Letter pubs.acs.org/NanoLett
Measuring Nanoscale Forces with Living Probes S. N. Olof,†,‡,§ J. A. Grieve,† D. B. Phillips,† H. Rosenkranz,∥ M. L. Yallop,∥ M. J. Miles,† A. J. Patil,§ S. Mann,*,§ and D. M. Carberry*,† †
H. H. Wills Physics Laboratory, ‡Bristol Centre for Functional Nanomaterials, and §Centre for Organized Matter Chemistry, School of Chemistry, University of Bristol, Bristol, BS8 1TL, United Kingdom ∥ School of Biological Sciences, University of Bristol, Bristol, BS8 1UG, United Kingdom S Supporting Information *
ABSTRACT: Optical trapping techniques have been used to investigate fundamental biological processes ranging from the identification of the processive mechanisms of kinesin and myosin to understanding the mechanics of DNA. To date, these investigations have relied almost exclusively on the use of isotropic probes based on colloidal microspheres. However, there are many potential advantages in utilizing more complex probe morphologies: use of multiple trapping points enables control of the interaction volume; increasing the distance between the optical trap and the sample minimizes photodamage in sensitive biological materials; and geometric anisotropy introduces the potential for asymmetric surface chemistry and multifunctional probes. Here we demonstrate that living cells of the freshwater diatom Nitzschia subacicularis Hustedt can be exploited as advanced probes for holographic optical tweezing applications. We characterize the optical and material properties associated with the high shape anisotropy of the silica frustule, examine the trapping behavior of the living algal cells, and demonstrate how the diatoms can be calibrated for use as force sensors and as force probes in the presence of rat B-cell hybridoma (11B11) cells. KEYWORDS: Bioinspired, diatoms, holographic optical tweezers, nanoscale forces, surface functionalization
T
Diatoms (Bacillariophyta) are unicellular, photosynthetic algae comprising an elaborately structured silicified cell wall (frustule) that typically demonstrate species-specific microarchitecture. They play a vital role in aquatic ecosystems and in the global carbon and silica cycles and exhibit clonal, vegetative reproduction that ensures conservation of the frustule shape over many generations when cultured in the laboratory. Significantly, the relative ease of culturing species-specific frustule architectures provides a direct biosynthetic route to the replication of large numbers of near-identical amorphous silica microstructures with complex structure and microscale morphology. Moreover, the high morphological diversity among different diatom species opens up the possibility of judiciously matching selective features of the biomineralized form to appropriate technological functions. To date, utilization of the glassy porous silica shells of diatoms in nanotechnology has been mainly limited to the fabrication of functional organic−inorganic hybrid nanomaterials;12 the notion that diatoms could be used as optical trapping probes has not been considered. In this regard, we chose to exploit the frustules of
he foundations of life at the cellular level are defined by fundamental biomolecular interactions that occur within the picoNewton force regime. Investigations of these mechanical and physical processes in their local environments are of paramount importance to our understanding of cellular interactions and depend critically on instruments that are extremely sensitive to low forces. One of the primary techniques capable of reaching this level of sensitivity is optical trapping using spherical probes. While several groups are currently investigating the use of nonspherical probes, such as nanorods1−3,6−8 and synthetic microtools,9−11 the fabrication and application potential of these advanced components are often limited by low yields of production and the necessity for specialized instruments. There is therefore an urgent need for a nonspherical optical probe with uniform and reproducible morphology that can be produced in high yields by facile methods, stored at high concentrations without loss of stability in low ionic strength stock solutions, and accessible to chemically mediated surface functionalization. In response to these challenges, here we describe the use of living cells of the freshwater diatom, Nitzschia subacicularis as high fidelity optical tweezer probes with remarkable trapping properties and picoNewton force sensitivity. © 2012 American Chemical Society
Received: September 26, 2012 Revised: October 11, 2012 Published: October 24, 2012 6018
dx.doi.org/10.1021/nl303585w | Nano Lett. 2012, 12, 6018−6023
Nano Letters
Letter
Figure 1. (A) Phase contrast optical microscope image of a single diatom recorded with a 100× oil immersion objective. The two white organelles are lipid globules that are used as optical trapping handles [scale bar: 5 μm]. (B) Fluorescence optical microscope image recorded at an excitation wavelength band of 450−490 nm and showing red emission characteristic of chlorophyll pigments [scale bar: 5 μm]. (C,D) SEM micrographs of the silica frustules of the diatom N. subacicularis showing detail of the mineralized tip regions (C) [scale bar: 100 nm] and microscopic morphology of the intact cells (D) [scale bar: 10 μm]. (E) Bright-field image showing four diatoms assembled into a cross-like pattern using holographic optical tweezers. The red circles denote the location of the optical traps [scale bar: 10 μm; 100× 1.4 NA oil immersion objective].
