SINGLE MOLECULE NANO-BIOSCIENCE

Toshio Yanagida

Graduate School of Frontier Bioscience and Graduate School of Medicine, Osaka University, 1-3, Yamadaoka, Suita, Osaka, 565-0871, Japan

1. Introduction

Biomolecules assemble to form molecular machines such as molecular motors, cell signal processors, DNA transcription processors and protein synthesizers to fulfill their functions. Their collaboration allows the activity of biological systems. The reactions and behaviors of molecular machines vary flexibly while responding to their surroundings. This flexibility is essential for biological organisms. The underlying mechanism of molecular machines is not as simple as that expected from analogy with man-made machines. Since molecular machines are only nanometers in size and has a flexible structure, it is very prone to thermal agitation. Furthermore, the input energy level is not much difference from average thermal energy, kBT. Molecular machines can thus operate under the strong influence of this thermal noise, with a high efficiency of energy conversion. They would not overcome thermal noise but effectively use it for their functions. This is in sharp contrast to man-made machines that operate at energies much higher than the thermal noise. In recent years, the single molecule detection (SMD) and nano-technologies have rapidly been expanding to include a wide range of life science. The dynamic properties of biomolecules and the unique operations of molecular machines, which were previously hidden in averaged ensemble measurements, have now been unveiled. The aim of our research is to approach the engineering principle of adaptive biological system by uncovering the unique operation of biological molecular machines. I survey our SMD experiments designed to investigate molecular motors, enzyme reactions, protein dynamics and cell signaling, and discuss the mechanism of biological molecular machines.

2. Single-molecule detection (SMD) techniques

How advantageous is SMD for investigations? Observing and manipulating biomolecules allows their dynamic behaviors to be directly revealed, as has been demonstrated for motor proteins. Reactions of biological molecules are generally stochastic. Therefore, even if the reactions of molecules are initiated at the same time, they cannot be synchronized, so the dynamic behaviors of individual molecules are averaged and hidden in ensemble-averaged measurements. In vivo, biomolecules work in dynamic and complicated heterogeneous systems, involving different kinds of molecules such as cell-signaling proteins. It is difficult to quantitatively detect dynamic behaviors of target molecules in such systems by ensemble-averaged measurements. The SMD techniques are expected to overcome these difficulties and have already been successfully applied to study the dynamic properties of biological molecules such as motor proteins, enzymes, RNA polymerase and cell-signaling proteins.

The SMD techniques are based on two key technologies for single-molecule imaging and single-molecule nanomanipulation. First, the imaging technique will be explained. The size of biomolecules and even their assemblies are in the order of nanometers, so they are too small to observe by optical microscopy. To overcome this problem, biomolecules can be fluorescently labeled and visualized using fluorescence microscopy. Single fluorophores in aqueous solution were first observed in 1995 by using total internal reflection fluorescence microscopy (TIRFM) and conventional inverted fluorescence microscopy. The major problem to overcome when visualizing single fluorophores in aqueous solution is the huge background noise, which can be caused by Raman scattering from water molecules, incident light breaking through filters, luminescence arising from the objective lens, immersion oil and dust. In this system, the evanescent field was formed when the laser beam was totally reflected by the interface between the solution and the glass. The evanescent field was not restricted to the diffraction limit of light, thus it could be localized close to the glass surface, which resulted in the penetration depth (~150 nm) being several-fold shorter than the wavelength of light. Therefore, the illumination was restricted to fluorophores either bound to the glass surface or located close by, thereby reducing the background light. Furthermore, by careful selection of optical elements, the background noise could be reduced by 2000-fold compared with that of conventional fluorescence microscopy. This made it possible to clearly observe single fluorophores in aqueous solution. Fluorescence measurements from fluorophores attached to biomolecules and ligands allow the detection of, for example, the movements, conformational changes, enzymatic reactions and cell-signal processes of biomolecules at the single molecule level.

The second key technology is single-molecule nanomanipulation. Biomolecules and even single molecules can be captured by a glass needle or by beads trapped by optical tweezers. The optical tweezers are the tool to trap and manipulate particles of 25 nm to 25 µm in diameter by the force of laser radiation pressure. The particle is trapped near the focus of laser light when focused by a microscope objective with a high numerical aperture. The optical tweezers produce forces in the piconewton range on the particles. Biomolecules are too small to be directly trapped by the optical tweezers, so they are generally attached to an optically -trapped bead. Microneedles or a bead trapped by a laser act as a spring that expands in proportion to the applied force. Thus, the force and the displacement caused by the biomolecules can be measured. The displacement of a microneedle and a bead has been determined with a subnanometer accuracy, much less than the diffraction limit of an optical measurement. This accuracy of displacement corresponds to the sub-piconewton accuracy in the force measurements. Thus, the mechanical property of biomolecules can be determined directly at the single-molecule level. Furthermore, combined with the single-molecule imaging technique, simultaneous measurements of mechanical and chemical reactions of single biomolecules are possible.

