Shedding light on the elegant mechanisms that control the push and shove of cells in living organisms

Shedding light on the elegant mechanisms that control the push and shove of cells in living organisms

Enlarge

Figure 1. a-Catenin and myosin in epithelial cells. a-Catenin (blue) is a protein that forms part of an ?adhesion apparatus? for joining adjacent cells and is uniformly distributed at cell boundaries. Myosin (green) is another protein that produces a force that deforms the inner structure. The areas undergoing the resultant structural change are shown in pink.

The many amazing photomicrographs on display at the entrance to the Electron Microscope Laboratory in the RIKEN Center for Developmental Biology (CDB) were all taken by laboratory leader Shigenobu Yonemura and his colleagues. In addition to its own research program, the laboratory also offers technical assistance with electron microscopic analysis to researchers at other laboratories in the CDB. “Moving cells seem to have a life of their own. I have long been fascinated by the motion force generated by cells,” says Yonemura. His team’s recent clarification of the mechanism by which adjacent cells pull and pull against each other has attracted considerable attention.

Yonemura clearly remembers his days as a graduate student examining the development of the sea urchin. “The fertilized egg of a sea urchin cleaves and replicates, doubling the number of cells in sequence to build up a spherical body. At one point, part of the sphere suddenly sinks, taking surface cells into the body itself. This process is known as archenteron invagination. I was fascinated by the change, in which it looks as if the surface cells move under their own volition. Since then, I have been interested in the force generated by cells.”

That force is produced by two proteins: actin and myosin. Individual actins occur in spherical form, with many actins joining together to form actin filaments. As myosins move on the filaments, a force that pulls the filaments is produced, which in turn allows the cells to move or change their form. “It is not necessary for the cells to be in the process of morphogenesis, such as during archenteron invagination, for them to generate a force. For example, adjacent epithelial cells are constantly pulling at each other.”

Epithelial cells play an important role in separating the internals of a living body from the outer world by arranging themselves in a sheet to form the surface of the skin and the gastrointestinal tract. Adjacent epithelial cells are linked together by adhesion apparatus (Figs 1, 2). The key component of this apparatus is cadherin, a protein discovered in 1982 by Masatoshi Takeichi, director of the CDB. Cadherin penetrates the cell membrane, with one end exposed to the cell surface, and the other end protruding into the cell. Bound to the end of the cell are a-catenin and ß-catenin. As the actin filaments bind to the a-catenin and the myosin acts on the filaments, a pulling force is produced.

Enlarge

Figure 2. Mechanism producing tensile forces among adjacent cells. Cadherin binds to actin filaments via a-catenin and ß-catenin, forming an adhesion apparatus. When not bound to a nearby cell, myosin does not act on the actin filament, and the a-catenin assumes a bent structure. When cadherin is bound to that in an adjoining cell, it is pulled strongly, inducing myosin to act on the actin filament. As both ends of the a-catenin are pulled in opposite directions, the bent portion of the a-catenin stretches. Vinculin then binds to the a-catenin, causing the actin filament to bind even more strongly to the a-catenin. The adhesion apparatus is thus enhanced, producing a pull-back force. enlarge image

“As cadherin adheres to the same protein in an adjoining cell, the adjoining cell is pulled by tensile force. If that pulling force continues, however, the cell is deformed. A good balance cannot be achieved unless the adjoining cell pulls back. When pulled strongly, the cell pulls back strongly, and when pulled weakly, it pulls back weakly. Believing in the existence of a mechanism behind this, I have been trying to find what it is, and recently succeeded in clarifying it. The key player proved to be a-catenin."

The mechanism is remarkable. Each cell contains a-catenin, which assumes a bent structure (Fig. 2). When pulled by an adjoining cell, the bent portion stretches. The vinculin molecule then binds to the a-catenin, and as the actin filaments continue to bind to the vinculin, the adhesion apparatus is enhanced and a pulling-back force is produced.

