2.1.

Admissible Size Range of Gripped Object

Optical tweezers have been shown to be very useful in trapping objects that are between a few hundred nanometers to about ten micrometers in diameter. Thermal or Langevin forces dominate below this size range such that high laser intensities are required to provide sufficient counteracting trapping forces. Gravity and viscous drag (drag only for moving particles) forces are dominant for larger sized objects such that high laser intensities are again required to ensure stable trapping. Now, even though laser beams are not focused directly on the biological objects in the case of indirect optical manipulation, some amount of light is always incident on the objects. Hence, high laser intensities are potentially damaging for the biological objects that we are trying to manipulate. So, the size of the gripper beads is usually restricted to lie within this range where tweezers can work satisfactorily at reasonable intensities.

2.2.

Role of Gripped Object Size

There is a strong correlation between the size of the gripped biological object and the number and size of the gripper finger beads. Just as larger, stronger robotic grippers are required to grasp bigger objects, relatively larger and a greater number of beads are necessary to indirectly manipulate the cells as compared to the much smaller nucleic acids and motor proteins (in terms of axial diameter or neck thickness). This can be easily explained from the fact that larger and a greater number of beads experience stronger optical trapping forces and, hence, can exert stronger contact forces to manipulate the gripped object. We should note here that although thermal and drag forces are smaller for larger objects, the enhanced effect of gravity more than counterbalances this decrease.

2.3.

Impact of Gripped Object Shape and Manipulation Type

The shape of the object and the desired form of manipulation play a significant role in governing the number of beads and the type of localization force. Since nucleic acid molecules are axially elongated, either one or two beads need to be attached at the ends of the molecules to manipulate them. Both the beads can be optically trapped or one of the beads can be conveniently held in a micropipette by means of suction force or even tethered to the coverslip surface. Two alternative arrangements are shown for DNA molecules in Fig. 1. Motor proteins are typically so small that just a single bead is sufficient for manipulation. However, if the objective of the experiment is to study the interaction between the motor protein and the walking medium, then three beads are often used. One is attached to the molecule itself, while the other two are attached to the two ends of the axial microtubule or actin filament, thus holding a piece of the scaffolding on which the motor proteins move. For manipulation of a suspended cell, different multibead arrangements are possible as depicted in Fig. 2. A two-bead arrangement works well if the purpose is to stretch the cell from two sides. On the other hand, a four or six bead arrangement is more suitable if the aim is to grasp a cell strongly and transport it at a reasonable speed. As cells are commonly translated and rotated instead of being only stretched, the beads are not usually held in micropipettes or tethered to coverslip surfaces, and are always optically trapped.

Fig. 1

Schematic illustration of two bead arrangement to manipulate DNA molecule; (a) (adapted from Ref. 25) shows a setup where both the beads are optically trapped, whereas (b) (adapted from Ref. 26) depicts a setup where one of the beads is held in a micropipette. Note that the figure is not drawn to scale.

Fig. 2

Schematic illustration of two and six bead arrangements to manipulate cells; (a) (adapted from Ref. 27) is useful for stretching red blood cells, while (b)(adapted from Ref. 28) is useful for transporting cells. Note that the figure is not drawn to scale.

Based on these observations, we can summarize some general principles for designing indirect optical manipulation setups as follows:

These principles provide justification for the successful experimental setups that have been designed so far as shown in the next three review sections. We also believe that they will prove to be very useful to future researchers who are planning to develop their own systems. As an example, if one wants to build a system for indirectly manipulating cell organelles such as vesicular networks of endoplasmic reticulum, the person can follow the basic principles that we stated above and select a small size of the beads (roughly the same as that used for motor proteins), and attach two properly-coated beads to the ends of the network strands. It turns out that this is exactly the arrangement which is used in Ref. 29. Thus, our simple analysis provides the foundation for a systematic and informed design of manipulation setup for any kind of biological object that resembles the geometry of the three commonly-studied systems presented in the paper.

3.1.

