Abstract—Despite the benefits of miniaturized devices, handling of tiny amount of molecules became a great challenge. Direct transport, similar to the one in intracellular transport, is a way to cope with the problem of transporting tiny amount of materials. Using motor proteins, i.e. kinesin, and the corresponding rail structures, i.e. microtubules, provides powerful building blocks for a nanotransport system. However, it is extremely difficult to construct a complex transport network to allow arbitrary design. Here, we have developed a MEMS-based molecular handling system for the pick-and-place assembly of individual microtubules. The technology enabled us to build a bio/MEMS hybrid system composed of multidirectional microtubule networks along which kinesin molecules move. Using MEMS tweezers to manipulate a single microtubule enables precise positioning and full control in functional orientation and stacking order. As a result, kinesin-coated beads can be transported along the route specifically assembled for them to follow. Direct transport of target molecules can be achieved by loading the beads with the molecules to carry. “Bottom-up” functionalities of biomaterials are incorporated with micro fluidic devices by the handling capabilities of the “top-down” approach. I. INTRODUCTION N the last decades, with the developments in the miniaturization of devices, smaller, faster and cheaper equipments have become available. Integration of biological materials with the miniaturized devices is extremely promising for healthcare and environmental monitoring. With smaller dimensions, necessary amounts of sample volume can be decreased. This is especially important in experiments where the target molecules are scarce. However, conventional continuous-flow-type micro fluidic devices have limitations in handling small amount of molecules due to diffusion in flowing medium against extremely high viscous drag in sub-micrometer channels [1]. Direct transport of molecules is a possible way to cope with the problems caused by miniaturization. In living Manuscript received March 10, 2010. M. C. Tarhan, D. Collard and H. Fujita are with CIRMM, Institute of Industrial Science, The University of Tokyo, Tokyo, 153-8505, Japan (phone: 81 3 5452 6249; fax: 81 3 5452 6250; e-mail: mctarhan@iis.u-tokyo.ac.jp, fujita@iis.u-tokyo.ac.jp). R. Yokokawa is with Department of Microengineering, Kyoto University, Kyoto, 606-8501, Japan, and PRESTO, JST, Kawaguchi, Saitama 332-0012, Japan (e-mail: ryuji@me.kyoto-u.ac.jp). L. Jalabert and D. Collard are with Laboratory for Integrated Micro Mechatronic Systems/CNRS-IIS, University of Tokyo, Tokyo, 153-8505, Japan (e-mail: jalabert@iis.u-tokyo.ac.jp, collard@iis.u-tokyo.ac.jp). organisms, intracellular transport widely uses direct transport provided by the motor proteins moving along rail structures. Direct transport has advantages on carrying small amount of molecules, because no liquid manipulation is necessary as in the case of microfluidics. Target molecules can be captured on the carriers via highly selective molecular recognition. Then, the carriers can be directly, and separately if needed, transported to the desired location for detection or further processing. Being the main components of cell cytoskeleton, microtubules have very important roles in intracellular transport as rail structures (Fig. 1). The polymeric structure of microtubules is assembled from tubulin monomers [2]. The polymerization processes with a fast growing (+) end and a slow growing (-) end. Once completed the polymerization process, microtubules can be found as hollow tubes of tens of micrometers long with a diameter of about 25 nm; they serve as a rail for kinesin molecules. Kinesin, a linear motor protein, moves along microtubules by hydrolyzing adeneosine tri-phosphate (ATP) [3]. Having two heads to bind to microtubules, conventional kinesin can move unidirectionaly according to the functional polarity of the microtubules from (-) end toward (+) end (Fig. 1). The tail of kinesin can be used to attach target molecules directly or through intermidiate carriers [4]. There have been several attempts to build transport systems based on motor proteins [5-8]. One of the main issues is controlling the direction of kinesin motion along immobilized microtubules. As kinesin motion is unidirectional depending on the polarity of underlying microtubules, functional orientation of microtubule polarity determines the direction of motion. However, orientation process is quite challenging and requires complex procedure with additional setup [9-12]. Even with the existing techniques, it is extremely challenging (if not impossible) to build a rail network with multidirectional and multilayered Biomotor-based Nanotransport System Constructed by Pick-and-Place Assembly of Individual Molecules Mehmet Cagatay Tarhan, Ryuji Yokokawa, Member, IEEE, Laurent Jalabert, Dominique Collard, Member, IEEE and Hiroyuki Fujita, Member, IEEE I Figure 1: Schematic view of a kinesin molecule moving along a microtubule from (-) end to (+) end. The 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems October 18-22, 2010, Taipei, Taiwan 978-1-4244-6676-4/10/$25.00 ©2010 IEEE 5628