Live-cell imaging of G-actin dynamics using sequential FDAP

Abstract

Actin filament dynamics and architecture in cells are highly organized in space and time. Cells integrate intracellular and extracellular signals to regulate the activities of various actin regulatory proteins and actin filament reorganization, which leads to cell migration, morphogenesis and division. To understand the mechanisms that spatiotemporally regulate actin filament dynamics and architecture in cells, various microscopic techniques have been developed. These techniques have provided quantitative data on the kinetics of actin filament dynamics that are useful for mathematical model analysis.^1,2 In particular, the technique of fluorescence single-molecule speckle imaging, which can be used to observe directly the behavior of single fluorescence-labeled molecules attached to actin filaments, has provided extensive kinetic data on the processes of actin filament assembly and disassembly in live cells.²

During migration, cells extend F-actin-rich lamellipodia at their leading edge. In lamellipodia, dendritic arrays of actin filaments are constructed with fast-growing barbed ends directed to the plasma membrane. Actin polymerization at the barbed end generates a physical force to push the plasma membrane forward, resulting in lamellipodium extension. Next, new focal adhesions are formed in the lamellipodium, and the actomyosin contraction force pulls the cell body forward. In principle, cells migrate by cycling through these steps. Actin polymerization-based lamellipodium protrusion is the first and most important step in cell migration. Lamellipodia sense environmental cues around the cell and determine the potency and directionality of cell migration by dynamically changing their morphology in response to these cues.^3,4

Fluorescence single-molecule speckle imaging has provided many kinetic parameters of actin and actin-binding proteins on F-actin, which underlie the dynamic behavior of lamellipodia, including the rate of mDia (FH1-FH2)-mediated F-actin elongation (2 μm s⁻¹), the rates of dissociation of Arp2/3 (0.048 sec⁻¹) and capping protein (0.58 sec⁻¹) from F-actin, and the rate of F-actin disassembly (0.03 sec⁻¹) in lamellipodia.^5-7 Thus, substantial kinetic data on actin and actin-binding proteins in the processes from actin assembly to disassembly on F-actin have been accumulated. In contrast, kinetic data on G-actin and its binding proteins in the processes after actin disassembly and before actin assembly (which occur primarily in the cytosol) have remained insufficiently characterized, owing to a lack of appropriate methods to measure the changes in G-actin concentration quantitatively in live cells.

In most cells, the cytoplasmic G-actin concentration (~100 μM) is much higher than the critical concentration at the barbed end of an actin filament (~0.1 μM).⁸ The high concentration of cytoplasmic G-actin is maintained by several factors: 1) the supply of actin monomers via actin-depolymerizing factor (ADF)/cofilin-mediated F-actin disassembly; 2) the prevention of spontaneous actin nucleation by G-actin-sequestering proteins, profilin and thymosin-β4; and 3) the interruption of actin assembly by F-actin-capping proteins. To evaluate the contribution of each actin regulatory protein to the maintenance of high cytoplasmic G-actin, it is essential to quantify the cellular G-actin concentration. To assess the cytoplasmic G-actin concentration in live cells, time-lapse analysis of fluorescence decay after photoactivation (FDAP) of Dronpa (Dp)-labeled actin was used (Fig. 1A).¹⁰

Open in a separate window

Figure 1. Outline of s-FDAP analysis.⁹ (A) Time course of Dp-actin fluorescence decay after single photoactivation (FDAP). Dp-actin expressed in the cell was prebleached to the background level, and a DIC image was acquired. After photoactivation, time-lapse images were acquired. The fluorescence decay was fitted to calculate the mobile and immobile fractions and diffusion coefficient of Dp-actin.⁹ (B) s-FDAP analysis. Fluorescence images were acquired before (blue) and at 0 (green) and 40 ms (red) after photoactivation. Each set of procedures was repeated 84 times at 10 sec intervals. FDAP values are expressed as the difference between the fluorescence intensities at 0 and 40 ms after photoactivation.

Dp is a GFP-like photochromic protein whose fluorescence can be reversibly and repeatedly photoactivated and photobleached by irradiation with light.¹¹ After photobleaching the whole cell, Dp-actin was photoactivated in a small region of the cytoplasm, and the time-dependent change in fluorescence intensity in the photoactivated region was measured (Fig. 1A). Fluorescence intensity in the photoactivated region decreased quickly due to the diffusion of Dp-labeled G-actin. Because the rate of the fluorescence decay of Dp-actin was much faster than the rate of F-actin disassembly, the FDAP value primarily reflects the amount of G-actin (mobile fraction of Dp-actin) in the local area of the cytoplasm. The contribution of F-actin disassembly to the FDAP value is negligible at least in the initial phase after photoactivation.

