Welcome to my homepage.
Dental Pulp Stem Cells
Inside your teeth, there is a small piece of soft tissue called dental pulp. You may not feel the presence of such a small tissue in daily life, but it plays quite an important role in tooth maineinance.
We prepared cells from dental pulps of discarded teeth, and expanded them. Those cells are called dental pulp stem cells or simply dental pulp cells. Unexpectedly, those cells grew very well and we could make a large human cell collection (more than 300 lines!) from medical waste. We colaborated with Prof. Yamanaka, a famous Nobel Prizest, and succeeded to genreate pluripotent stem cells with his innovative tecnique.
Currently, we are collecting dental pulp cells from donors having gentically homophilic human leukocyte antigen haplotypes (HLA haplotype homo donor, HHH donor). Japan is a unique country with racially homogeneous population geographically isolated from other continents by ocean. Through our experiments, we are starting to find HHH donors more frequently than expected.
We are not sure how these cell resource can be applied to future regenerative medicine. But, collecting as many variation of HHH cells as possible will surely be beneficial in application for such next generation technology. We may also create a functional artificial organs using them, and, for such purpose some kind of theolitical background will be needed to challenge such kind of innovative field.
What is iBone?
iBone is a computer simulation. In the year 2000, I was looking for a new strategy for the future bone biology research. At that time, I was using molecular biological techniques to understand the mechanism of bone physiology. We have already succeeded to clone several important molecules, such as OSF-1 (Pleiotrophin), Cathepsin K, and POEM (Nephronectin), however, we could not obtain a bird's-eye view of the bone physiology. It was very painful to feel that I was not going to see something I want to see.
Then, I started to learn about complexity. Complexity is a new field of science, by which scientists use computer simulation to understand complex system made with simple rules, like biological organisms. One of the models which I was most impressed was the Turing's reaction-diffusion system. You may remember the fish skin pattern simulated by this model published in Nature (Kondo and Asai, Nature 1995). Soon, I have met Prof. Kikuchi, who is a specialist in destructive mechanics field. He used finite element method (FEM) to calculate and simulate how materials are broken. Bone is a structure which do not break for a long time (my bone have never broken in my life!). He was interested in my project trying to create such a sturucture, and iBone project started.
At first, I did not know anything about finite element method. Even worse, I have almost forgotten mathematics and physics I have learned when I was a student. But, Prof. Kikuchi said to me, "Do not read textbooks of FEM. Those are too difficult and hard to be understood by the biginners. Just play with some of my programs." Then, I was given a set of FEM program and pre-post prosessors, and I bought my first intel CPU machine equipped with Pentium III 933MHz to install them. The very first iBone machine (iBone1) was born.
How iBone works.
The very first objective of iBone project was to create a system which behaves just like bone. Then, what makes bone look like a piece of bone? I think the answer is shape. Each piece of bone has a very unique shape. Then, how these shapes are created? In 19th century, a german surgeon, Dr. Julius Wolff (1892), wrote a book named "The law of bone remodeling." In this book, Wolff concluded that (trabecular) bone is present only along the mathematical stress trajectories. This means internal structure of bone is adapted to the mechanical environment. This rule is very simple, and looked very attractive to me. Yes, bone is a simple organ, whose primary function is to adapt to the mechanical stress. Bone is a system that translate mechanical stress to the shape.....
Then, I started to create a bone remodeling simulation. I knew that bone changes shapes by the activities of two kinds of cells, osteoblasts and osteoclasts. What I needed was some sort of system which can change the given shape to one which was more adapted to the given stress. At that time, in the field of mechanics, there were several methods to obtain optimal shapes for given stress (by Kikuchi N., Azegami H. etc..). Even in the bone biology field, a few groups had already proposed bone remodeling models (Huiskes et al. Nature 2000, Adachi et al. J Biomech. Eng.). Each of these method use different complicated calculation, however, basically all of them took similar strategy to obtain adapted shapes. Put more marterials where the stress concentrates. Remove materials from where the stress does not concentrate.
As I mentioned above, I was very interested in Turing's reaction-diffusion system. So, I decided to use this system for my biological simulation. Turing used two kinds of hypothetical materials with opposing effects, activator and inhibitor. Bone remodeling is also performed by two kinds of cells with opposing functions, osteoblasts and osteoclasts. Ok let's define as, activator = osteoblast, and inhibitor = osteoclasts (this hypothesis has been changed a bit in the current model). Then, where is the stress? To my knowledge, osteoblasts respond to stress, and so activator should be produced in response to local stress..... The simulation was very successful. Basically, I could adapt any shapes to the given stress with this simple reaction-diffusion system as shown in the following paragraph.
iBone can make something like bone.
Please just have a look of this movie. You may need to have quicktime installed in your PC. As you can see, at the beginning, the internal structure of model bone is filled with uniform structure with holes like a piece of cartilage (cartilage is a primordial tissue for most of the weight baring bones). Stress mimicing the body weight added on the femoral head and a strong muscle which pulls up the greater trochanger was given to the model bone. Then, the reaction-diffusion system started to calculate the concentration of activator and inhibitor depending on the local stress value provided by FEM simulation. Where the stress concentrates, more activator were produced. At the circumferential region of stress center, inhibitor became dominant. Then, iBone program put new bone material where the activator accumulated, while removing material from the place where inhibitor showed relatively high concentration in comparison with that of activator.
After 30 cycles of calculation, there was no stress concentration anywhere, and the internal structure of the model bone looked just like a piece of bone! The thin beams of bone inside the model are called trabecular bones. The thick area observed at the surface of bone shaft is called cortical bone. I did not intended to create these distinct structures. In other words, I have never put any instructions in the program to create these two kinds of structures in my iBone software. But, it was just formed!
Three dimension iBone.