“This can be compared to a tug-of-war in which one team is about to lose the game as it is pulled with a strong force by its opponent, then many helpers come to join and pull back strongly,” says Yonemura. When the opponent's pulling force weakens, the helpers are no longer necessary. The a-catenin then bends again, and the vinculin and actin filaments leave the adhesion apparatus, restoring the original state. “From this system, I realized that a-catenin is responsible for two tasks: sensing the forces and enhancing the adhesion apparatus.”

When a cell is pulled by an adjacent cell, it must immediately pull back at the same location. However, transducing information about such a response by the release of a signal transmitter, as occurs in many biological processes, would not provide the responsiveness required to balance rapid changes in pulling location and strength. “It is reasonable that one molecule plays the dual roles of sensing the forces and enhancing the adhesion apparatus. We have arrived at a definite answer to the inveterate problem of how cells perceive forces and respond to them.”

Yonemura is planning to make observations of how the a-catenin structure changes to allow binding to the vinculin, and to quantify the pulling force required to cause morphological changes in the cells.

“Pulling at each other is very important for cells. I think that mutual pulling is utilized in, for example, sensing whether an adjoining cell is alive or dead,” says Yonemura.

Dead epithelial cells are quickly eliminated from the sheet structure. “The mechanism by which cells sense the death of adjoining cells remains unclear. Although it is said that dead cell detection may be achieved by molecular exchanges on the cell surfaces, I do not think that’s the whole story.”

To investigate the mechanism behind the sensing of dead cells, Yonemura artificially damaged epithelial cells and examined them in the process of repair. Similar experiments undertaken before had not yielded good results because the experimentally inflicted damage affected a larger range than intended due to cell scratching. Yonemura succeeded in selectively killing a target cell by laser irradiation. His technique has made it possible to make extensive observations of the behavior of cells adjacent to the dead cell (Fig. 3).

“When a cell dies, a change begins to occur in the cells adjacent to the dead cell. Actin filaments and myosins gather on the surfaces in contact with the dead cell as if they were joining hands to form a ring surrounding the dead cell. The ring gradually shrinks, and eventually the dead cell is ejected from the sheet.”

This accumulation of actin filaments and myosins begins as soon as a cell dies. “The repair of epithelial cells cannot begin immediately as it does unless actin filaments and myosins are involved in sensing cellular death,” says Yonemura. “Cells are constantly pulling at each other. When an adjacent cell has died, however, the living cell in question continues to pull the dead cell, but is no longer pulled back. It is through this process that living cells may sense the death of an adjacent cell. Such a process would ensure that death signals are detected earlier and more accurately, allowing the elimination of the dead cell to start more quickly than if cellular death were detected by molecular exchange on cell surfaces.”

Enlarge

Figure 3. Elimination of dead epithelial cells. Epithelial cells visualized by transmission microscopy (upper) and fluorescence microscopy (lower) showing intracellular myosin. When an epithelial cell is damaged (center of each plate) by laser irradiation, actin filaments and myosins in adjoining cells gather to form a ring around the dead cell. The ring then gradually shrinks to expel the dead cell out of the sheet. enlarge image

However, that may not be the full story. “When actin filaments and myosins have formed a ring surrounding the dead cell, the forces remain balanced. It is natural that the process ceases in that state. In actuality, however, the ring continues to shrink further to eliminate the dead cell. There must be another, unknown mechanism that breaks the balance of forces. Clarification of the mechanism behind the series of processes from the sensing of cellular death to its elimination is an issue that remains to be addressed.”

Of course, gene expression and molecular signal transduction are vitally important to the functioning of cells. “Although this is true, there is also a wide variety of mechanisms that act like machinery in cells. I am much more interested in those types of machinery. I want to discover these exquisite mechanisms, which are remarkably simple and controlled by feedback,” says Yonemura.

In addition to its own research program, the Electron Microscope Laboratory provides technical support for electron microscopic analysis to researchers at the CDB. Yonemura also provided such support while conducting his own research at the Laboratory for Cellular Morphogenesis in the CDB, for which he served as team leader between 2001 and 2006. Even after the team was renamed the Electron Microscope Laboratory in April 2007, its tasks have continued basically unchanged.

