are immotile (Supp et al., 2000).
This motivated Nonaka and coworkers to look at the relation-
ship between these monocilia and the distribution of Nodal around
Hensen™s node in the mouse embryo (Nonaka et al., 1998). They found
that the vortical motion of the monocilia in Hensen™s node is im-
plicated in left--right symmetry breaking. The clockwise rotational
movement of several hundred nodal monocilia generates leftward
(i.e., anticlockwise) ¬‚uid currents (nodal ¬‚ow) that bias the distribu-
tion of nodal and thus determine the polarity of the broken symmetry
(Nonaka et al., 2002). In a striking con¬rmation of their model for axis
symmetry breaking, when Nonaka et al. arti¬cially reversed the direc-
tion of nodal ¬‚ow in wild-type embryos, or subjected the iv mutant
to leftward ¬‚ow, they were able to induce the normal arrangement
of internal organs.
These results have a ready interpretation in terms of the reaction--
diffusion mechanism for embryonic axis generation described above
(Solnicka-Krezel, 2003). Reaction--diffusion mechanisms in nonliving
systems are indifferent to which side is left and which side is right,
the local initiators of activation typically arising by random ¬‚uctu-
ations (see, for example, Castets et al., 1990; Ouyang and Swinney,
1991). In embryonic systems, where there is clearly a premium on
having pattern-forming mechanisms that generate the same results
in each successive generation, molecular or structural cues may act
as guides, biasing a mechanism capable of producing multiple out-
comes to generate one that is biologically functional. The monocilia
near Hensen™s node in the mouse and the embryonic organizers of
other vertebrate species (Essner et al., 2002) appear to play such a
role. Here the inherent pattern-forming potential of one physical
7 PATTERN FORMATION: SEGMENTATION, AXES, AND ASYMMETRY 187


system (i.e., the axis-forming reaction--diffusion process comprising
the cells and various gene products described above) is channeled and
restricted by its interaction with another set of physical systems (i.e.,
the motile monocilia). Although these systems are intrinsically dif-
ferent in scale, the collective behavior of the appropriately arranged
monocilia (a product of evolution; see Chapter 10 for a discussion of
the evolution of developmental mechanisms) tips the balance in the
otherwise indifferent pattern-forming mechanism in such a way as
to reliably produce a biologically functional macroscopic outcome.


Perspective
Embryos employ a strikingly wide range of physical mechanisms to
establish the characteristic spatial patterns of cell differentiation.
Most of these mechanisms -- adhesion-based cell sorting, juxtacrine
feedback loops, the establishment of gradients over the multicellular
spatial scale by diffusion -- are virtual inevitabilities in simple ag-
gregates composed of cells capable of differentiation. Epithelia, con-
taining cells in direct contact with one another, or mesenchymes,
containing ECMs permeable to diffusible factors, can only employ
physically based mechanisms arising from, or compatible with, their
material nature. Living tissues thus have an inherent capacity to gen-
erate spatially heterogeneous developmental outcomes. When the rel-
evant physical mechanisms are embodied in, or integrated with, ¬ne-
tuned molecular signal-response mechanisms (e.g., the Notch--Delta
or Nodal--Lefty couples), they become indispensable components of
reliable developmental pattern-formation mechanisms.
Chapter 8




Organogenesis

As a consequence of generating and reshaping the epithelia of their
ectodermal and endodermal germ layers and (for those species that
have them) the mesenchymes that make up the mesodermal germ
layer and neural crest, embryos take on one of approximately three
dozen stereotypical ˜˜body plans.” These body plans can be asymmetric
(sponges), radially symmetric (e.g., hydra and sea urchins), bilaterally
symmetric (e.g., planaria, insects), or ˜˜bilaterally asymmetric” (verte-
brates; see the last section of Chapter 7). While it is obvious that
nothing in development, or in any other domain of biology for that
matter, can occur without the participation of physical mechanisms,
the high degree of structural and dynamical complexity of most liv-
ing systems makes it exceedingly dif¬cult, in general, to follow the
workings of basic physical principles or appreciate their roles in gen-
erating characteristic biological phenomena. It is therefore remark-
able to how great an extent (as the preceding chapters have shown)
the tissue generation and reshaping processes leading to embryonic
body plans can be accounted for by physical forces and properties --
adhesion, viscoelasticity, dynamical multistability, network forma-
tion, positive and negative feedback dynamics, diffusion -- that also
govern the behaviors of nonliving condensed materials, that is, by
˜˜generic” physical mechanisms.
The ability to understand aspects of early development in generic
physical terms can be attributed, in part, to the likelihood that the
original multicellular organisms were simple, loosely organized, cell
masses, whose forms were determined to a great extent by their inher-
ent physical properties (see Chapter 10). Early-stage modern-day em-
bryos, however ˜˜programmed” they may be by sophisticated genetic
mechanisms of pattern formation that have evolved over the last half-
billion years, are likely still to retain many of these presumed origi-
nal physical determinants of form and pattern (Newman and M¨ ller, u
2000).
Attempts to use basic physical models to understand changes at
later developmental stages encounter greater dif¬culties. The simul-
taneous actions of multiple morphogenetic and patterning processes
8 ORGANOGENESIS 189