N. subacicularis initially because of their regular shape anisotropy and found that, unlike many anisotropic probes,5 the living diatoms could be readily trapped at arbitrary angles with respect to the focal plane (Figure 1). This was attributed to the presence of two optically dense intracellaular lipid globules that were located symmetrically within the diatom cell (Figure 1a) and which represented foci of high refractive index (typically 1.43−1.47 in value)13 that were indispensible as optical trapping handles. Chlorophyll pigments were also observed in the protoplasm (Figure 1b). SEM images of N. subacicularis showed cells with lengths between 20 and 40 μm and which had distinctive highly elongated frustules with regular bilateral symmetry and isopolar extensions that were prolonged and terminated by hemispherical tips of amorphous silica (Figure 1c,d). The narrowly pointed tips ranged between 500 nm and 2 μm in width and were morphologically welldefined. In essence, the presence of the lipid (oil) globules and silica encasement provides an effective core−shell system for increasing the optical trapping efficiency and reducing surface reflections,14 and as a consequence the frustules could be readily optically manipulated to produce organized arrays of living cells (Figure 1e). Significantly, the diatoms remained alive and stably trapped for several hours at 21 °C when the pair of lipid organelles in each cell was used to trap the frustules away from the surface. The above observations suggested that optically trapped N. subacicularis cells could be developed as ideal living probes for force spectroscopy, for example, by cotrapping and aligning a single diatom and rat 11B11 cell and exploiting the interactions between the polar tips of the frustule and cell outer membrane (Figure 2a). For this, an accurate method to infer the force from the position of the diatom is required. For a spherical probe, a small displacement from the center of the optical trap is linearly proportional to the force that the sphere experiences. Using the same argument but allowing for coupling between
translations and rotations, the force (F) acting on a diatom can be inferred using F = κ·q, where q ≡ (x, y, z, α, β) represents the positional and rotational coordinates of the diatom (Figure 2b) and κ is the trapping stiffness matrix. Determination of κ is therefore of central importance to any application where accurate forces are required. Calibration of κ is well established for spherical particles and can be achieved using equipartition, power spectral density (PSD), and Stokes drag methods.15 However, nonspherical particles possess rotational degrees of freedom and may promote coupling between translational and rotational motions, thereby adding extra layers of complexity to the calibration process. In this regard, a generalized calibration procedure for any optically trapped object was recently described4 and experimentally demonstrated,11 although the latter system was designed such that cross-correlation between coordinates could be neglected and the hydrodynamic resistance terms, ξ, robustly modeled. This resulted in the hydrodynamic and covariance matrices being diagonal, such that the spring constants could be calibrated analogous to a spherical probe. However, these conditions cannot be assumed for a biologically derived moiety. The shape of the frustules of N. subacicularis contains sufficient uncertainty and variation that modeling can only yield approximate hydrodynamic resistances, particularly as it is possible that structural irregularities will produce coupling between various coordinate axes. Together, these factors imply that the hydrodynamic resistance matrix cannot be well-defined, and as a consequence, calibration using the generalized Stokes drag and generalized PSD techniques is compromised. However, the generalized equipartition method4 can still be used, provided the laser power is low and a good signal-to-noise ratio can be maintained. Given these conditions, evaluation of κ is performed by calculating the inverse of the position covariance matrix: κ = kBT(q ⊗ q)−1 6019