Representative Work in Indirect Manipulation

Many researchers have investigated the properties of cells using indirect optical manipulation. Laurent ³⁸ measured the viscoelastic properties of alveolar epithelial cells using magnetic twisting cytometry and optical tweezers. Li ²⁷ studied the deformation of the erythrocyte cells by stretching them through optically trapped beads. Fontes ³⁹ developed a new method to measure mechanical and electrical properties of red blood cell (RBC) rouleux using double optical tweezers. Wei ⁴⁰ used a micro, rheometer based on oscillatory optical tweezer to measure the extracellular and intracellular complex shear modulus for alveolar epithelial cell. Python ⁴¹ studied the viscoelastic properties of microvilli using optical tweezers.

In order to measure the mechanical force constants (i.e., perform tensile tests), cells need to be held from one side while applying force on the other side. Indirect manipulation systems based on optical tweezers have proven to be appropriate for these kinds of experiments because of their ability to localize a bead on the surface of the cell precisely to hold it from one side or apply force. Henon ⁴² used optical tweezers to measure the shear modulus of RBC. Using a two bead arrangement, Li and Liu⁴³ measured the transverse and longitudinal strains of RBCs both experimentally using the optical tweezer system and theoretically using the finite element analysis (FEA) model analysis. The best matched results were then used to calculate the elastic constants of RBCs. In a similar experiment, Wu ⁴⁴ and Sleep ⁴⁵ measured the elasticity of RBCs with increasing osmotic pressure. Another research group led by Li ⁴⁶ attached one side of RBC with the coverslip and applied force on the other side using an optically trapped bead to measure the mechanical properties. Tan ⁴⁷ used a similar procedure for mechanical characterization of RBCs.

Some researchers have investigated the response of cells to external stimuli using optical tweezers. Miyata ⁴⁸ studied the effect of temperature and opposing force on the gliding speed of the bacteria Mycoplasma mobile. Kress ⁴⁹ investigated the binding mechanism of cells during phagocytosis using an optically-trapped bead as a local probe. Taka ⁵⁰ studied the dynamic behavior of a fibroblast cell membranes. Pozzo ⁵¹ used optical tweezers to study the chemotaxis behavior of a flagellated micro-organism when exposed to a gradient of attractive chemical substance. In order to understand the role of the pili of E. coli during adhesion to the host tissues, Andersson ⁵² studied the biomolecular properties of pilis.

Some researchers have also experimented with new optical tweezer setups. For example, Ferrari ⁵³ used two different setups to create multiple traps for indirect manipulation of biological objects. One of the setups used acousto-optic deflectors to achieve deflection of laser fast enough to maintain multiple traps by sequential sharing of the laser beam. However, this setup could only provide planar trapping configuration. The second setup used diffractive optical elements. The optical tweezer setup developed by Mejean ⁵⁴ was capable of measuring the mechanical coupling force between the cytoskeleton of Aplysia bag cell and neutron cell adhesion molecule.

Some researchers have also used objects other than microspheres as handles. For example, Sun ⁵⁵ used an irregularly shaped diamond as handles for the controlled rotation and translation of cells. Ichikawa ⁵⁶ proposed a new method for manipulation of micro-organisms by instantly creating and destroying the microtool. The microtool was formed by local thermal gelation using the laser power. After manipulation the microtool was dissolved by stopping the laser. Zhang ⁵⁷ successfully manipulated RBCs under various physiological flow conditions by attaching microbeads using optical tweezers.

Many researchers have been interested in cell sorting using optical tweezer systems. Dholakia ⁵⁸ performed passive cell sorting operation inside the microfluidic chamber by applying optical forces. Cells were tagged with microspheres to provide variations in refractive indices which enhanced the speed of the sorting process. Paterson ⁵⁹ used the same idea of tagging cells with microspheres in order to sort them using Bessel light beams. Mthunzi ⁶⁰ tagged mammalian cells with microspheres in order to improve the manipulation forces.

3.2.

Comparison with Other Approaches

As cell manipulation is an important area both for medical applications and making fundamental advances in biological sciences, several different techniques have been developed for manipulating cells. Dielectrophoresis has been successfully used to manipulate cells.⁶¹ Magnetic manipulation involves tagging cells by magnetic particles and then using the time varying magnetic field to move the particles, and, hence, the cells.⁶² Both of these methods impose restrictions on the type of cells that can be manipulated by these methods and the environments in which the cells should be manipulated. Moreover, it is very difficult to achieve an independent placement control over multiple cells concurrently.