The fluorescence recovery after photobleaching (FRAP) of GFP-actin was also utilized to assess the local amount of G-actin.¹² However, FDAP analysis has an advantage over FRAP analysis for quantifying the amount of G-actin with high signal-to-noise ratio: the fluorescence signal in the region of interest (ROI) is maximal at the beginning of FDAP measurements but minimal at the beginning of FRAP measurements. The FDAP analysis revealed that the cytoplasmic G-actin concentration was decreased by the inactivation or depletion of ADF/cofilin to less than half of the concentration in control cells.¹⁰ This result provides clear evidence that ADF and cofilin play a central role in the production and maintenance of high concentration of cytoplasmic G-actin.¹⁰ A similar approach can be used to evaluate the contributions of other actin regulatory proteins to the control of cytoplasmic G-actin concentration.

In the cytoplasm, the ability of G-actin to polymerize to F-actin is controlled by its state of association with G-actin-binding proteins. Major components of the cytoplasmic G-actin pool are free ATP-G-actin, profilin-bound and thymosin-β4-bound ATP-G-actin, and ADF/cofilin-bound ADP-G-actin. Free ATP-G-actin and profilin-bound ATP-G-actin are polymerizable, whereas thymosin-β4-bound ATP-G-actin and ADF/cofilin-bound ADP-G-actin are nonpolymerizable.^13-15 Although most cytoplasmic G-actin consists of nonpolymerizable thymosin-β4-bound ATP-G-actin, this complex is readily converted to free ATP-G-actin when the free G-actin concentration is reduced by an increase in the actin assembly rate or a decrease in the actin disassembly rate.¹⁶ Thus, thymosin-β4-bound ATP-G-actin also contributes to actin assembly by acting as a reservoir of ATP-G-actin. Kinetic data on these biochemical reactions in actin dynamics have been extensively studied in the cell-free system.^14,16

To investigate the mechanisms of intracellular reactions of G-actin dynamics, such as G-actin disassembly from F-actin, G-actin assembly to F-actin, and the interactions between G-actin and its regulating proteins, it is essential to measure time-dependent changes in cytoplasmic G-actin concentration in live cells quantitatively and with high temporal resolution. As described above, the amount of cytoplasmic G-actin can be assessed by FDAP analysis of Dp-actin. Because Dp is a unique protein whose fluorescence can be reversibly and repeatedly switched on and off, Dp-actin is a useful probe for sequential measurements of FDAP. These characteristics allow the quantitative analysis of time-dependent changes in the G-actin concentration in live cells.⁹

Sequential FDAP (s-FDAP) analysis consists of repetitions of a set of procedures: differential interference contrast (DIC) image acquisition, photobleaching of the whole cell, and photoactivation of a small (1.8 μm diameter circle) region of the cytoplasm (Fig. 1B).⁹ In each cycle, a set of three fluorescence images of the photoactivated region was acquired before and at 0 and 40 ms after photoactivation. The FDAP value, which reflects the G-actin concentration, was calculated as the difference between the fluorescence intensities at 0 and 40 ms after photoactivation. The time point for image acquisition (40 ms after photoactivation) was set on the basis of the half-life of the mobile fraction of Dp-actin (41 ms) measured by a single FDAP analysis under similar conditions. The short-term measurement of FDAP enables s-FDAP analysis to measure temporal changes in the G-actin concentration with high temporal resolution. As described above, the effect of F-actin disassembly on s-FDAP analysis is negligible.^9,10

Fluorescence correlation spectroscopy (FCS) has been developed to measure the cytoplasmic concentration of diffusive molecules.^17,18 FCS detects the number and diffusion time of fluorescence-labeled single molecules in the observed volume. Its autocorrelation function is calculated to determine the concentration and diffusion coefficient of the fluorescent molecule in the cytoplasm. However, FCS can only measure nanomolar concentrations of a fluorescent molecule; high concentrations of fluorescent molecules make it difficult to detect each fluorescent molecule separately in the observed volume.¹⁸ Thus, FCS has an upper limit of the fluorescent signal intensity and requires a relatively long measuring time (typically 10 sec × 5~10 times).¹⁸ In addition, FCS is based on the premise that the concentration of fluorescent molecules does not change during the measuring time. Thus, FCS is unfavorable for measuring temporal changes in cytoplasmic G-actin concentration because the changes in G-actin concentration are very fast in cells and a high temporal resolution is required to measure them.⁹