3. Movement and ATPase turnovers of biological molecular motors

The SDM techniques were first used to study molecular motors. The example discussed here is that of a microtubule-based kinesin motor, which transports organelles along a microtubule in cells. Kinesin is composed of two heavy chains, each consisting of a force-generating globular domain (head), a long α-helical coiled -coil and a small globular C-terminal domain (tail). Microtubules are cylinders comprising parallel protofilaments, which usually number 13 or 14 when reassembled in vitro. Movement of single kinesin molecules along a microtubule has been directly observed by TIRFM. Kinesin molecules, fluorescently labeled at the tail-end without damage, were added to microtubules adsorbed onto a glass surface in the presence of ATP (adenosine triphosphate). This demonstrated directly that a single molecule of kinesin could processively move for long distances along a microtubule without dissociating.

In general, the motions of molecular motors and operations of other biomolecules are fueled by the chemical energy released from ATP hydrolysis. Kinesin and myosin are both motor proteins and ATPases. To uncover how the biomolecules work using the chemical energy from ATP, it is crucial to observe the individual cycles of ATP hydrolysis by single ATPase molecules. This has been achieved using the single-molecule imaging technique TIRFM in combination with the fluorescent ATP analog, Cy3-ATP . This method was first applied to an actin-based myosin motor that is involved in muscle contraction and other cellular motility. Myosin has a similar structure to kinesin, although it is twofold larger in size. Cy3-ATP is hydrolyzed by myosin in the same way as ATP . The rate of the biochemical cycle of ATP hydrolysis averaged for many events or individual myosin molecules was consistent with that obtained by a conventional biochemical method using a suspension of myosin.

4. Simultaneous observation of the ATPase turnover and mechanical events of an actin-based myosin motor

To investigate how the mechanical event of myosin corresponds to the ATPase cycle, the single-molecule imaging technique was combined with optical-trapping nanometry to simultaneously measure individual ATPase cycles and mechanical events of a single myosin molecule (Fig. 1). Dissociation of the myosin head from actin corresponded to the binding of ATP, and association of the myosin head with actin and generation of displacement were followed by dissociation of a nucleotide (most likely ADP). Each displacement corresponded to a single ATP molecule.

Understanding the process of a displacement became essential to investigating how myosin works using the chemical energy from ATP. Optical-trapping nanometry cannot resolve this process, because the displacement is determined indirectly through a long actin filament and optically-trapped beads, which have an unknown elasticity. Thus, the signal to noise ratio is not high enough to resolve the process. To overcome this problem, a more direct method combining scanning probe and single-molecule techniques has been developed. A single myosin head was attached to the tip of a scanning probe and the process of a displacement was resolved by measuring the displacement of a scanning probe with nanometer and millisecond accuracies. The results showed that a myosin head moved along an actin filament with regular 5.5 nm steps and underwent five steps to produce a maximum displacement of 30 nm per displacement (i.e. representing the ATPase cycle). As the step size coincides with the actin monomer repeat (5.5nm) and each 5.5 nm step is not directly coupled to the ATPase cycle (loose coupling), the results strongly indicate that the myosin head walks along the actin monomer repeat using biased Brownian motion. This idea challenges the widely accepted view that the movement is caused by a large conformational change in the myosin head, tightly coupled to the ATPase cycle in a one-to-one fashion (tight coupling).

5. DNA transcription

The initial steps of gene expression include the binding of RNA polymerase (RNAP) to DNA, the search for a promoter in the DNA sequence and the synthesis of RNA based on the information encoded by the DNA. These steps are central regulatory mechanisms of gene expression and have been extensively investigated. Harada et al. have observed single, fluorescently labeled RNAP molecules interacting with a single molecule of DNA suspended in solution using optical traps. The kinetic studies have proposed some mechanisms for promoter searching based on the results of the binding of RNAP molecules to specific and nonspecific sites on the DNA: sliding, intersegment transfer and simple dissociation and/or association reactions. This observation provides direct evidence that a sliding motion is a mechanism used for the search of promoters. The association and dissociation rate constants of RNAP could be also determined, depending on the sequence of DNA and on the mechanical strain exerted on the DNA. These values proved difficult to determine in solution because DNA molecules aggregate to form a network structure. The transcription process was directly monitored by measuring the displacement or rotation of DNA during the interaction with RNAP by manipulating DNA with optical and magnetic tweezers. One end of DNA is attached to a magnetically-trapped bead and the other is interacting with a RNAP adsorbed onto the glass surface. Rotation of DNA is determined by monitoring the rotation of the bead. The results showed that the rotation rate is consistent with high-fidelity tracking. The rotation per base pair is as much as 35[] and should, in principle, be detectable. Therefore, this method could resolve individual steps of transcription in real time.

