Correlative Super-Resolution and AFM
Imaging the Structure and Mechanics of Cell Adhesions
- Fig. 1: A schematic diagram of the correlative AFM + localization microscopy setup.
- Fig. 2: Podosomes in a fixed THP-1 cell imaged with correlative 2-colour STORM and AFM. (a) Localization microscopy improves resolution significantly over (b) conventional wide-field imaging, showing the actin-rich podosome cores (red) surrounded by a talin ring (green). In the (c) AFM height and (d) stiffness images the podosomes can be seen as higher and stiffer areas.
- Fig. 3: Live cell localization microscopy images of podosomes in a THP-1 cell, where adhesion-related protein talin was labelled with mEOS3.2. (a) HAWK-processed image shows the podosome ring structure, (b) image without HAWK processing shows typical high density processing artefacts (sharpening, missing structures).
Fluorescence localization microscopy techniques (STORM/PALM) and atomic force microscopy (AFM) are both capable of imaging living cells at a resolution of a few tens of nanometers. We have constructed a hybrid system combining these techniques and apply it for functional imaging of cell adhesions, where we employ STORM/PALM to image the location of fluorescently labelled molecules, while at the same time mapping the spatial variation of height and stiffness with AFM.
Fluorescence and atomic force microscopy (AFM) are both capable of imaging subcellular features in living cells. The type of information obtained by these techniques is very different: AFM generates height and/or mechanical property maps of the sample surface, while fluorescence microscopy provides the location of specific molecules, which have been labelled with fluorophores, inside the cell. The complementarity of the information provided makes a combination of these techniques a desirable tool in cell biology research.
In the past, the correlation of these techniques was made difficult by the diffraction limit in light microscopy, which restricted the resolution of fluorescence microscopy to two orders of magnitude more than AFM. Recently developed localization super-resolution microscopy techniques (STORM/PALM) [1,2] have brought the resolution of light microscopy down to a few tens of nanometers, a similar scale to the typical lateral resolution of AFM when imaging soft biological samples .
Localization microscopy requires fluorophores that switch between a bright and a dark state, so that many images of the sample can be taken, in each of which only a few fluorophores are emitting light. STORM is most commonly performed with the cyanine dye Alexa-647, which blinks well in a buffer containing an enzymatic oxygen scavenger [4,5]. Unfortunately, when AFM cantilevers are immersed in this buffer, some of the buffer components crystallize on the cantilever, making AFM image acquisition impossible. Consequently, the combination of AFM with STORM was hindered by the need to change buffers between imaging modes.
Here correlative AFM and STORM on fixed cell samples is demonstrated without the need to change buffers by using a new fluorescent dye, iFluor-647, that enables STORM imaging in a buffer that is also compatible with AFM.
Also the photoswitchable fluorescent protein mEOS3.2  is employed and correlative 2-colour localization microscopy with AFM is demonstrated.
To investigate the function and mechanics of cells, live cell imaging is essential. The use of localization microscopy methods has been mostly limited to fixed samples due to long image acquisition times. Acquisition time can be shortened by acquiring higher emitter density data, but processing this kind of data with conventional methods usually leads to artefacts, such as artificial sharpening and missing structures in the final image. A new localization microscopy technique termed HAWK has been demonstrated  which can separate molecules based on their blinking statistics, enabling artefact-free localization microscopy of living cells.
These correlative techniques have been applied to simultaneously image the molecular structure and mechanical properties of cell adhesions. The formation of podosomes — micron-sized circular adhesion/invasion structures on the cell surface — is linked to cell motility and pathologies such as cancer cell invasion and chronic inflammation [9,10]. To build a dynamic molecular model of podosomes, localization microscopy techniques have been employed to image the location of fluorescently labelled podosome components, while at the same time mapping the spatial variation of stiffness with AFM. We also demonstrate HAWK imaging of mEOS3.2-labelled podosome components in living cells.
The setup for combined localization microscopy and AFM was built around a standard inverted microscope (fig. 1). The microscope was equipped with a LightHUB-6 laser combiner (Omicron, Germany) with 405 nm, 488 nm, 561 nm and 647 nm lasers for fluorescence excitation, an EMCCD (Andor iXon Ultra) for fluorescence data collection, and a JPK Nanowizard 3 for AFM imaging.
THP-1 cells stably expressing mEOS3.2-talin  were seeded on fibronectin-coated dishes with TGF-β to induce podosome formation, and incubated for 24 hours. For STORM imaging, the cells were fixed with formaldehyde and labelled with iFluor-647-phalloidin. STORM imaging was performed in TN buffer (H2O + 50 mM Tris pH 8.0 + 10 mM NaCl) with 50 mM MEA. Live cell imaging was performed in the growth medium.
Figure 2 shows an example of podosomes imaged with correlative 2-color STORM and AFM. Localization microscopy (fig. 2a) improves resolution significantly over conventional wide-field imaging (fig. 2b), showing the actin-rich podosome cores (red) surrounded by a talin ring (green). In the AFM height (fig. 2c) and stiffness (fig 2d) images the podosomes can be seen as higher and stiffer areas. The iFluor-647 (red) localization image was acquired first, then the mEOS3.2 (green) localization image, and the AFM scan directly afterwards without buffer change.