(1)
dx.doi.org/10.1021/nl303585w | Nano Lett. 2012, 12, 6018−6023
Nano Letters
Letter
Figure 2. (A) Optical microscope image showing a single diatom probe interacting with the surface of an 11B11 cell [scale: 2.5 μm, 100× oil objective]. (B) Schematic illustration of a trapped diatom, defining the principle axes used in the analysis. The use of stereoimaging permits all degrees of freedom, excluding roll about the major axis γ, to be measured. (C) Illustration showing the application of force to a cell using a diatom probe. Initially the diatom is not in contact and is brought toward the cell. Upon contact the diatom is able to translate and/or rotate. (D) Plot of force against trap position showing the onset of interaction between a diatom and a rat 11B11 cell. Solid symbols (approach) represent forces calculated for translations of the diatom toward the B cell, while open symbols (retract) represent forces calculated for translations away from the B cell. The blue circles represent forces along the x-axis, and red squares show forces along the y-axis, while the green diamonds show z, out-of-plane, forces. The observed hysteresis is due to rotation of the 11B11 cell on the substrate. (E) Plot showing the torque acting on the center of the diatom as a function of trap position [solid (approach) and open (retract) symbols]. The transparent gray line linking parts (C−E) indicates the contact area; this has been widened in the graphic to indicate the convolution of the diatom tip with the sample. The components after the gray line indicate that rotations are occurring as a result of the contact between the probe and the sample.
where ⊗ is used to represent the dyadic product of two vectors, kB is the Boltzmann’s constant, T is the temperature, and κ is the stiffness matrix (the 5 × 5 optical trapping constant). To demonstrate the use of a living probe to measure and exert forces on a cell in its natural environment, a sample chamber was filled with aliquots of a suspension containing N. subacicularis diatoms and rat B-hybridoma (11B11) cells. The latter, which are used conventionally as a cell line in antibody production, were chosen for their nonadherence, size, and ease
of culture. Based on the elastic moduli of typical cells and the picoNewton forces exerted by optical traps, the 11B11 cells act as rigid objects with only surface effects being detectable. Diatoms were optically trapped and positioned such that the diatom was oriented perpendicularly to the cell at a tip distance of approximately 5 μm (Figure 2a,c). The stiffness matrix was evaluated, and the optical traps were stepped toward the cell in 100 nm increments at a rate corresponding to the recording of 200 image frames (approximately 0.5 s per increment). The 6020
dx.doi.org/10.1021/nl303585w | Nano Lett. 2012, 12, 6018−6023
Nano Letters
Letter
Figure 3. Images of surface-modified diatoms recorded using a 100× oil-immersion objective. (A) Bright-field image of a diatom incubated with aqueous ANS. (B) The same diatom with excitation at 355−425 nm, showing accumulation of ANS on the diatom surface. (C) Single diatom excited at 515−560 nm, showing intact central intracellular organelles. The emission is characteristic of chlorophyll pigments. (D) Fluorescence image recorded at an excitation of 450−490 nm of a diatom grown in culture media containing GFP. The GFP stain is only visible on the surface of the diatom probe. (E) Excitation at 340−380 nm of the same diatom shown in (D) showing blue emission from the probe surface due to accumulated GFP. Use of a lower wavelength excitation allows the red emission from the organelles to be viewed simultaneously. (F,G) Optical fluorescence images recorded using combined fluorescent holographic optical tweezers and showing four GFP-modified diatoms being assembled into a cross-like pattern. The surface adhesion of the target molecules does not interfere with optical manipulation. (H) Bright-field image of the assembled cross-like structure. Scale bars in A−E are 5 μm and in F−H are 15 μm.
