Back

Hiroyuki Takeda

Segmentation: a successful evolutionary innovation

Somite formation during vertebrate embryogenesis

Fig 1. Somite formation during vertebrate embryogenesis

The fact that the body plan of several metazoan phyla, such as annelids, arthropods (the most successful group of animals on our planet, by the way) and chordates shows segmental organization is by no means coincidental: segmentation is a major developmental strategy by which to build and diversify different body regions. And though the metameric organization of an adult vertebrate animal might not be obvious to the untrained eye - since multiple segments are grouped into tagmata: functional morphological units - it is still manifest in complex anatomical structures such as the branchial arches, the rib cage or the vertebral column. As a matter of fact, one of the most striking features of the vertebral column is precisely its organization in a segmental pattern. This process of segmentation is initiated during early embryogenesis and involves the generation of somites: bilaterally paired blocks of mesoderm that form along the anterior-posterior axis of the embryo and which contribute to multiple tissues, such as the axial skeleton, skeletal and smooth muscles, dorsal dermis, tendons, ligaments, cartilage and adipose tissue. Disruption of this segmentation process can result in multiple anatomical malformations, e.g. fusion of the ribs, spinal deformities and/or truncations.

The segmentation clock

Somites progressively bud off in pairs from the anterior end of the embryo's unsegmented pre-somitic mesoderm (PSM), a tissue which is replenished by continuous recruitment of cells from the tail-bud, a region located at the posterior end of the embryo. Characteristically, this process follows a strict developmental program: in zebrafish, every 30 min a pair of somites bud off from the PSM, until the final number of somites (around 35) is reached. Somitogenesis, in other words, is under tight temporal control. To trigger this rhythmic production of somites, several signaling pathways are integrated into a molecular network that generates a traveling wave of gene expression along the embryonic axis: the segmentation clock.

This segmentation clock consists of numerous individual cellular oscillators (an oscillator is a pattern that returns to its original state, after a finite number of generations). And although the properties of a single cellular oscillator have been studied, the system-level properties of the segmentation clock have remained largely unknown. During the past years, Prof. Takeda has been trying to understand this aspect of the segmentation clock, in trying to answer the question: “How does intercellular communication coordinate individual oscillators, such that a global oscillation pattern during somitogenesis is maintained?”

Who is making all that noise?

The period and the phase of individual cellular oscillators are likely to fluctuate for several reasons. On the one hand, rhythm-generating reactions, such as transcription, translation and the translocation of key components are in themselves stochastic processes at the molecular level, which can be affected by growth and mitosis. On the other hand, changes in cell shape (e.g. during cell division) may also alter the molecular and biomechanical settings of contact-dependent intercellular communication, which could have a profound impact on the oscillation period during somitogenesis. As a matter of fact, Prof. Takeda and his team have indeed observed active cell proliferation in the synchronized oscillation zone and could demonstrate that the cyclic expression of certain segmentation genes (e.g. hairy-related gene1) actually fluctuates in individual cells: some cells in the field are slightly advanced, whereas other cells are slightly delayed. As such, their data indicate that internal noise caused by stochastic gene expression and cell division affects the oscillation period of unit oscillators. Consequently, a comprehensive understanding of how cellular oscillations are synchronized and how the segmentation clock - a complex network of genes that functions both intra- and intercellularly - is coordinated, has to incorporate a thorough analysis of the phenomenon “biological noise”.1-3

Stay tuned: intercellular crosstalk ensures synchronization

The properties of the segmentation clock

Fig 2. The properties of the segmentation clock

How does the embryo manage to perform a strict morphogenetic program under such noise stress? Obviously the effects of noise must be minimized, as to maintain coherent oscillation. Tackling this problem from both an experimental and a theoretical angle - joining forces with Prof. Shigeru Kondo (Osaka University), involved in the development of a computer program that simulates the essential features of somite formation - it could be convincingly demonstrated that the segmentation clock is indeed strongly buffered against noise, at the molecular level. Indeed, a major outcome of Prof. Takeda's research is that the segmentation clock behaves as so-called “coupled oscillators”, by which the oscillation of neighbouring cells is synchronized. In other words: cells gradually adjust their oscillatory phase (e.g. dynamic RNA levels of key regulatory genes, such as hairy-related gene1) to their surroundings. This synchronization phenomenon is dependent on the Notch signaling pathway that controls key aspects of an intercellular oscillatory circuit (involving a feed-back loop mediated by hairy, encoding a transcriptional repressor). Or in other words: Notch-dependent intercellular coupling is crucial for the maintenance of synchronized oscillation, by reducing the effects of internal noise in the pre-somitic mesoderm.

How do cells collectively make a 3D shape?

Dynamic 3D morphogenesis of a zebrafish somite

Fig 3. Dynamic 3D morphogenesis of a zebrafish somite

Multicellular organs have a particular structure, dynamically formed during embryogenesis and determined by functional and anatomical constraints. One of the major unsolved questions in Developmental Biology is how individual cells “work together” to create such a complex 3D shape. Prof. Takeda and his team are devoting themselves to answering that question - again, using an experimental, as well as a theoretical approach - by studying somite morphogenesis in zebrafish. Originally the form of a newly formed somite is cuboidal, but within a period of 3-4 hours it elongates along the DV axis and gradually adopts a chevron (V) shape. How is this possible? How do individual cells communicate to collectively create such a multicellular 3D shape? And how are the mechanical forces generated that cause such dramatic tissue deformations?