“Only a few researchers are good at using electron microscopes,” says Yonemura. Electron microscopes have a resolution of as fine as 0.1 nanometers, compared with the much coarser resolution of 200 nanometers achievable using optical microscopes, making it possible to observe much smaller features. “It is likely that users realize the performance of the electron microscope but tend to view it as something that is difficult to operate, possibly because of the large amount of instrumentation, so they are reluctant to use it. Myself and two technical staff members provide technical assistance on their use.”

At the Electron Microscope Laboratory, members manage two transmission electron microscopes and one scanning electron microscope, as well as peripheral equipment including two microtomes for preparing sample sections. “We want researchers to be able to examine as many samples as possible, so we chose to install more pieces of standard equipment, rather than one or two highly specialized, high-performance units.”

Currently, the laboratory provides assistance to about 20 research projects annually. In some cases, support is provided for all steps from the preparation of resin-fixed sectional samples to examination and photographic recording. In other cases, it is only necessary to provide instructions on how to use the instruments. Many different types of materials are brought into the laboratory, including mice, drosophila fruit flies, nematodes, zebrafish and chickens. The types of tissue examined are also diverse, and include embryos, nerves and kidneys. “New types of samples can be quite difficult to handle the first time, so it can take a lot of time to prepare them properly. However, since all trial-and-error experimentation represents the accumulation of knowhow, we always endeavor to work steadily without rushing. State-of-the-art techniques are not the only key to success in this field. We must have a broad range of knowledge about the wide variety of tissues of various organisms, and be able to hold scientific discussions with the users.”

When providing support, Yonemura discusses the project and desired goals with the requestor. “We do this to determine whether the target is in fact suitable for analysis using an electron microscope. Electron microscopy only allows a relatively narrow range of examination compared with optical microscopy. The sample must also be fixed in resin, and this is a time-consuming procedure. Our discussions sometime lead us to recommend use powerful optical microscopes. To ensure the best outcomes, we make it a rule to first discuss things in detail with the requestor.”

Yonemura’s support role also has its own benefits. “The support work allows me to get involved in many kinds of research that I would otherwise never do myself, and it provides me with opportunities to see the very latest achievements before they are published. For me, the greatest fascination is in meeting many different people. It makes me very happy to be able to continue to do my own research and at the same time balance that with our support work.”

Unearthing the mechanisms controlling plant size

Unearthing the mechanisms controlling plant size

Enlarge

Mechanism underlying increases in plant size. Photograph showing a cross-section of Arabidopsis roots. Plant cells divide mainly in the meristems at the tips of roots and stems. After dividing several times, cells begin to increase their size due to endoreduplication. When the cell reaches a certain size, endoreduplication and cell growth cease. The HPY2 gene controls the transition from cell division to endoreduplication, whereas the GTL1 gene stops cell growth. The green fluorescence indicates the HPY2 expression.

Plants have been cultivated and studied from the earliest days of human civilization, yet much remains unknown about them. A good example is the mechanism by which the size of plant cells is determined. Keiko Sugimoto, leader of the Cell Function Research Unit in RIKEN's Plant Science Center, in working to elucidate this mechanism has discovered a series of genes that control cell division or cell growth, attracting the attention of researchers and companies worldwide. "Our research focuses on the cellular aspects of plants," says Sugimoto.

A garden of lilies

While a high-school student, Sugimoto noticed that a single lily that had blossomed in her garden the year before had become three lilies a year later, followed by ten the next year and as many as 100 the year after. “But what impressed me most was that all of those flowers were the same size, and had the same color and same shape every year,” she says. “Although I knew that this was a manifestation of heredity, which I had learned at school, I was fascinated. I wanted to understand the mystery of plants, and this led me into research.”

After completing her master’s course in Japan, Sugimoto gained her PhD in plant science at the Australian National University. She then went to work at the John Innes Center in the UK—a Mecca for researchers studying plant biology—and in 2007 she set up the Cell Function Research Unit in RIKEN’s Plant Science Center. “How is plant cell size controlled? We are now working to solve this difficult problem.”

