Control of microtubule dynamics using an optogenetic microtubule plus end-F-actin cross-linker

Abstract

We developed a novel optogenetic tool, SxIP-improved light-inducible dimer (iLID), to facilitate the reversible recruitment of factors to microtubule (MT) plus ends in an end-binding protein-dependent manner using blue light. We show that SxIP-iLID can track MT plus ends and recruit tgRFP-SspB upon blue light activation. We used this system to investigate the effects of cross-linking MT plus ends and F-actin in Drosophila melanogaster S2 cells to gain insight into spectraplakin function and mechanism. We show that SxIP-iLID can be used to temporally recruit an F-actin binding domain to MT plus ends and cross-link the MT and F-actin networks. Cross-linking decreases MT growth velocities and generates a peripheral MT exclusion zone. SxIP-iLID facilitates the general recruitment of specific factors to MT plus ends with temporal control enabling researchers to systematically regulate MT plus end dynamics and probe MT plus end function in many biological processes.

Graphical abstract

Figure 1.

SxIP-iLID constructs track MT plus ends and do not dramatically perturb MT comet velocities. (A) Schematic of EGFP-labeled SxIP-iLID constructs. (B) Cartoon diagraming EGFP-SKIP-LZ-iLID at a MT plus end. (C) Representative images of EGFP-SxIP-iLID constructs in S2 cells. Bars: 5 µm; (inset) 1 µm. Kymographs below each image show a representative EGFP-SxIP-iLID MT plus end comet. See accompanying Video 1. (D) The ratio of mean EGFP-SxIP-iLID construct intensity on a MT plus end per area relative to the mean intensity per area in an adjacent cytoplasmic region. (E) The number of EGFP positive comets per μm². (F) SxIP-iLID constructs show some variation in EGFP comet velocities; however, mean comet velocities of the SKIP-LZ-iLID construct and the EB1-GFP control are not significantly different. Error bars indicate the SD. Numbers in parentheses indicate (number of experiments, total number of cells quantified). P-values were determined by two-way unpaired Student’s t test. *, P < 0.05; **, P < 0.005; ****, P < 0.0001.

Figure 2.

Photoactivated SxIP-iLID constructs rapidly recruit tgRFP-SspB to MT plus ends without altering MT comet velocities. (A) Diagram of SKIP-LZ-iLID and tgRFP-SspB constructs. (B) Cartoon diagram of tgRFP-SspB binding to SKIP-LZ-iLID upon blue light activation (light: hν). SKIP-LZ-iLID localizes to MT plus ends via association with EB dimers. (C) Representative images of S2 cells cotransfected with either a tandem SxIP-iLID construct or a dimeric SKIP-LZ-iLID construct and a tgRFP-SspB control construct, repeatedly pulsed with blue light (700 ms) every 3 s at 488 nm. Images show tgRFP-SspB localization before and after photoactivation (see accompanying Video 2). The SKIP-LZ-iLID^Lit construct is constitutively competent to bind SspB in the absence of blue light. Bars: (left) 5 µm; (inset) 1 µm. Kymograph bars: 2 µm, 25 s. (D) The postactivation/preactivation ratio of mean tgRFP comet intensity demonstrating effective tgRFP-SspB recruitment. tgRFP comet intensity per area on a MT plus end was calculated relative to the mean intensity per area in an adjacent cytoplasmic region. (E) SxIP-iLID-based recruitment of tgRFP-SspB to MT plus ends does not significantly alter mean MT plus end comet velocities compared with control cells cotransfected with EB1-GFP and tgRFP-SspB. Error bars indicate SD. Numbers in parentheses indicate (number of experiments, total number of cells quantified). P-values were determined by two-way unpaired Student’s t test. **, P < 0.005; ***, P < 0.0005.

Figure 3.