Recent advances in silicon and polymer-based micro-electromechanical systems have been exploited to develop microscale grippers that can hold individual cells and arrays of cells.⁶³ These methods utilize customized grippers to grasp the cells. These grippers are used in conjunction with mechanical micromanipulators to move the cells. These grippers are not reconfigurable to allow for changes in the cell shapes. Moreover, only a limited field of view is available for imaging while the gripper is holding the cell. Integrating multiple mechanical manipulators together to perform multiple independent operations is challenging due to workspace limitations.

Microfluidics, when combined with, e.g., electro-osmotic actuation, can be a powerful tool to steer a small number of objects. It has been shown to be a useful technology for cell manipulation.⁶⁴ However, fluids are incompressible, making it harder to aggregate cells than optical traps. Microfluidics also generally requires a closed system for controlled flows and, thus, makes further manipulation of the sample by inserting a micropipette or a chemoattractant difficult unless they are integrated with the microfluidics device.

4.1.

Manipulation of DNA

DNA strands are too thin for simple direct manipulation, and, hence, virtually all work on DNA involves indirect manipulation. As is done throughout this paper, the focus is on the experimental setups being used and readers are advised to go through the review articles mentioned in Sec. 1 for details on the scientific breakthroughs in this active field of research.

Several methods have been used to apply forces to DNA strands at multiple points. Perkins ⁶⁵ first studied the relaxation of single DNA molecules by attaching optically trapped 1 μm latex microspheres to one end of the molecule and pulling the other end via fluid flow. A feedback stabilized motor was used to move the microscopic stage at a constant speed to generate fluid flow around the stationary, trapped beads that stretched the molecules. Smith ⁶⁶ attached microscopic latex beads to both ends of DNA molecules, one of which was trapped by a laser tweezer and the other one was held by suction on a glass micropipette. The DNA was extended by moving the micropipette relative to the optical trap.

Shivashankar and Libchaber⁶⁷ grafted DNA-tethered beads onto silicon substrates such as AFM cantilevers using optical tweezers. The other end of the DNA molecules were attached to coverslip surfaces. Shivashankar ⁶⁸ also studied the flexibility of DNA molecules by using an optical tweezer instead of an AFM. A collinear red laser light beam was used to probe the fluctuations in bead position with nanometer accuracy based on the direction of backscattered light. Wang ⁶⁹ used a similar setup but moved the coverglass with respect to the optical trap using a piezo-driven stage, while the bead position was recorded with nanometer-scale precision. A feedback circuit was activated to prevent bead movement beyond a preset clamping point by modulating the light intensity, thereby altering the trap stiffness dynamically.

The interaction of DNA with proteins has also been investigated. Bennink ⁷⁰ studied the interaction of DNA with Rec A and YOYO-1 molecules by capturing the DNA molecule between two polystyrene beads using biotin-streptavidin linkers. Wuite ⁷¹ used a similar setup to study the relation between the DNA strand tension and the activity of polymerase proteins bound to DNA. In one of the more unconventional studies, Arai ⁷² studied the mechanical properties of DNA molecules by continuously controlling the radius of curvature of the molecular strand by tying a knot in it.

At larger scales, Cui and Bustamante⁷³ studied the forces responsible for maintaining the higher-order structure of chromatin fibers using optical tweezers. They connected the two ends of a single biotinylated chromatin fiber between two avidin-coated polystyrene beads inside a flow chamber. Identical to the setup used in Ref. 74, one bead was trapped in a dual-beam optical tweezer, and the other was held atop a glass micropipette.

Bocklemann ⁷⁵ performed force measurements on single DNA molecules using optical tweezers to study the high sequence sensitivity of strand breakage. They created a molecular construction, wherein both strands of the DNA molecule to be unzipped were prolonged by linker arms of 2.5-μm length each, consisting of double-stranded DNA with multiple, modified base pairs at the ends. Hirano ⁷⁶ showed that the ends of single DNA molecules could be gripped by clustering microparticles using optical tweezers. As many as 40 latex beads of 0.2-μm diameter were aggregated at 400-mW laser power and moved at a speed of 40 μm/s.

