Spatial sensing in fibroblasts mediated by 3' phosphoinositides

Abstract

The directed movement of fibroblasts towards locally released platelet-derived growth factor (PDGF) is a critical event in wound healing. Although recent studies have implicated polarized activation of phosphoinositide (PI) 3-kinase in G protein-mediated chemotaxis, the role of 3' PI lipids in tyrosine kinase-triggered chemotaxis is not well understood. Using evanescent wave microscopy and green fluorescent protein-tagged Akt pleckstrin homology domain (GFP-AktPH) as a molecular sensor, we show that application of a shallow PDGF gradient triggers a markedly steeper gradient in 3' PI lipids in the adhesion zone of fibroblasts. Polar GFP-AktPH gradients, as well as a new type of radial gradient, were measured from front to rear and from the periphery to the center of the adhesion zone, respectively. A strong spatial correlation between polarized 3' PI production and rapid membrane spreading implicates 3' PI lipids as a direct mediator of polarized migration. Analysis of the temporal changes of 3' PI gradients in the adhesion zone revealed a fast diffusion coefficient (0.5 microm(2)/s) and short lifetime of 3' PIs of <1 min. Together, this study suggests that the tyrosine kinase-coupled directional movement of fibroblasts and their radial membrane activity are controlled by local generation and rapid degradation of 3' PI second messengers.

Figure 1

Evanescent wave imaging of PDGF-induced GFP–AktPH plasma membrane translocation in living fibroblasts. (A) Schematic view of the modified evanescent wave microscope design (see Materials and Methods for details). In short, the microscope uses water immersion objectives to image adherent cells in an open chamber, leaving the cells accessible for live cell experimentation. The 488-nm excitation laser beam is coupled through a prism below the coverslip and utilizes an oil drop between the coverslip and prism for light coupling, enabling total internal reflection from the glass–water interface. The coverslip is fixed on two sides to a motor controlled stage so that the cells can be moved in the x and y directions. (B) Comparison between evanescent wave and epifluorescence illumination. The same GFP-expressing NIH 3T3 fibroblasts were imaged using epifluorescence (top) and evanescent wave excitation (bottom). Bars, 20 μm. (C) Comparison of evanescent wave excited fluorescence intensities between soluble and plasma membrane-localized GFP. Low magnification images are shown of fibroblasts expressing GFP (left) or membrane-targeted GFP (Lyn-GFP, right). (D) Evanescent wave images of NIH 3T3 fibroblasts expressing a GFP fusion of the Akt PH domain (GFP–AktPH) captured before (left) and 5 min after (right) addition of PDGF-BB. Translocation was blocked by 100 μM LY294002, a specific inhibitor of PI 3-kinases (bottom). (E) Representative translocation responses of individual GFP–AktPH-transfected fibroblasts in response to 1–30 pM, 100–300 pM, or 1–10 nM PDGF-BB at room temperature. A second dose adjusted the PDGF concentration to 10 nM. (F) Dose response of PDGF-induced GFP–AktPH translocation, normalized by the response to a second, maximal dose. The responses of cells in the same field were compiled, and the data points are averaged over experiments on three separate days (± SEM, n = 3). (G) The fold increase in fluorescence intensity stimulated by a maximal PDGF dose is plotted for individual cells, as a function of the prestimulus fluorescence intensity normalized by the beam power*exposure time (mJ). The symbols signify experiments performed on different days.

Figure 2

Induction of 3′ PI lipid gradients and polarized membrane spreading in response to PDGF gradients. (A) Time series of evanescent wave images showing GFP–AktPH-transfected NIH 3T3 fibroblasts responding to a transient PDGF-BB gradient at 37°C (PDGF diffusing in from the right). The arrows in the bottom panels mark a membrane extension event on the side of a cell that was initially oriented towards the bottom of the field. (B) Magnification of the cell in the upper right of A, 3-min time point. (C) Surface-proximal fluorescence profiles for the cell in A and B at various time points after PDGF addition. The arrow in the first panel of A shows the position and direction of the line scan. (D) Comparison of the time course of GFP–AktPH translocation and membrane spreading. GFP–AktPH fluorescence increase at the leading edge was compared with the increase in total contact area spreading (for the top cell described in A–C). (E) Time series of localized translocation as in A, except PDGF is diffusing from the left. The arrows mark an apparent lamellipodial extension. A and E are representative of six separate experiments. Bars, 20 μm.

