Elsevier

Methods in Enzymology

Volume 505, 2012, Pages 219-232
Methods in Enzymology

Chapter twelve - Fluorescence Single-Molecule Imaging of Actin Turnover and Regulatory Mechanisms

https://doi.org/10.1016/B978-0-12-388448-0.00020-6Get rights and content

Abstract

Cells must rapidly remodel the actin filament network to achieve various cellular functions. Actin filament turnover is a dynamic process that plays crucial roles in cell adhesion, locomotion, cytokinesis, endocytosis, phagocytosis, tissue remodeling, etc., and is regulated by cell signaling cascades. Success in elucidating dynamic biological processes such as actin-based motility relies on the means enabling real time monitoring of the process. The invention of live-cell fluorescence single-molecule imaging has opened a window for direct viewing of various actin remodeling processes. In general, assembly and dissociation of actin and its regulators turned out to occur at the faster rates than previously estimated by biochemical and structural analyses. Cells undergo such fast continuous exchange of the components perhaps not only to drive actin remodeling but also to facilitate rapid response in many other cell mechanics and signaling cascades. This chapter describes how epifluorescence single-molecule imaging which visualizes deeper area than the TIRF microscopy is achieved in XTC cells, the currently best platform for this approach.

Introduction

Motile cells form actin-based pseudopods called lamellipodia and filopodia at the leading edge, which guide cell movement sensing the outer environment and extracellular stimuli. The mechanisms controlling protruding activities of the leading edge structures have been the major problem in cell biology. The electron micrograph demonstrates that actin filaments direct their fast growing barbed ends toward the cell edge (Small et al., 1978). Actin polymerization indeed pushes the membrane forward as pharmacological perturbation of actin polymerization ceases the protrusive activity of neuronal growth cone (Forscher and Smith, 1988). To investigate the live-cell dynamics of actin remodeling in 1980s and 1990s, two methods were employed. One observes the site of incorporation of microinjected labeled actin (Okabe and Hirokawa, 1989, Symons and Mitchison, 1991). These studies revealed that F-actin assembles the fastest at the leading edge. The other method is photoactivation of fluorescence (PAF) of caged fluorescent actin or fluorescence recovery after photobleaching (FRAP) of fluorescently labeled actin. In an early FRAP study (Wang, 1985), the retrograde actin flow was discovered.

There still remains debate over the question as to where F-actin is assembled in lamellipodia. The array treadmilling model in which actin polymerizes exclusively at the leading edge is supported by studies employing FRAP (Lai et al., 2008). On the other hand, the ubiquitous actin polymerization throughout lamellipodia was demonstrated by the fast disappearance of photoactivated caged actin in fish keratocyte (Theriot and Mitchison, 1991). The frequent polymerization in the lamellipodium body is congruent with the idea that cofilin vigorously generates new barbed ends through its filament severing activity in lamellipodia (Ghosh et al., 2004).

The discrepancy between the two distinct models may simply reflect difference in the cell types used. However, this can also arise from the insufficiency of FRAP and PAF data in precisely revealing the filament disassembly kinetics. As I discussed (Watanabe, 2010), fluorescence single-molecule imaging has several advantages over the FRAP and PAF experiments under certain circumstances where these methods have intrinsic problems.

First, the photobleached or photoactivated area is normally at least a few microns wide. With this resolution, reincorporation of disassembled molecules into the same labeled area may frequently occur under certain conditions. This reincorporation problem has been analyzed in a mathematical simulation (Tardy et al., 1995), which revealed that the decay in the FRAP and PAF label becomes substantially slower than the disassembly rate when actin disassembly is fast and the ratio between F- and G-actin is high (Tardy et al., 1995, Watanabe, 2010). Second, single-molecule observation is superior to FRAP and PAF experiments in precisely detecting the frequency of the short- and long-lived populations. In FRAP and PAF experiments, it is difficult to know the precise distribution of “lifetime” or duration time of the molecule in the system, and the data are often interpreted by fitting the decay with a single exponential. The single-molecule analysis of actin turnover in lamellipodia revealed the complex distribution of F-actin lifetime (Watanabe and Mitchison, 2002) which should occur naturally given the high order polymer structure. Insufficiency of FRAP and PAF to capture lifetime distribution of F-actin may in part explain how researchers have attained distinct interpretations by FRAP and PAF analyzes (Watanabe, 2010).

Overall, the single-molecule approach reveals previously unknown properties of actin remodeling machineries as it directly observes assembly, duration time, and movement of the molecule. It can also follow the change in the molecular behavior overtime, which is useful for monitoring the effects of bioactive compounds. My research group has applied this method not only to the actin turnover mechanisms (Miyoshi et al., 2006, Tsuji et al., 2009, Watanabe and Mitchison, 2002) but also to revealing actin nucleation and the long-range processive actin elongation by formin homology proteins (Higashida et al., 2004, Higashida et al., 2008) and the inhibitor-induced translocation of Abelson kinase (Fujita et al., 2009). The direct viewing of the molecular behavior is also advantageous over the technique called quantitative fluorescent speckle microscopy (qFSM) which computes the degree of assembly and disassembly from images composed of dense clusters of fluorescent molecules (Ponti et al., 2004). Although qFSM provides the better statistical power than single-molecule observation, the output by qFSM may contain intrinsic errors as argued recently (Vallotton and Small, 2009). It might, therefore, be important to reinvestigate the conclusions drawn by the methods including FRAP, PAF, and qFSM by fluorescence single-molecule observation even though it may still be difficult to apply to various cell types.