force and torque responses of the diatom’s center-of-mass are shown in Figure 2d,e. Each point on the graph represents an average of 200 data points. Two distinct regions are evident in Figure 2d: a zero-force plateau, which occurred before contact, and the force responses after contact. The ±0.5 pN and ±1 pN·μm distributions of points demonstrated in the plateau of the force and torque curves, respectively, were indicative of the time over which the diatom probe’s Brownian motion was averaged at each point (0.5 s). This represented the force and torque resolution of the living probe, which could be increased by averaging for longer. Techniques such as holographic position clamping16 could also be employed to further reduce the Brownian motion. Typically, following contact, the probe experienced a force acting along the length of the diatom, displacing it from its equilibrium position (Figure 2d, open blue circles). Statistical analysis showed no measurable forces perpendicular to the probe after contact occurs (Figure 2d, open squares and triangles), confirming that the diatom was orthogonal to the three-dimensional curved surface of the outer membrane of the 11B11 cell. Further advances in the optical trap locations resulted in larger forces along the probe until the 11B11 cell slightly moved in the scan direction (this occurred at 5.8 μm from the scan starting point in Figure 2d), producing a slight plateau between 5.8−6.0 μm and hysteresis in the approach and retract scans. The retract curve (Figure 2d, solid circle) showed behavior consistent with an elastic interaction (due to the optical traps) involving a hard, nondeformable surface. The torque curves (Figure 2e) were calculated from the probe orientation and contained information that could not be measured with confidence in the y and z force channels. Analysis of the torque signal showed that it was significantly
more sensitive to forces applied to the probe tip, exhibiting measurable deflections at forces that were entirely obscured by noise. In parallel experiments, we demonstrated that the organic coatings and additional extracellular polymeric substances on the surface of the N. subacicularis diatom frustule could be used to anchor a variety of inorganic, organic, and biological functional species onto the probe surface. To achieve this, an aqueous suspension comprising nanoparticles of aminopropylfunctionalized magnesium phyllosilicate clay (see Supporting Information), 8-anilino-1-naphthalenesulfonic acid (ANS) dye or green fluorescent protein (GFP) were added to a diatom culture media and vortexed gently for 2 h. In each case, optical micrographs of the surface-modified diatoms (Figure 3a−e) showed that the silica frustules and internal organelles remained intact, indicating that the diatom cells were relatively unaffected by the addition of the functional molecules to the growth media. Moreover, as demonstrated from time-elapsed images (Figure S1c−e), the diatoms remained motile after modification, indicating that the extrusion process involving mucilage secretion was not perturbed by surface immobilization of the nanoparticles, dye, or protein molecules. Thus, subtle functionalization of the diatoms could be undertaken without disruption of the biological structure, suggesting that the modified frustules would retain their capacity to act as optically actuated probes. This was confirmed by assembling the surfacefunctionalized diatoms into range of a ordered structures (Figure 3f−h). Conclusions. We have demonstrated the manipulation, calibration, utilization, and functionalization of a biologically derived, living, nonspherical optical trapping probe. While more 6021
dx.doi.org/10.1021/nl303585w | Nano Lett. 2012, 12, 6018−6023
Nano Letters
Letter
captions. Exposure times were modified between samples, to illustrate the presence of functional molecules and therefore cannot be used for abundance assessment.