We are going to publish the result described above in the next year's Journal of Bone and Mineral Metabolism. That paper will include several interesting results obtained with iBone simulation. Then, what's next? We are now creating a new powerful computer called, iBone cluster, to calculate large 3-dimensional bone models. This is very important to apply our iBone simulation to clinical use. The real world is 3-dimension, not two. Yes, our bones are 3-dimensional. So, to calculate our bones with iBone simulation, we have to move a step forward. But, that is not very easy. To calculate stress distriibution of a 0.75 million element-3D-model, it takes about 2 hours by single Pentium 4 (3GHz) machine. iBone uses at least 2 FEM calculation with different loading conditions for each cycles of calculation, and typically 30 cycles of calculation are performed in one simulation. Therefore, single iBone simulation will take more than 120 hours = 5 days of continuous calculation! To obtain one meaningful result, usually more than 10 simulations are needed. So.... you may understand the reason why we need a powerful computer. The cheapest way to obtain such a computer is clustering PCs. That is called PC cluster. Look at this picture. We bought 11 Pentium 4 machines, and connected them with Gigabit ethernet. We found the 22-inch monitor in the dumping place of our university. The total cost to create this cluster was about $30000, and this machine can calculate the above model with about 700 seconds. Now, we can perform 3-dimensional simulation just overnight.
To be continued....
Recent Publications
Naritaka Tamaoki, Kazutoshi Takahashi, Hitomi Aoki, Kazuki Iida, Tomoko Kawaguchi, Daijirou Hatakeyama, Masatoshi Inden, Naoyuki Chosa, Akira Ishisaki, Takahiro Kunisada, Toshiyuki Shibata, Naoki Goshima, Shinya Yamanaka, and Ken-ichi Tezukaa: The homeobox gene DLX4 promotes generation of human induced pluripotent stem cells. Sci Rep. 2014; 4: 7283, 2014 K. Iida, T. Kawaguchi, M. Hada, M. Yuriguchi, H. Aoki, N. Tamaoki, D. Hatakeyama, T. Kunisada, T. Shibata, and K. Tezuka: Hypoxia-enhanced Derivation of iPSCs from Human Dental Pulp Cells. J. Dental Res. 92: 905-910, 2013 Keisuke Okita, Yasuko Matsumura, Yoshiko Sato, Aki Okada, Asuka Morizane, Satoshi Okamoto, Hyenjong Hong, Masato Nakagawa, Koji Tanabe, Ken-ichi Tezuka, Toshiyuki Shibata, Takahiro Kunisada, Masayo Takahashi, Jun Takahashi, Hiroh Saji, and Shinya Yamanaka: Efficient and Simple Method to Generate Integration-Free Human iPS Cells. Nature Methods. 8, 409-412, 2011. Kazuki Iida, Tomoko Takeda-Kawaguchi, Yoko Tezuka, Takahiro Kunisada, Toshiyuki Shibata, Ken-ichi Tezuka: Hypoxia enhances colony formation and proliferation but inhibits differentiation of human dental pulp cells. Arch. Oral Biol. 55, 648-654, 2010. N. Tamaoki, K. Takahashi, T. Tanaka, T. Ichisaka, H. Aoki, T. Takeda-Kawaguchi, K. Iida, T. Kunisada, T. Shibata, S. Yamanaka, K. Tezuka: Dental Pulp Cells for Induced Pluripotent Stem Cell Banking J.Dent. Res. 89, 773-778, 2010. T. Takeda, Y. Tezuka, M. Horiuchi, K. Hosono, K. Iida, D. Hatakeyama, S. Miyaki, T. Kunisada, T. Shibata, K. Tezuka: Characterization of Dental Pulp Stem Cells of Human Tooth Germs. J.Dent. Res. 87, 676-681, 2008. Ken-ichi Tezuka, Tomoko Takeda, Yoshitaka Wada, Akiyuki Takahashi, and Masanori Kikuchi: Computer simulation of human mandible bone structure by iBone, a novel reaction-diffusion bone remodeling model. Key Engineering Materials, vols. 306-308, pp. 1277-1282, 2006. Yasuko Sugawara, Hiroshi Kamioka, Tadashi Honjo, Ken-ichi Tezuka, Teruko Takano-Yamamoto; Three-dimensional reconstruction of chick calvarial osteocytes and their cell process using confocal microscopy. Bone, 36, 877-883, 2005. Ken-ichi Tezuka, Yoshitaka Wada, Akiyuki Takahashi, and Masanori Kikuchi: Computer-simulated bone architecture in a simple bone-remodeling model based on a reaction-diffusion system. J. Bone Miner. Metabol. 23, 1-7, 2005. Ken-ichi Tezuka, Yoshitaka Wada, and Masanori Kikuchi: iBone: a Reaction Diffusion Based Shape Optimization Method. Progress in Experimental and Computational Mechanics in Engineering, Key Engineering Materials Vols. 243-244, pp601-606, 2003.
CV
1987 B.S. Biology, Kyoto University
1987-91 Researcher, Department of Molecular Biology, Hoechst Japan Research Laboratories
1991-94 Research Assistant, Department of Oral Anatomy, Meikai Univerisity
1994 Ph.D. Dentistry, Meikai Univerisity
1994-96 Postdoctoral Scientist, Department of Bone Biology and Osteoporosis Research, Merck Research Laboratories
1997-2002 Assistant Professor, Research Institute for Biological Sciences, Tokyo University of Science
2002- Associate Professor, Department of Tissue and Organ Development, Gifu University
2004- Investigator, PRESTO, Japan Science and Technology Agency
Ken-ichi Tezuka, Ph.D., Associate Professor
Department of Tissue and Oragan Development
Gifu University Graduate School of Medicine
Gifu-city, Gifu, JAPAN
