Letter pubs.acs.org/NanoLett
Imaging Single Nanoparticle Interactions with Human Lung Cells Using Fast Ion Conductance Microscopy Pavel Novak,*,†,‡ Andrew Shevchuk,‡ Pakatip Ruenraroengsak,§ Michele Miragoli,∥,⊥ Andrew J. Thorley,§ David Klenerman,# Max J. Lab,∇ Teresa D. Tetley,*,§ Julia Gorelik,*,∇ and Yuri E. Korchev*,‡ †
School of Engineering and Materials Science, Queen Mary University of London, Mile End Rd, London E1 4NS, United Kingdom Department of Medicine, Imperial College London, Du Cane Road, London W12 0NN, United Kingdom § Lung Cell Biology, Section of Pharmacology and Toxicology, National Heart and Lung Institute, Imperial College London, Cale Street, London SW3 6LY, United Kingdom ∥ INAIL Research, Centro di Eccellenza per la Ricerca Tossicologica, University of Parma, Strada Dell’Universitá, 12, 43100 Parma, Italy ⊥ Humanitas Clinical and Research Center, Via Manzoni 56, 20089 Rozzano, Milano, Italy # Department of Chemistry, Cambridge University, Trinity Ln, Cambridge CB2 1TN, United Kingdom ∇ Department of Cardiac Medicine, National Heart and Lung Institute, Imperial College London, Du Cane Road, London W12 0NN, United Kingdom ‡
S Supporting Information *
ABSTRACT: Experimental data on dynamic interactions between individual nanoparticles and membrane processes at nanoscale, essential for biomedical applications of nanoparticles, remain scarce due to limitations of imaging techniques. We were able to follow single 200 nm carboxyl-modified particles interacting with identified membrane structures at the rate of 15 s/frame using a scanning ion conductance microscope modified for simultaneous high-speed topographical and fluorescence imaging. The imaging approach demonstrated here opens a new window into the complexity of nanoparticle−cell interactions. KEYWORDS: Nanoparticles, scanning ion conductance microscopy, alveolar epithelial cells, clathrin, actin
M
at the level of individual nanoparticles under physiological conditions and in real time.3 Here we have employed a noncontact scanning probe microscopy (SPM) technique called scanning ion conductance microscopy (SICM),10,11 based on an electrolyte-filled glass nanopipette sensing the proximity of a cell sample surface via reduction in the pipet current. SICM can be applied to study the dynamics of live cellular structures at the nanoscale12−15 as well as effects of exposure to nanoparticles.16,17 The hopping mode14 of SICM, where the pipet approaches the cell surface only at selected imaging points and retracts to a safe distance, avoiding spurious contacts, before moving to another imaging point, was shown to produce markedly less imaging artifacts in fine cellular processes and tall overhanging structures than conventional AFM.18 However, despite performance improvements introduced in this hopping mode of SICM,14 the scanning speed remains one of the limiting factors in imaging dynamic and convoluted cellular structures at the nanoscale.
ost of the existing experimental studies on the interaction between live cells and nanopaticles explore only the selected points of the interaction, for example, membrane-bound early stage or vesicle-encapsulated, usually following exposure to large numbers of nanoparticles.1−3 The influence of the individual circumstances faced by each nanoparticle upon contact with a heterogeneous cell surface is lost in the average outcome of such a large assay. Moreover, the identification of mechanisms involved in the internalization of nanoparticles in this type of study requires the use of specific pharmacological inhibitors or specific fluorescent labels which are not available for some mechanisms of internalization such as macropinocytosis.4,5 The use of alternative label-free experimental techniques with nanoscale resolution has so far been limited to a few studies utilizing principles of atomic force microscopy (AFM) to visualize nanoparticles and their effect on generic membranes6−8 or to measure the force of interaction between nanoparticles and the membrane.9 Thus, there is a continuing need for new live imaging techniques complementary to the existing label-based techniques capable of studying the process of nanoparticle−living cell interactions © 2014 American Chemical Society
Received: November 1, 2013 Revised: February 7, 2014 Published: February 20, 2014 1202
dx.doi.org/10.1021/nl404068p | Nano Lett. 2014, 14, 1202−1207
Nano Letters
Letter