Davenport ⁷⁷ studied transcriptional pausing and arrest by E. coli RNA polymerase in real time over large template distances using a combination of optical tweezer and flow-control video microscopy. The RNA polymerase-DNA complex was tethered between two 2.2-μm streptavidin-coated beads and kept in a continuous buffer flow. Soni ⁷⁸ combined optical tweezers with micropipettes to develop a setup that was capable of operating autonomously at constant force, constant velocity, or constant position. The authors conducted three different experiments that had sub-pN force sensitivity and a nanometer scale positioning accuracy to highlight the usefulness of the system.

Larson ⁷⁹ investigated the mechanism of transcription termination of bacterial RNA polymerase by creating a two-bead assay with unequal size of the optically-trapped polystyrene beads; the polymerase was attached to the smaller bead via a biotin-avidin linkage and the DNA template was attached to the larger bead via a digoxigenin–antidigoxigenin linkage. Terao ⁸⁰ manipulated single choromosomal DNA molecules by using a microhook and a microbobbin structure, both of which were driven using optical tweezers. The microhook was used to capture a molecule at any desired point and two microbobbins (one revolving around the other) were used to wind the molecule.

More recently, Galburt ⁸¹ studied the dynamics of transcriptional elongation of RNA polymerase II by using a dual-trap optical tweezer setup similar to the one described in Ref. 79 except for the fact that they used beads of identical size. Kleimann ⁸² investigated the binding kinetics of Triostin-A to λ-DNA by measuring the force-extension curves of the DNA-ligand complex. Landry ⁸³ characterized the damage caused by optical traps on DNA tethered between optically-trapped polystyrene beads. Lin ⁸⁴ developed an optically-induced dielectrophoresis platform to elongate and rotate single DNA molecules by tethering one end of the molecule to the substrate and binding the other end to a polystyrene bead.

Mameren used a combination of fluorescence microscopy, optical tweezers, and microfluidics to resolve the structural basis of DNA overstretching. They held and extended double-stranded DNA molecule by attaching both ends to an optically-trapped microsphere. The same setup⁸⁵ was also used to study the DNA strand tensions needed to disassemble (or shed) aggregates of proteins that can form around DNA strands, e.g., filamentous aggregates of RAD51. Mossa ⁸⁶ evaluated the equilibrium free energy differences in pulling DNA hairpins by adopting the common procedure of attaching beads at the two ends; one of them was held in a pipette, whereas the other one was placed inside a moving optical trap. Murade ⁸⁷ studied the force extension of DNA molecules in the presence of oxazole yellow dyes by integrating fluorescence microscopy imaging in an optical tweezer setup.

Carter ⁸⁸ developed an optical trapping assay with one DNA basepair resolution by employing active stabilization techniques to reduce the 3D surface motion and the effect of different sources of laser trap noise. Fuji ⁸⁹ presented a technique for fabricating a single DNA nanowire by using laser local heating at Au/water interface, wherein an optical tweezer was used to compress a bead attached to the DNA molecule to the solid surface, thereby resulting in pinning of DNA. Ommering ⁹⁰ characterized the bonding between streptavidin-coated polystyrene and superparamagnetic particles and a biosensor surface by tethering the particles with double-stranded DNA molecules and manipulating them using magnetic fields in the presence of a laser beam. Zohar ⁹¹ demonstrated the usefulness of modified peptide nucleic acids (PNAs) in manipulating DNA molecules by attaching one end of the PNA-DNA-digoxigenin complex to an optically-trapped, antidigoxigenin-coated polystyrene bead and the other end to a streptavidin-coated bead that was held in a micropipette.

4.2.

Manipulation of RNA

Single stranded RNA comes in a wide variety of arrangements in terms of secondary and tertiary structures, which have been investigated in depth. Liphardt ⁹² first applied mechanical force to induce unfolding and refolding of single RNA molecules by attaching them to polystyrene beads using DNA-RNA hybrid handles. Similar to the setup used in much of the reported work, one of the beads was held in a force-measuring optical trap and the other one was linked to a piezo-electric actuator through a micropipette. Three different RNA secondary structures were investigated, namely an RNA hairpin, a three-helix junction, and the P5abc domain.