complex to calibrate than a simple microsphere, the advantages of removing the laser focus from the sample and the exertion and measurement of sub-picoNewton scale forces greatly outweigh this challenge. Furthermore, the torque and coupling signals are more sensitive than the applied forces and could readily be utilized for optical force microscopy applications similar to that described in Phillips et al.17 but at much lower force interactions. The ease of surface modification by chemically tethering a range of functional molecules to the polysaccharide outer layer of the diatom frustule further extends the application of diatoms as living probes. Indeed, it should be possible to design a range of functional probes based on diatom surface chemistries that could find use in diverse nanobiotechnological applications. For example, we envisage that these living probes will find substantial use for the investigation of molecular recognition and adhesion properties relating to protein−antibody interactions, highly deformable cells, and cellular systems highly sensitive to local effects mediated by the application of laser intensity. Materials and Methods. Cell cultures. Clonal cultures of the diatom N. subacicularis, isolated from the River Frome, Dorset, U.K., were grown in Diatom Medium 18 and characterized using optical, fluorescence, and scanning electron microscopy (SEM) techniques. A rat 11B11 B-cell hybridoma cell line (ATCC) was cultured in complete RPMI (Lonza) supplemented with 10% fetal calf serum (Biosera) and 2 mM Lglutamine (Biowhittaker). For experiments requiring cells to be adhered to cover glasses, the latter were pretreated with 1% poly-L-lysine. Cell suspensions were then placed onto coverslips and incubated for 30 min at 37 °C to ensure adherence. Holographic Optical Tweezers. The holographic optical tweezers were slightly modified from that reported in Gibson.19 The equipment consisted of an inverted microscope (Zeiss, Axiovert 200) with a 1.4 NA, 100× objective (Zeiss, PlanNeofluor) and a motorized xyz stage (Applied Scientific Instruments, MS-2000). The trapping beam was provided by a Ti:Sapphire laser (Coherent, 899) emitting up to 4 W at 800 nm. The beam was expanded to fill a spatial light modulator (Boulder Nonlinear Systems, P512- 0785), passed through a polarizing beam splitter, and imaged onto the back aperture of the objective lens. The beam splitter also directed the light from a custom-made stereoillumination system20 onto a high-speed CMOS camera (Prosilica, GC640). The resulting stereoscopic images were processed such that accurate 3D position tracking was achieved. Imaging was performed using exposure times of 1 ms and regions-of-interest measuring 32.4 × 20.3 μm (400 × 250 pixels) at a frame rate of ≈400 Hz. Optical tweezers sample chambers were prepared by establishing a capillary between a microscope slide and UV glued coverslips and then allowing capillary action to draw the sample inside the chamber. The edges were sealed, and the chamber placed onto the holographic optical tweezers system. Diatoms were optically trapped using the lipid globules as handles. SEM. Diatom samples were prepared by air drying onto carbon pads (Agar Scientific Ltd.). Samples were transferred to aluminum stubs and coated with a 15 nm gold film using a sputter coater. Scanning electron micrographs were obtained using a JEOL JSM 5600LV SEM. Fluorescence Optical Microscopy. Fluorescence optical microscopy was performed using either a manual inverted microscope (Leica DMI3000 B) or an alternate camera port on the holographic optical tweezers (Rolera EMCCD, Q-Imaging). Excitation and emission wavelengths are as described in figure
■
ASSOCIATED CONTENT
S Supporting Information *
Experimental details for trapping properties of live and senescing diatoms, organoclay fuctionalization of diatoms, stiffness matrix correlations, trajectory control, correction, and analysis. This information is available free of charge via the Internet at http://pubs.acs.org
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected];
[email protected] Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work is funded by a Basic Technology Translation Grant through the Research Councils of the United Kingdom. M.J.M. and S.M. acknowledge Royal Society Wolfson Merit Awards. This work was carried out with the support of the Bristol Centre for Nanoscience and Quantum Information. We would also like to thank G. Britton for his invaluable assistance with the culture and handling of the 11B11 cells, D. Fagan for diatom advice and discussions, and J. Armstrong for providing the GFP stain. R. Krishna Kumar is acknowledged for help with fluorescence microscopy. D.M.C. and M.J.M. devised and developed the physical probe concepts. A.J.P. devised and developed the functionalisation methods. M.Y. selected the diatom species and advised on aspects of diatom physiology. H.R. obtained and isolated the original diatom culture. S.N.O. identified diatoms as ideal probes and characterised them with assistance from H.R. S.N.O. performed the majority of the calibration and probing experiments and analyzed the data, with the assistance of D.B.P. and J.A.G. D.B.P. and J.A.G. also wrote the analysis software for the probing experiments. D.M.C., A.J.P., and S.M. managed the project and wrote the manuscript, using contributions from all authors.