In this chapter, I explain how single-molecule fluorescence observation in live cells is carried out with epifluorescence microscopy. The question over the turnover of F-actin which moves as an array with lifetime of several to hundreds of seconds has given me the best opportunity to learn how to compromise the data quality with the photodamage to the cell and the probes. In my experience so far, XTC cells provide the superior platform for this approach to other common mammalian cell lines for unidentified reasons. The information should be useful for those who intend to apply the live-cell fluorescence single-molecule observation to various molecular mechanisms and organisms.

Section snippets

Introducing a Low Density of Fluorescent Proteins in XTC Cells

In this part, I provide the optimal conditions and procedures for expression of GFP-tagged actin and its regulators at the low level suitable for single-molecule observation in Xenopus XTC cells (Watanabe and Mitchison, 2002). The protocols include maintenance of XTC cells, transfection procedures, and the use of the defective CMV promoter optimized for slow expression of fluorescent proteins. XTC cells are suitable for live-cell imaging as they grow at room temperature in ambient atmosphere.

Observation of XTC Cells Spreading on Poly-l-lysine (PLL)-coated Glass Coverslips

XTC cells spread on PLL-coated glass coverslips rapidly and form wide flat lamellipodia, which allows long-range single-molecule tracking in a single focal plane. PLL helps adhesion of XTC cells better than fibronectin. Removal of FCS from the medium ensures efficient cell spreading and the formation of flat morphology of spreading XTC cells. In some experiments for long-term observation, the concentration of PLL in the coating solution is reduced to ∼ 10 μg/ml and 1 or 2 h after cell spreading

Data Analysis

In early studies, tracking was carried out manually and recorded on the printed images, and the data were input in the spreadsheet with 10 keys. Automatic tracking software currently available from various resources can track individual molecules fairly precisely in ideal situations, but the frequency of tracking errors often become intolerable, influenced by the variations in clarity and density of single-molecule images. Optimizing acquisition conditions for complete automatic tracking is not

Acknowledgments

This work was supported in part by the Cabinet Office, Government of Japan through its Funding Program for Next Generation World-Leading Researchers (LS013), and the grants from the Human Frontier Science Program and the Takeda Science Foundation.

References (22)

  • M.B. Smith et al.

    Biophys. J

    (2011)
  • Y. Tardy et al.

    Biophys. J.

    (1995)
  • P. Forscher et al.

    J. Cell Biol.

    (1988)
  • A. Fujita et al.

    Mol. Pharmacol.

    (2009)
  • M. Ghosh et al.

    Science

    (2004)
  • C. Higashida et al.

    Science

    (2004)
  • C. Higashida et al.

    J. Cell Sci.

    (2008)
  • K.E. Kubow et al.

    Nat. Cell Biol.

    (2011)
  • F.P. Lai et al.

    EMBO J.

    (2008)
  • T. Miyoshi et al.

    J. Cell Biol.

    (2006)
  • H. Mizuno et al.

    Science

    (2011)
  • Cited by (27)

    • Actin Assembly Dynamics and Its Regulation in Motile and Morphogenetic Processes

      2022, Encyclopedia of Cell Biology: Volume 1-6, Second Edition
    • Paradoxical activation of c-Src as a drug-resistant mechanism

      2021, Cell Reports
      Citation Excerpt :

      HEK293F, MCF-7 cells and MCF-7-T341I cells were maintained in DMEM (nacalai tesque, 08459-35) containing 10% FCS at 37°C in humidified 95% CO2 incubator. Plasmid transfection was carried out with polyethyleneimine (PEI) (Polysciences, 23966-1) (Fujita et al., 2009; Watanabe, 2012; Yamashiro and Watanabe, 2014). MCF-7-T341I cells were generated using CRISPR technology.

    • Convection-Induced Biased Distribution of Actin Probes in Live Cells

      2019, Biophysical Journal
      Citation Excerpt :

      The expression vector for miRFP703-actin was generated by substituting miRFP703 cDNA for the coding sequence of EGFP in delCMV-EGFP-actin. Xenopus laevis XTC cells were maintained as described previously (13,17). The expression constructs were transfected into XTC cells by using polyethyleneimine as described previously (17).

    • Mechanostress resistance involving formin homology proteins: G- and F-actin homeostasis-driven filament nucleation and helical polymerization-mediated actin polymer stabilization

      2018, Biochemical and Biophysical Research Communications
      Citation Excerpt :

      Actin filaments (F-actin) in cultured cells exhibit a wide range of stability [8], which can now be precisely elucidated by our improved live-cell single-molecule speckle microscopy called eSiMS (easy, efficient and electroporation-based SiMS) [9–11]. eSiMS microscopy employs electroporation of fluorescently-labeled proteins instead of the low-level expression of GFP-tagged proteins under the control of the defective, truncated CMV promoter [12,13]. This eSiMS method has several advantages over the previous method.

    • Actin Cross-Linking Toxin Is a Universal Inhibitor of Tandem-Organized and Oligomeric G-Actin Binding Proteins

      2018, Current Biology
      Citation Excerpt :

      The accumulation of ACD-cross-linked actin species was quantified from three independent replicates using a densitometry plug-in of the ImageJ software package [51]. EGFP-constructs under defective CMV promoter optimized for low levels of expression of fluorescent proteins [50] were a generous gift from Dr. Naoki Watanabe [17, 20, 22, 49, 50], except for human full-length Spire and Nt-Spire (corresponding to aa 1-396), which were amplified by PCR from HeLa cells cDNA and cloned into EGFP-vector with the defective CMV promoter. Transfections of XTC cells were performed using TurboFect transfection reagent (ThermoFisher Scientific).

    View all citing articles on Scopus
    View full text