over time, along with the acquisition by cells of speci¬c differentiated
properties (e.g., the contractility of muscle cells, the electrical ex-
citability of nerve cells, the secretion of solid matrices by cartilage
and bone cells) causes the developing body to become increasingly
complex, both structurally and functionally. And once an active cir-
culatory system is in place (in species that have one), and ˜˜endocrine”
tissue types have been generated that are capable of transmitting
tissue-speci¬c regulatory signals through this network of branching
and anastomosing (i.e., intercommunicating) conduits, the hope of
accounting for coordinated changes in body shape and form by one
or two generic physical mechanisms must be abandoned.
Nonetheless, in most animal species the various specialized organs
develop in a semi-autonomous fashion from distinct clusters of cells
(˜˜organ rudiments”) during a period after the establishment of the
body plan but prior to the extensive elaboration and integration of
the embryo™s or fetus™s anatomical and physiological systems. Organo-
genesis, the formation of bodily organs, therefore can make use of all
the generic physical mechanisms discussed previously in relation to
body-plan formation and in certain cases utilize some novel ones. A
physical mechanism that generates a branched network of tubules, for
example, while not a likely determinant of any known animal body
plan can come into play at the level of organogenesis, during the for-
mation of insect tracheal systems or the lungs, kidneys, and vascular
systems of vertebrates. Branched tubes can also provide the basis for
external appendages such as limbs, in animals with exoskeletons (e.g.,
insects), as we saw when we considered the Mittenthal--Mazo model
in chapter 5.
We considered the physical properties of epithelial and mesenchy-
mal tissues separately in Chapters 4, 5, and 6 and, indeed, during
development of the germ layers and body plans these tissue types of-
ten act independently of one another, as, for example, in neural crest
dispersion. In other cases, as happens with convergent extension in
vertebrates and archenteron elongation in sea urchins, the two tis-
sue types behave in a coordinate or parallel fashion, as if they were
subject to similar forces. During organogenesis, however, epithelial--
mesenchymal interactions, in which the speci¬c biomechanical proper-
ties of the two components act in a synergistic fashion in producing
an outcome, become a major determinant of morphogenesis.
We saw hints of this capability in the description of neurulation
in Chapter 5. In that case, the mesenchymally derived notochord (by
that stage a stiff rod) induced the overlying epithelial sheet to form
neural folds. However, in organ rudiments not constrained to produce
a structure with a central axis the mesenchymal component will typi-
cally retain its ˜˜liquid” state during morphogenesis, the tissue spread-
ing as a coherent mass and the individual cells moving relative
to one another through the ECM. As we will see below, epithelial--
mesenchymal interactions under these more general conditions can
lead to the formation of branched and lobulated structures, such as
190 BIOLOGICAL PHYSICS OF THE DEVELOPING EMBRYO


those seen in the pancreas and salivary glands, or of arrays of struc-
tural elements that are repetitive along more than one axis, as seen in
bird feather tracts or the vertebrate limb skeleton. And while we saw
examples of epithelium-to-mesenchymal transformation in mesoderm
generation during gastrulation (Chapter 5) and during the formation
of the neural crest (Chapter 6), organogenesis is often accompanied
by a transformation of mesenchyme to epithelium, with associated
physically induced morphogenetic consequences.
In this chapter we will describe several examples of organ forma-
tion, in particular, assembly of the blood vessel network (vasculoge-
nesis), gland development (branching morphogenesis), and skeletoge-
nesis in the vertebrate limb. As earlier, we will introduce, as far as
possible, physical concepts not encountered previously along with the
biological topics covered.


Development of the cardiovascular system
The cardiovascular system of vertebrate organisms is a network of
tubes that de¬nes a single, enclosed, blood-¬lled space. The heart,
an initially tubular structure that undergoes extensive modi¬cations
during development, is interconnected with millions of tubular blood
vessels, beginning with the pulmonary vein and artery and the large-
caliber aorta and vena cava (two large veins), which extend directly
from the heart, continuing through the branched arteries and veins
of intermediate caliber, on to the small-caliber arterioles and venules
and minuscule capillaries responsible for local blood supply. In the
systemic circulation, oxygenated blood is pumped from the heart
through the aorta and is carried by the arteries to the body™s or-
gans and tissues; oxygen-depleted blood from these peripheral sites
is returned to the heart by way of the veins and ultimately the vena
cava. Reoxygenation of the blood takes place by means of the pul-
monary circulation, in which the pulmonary artery conveys deoxy-
genated blood from the heart to the lungs and the pulmonary vein
returns it to the heart in oxygenated form. (Thus in this artery--vein
pair alone, the normal oxygenation status is reversed.) Since the circu-
latory system is essentially closed (with rare exceptions, as in lymph
nodes, where white blood cells of the immune system can squeeze
through the walls of small blood vessels, or in the liver, where the
capillary walls are discontinuous), oxygen and soluble nutrients are
supplied to the tissues, and waste materials conveyed away, by diffu-
sion across the membranes of the endothelial cells, which line all the
vessels and comprise the major structural component of the capillary

<<

. 37
( 66 .)



>>