Figure 3 shows HAWK image of podosomes in a living THP-1 cell, where the ring protein talin was labelled with mEOS3.2. HAWK (fig. 3a) produces an image that shows the expected podosome ring structure and a lower concentration of talin distributed throughout the cell, while traditional processing (fig. 3b) causes artefacts related to high molecule density in the raw data, e.g. artificial sharpening, contraction of the ring structures and missing features.
Correlative AFM+STORM provides both structural and mechanical information of a sample and is a valuable tool especially in mechanobiological studies. Here an easy and straightforward method for correlative AFM + STORM imaging has been presented using iFluor-647 dye in a simple buffer without an enzymatic oxygen scavenger, such that no buffer change is required between the imaging modalities, simplifying the correlative imaging process and eliminating artefacts. The use of this buffer also enables long term STORM imaging, unlike buffers with enzymatic oxygen scavengers, where the pH change limits imaging time typically to a couple of hours, benefitting any applications where longer term imaging of the sample is required, or sample damage caused by the STORM buffer is a concern . The use of endogenous fluorescent proteins offers an alternative to antibody labelling, and enables live-cell imaging which can be difficult with dye labels. mEOS3.2 was combined with iFluor-647 for labelling podosome molecular components in fixed THP-1 cells, and 2-colour localization microscopy + AFM imaging in one setup was demonstrated.
Live-cell localization microscopy of podosomes was demonstrated using a new localization microscopy method HAWK , which speeds up the localization microscopy data collection time from 10s of minutes to a few seconds — ideal for many cell biology studies. This work is currently being expanded to perform simultaneous AFM and localization microscopy measurements in living cells to determine whether podosome stiffness fluctuations are associated with changes in their molecular composition.
Besides podosomes, the correlative methods introduced here can be applied to the study of a wide range of biophysical processes in many cell biology studies, while HAWK will advance a wide variety of biological imaging applications where the speed of data acquisition is a critical parameter and resolution below the diffraction limit is needed.
Support from Human Frontier Science Program is gratefully acknowledged.
Liisa M. Hirvonen1, Richard J. Marsh1, Gareth E. Jones1, Susan Cox1
1Randall Centre for Cell and Molecular Biophysics, King’s College London, United Kingdom
Dr. Liisa M. Hirvonen
Randall Centre for Cell and Molecular Biophysics
King’s College London
London, United Kingdom
 E. Betzig, G. H. Patterson, R. Sougrat, O. W. Lindwasser, S. Olenych, J. S. Bonifacino, M. W. Davidson, J. Lippincott-Schwartz, and H. F. Hess: Imaging intracellular fluorescent proteins at nanometer resolution, Science, 313:1642-1645 (2006) doi: 10.1126/science.1127344
 M. J. Rust, M. Bates, and X. Zhuang: Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM), Nat Methods, 3:793-796 (2006) doi: org/10.1038/nmeth929
 H. Miller, Z. Zhou, J. Shepherd, A. Wollman, and M. Leake: Single-molecule techniques in biophysics: a review of the progress in methods and applications, Rep Prog Phys, 81:024601 (2017) doi: 10.1088/1361-6633/aa8a02
 M. Heilemann, S. van de Linde, M. Schuttpelz, R. Kasper, B. Seefeldt, A. Mukherjee, P. Tinnefeld, and M. Sauer: Subdiffraction-resolution fluorescence imaging with conventional fluorescent probes, Angew Chem Int Ed, 47:6172-6176 (2008) doi: 10.1002/anie.200802376
 S. van de Linde, A. Loschberger, T. Klein, M. Heidbreder, S. Wolter, M. Heilemann, and M. Sauer: Direct stochastic optical reconstruction microscopy with standard fluorescent probes, Nat Protoc, 6:991-1009 (2011) doi: org/10.1038/nprot.2011.336
 L. M. Hirvonen & S. Cox: STORM without enzymatic oxygen scavenging for correlative atomic force and fluorescence superresolution microscopy, Methods Appl Fluoresc, 6:045002 (2018) doi: org/10.1088/2050-6120/aad018
 M. Zhang, H. Chang, Y. Zhang, J. Yu, L. Wu, W. Ji, J. Chen, B. Liu, J. Lu, Y. Liu, J. Zhang, P. Xu, and T. Xu: Rational design of true monomeric and bright photoactivatable fluorescent proteins, Nat Methods, 9:727-729 (2012) doi: 10.1038/nmeth.2021
 R. J. Marsh, K. Pfisterer, P. Bennett, L. M. Hirvonen, M. Gautel, G. E. Jones, and S. Cox: Artefact-free high density localisation microscopy, Nat Methods, 15:689-692 (2018) doi: 10.1038/s41592-018-0072-5
 I. Maridonneau-Parini: Control of macrophage 3D migration: a therapeutic challenge to limit tissue infiltration, Immunol Rev, 262:216-231 (2014) doi: 10.1111/imr.12214
 E. K. Paterson and S. A. Courtneidge: Invadosomes are coming: new insights into function and disease relevance, FEBS J, 285:8-27 (2018) doi: 10.1111/febs.14123
 V. Vijayakumar, J. Monypenny, X. J. Chen, L. M. Machesky, S. Lilla, A. J. Thrasher, I. M. Antón, Y. Calle, and G. E. Jones: Tyrosine phosphorylation of WIP releases bound WASP and impairs podosome assembly in macrophages, J Cell Sci, 128:251-265 (2015) doi: 10.1242/jcs.154880