The aspect that had impressed Sugimoto most as a high-school student was that the sizes of flowers, leaves, seeds and other plant organs depend roughly on the species of plant. Each organ grows as its cells self-divide and increase in number, and each cell expands. However, plant organs do not continue to grow infinitely. “Flowers and leaves stop growing when they reach a certain size. Research has shown that plant hormones such as auxin and cytokinin are involved in plant growth, but we still don’t know how plant hormones control cell division and cell expansion to determine ultimate organ size,” says Sugimoto. “Plant size cannot be understood without knowing what is happening in cells. We are conducting research focusing on the cellular aspects of plants, which is a unique approach.”

Determinants of plant growth

“Every time I cut radishes or carrots into long, thin strips for cooking, I cannot help admiring the slices for a moment,” says Sugimoto with a smile. “If you look closely at a slice, you can see a finely textured portion near the tip. This is called the meristem. It is dividing tissue where the cells self-divide. Try taking a look next time you’re preparing a meal.”

Enlarge

Endoreduplication and cell size in Arabidopsis A scanning electron microscopy image of the surface of an Arabidopsis leaf showing a trichome. Most cells have 2C nuclear DNA content, but some cells have increased nuclear DNA contents of 4C, 8C and 32C due to endoreduplication. The four insets show the correlation between cell size and the amount of nuclear DNA. The trichome has a nuclear DNA content of 32C.

In plants, meristems are found only in the tips of roots and stems. Plant growth is the result of cell division and proliferation at the meristem, which is followed by cell expansion. “There are two key time points in plant growth. One is a turning point when cell division switches to cell expansion. Once a cell begins expanding, it cannot return to the stage of proliferation by division. The other is the point when cell growth stops and cells no longer expand. Plants cannot grow normally unless these two points are strictly controlled.” Sugimoto and her colleagues have attracted global attention for their discovery of the genes that control these two growth points.

Genes control cell division and endoreduplication

Sugimoto used a new technique to discover genes that control the point at which cells stop dividing and begin increasing in size. “Many researchers have tried to look for genes that control cell size, but most of them were trying to find mutants with altered cell size. Mutants have been identified merely based on the appearance of cells. We developed a method to accurately measure nuclear DNA content and isolate mutants with altered DNA levels.”

The cells of Arabidopsis, a commonly used experimental material in plant science, just like human cells, have two sets of chromosomes, one from the mother and the other from the father. The DNA of a cell having two sets of chromosomes is denoted 2C. When a 2C cell divides, its DNA is first replicated to produce 4C, which is then equally distributed into the next two dividing cells, resulting in two 2C daughter cells. In Arabidopsis, however, 2C and 4C cells are not the only cell types to be found. Gametes (pollen, ovules) that have undergone meiosis, a special process of cell division that results in half the number of chromosomes as found in somatic cells, are 1C cells, but there are also 8C, 16C and 32C cells. “In plant cells, DNA replication is sometimes followed by doubling in DNA without mitosis,” says Sugimoto. “This phenomenon is called endoreduplication, which results in 8C, 16C and 32C cells. The nuclear DNA and cell size are correlated; cells expand as their nuclear DNA increases.”

Together with Takashi Ishida, a postdoc in her lab, Sugimoto examined the nuclear DNA content of Arabidopsis cells and discovered a mutant having fewer 2C and 4C cells and more 32C, 64C and 128C cells. “Usually in plants, 2C and 4C cells in meristems continue to divide at a constant rate. We assume that in the mutant we discovered, these meristematic cells have undergone endoreduplication and switched into cell expansion prematurely. A more detailed investigation revealed that this mutant had lost the function of the HPY2 gene. Hence, HPY2 plays a role in controlling the point of switching to endoreduplication, where cells stop dividing and instead grow in size.”

This achievement was announced in August 2009, drawing attention not only from plant biologists, but also from researchers studying a wide variety of other organisms. “This is because HPY2 is involved in the function of a small peptide known as SUMO, a small ubiquitin-like modifier. SUMO is found in a broad range of species, from humans to plants and yeasts. It binds to other proteins to enhance or weaken their functions, and to regulate the diverse functions of cells. The reason why my result attracted so much attention is that researchers studying diverse ranges of organisms have been interested in SUMO.”