The s-FDAP analysis was used to measure time-dependent changes in cytoplasmic G-actin concentration after the treatment of cells with an F-actin-stabilizing drug, jasplakinolide (Jasp).⁹ The FDAP value gradually decreased after Jasp treatment in a time- and dose-dependent manner. The high temporal resolution of s-FDAP analysis was sufficient to use the data of Jasp-induced changes in G-actin concentration for model analysis. In the best-fit model, the cellular concentrations of free G-actin, profilin-bound and thymosin-β4-bound G-actin, and free barbed and pointed ends of actin filaments in the untreated cell were estimated to be 1.98 μM, 9.52 μM, 88.5 μM, 4.8 nM and 5.52 nM, respectively, assuming that the total concentrations of G-actin, profilin and thymosin-β4 were 100, 10 and 200 μM, respectively.⁹ Based on these values, time courses of the concentrations of each component after Jasp treatment were estimated, as shown in Figure 2. Thus, the high temporal resolution and quantitative features of s-FDAP analysis make it useful for estimating the cellular concentrations of key elements involved in the regulation of actin dynamics.

Open in a separate window

Figure 2. Model analysis of Jasp-induced changes in the concentrations of free pointed ends (A), free G-actin (B), profilin-bound G-actin (C) and thymosin-β4-bound G-actin (D). Temporal changes in FDAP values were measured after the treatment of MCF-7 cells with Jasp at various concentrations. Cellular concentrations of free G-actin, profilin-bound G-actin, thymosin-β4-bound G-actin, and the free barbed and pointed ends of F-actin were estimated by a differential equation model.⁹ With these concentrations and the chemical reaction equations in reference ⁹, the time course of the concentration of each component was plotted by KINSIM. The free barbed end concentration is constant in the chemical reaction equations.

The s-FDAP analysis was also applied to measure the temporal changes in G-actin concentration after cell stimulation with neuregulin (NRG), a growth factor that induces lamellipodium extension and cell migration.⁹ Results from s-FDAP analysis demonstrated that the G-actin concentration decreased by ~40% at 2–4 min after NRG stimulation and then reached a plateau. This result indicates that ~40% of G-actin in the cytoplasm is polymerized to F-actin in response to NRG stimulation.⁹

The s-FDAP analysis can be used to measure temporal changes in cell morphology by DIC image acquisition, simultaneously with changes in G-actin concentration, in individual live cells. The DIC image analysis showed that the cell area increased about 1.6-fold at 6–8 min after NRG stimulation, with a lag time of 2–4 min after the decrease in cytoplasmic G-actin concentration. Data on the time-lapse fluorescence analyses of YFP-actin localization and measurements of free barbed-end formation in NRG-stimulated cells suggest that the cytoplasmic G-actin is used for F-actin assembly and Arp2/3 complex-mediated dendritic formation of barbed ends at the cell periphery in the initial phase after NRG stimulation, and thereafter the cell area begins to extend.⁹

Because the rate of actin polymerization is proportional to the G-actin concentration in cell-free assays,¹⁹ it is reasonable to assume that the cytoplasmic G-actin concentration is a critical parameter for the intracellular rate of actin polymerization. However, some studies have paid little attention to the role of G-actin concentration in stimulus-induced actin assembly and lamellipodium extension, probably because the cytoplasmic G-actin is more than enough for actin assembly. Thus, it remains unclear to what extent the cytoplasmic G-actin concentration contributes to stimulus-induced actin assembly and cell extension.

To address this issue, the G-actin concentration in cells was artificially reduced by Jasp treatment, and its effects on NRG-induced actin assembly and cell size extension were analyzed by s-FDAP.⁹ The extent of NRG-induced actin assembly (measured as the decrease in the cytoplasmic G-actin concentration) and cell size extension exhibited a linear correlation with the G-actin concentration in the prestimulated cell.⁹ Strong correlations over a wide range of G-actin concentrations indicate that the extent of actin assembly and cell size extension after cell stimulation strongly depend on the cytoplasmic G-actin concentration before cell stimulation over a wide range of G-actin concentrations.