Figure 1. Imaging interactions between single nanoparticles and alveolar epithelial cells. (A) Typical gray scale height-coded topography image obtained with the hopping mode of SICM, of cultured, live AT1-like lung epithelial cells. The luminance of each pixel reflects its z-coordinate (height). (B) Simultaneous hopping mode SICM and surface fluorescence22 scan of 6.4 × 6.4 μm2 area on AT1-like cell membrane exposed to 0.05% of 200 nm CMP for 15 min and fixed afterward. SICM topography (top left) and calculated first derivative of the topography is shown as a slopecoded color image (top right). Fluorescence confocal images of red fluorescent 200 nm nanoparticles (bottom left) and overlay with first derivative (bottom right) show colocalization of nanoparticles (yellow arrowheads) with membrane protrusions (white arrowheads). Some particles appear to be halfway through the process of internalization, covered by membrane protrusions (blue arrowheads), while others are already fully internalized (green arrowheads) without any remaining topographical feature left on the cell surface. (C) Left, high-resolution hopping mode SICM scan of 1.5 × 1.5 μm2 area showing a single 200 nm CMP (yellow arrowhead) partially covered by membrane protrusions (white arrowheads). Right, scanning electron microscopy (SEM) image shows similar interaction of a single 200 nm CMP (yellow arrowhead) with a membrane protrusion (white arrowhead) in a different AT1-like cell sample. Note that the luminance of each pixel in the SICM image reflects its z-coordinate (height), while the luminance of the SEM image is determined by scattering of electrons and does not, in principle, reflect the height of the features. (D) Stop-frame images of a movie of 6.4 × 6.4 μm2 area on the membrane of live AT1-like cells imaged using conventional hopping mode of SICM at a rate of 2 min per frame and effective pixel size of 100 nm. Note that the dynamics of some membrane processes (white arrowheads) is clearly faster than the frame rate of the conventional hopping mode SICM setup. Scale bars 10 μm (A), 400 nm (B), 200 nm (C), and 2 μm (D).
of these membrane protrusions, demonstrated on the macroscale as an increase in cell surface roughness, appears to be triggered by exposure to nanoparticles (Supporting Information, Figure S2A,B). High-resolution images reveal that these protrusions often form physical contacts with nanoparticles (yellow arrowheads in Figure 1B,C and Supporting Information, Figure S2C,D) and display dynamics faster than the temporal resolution of the current design of hopping mode SICM (white arrowheads in Figure 1D). The overall scanning speed of the hopping mode is dictated by the speed at which the scanning nanopipette can repeatedly approach the sample surface without touching it, the approach rate referred to as the fall rate, and the step response of x−y piezo stage. In the original implementation of the hopping mode, the fall rate was limited by the delay in the execution of the pipet withdrawal command (the z-piezo latency). Since the command is issued when the pipet approaches the surface to a distance typically equal to the pipet tip radius, the fall rate of the setup, built around 25 μm z-axis flexure piezo actuator with resonance frequency of 3.7 kHz operating in closed-loop, was practically limited to approximately one pipet tip radius per millisecond. To reduce the z-axis latency without sacrificing the z-axis travel range needed for imaging complex biological samples, we mounted an additional fast shear piezo-actuator with the resonance frequency of 150 kHz and travel range of
To address specific challenges of imaging single-nanoparticle interactions with live membrane processes, we used a culture of immortalized human alveolar epithelial type 1-like (AT1) cells exposed to commercially available 200 nm red fluorescent carboxyl-modified latex particles (CMPs, see Supporting Information for further details). Human alveolar epithelial cells are regularly exposed to airborne nanoparticles during air pollution episodes,19 and the cells will also be targeted in future by nanoparticles engineered for pulmonary drug delivery.20 The size of particles was selected to be similar to the diameter of membrane protrusions observed in AT1 cells as this is expected to “encourage” particle−protrusion interactions and create a dynamic sample challenging enough to test imaging capabilities of SICM. Reflecting the naturally low protein content of lung fluid,21 we did all of the experiments in serum-free L15 medium where CMPs remain monodispersed (polydispersity index of 0.040 ± 0.019) and display negative surface charge (zeta potential of −31.1 ± 2.5 mV) (Supporting Information, Figure S1). As illustrated by our 3D topographic images of AT1-like cells exposed to 200 nm CMPs (Figure 1A−C), imaging nanoparticle−membrane interactions poses a challenge in the form of fine (