6. Future perspectives

Recently, life science has made remarkable progress. This progress has been possible because of the identification of functional proteins and studies on the characteristic (structural and functional) properties of proteins as revealed by molecular cell biology, structural biology and molecular genetic approaches. The DNA sequences of not only invertebrates but also of vertebrates (including human) are now available. Thus, research is charging into the post-genomic era. In this era, we tend to emphasize the concept that the structure explains the function and we endeavor to understand the mechanisms of the proteins and molecular machines based on this concept. However, knowing the function of proteins and molecular machines is not a simple task. Moreover, we could not understand the function of these proteins even if we knew their structures. Proteins and molecular machines are not simple and their function cannot be learned in analogy to artificial machines. Proteins and molecular machines have a size in the nanometer range, and a dynamic and soft structure. In addition, the input energy to the molecular machines is comparable to thermal energy. Molecular machines function at a very high efficiency when exposed to thermal agitation. This is contrasted by artificial machines, which use much higher energy than thermal energy to work rapidly, accurately and deterministically. For such reasons, it is necessary to understand the dynamic properties of proteins themselves and their interactions to each other. The SMD techniques have been developed as techniques to directly monitor the dynamics of proteins and molecular machines and have rapidly expanded to include a wide field of biological sciences. Combined with nanotechnology from the engineering field, the SMD techniques will prove more powerful in the future. Thus, the SMD will govern and lead the future direction for research in proteins and molecular machines.

HIGH-SPEED PARTICLE SORTING: COMBINING DIELECTROPHORESIS AND FLUID FLOW

David Holmes, Mairi E. Sandison, Nicolas G. Green and Hywel Morgan

School of Electronics and Computing Science, University of Southampton, Highfield, Southampton, SO17 1BJ, UK.

Abstract

We present a high-speed particle sorting and deflection system which is an integral part of a micro flow-cytometer chip capable of high speed detection and sorting of micron-sized particles. The device sorts particles using a combination of DEP and hydrodynamic forces. DEP focusing of particles is used to axially centre particles in a channel. Negative dielectrophoresis, together with hydrodynamic flow is used to achieve high speed particle sorting at a microfluidic T-junction.

Keywords: particle sorting, dielectrophoresis, AC-electrokinetics, micro-flow cytometry

1. Introduction

A number of particle sorting and detection devices have appeared in the µTAS literature in recent years [e.g.]. Many of these devices mimic the operating principle of a bench-top flow -cytometer but on the micro-scale. Flow cytometers (or Fluorescent Activated Cell Sorters, FACS) allow counting and sorting of micron -sized particles such as cells and latex beads. The disadvantages of conventional FACS machines are the requirement for relatively large sample volumes (100µl), skilled operators and the high cost of such systems. Many of the functionalities of a FACS machine have been implemented into micro devices such as detection of fluorescence, or scattering from particles. Impedance detection has also been incorporated into micro-devices for single cell analysis. Although particles can be counted at relatively high speed in these micro-devices, particle sorting speeds are generally lower.

In this paper we present a novel approach to achieve rapid particle sorting. The method uses a combination of hydrodynamic forces and dielectrophoresis to achieve high speed particle deflection and sorting. Fuhr and co-workers presented DEP based cell sorting devices which used bar-shaped electrodes that deflected particles across a channel to one or other of an outlet. This method is relatively slow as the particle may have to be moved a relatively large distance across the channel to the outlet. Our device differs from previous designs in that it uses a particle positioning system to focus particles into the centre of the channel prior to the sorting junction as shown in Fig. 1. The operating principle of the device is shown in Fig. 2. Particles are centred along the axis of the channel, and arrive at the T-junction with equal probability of flowing into either outlet. A small negative-DEP deflection force is provided by the electric field configuration of the three electrodes at the gate, pushing a particle from the central stream line, so that it passes into one or other of the outlets.