Figure 4

Demonstration of fibroblast orientation and 3′ PI gradients in wounded monolayers. Evanescent wave images showing GFP–AktPH-transfected fibroblasts at 37°C with basal gradients at the edge of a wounded monolayer. The response to a uniform dose (5 nM) of PDGF-BB is shown (representative of five separate experiments). Epifluorescence and bright-field images are shown in the last two panels. The acellular area generated by the wounding protocol is to the right of the field. Bar, 20 μm.

Figure 3

Basal gradients in 3′ PI lipids correlate with cell orientation. (A) Evanescent wave images of GFP–AktPH-transfected NIH 3T3 fibroblasts at 37°C, showing basal 3′ PI gradients and a more symmetrical response after uniform PDGF-BB stimulation (5 nM). PI 3-kinase activity was rapidly inhibited after 3 min by the addition of wortmannin (Wort). (B) Spatial fluorescence profiles for the cell on the right in A. (C) Time course, average fluorescence intensity in the contact area for the cell in A and B. (D) Evanescent wave images of a representative GFP–AktPH-transfected fibroblast at low cell density, stimulated with a uniform dose (5 nM) of PDGF-BB at 37°C. (E) Spatial fluorescence profiles for the cell in D. All arrows indicate the position and direction of the line scans used to generate spatial profiles. Bars, 20 μm.

Figure 7

Proposed dual roles of 3′ PIs in fibroblast motility. The model assumes that multiple spatial cues, including gradients of PDGF and immobilized ligands that mediate adhesion, are integrated over space and time at the level of 3′ PI distribution in the membrane to generate polarized cell migration. Polar (front-back) asymmetry in 3′ PI generation leads to a bias in the orientation of membrane extension. At the same time, cell adhesion regulates signaling from PDGF receptors, leading to radial asymmetry in 3′ PI concentration within the contact area. The important parameter is the ratio of 3′ PI diffusion to degradation, which determines the range of the second messenger. This suggests a shaping mechanism in which 3′ PI level contributes to cell spreading, but the increase in adhesion limits signal propagation. These two modes of spatial asymmetry would cooperate to focus 3′ PI distribution at the leading edge for optimal migration.

Figure 5

PDGF-stimulated radial 3′ PI gradients in the cell contact area. (A) Representative translocation response of GFP–AktPH-transfected NIH 3T3 fibroblasts to uniformly maximal PDGF-BB stimulation at room temperature, showing radial gradients in GFP–AktPH translocation with low intensities in the center and high intensities at the periphery. (B) Control experiments with soluble and membrane-targeted GFP show uniform fluorescence intensities that are not significantly changing with PDGF stimulation. Representative evanescent wave images of fibroblasts expressing GFP (top) or Lyn-GFP (bottom) were captured before (left) and 5 min after (right) addition of PDGF-BB. (C) Confocal fluorescence images of unstimulated NIH 3T3 fibroblasts cotransfected with GFP–AktPH and either a control vector (pΔC1) or G12V Ha-Ras. The coexpressed activator of PI-3 kinase induces a strong localization of GFP–AktPH in unstimulated cells. (D) Evanescent wave images showing fibroblasts, cotransfected with GFP–AktPH and G12V Ha-Ras, and responding to PDGF stimulation at 37°C show a marked initial decrease in fluorescence intensity. (E) Time course of the average fluorescence intensity for each of the cells in D. Bars, 20 μm.

Figure 6

Model calculations enable a determination of the diffusion coefficient and lifetime of 3′ PI lipids. (A) Model schematic. The mathematical model accounts for lipid generation, characterized by a reaction velocity V, lipid diffusion, characterized by a diffusion coefficient D, and lipid consumption, characterized by the rate constant k_c. The expression level and equilibrium dissociation constant K _D of the fluorescent probe are also accounted for. (B) Experimental fluorescence profiles measured for the cell on the right in Fig. 5 A. (C) Theoretical fluorescence profiles, assuming that no 3′ PI lipids are generated in the contact area (V_b/V_t = 0). Other model parameters were optimized to match various aspects of the experimental profiles in B (Materials and Methods): D = 0.5 μm²/s, k_c = 1 min⁻¹ (Da = k_cR²/D = 3), κ = 1/3, γ = 3, σ = 14. In accord with the early time points of the series in Fig. 5 A (not shown), a time lag of 25 s was built into the model. (D) Experimental determination of the degradation rate constant k_c by rapidly inhibiting PI 3-kinase activity. GFP–AktPH-expressing fibroblasts were maximally stimulated with PDGF-BB for 5 min, after which 250 μM LY294002 was added (time zero). (E) An alternative model assuming significant second messenger generation in the contact area yields poor agreement with experiment. Computed profiles are shown for different normalized times τ = k_c*t, with the parameter values in C except Da = 30, V_b/V_t = 0.35.