Enlarge

Regulation of cell division by the HPY2 gene. (Upper) Nuclear DNA content of a cell population. The hpy2 mutant has lower ratios of 2C and 4C and higher ratios of 32C, 64C and 128C compared to the wild-type control. (Lower) Photographs of wild-type (left) and hpy2-mutant (right) plants ten days after germination. The mutant has very small roots and leaves. The Blue staining indicates defective cell division in the mutant.

Sugimoto’s group demonstrated that the protein produced by HPY2 mediates the binding of SUMO to other proteins, resulting in the regulation of cell division. This was the first report of SUMO being associated with the regulation of cell division in multicellular organisms. “I never thought that my studies on the mechanism of plant cell size control would lead to SUMO. Research is fascinating because it can lead to unexpected results.”

Sugimoto has also discovered three other genes that control the switch into endoreduplication like HPY2. Her next task is to clarify the differences in their functions.

A gene terminating cell growth

In September 2009, following the discovery of HPY2, Sugimoto’s group discovered a gene involved in the second point—when plant cells stop growing in size. “It began with the discovery of a mutant having very large trichomes by Christian Breuer, a posdoc in my lab, who was searching for mutants with abnormal cell size.”

Trichomes are hair-like outgrowths that cover the surfaces of Arabidopsis leaves to protect them from insects, pathogens and even ultraviolet radiation. “Each trichome comprises a single epidermal cell in Arabidopsis. It is large enough to be seen macroscopically.” While even a normal-sized trichome is 500 times larger than an ordinary cell, the mutant discovered by Breuer has trichomes that are more than twice this size.

The mutant was found to have the GTL1 gene partially modified and expressed in excess. When the function of GTL1 was artificially suppressed, the mutant’s trichomes became more than twice the size of wild-type trichomes. Based on these experiments, Sugimoto’s group hypothesized that GTL1 functions to terminate cell growth. To test the hypothesis, they examined when and where GTL1 is expressed. It was found not to be expressed in smaller trichomes in the early stage of growth or trichomes that had stopped growing, but to be expressed only in trichomes that have just expanded to maximum size.

Previously, it had been thought that cell growth ceases when the supply of cellulose and other components of the cell wall is stopped, or when water absorption in vacuoles ceases. However, the discovery of GTL1 shows that plants have an intrinsic mechanism for actively stopping cell growth. The discovery is groundbreaking, overturning the traditional concept of plant growth.

Trichomes are cells undergoing endoreduplication, which is known to cease at 32C. Sugimoto’s group is conducting research on the hypothesis that GTL1 may control endoreduplication. It is already known that the function of the gene necessary for endoreduplication is activated in mutants lacking the function of GTL1. “GTL1 produces a protein known as a transcription factor, which binds to the DNA of a certain gene to promote or suppress its transcription to RNA. In the future, I want to clarify how GTL1 controls transcription and of which genes, and to discover the mechanism of endoreduplication.”

Giant prospects

Since the announcement of the discovery of GTL1, Sugimoto has received a flood of offers for joint research, including many inquiries from industry, who have great expectations for creating larger fruits and vegetables by suppressing the function of GTL1.

Some cultivars are already available with increased yields thanks to artificial duplication of nuclear DNA with chemical agents. However, this chemical treatment unavoidably duplicates the nuclear DNA in all cells constituting the plant body, which in turn makes the plant unable to produce seeds. “Advanced research on GTL1 may allow us to promote endoreduplication at desirable portions of plants, such as fruits, flowers and leaves, or whenever needed, to change their sizes without preventing seed production,” says Sugimoto, who is keen to conduct joint research with industry.

“Now is the most enjoyable time in my academic career,” declares Sugimoto. However, she is not satisfied with just discovering the genes that control plant growth. Further extensive investigation of the functions of individual genes is needed. It is also necessary to identify the targets of HPY2 and GTL1 to determine on which genes and proteins they act. She is also interested in the relationship between HPY2 and GTL1, and their association with plant hormones. “Much remains to be done, and I have not found the answer to my question about lilies when I was a high school student. In the plant kingdom, there are so many unanswered questions. This is why I am fascinated by plant research.”