Inactivation or depletion of ADF/cofilin suppresses stimulus-induced actin assembly and lamellipodium extension, indicating that ADF/cofilin plays a critical role in these processes. In general, the rate of actin assembly depends on the concentrations of both G-actin and the free barbed ends of F-actin. Based on the potency of ADF/cofilin to increase both the G-actin concentration and free barbed ends, two models for the roles of ADF/cofilin have been proposed: ADF/cofilin contributes to stimulus-induced actin assembly and lamellipodium extension by (1) abundantly supplying G-actin to maintain a high concentration of cytoplasmic G-actin for polymerization,^9,10 or (2) severing F-actin to create free barbed ends that are used as nucleation sites for actin polymerization.^20,21 We showed that the (a) inactivation or depletion of ADF/cofilin remarkably decreases the cytoplasmic G-actin concentration, indicating that ADF/cofilin primarily contributes to the maintenance of high cytoplasmic concentrations of G-actin; (b) stimulus-induced actin assembly and cell extension strongly depend on the cytoplasmic G-actin concentration in the prestimulated cell; and (c) microinjection of actin monomers into ADF/cofilin-inactivated cells rescues actin assembly and cell extension. These results suggest that ADF/cofilin contributes to stimulus-induced actin assembly and cell extension by supplying a large amount of G-actin into the cytoplasm, thereby maintaining a high concentration of cytoplasmic G-actin for polymerization. Although ADF/cofilin is required for stimulus-induced barbed-end formation at the cell periphery, it probably indirectly supports the creation of barbed ends by increasing the cytoplasmic G-actin concentration (which, in turn, accelerates the rate of actin polymerization and increases the Arp2/3 complex-mediated branching and barbed-end formation of actin filaments), rather than by directly creating barbed ends via their F-actin-severing activity.

The FDAP technique was also applied to measure the spatial distribution of G-actin concentration by sequentially performing single FDAP analyses at multiple points in a live cell.⁹ Multipoint FDAP analysis of N1E-115 cells extending stable lamellipodia has shown that the G-actin concentration in lamellipodia is comparable to that in the cell body.⁹ This finding indicates that the G-actin concentration in lamellipodia is sufficiently high to support F-actin polymerization at the leading edge. Time-lapse fluorescence imaging has shown that G-actin disassembled from the rear of the lamellipodia rapidly diffuses into the cytoplasm, and that cytoplasmic G-actin is effectively incorporated into F-actin in the lamellipodia.¹⁰ Thus, it is likely that the rapid diffusion of a large amount of cytoplasmic G-actin, which is much greater than the amount of G-actin generated by lamellipodial F-actin disassembly in a given period of time, equalizes the G-actin concentrations in the lamellipodia and cell body. It will be interesting to investigate whether or not the spatial distribution of G-actin concentration differs in various types of migrating or polarized cells.

In motile cells, the lamellipodium extends and retracts repeatedly.^22,23 In the spontaneous polarization of isotropically spreading fibroblasts and keratocytes, membrane retraction in the prospective rear of the cell was initiated prior to directional cell movement.^24,25 Actin assembly and disassembly are local biochemical reactions regulated by different reaction components. However, spatiotemporal linkage between these reactions suggests their coordination in the entire cell.

We have shown that the change in G-actin concentration in the cytoplasm is directly correlated with stimulus-induced actin assembly and cell extension.⁹ Changes in the total G-actin concentration probably alter the biochemical reactions between G-actin and its regulatory proteins in the cytoplasm. In addition, the mDia1-mediated nucleation of actin filaments is presumably regulated by fluctuation in the G-actin concentration,²⁶ and the rate of G-actin incorporation at the tip of the lamellipodium is linearly correlated with the cytoplasmic G-actin concentration.¹⁰ Thus, the quantitative analysis of the correlations between cytoplasmic G-actin concentration and kinetics of local actin assembly/disassembly will shed light on the molecular mechanisms by which motile cells coordinate local actin assembly and disassembly in the entire cell.

The s-FDAP and multipoint FDAP analyses are applicable to measure temporal and spatial changes in the concentrations of mobile components of various cytoskeletal and membrane-bound proteins in live cells. These techniques will serve to investigate the roles of these proteins in diverse cell processes, including cell migration, morphological changes and cytokinesis.

Glossary

Notes

Kiuchi T, Nagai T, Ohashi K, Mizuno K. Measurements of spatiotemporal changes in G-actin concentration reveal its effect on stimulus-induced actin assembly and lamellipodium extensionJ Cell Biol201119336580

Footnotes

References