Digitizing somite morphogenesis in zebrafish

Fig 4. Digitizing somite morphogenesis in zebrafish

Using advanced imaging technology (e.g. light sheet fluorescence microscopy), Prof. Takeda and his co-workers are digging deeper and deeper into the zebrafish embryo, observing the behavior of individual cells at every stage in a developing somite. In doing so, they could demonstrate that the number of cells (+/-300) within a developing somite remains largely constant. Also the total volume of the somite does not change during that developmental time window. This means in other words that the physical force, needed to provoke the shape changes comes largely or exclusively from cell rearrangements. Recently, Prof. Takeda and his team could show unique behavior in the somitic cell population: cells migrate and rotate in the same direction, thereby collectively forcing the observed dynamic changes in 3D shape.

A look into the Future
Science Fiction
Science or Fiction?
Prof. Takeda, in which directions are your research projects developing?
Prof. Takeda:
A major project in our lab deals with unraveling the mechanisms of somite formation in zebrafish. We are trying to understand 3D morphogenesis, both at the cellular and the tissue level.
At the cellular level?
Prof. Takeda:
We are analyzing cell behavior during somite morphogenesis. We want to understand how individual cells produce the mechanical force necessary to induce the dramatic tissue deformations we observe. We have been able to show that somitic cells migrate - a moving cell produces mechanical force - and collectively rotate. So, it is not inconceivable that the mechanical forces produced by individual cells are transduced extracellularly and combined to provoke a morphogenetic event. This process by the way, shows remarkable similarities to what has been observed during Drosophila oogenesis: eggs undergo elongation from a sphere to an ellipsoid, a 3D shape change which is associated with collective migration and rotation of cells.
At the tissue level?
Prof. Takeda:
Indeed, the next question is. “How can 300 cells collectively produce mechanical force in a certain direction? ” Obviously somite morphogenesis is a highly coordinated - both in time and space - multicellular event. How is this process guided? Which mechanisms control intercellular crosstalk, necessary to change the 3D shape of a somite from cuboidal to a chevron form within a couple of hours?
How do you plan to tackle these problems?
Prof. Takeda:
We are combining two research approaches. On the one hand, we have been analyzing every step during somite morphogenesis: we have video-recorded the position of every single nucleus and every shape change at the cellular and the tissue level. In collaboration with Prof. Matsumoto at the Nagoya Institute of Technology, we are now using 3D printing technology to construct anatomical models of somites at different developmental stages.

On the other hand, we have started collaborating with Prof. Akiyama at Hokkaido University, who is developing a 3D model which simulates somite morphogenesis in zebrafish. It is not possible to imagine the collective behavior of 300 individual cells, so I really believe that constructing computer models will be helpful in our search for general principles that guide somite formation.
Would you expect that one or a few underlying principles guide morphogenesis?
Prof. Takeda:
I am tempting to believe so. What I do believe is that the key to understanding 3D morphogenesis is analyzing individual and collective cell behavior and characterizing all mechanical forces that come into play during a given developmental event. For example, we need to know precisely how much and in which direction a mechanical force is generated by each individual cell in a tissue. This will be one of the major challenges for the future.
REFERENCES
  1. Horikawa, K., Ishimatsu, K., Yoshimoto, E., Kondo, S. Takeda, H. Noise-resistant and synchronized oscillation of the segmentation clock. Nature, 441, 719-23, 2006.
  2. Ishimatsu K, Horikawa K, Takeda H. Coupling cellular oscillators: a mechanism that maintains synchrony against developmental noise in the segmentation clock. Dev. Dyn. 236, 1416-21, 2007.
  3. Ishimatsu K., Takamatsu, A. Takeda, H. Emergence of traveling wave in the zebrafish segmentation clock. Development, 137, 1595-9, 2010.
Who is Who?

Prof. Takeda

Prof. Hiroyuki Takeda

Prof. Hiroyuki Takeda was born and raised in the Niigata area, at the northwest coast of Honshu, and studied Biology at the Graduate School of Science, University of Tokyo. His doctoral work, under supervision of Prof. Takeo Mizuno, dealt with epithelium-mesenchyme interactions during rodent development. During this period Prof. Takeda became interested in developmental endocrinology, more specifically in the process of androgen hormone induced mesenchyme interactions during prostate gland development. Following a period of more clinically oriented research in the field of prostate cancer, at the Strangeways laboratories (Cambridge, UK) and the University of Chicago, Prof. Takeda decided to enter the field of zebrafish development, being the first scientist in Japan to do so. After having done research at Riken Research Institute, Nagoya University and the National Institute of Genetics in Mishima, Prof. Takeda returned to the Graduate School of Science, University of Tokyo, where he presently leads a research team studying early developmental processes, using medaka and zebrafish as vertebrate model systems.

Back