Harlepp ⁹³ probed RNA secondary structures by combining single molecule stretching experiments with stochastic simulations. The RNA structures were hybridized to two double stranded DNA extensions, which were attached to beads and surfaces using biotin-streptavidin and digoxygenin-antidigoxigenin linkages. Mangeol ²⁵ probed the unfolding/refolding hysteresis behavior of RNA molecules after first validating their two-bead experimental setup by conducting experiments on stretching of DNA molecules. The mechanical unfolding, force-quench refolding, and the hopping rates of RNA hairpins were studied and a comparison of experimental and simulation results with theoretical analysis were conducted in Ref. 94.

Li ⁹⁵ investigated the mechanical folding kinetics of single RNA molecules by using dual-beam optical tweezers within the standard two bead setup, where the antidigoxigenin-coated polystyrene bead was held by suction in a micropipette and the other streptavidin-coated bead was held in an optical trap. Green ⁹⁶ characterized the mechanical unfolding of RNA pseudoknots formed from an infectious bronchitis virus using the same two bead, dual-beam optical tweezer setup. Wen ⁹⁷ also studied the effects of different experimental variables on the RNA folding/unfolding kinetics on a model RNA hairpin (P5ab) by using the two bead, dual-beam setup. In a companion article, Manosas ⁹⁸ applied a mescoscopic model to simulate the kinetics under comparable conditions and obtained good agreement with the experimental results.

5.1.

Manipulation of Kinesin

Block ⁹⁹ used optically-manipulated silica beads to measure movement of kinesin molecules along microtubules. The beads were coated with carrier proteins, exposed to varying kinesin concentrations, and individually manipulated by single-beam optical traps. Svoboda ¹⁰⁰ directly observed that kinesin moves with 8-nm steps. The motion was analyzed under varying laser power (and, consequently, trapping force) conditions using optical interferometry, which combined an optical tweezer with a dual-beam interferometer. The motion analysis was done by tracking the bead position, keeping the trap stationary after depositing the bead. Once the molecule would get released after traveling for a certain distance along the microtubule, the bead would return to the trap center, reattach, and fresh movement would start. Thus, an individual molecule motion could be studied for several minutes and up to hundreds of mechanochemical events, until the molecule failed to bind to the microtubule or got stuck irreversibly. Svoboda and Block¹⁰¹ also used the same principle to measure the force-velocity curve of single, silica bead-attached kinesin molecules moving on microtubules.

Kojima ¹⁰² studied the motion of kinesin molecules by adsorbing them onto optically-trapped latex beads, and then bringing them in contact with axenomes that were bound to a glass surface. Visscher ¹⁰³ studied how the chemical energy is coupled to mechanical displacements in single kinesin molecules by adsorbing them on to optically-trapped silica beads and moving them on immobilized microtubules. As the precision of the measurements increased, the position in the third dimension needed to be taken into account to compute forces and displacements. Jeney ¹⁰⁴ studied the mechanical properties of single kinesin molecules by recording the 3D positions of kinesin-coated, optically-trapped glass beads, tethered to cover-slip adsorbed microtubules, with a spatial precision in the nanometer range and a temporal resolution in the order of few tens of microseconds. Unlike in Ref. 103, where high loads were used to reduce Brownian motion, optical forces were kept to a minimum (of the order of fN) here to allow thermal fluctuations to dominate the probe measurements.

Carter and Cross¹⁰⁵ investigated the mechanics of kinesin stepping by attaching a single molecule to an optically-trapped spherical bead that was steered toward an immobilized microtubule. Bormuth ¹⁰⁶ used optical tweezers to characterize the frictional drag force of individual yeast kinesin-8 (Kip3p) molecules interacting with microtubule tracks in the presence of ADP. Kip3p-coated microspheres were dragged near immobilized microtubules at a low enough myosin concentration to ensure that only single molecules would interact with the microtubule. The microtubule was moved back and forth relative to the laser trap and the position of the microsphere was recorded. Guydosh and Block¹⁰⁷ provided useful insight on the interactions of the individual motor heads with the microtubule.