■
REFERENCES
(1) Ashkin, A.; Dziedzic, J. M.; Bjorkholm, J. E.; Chu, S. Observation of a single-beam gradient force optical trap for dielectric particles. Opt. Lett. 1986, 11, 288−290. (2) Greenleaf, W. J.; Woodside, M. T.; Block, S. M. High-resolution, single-molecule measurements of biomolecular motion. Annu. Rev. Biophys. Biomolec. Struct. 2007, 36, 171−190. (3) Bustamante, C.; Smith, S. B.; Liphardt, J.; Smith, D. Singlemolecule studies of DNA mechanics. Curr. Opin. Struc. Biol. 2000, 10, 279−285. (4) Simpson, S. H.; Hanna, S. Thermal motion of a holographically trapped SPM-like probe. Nanotechnology 2009, 20, 395−710. (5) Konig, K.; Liang, H.; Berns, M. W.; Tromberg, B. J. Cell damage by near-IR microbeams. Nature 1995, 377, 20−21. (6) Reicherter, M.; Haist, T.; Wagemann, E. U.; Tiziani, H. J. Optical particle trapping with computer-generated holograms written on a liquid-crystal display. Opt. Lett. 1999, 24, 608−610. (7) Dufresne, E. R.; et al. Computer-generated holographic optical tweezer arrays. Rev. Sci. Instrum. 2001, 72, 1810−1816. (8) Pauzauskie, P. J.; et al. Optical trapping and integration of semiconductor nanowire assemblies in water. Nat. Mater. 2006, 5, 97− 101.
6022
dx.doi.org/10.1021/nl303585w | Nano Lett. 2012, 12, 6018−6023
Nano Letters
Letter
(9) Friese, M. E. J.; et al. Optically driven micromachine elements. Appl. Phys. Lett. 2001, 78, 547−549. (10) Rodrigo, P. J.; et al. Actuation of microfabricated tools using multiple GPC-based counterpropagating-beam traps. Opt. Express 2005, 13, 6899−6904. (11) Carberry, D. M.; et al. Calibration of optically trapped nanotools. Nanotechnology 2010, 21, 175501. (12) Gordon, R.; et al. The Glass Menagerie: diatoms for novel applications in nanotechnology. Trends Biotechnol. 2009, 27, 116−127. (13) Craig, B. M. Refractive indices of some saturated and monoethenoid fatty acids and methyl esters. Can. J. Chem. 1953, 31, 499−504. (14) Viravathana, P.; Marr, D. W. M. Optical Trapping of Titania/ Silica Core-Shell Colloidal Particles. J. Colloid Interface Sci. 2000, 221, 301−307. (15) Neuman, K. C.; Block, S. M. Optical trapping. Rev. Sci. Instrum. 2004, 75, 2787−2809. (16) Phillips, D. B.; et al. Position clamping of optically trapped microscopic non-spherical probes. Opt. Express 2011, 19, 20622− 20627. (17) Phillips, D. B.; et al. Surface Imaging using Holographic Optical Tweezers. Nanotechnology 2011, 22, 285503. (18) Beakes, G. W.; Canter, H. M.; Jaworski, G. H. M. Zoospore ultrastructure of Zygorhizidium af f luens and Z. planktonicum, two chytrids parasitizing the diatom Asterionella formosa. Can. J. Bot. 1988, 66, 1054−1067. (19) Gibson, G.; et al. Holographic assembly workstation for optical manipulation. J. Opt. A: Pure Appl. Opt. 2008, 10, 044009. (20) Bowman, R.; Preece, D.; Gibson, G.; Padgett, M. Stereoscopic particle tracking for 3D touch, vision and closed-loop control in optical tweezers. J. Opt. 2011, 13, 044003.
6023
dx.doi.org/10.1021/nl303585w | Nano Lett. 2012, 12, 6018−6023