Gutierrez-Medina ¹⁰⁸ measured the torsional properties of kinesin molecules by attaching them with fluroscence-marked polystyrene beads and then trapping them in the solution medium using optical tweezers. The captured kinesin-bead complex was then placed near a microtubule that was immobilized on the coverglass surface. The optical trap was switched off after kinesin-microtubule binding took place, allowing free rotation of the tethered bead due to thermal forces. Bruunbauer ¹⁰⁹ investigated the regulation of heterodimeric kinesin-2 motor molecules by moving them on a microtubule tract that was attached to the coverslip surface. The molecules were coated on an optically-trapped polystyrene bead that was moved out and then pulled back into the trap focus during binding–unbinding with the tract, where the restoring force was provided by a piezoelectric stage clamped to the coverslip surface. Butterfield ¹¹⁰ conducted measurements of power strokes of kinesin-14 molecules using a three-bead geometry, where a biotin-coated microtubule was suspended between two streptavidin-labeled, optically-trapped silica beads. The microtubule was attached to the third, larger diameter bead that was sparsely coated with the motor molecules.

5.2.

Manipulation of Myosin

Finer ¹¹¹ measured the force and displacement resulting from the interaction of myosin with an actin filament where the substrate of myosin was micromanipulated. An actin filament was attached to polystyrene beads at each end and held in place by two optical traps. Just as in the kinesin studies, measurements were performed in the constant trap stiffness region by pulling the actin filament using one of the beads. Then it was brought close to the coverslip surface so that it could interact with one or a few myosin. Once contact was established, a quadrant photodiode was used for high resolution position detection of the other trapped bead that started moving along the direction of the filament.

Veigel ¹¹² used the three-bead setup described in Ref. 111 to measure the stiffness and working stroke of a single actomyosin structure. Wakayama ¹¹³ studied the motion of myosin actively sliding along actin filaments suspended between two immobilized microbeads which were trapped by double-beam optical tweezers. Clemen ¹¹⁴ used single beam optical tweezers with a force feedback that allowed for a large range of motion to study the stepping kinetics of myosin-V molecule under controlled forward and backward loads. Polystyrene beads exposed to myosin-V were optically trapped and positioned over fluorescently labeled, surface-anchored actin filaments. Cappello ¹¹⁵ used optical tweezers to bring a myosin-coated bead in contact with the actin filaments and the motion of the bead was recorded parallel and perpendicular to the filament axis with nanometer accuracy microseconds time resolution.

Instead of following the common three-bead setup, Arsenault ¹¹⁶ used dielectrophoresis to suspend actin filaments across a trench that was created between gold electrodes to study the helical motion of myosin molecules, which were attached to a bead held by an optical tweezer. One of the main advantages of this hybrid setup was to provide clearance beneath the filament to allow unhindered motion of the bead. Kaya and Higuchi¹¹⁷ measured the step size and stiffness of skeletal myosin molecules interacting with actin filaments that were suspended between two streptavidin-coated, optically-trapped beads. Streptavidin-coated quantum dots were also attached to the actin filaments to reduce the uncertainty in the linkage stiffness and single myosin molecules, embedded in myosin-rod co-filaments, were tightly bound to the filaments. Sellers and Veigel¹¹⁸ investigated the reversibility of the power stroke of myosin-Va motor heads; they made direct observations on the interactions of the myosin molecules present on surface-attached beads with F-actin filaments that were held between two optically-trapped polystyrene beads.

6.1.

Trends

6.2.

Challenges and Research Directions

Various approaches for indirect micromanipulation of cells and biomolecules are now well-established. In particular, some of the techniques are quite optimized for manipulating biomolecules. However, the slow speed of optical manipulation of cells, confinement to single-cell studies, and lack of widespread usage in cell biology laboratories and clinics indicate that a more systematic approach to design and control this complex system may be valuable for broader implementation. Hence, we believe that there are many promising areas of future research. We list them and briefly discuss how they may help in addressing the current challenges.