Dynamics of blue light–dependent tgRFP-SspB MT plus end recruitment and dissociation. (A) Montage of tgRFP-SspB before and after activation in a S2 cell cotransfected with EGFP-SKIP-LZ-iLID. Time in seconds is relative to t = 0, at which time the cell was activated with a single 200-ms pulse of 0.15 mW 488-nm light. The montage shows blue light–dependent tgRFP-SspB recruitment to MT plus ends as well as subsequent dissociation over time (see accompanying Video 3). Bar, 5 µm. (B) Plot displaying the apparent cellular kinetics of tgRFP-SspB MT plus end association and dissociation over time. The system was activated at t = 0 with a single pulse of 0.15 mW 488-nm light. n = 5 cells. In cells cotransfected with SKIP-LZ-iLID, tgRFP-SspB immediately associates with MT plus ends, reaches maximal recruitment within 30 s after activation, and dissociates within 90 s. Inset shows the general kinetic steps that enable tgRFP-SspB MT plus end association, the convolution of which yields the apparent kinetics observed. The apparent half life of tgRFP-SspB on the MT plus end after activation is 25.1 ± 6.7 s. (C) Multiple rounds of activation of cells cotransfected with SKIP-LZ-iLID and tgRFP-SspB show the ability of tgRFP-SspB to be recruited to MT plus ends multiple times. Cells were activated at t = 0, 150, and 300 s with a single pulse of 0.10 mW 488-nm light (see accompanying Video 4). n = 7 cells. For B and C, black points and gray area represent the mean and SEM, respectively.

Figure 4.

Optogenetically induced cytoskeletal cross-linking decreases MT comet velocities and increases the MT-void area. (A) Schematic of Drosophila (D.m.) Shot. Shot contains tandem N-terminal CH domains that bind F-actin, a C-terminal GAR domain that binds MTs, and a SxIP motif that confers MT plus end localization. We functionally parsed Shot’s F-actin– and MT-binding activity into the SxIP-iLID system, fusing Shot’s CH domains to tgRFP-SspB and using the SKIP-LZ-iLID construct, which contains a SxIP motif from the mammalian spectraplakin MACF2. (A’) Image of CH-CH-tgRFP-SspB in a transfected S2 cell (left). Bar, 10 µm. See accompanying Video 5. The kymograph at right represents a time-lapse of the red line scan at left, and shows retrograde CH-CH-tgRFP-SspB movement in the lamellar region. Kymograph bars: 1 µm, 25 s. (B) Representative images of S2 cells cotransfected with EGFP-SKIP-LZ-iLID and either a tgRFP-SspB control, or the F-actin–binding CH-CH-tgRFP-SspB construct, and repeatedly pulsed with blue light (250 ms every 3 s at 488 nm). Top and middle: Single images from the EGFP and RFP channels. Bar, 5 µm. Right: Kymographs show representative EGFP-SKIP-LZ-iLID plus end comets. Kymograph bars: 2 µm, 25 s. Bottom: Maximal projections of 60 frames (total 3 min) are shown, revealing the area of the cell traversed by EGFP-SKIP-LZ-iLID–containing MT plus ends. Areas void of MT plus ends can be seen in the cross-linked cell (arrowhead). Blue boxes: Blue light recruitment of tgRFP-SspB constructs. (C) Representative images of S2 cells cotransfected with EGFP-SKIP-LZ-iLID^Lit and either a tgRFP-SspB control or the F-actin–binding CH-CH-tgRFP-SspB construct. Top and middle: single images from the EGFP and RFP channels. Bar, 5 µm. Right: Kymographs show representative EGFP-SKIP-LZ-iLID^Lit plus end comets. Kymograph bars: 2 µm, 25 s. Bottom: Maximal projections of 60 frames (total 3 min) are shown, revealing the area of the cell traversed by EGFP-SKIP-LZ-iLID^Lit–containing MT plus ends. Areas void of MT plus ends can be seen in the constitutively cross-linked cell (see accompanying Video 6). (D) EGFP comet velocity/cell (μm/min) for control cells (iLID +SspB and iLID^Lit+SspB), light-activated cross-linked cells (iLID + CH-CH-SspB), and constitutively cross-linked cells (iLID^Lit + CH-CH-SspB). P-values were determined by two-way unpaired Student’s t test. (E) MT-void area as a percentage of the total cell area for control and cross-linked cells. Maximum projection images of 60 frames (spanning 3 min) were used to determine the area of the cell void of MTs (representative images used for quantification are shown in B and C and Video 6). P-values were determined by two-tailed nonparametric Mann–Whitney U test. For plots in D and E, numbers in parentheses indicate (number of experiments, total number of cells quantified). Line represents the mean, and error bars indicate SD. *, P < 0.05; **, P < 0.005; ****, P < 0.0001.

Figure 5.

Optogenetically induced cross-linking decreases MT comet velocities and increases the area void of MT plus ends in an F-actin–dependent manner. (A) Representative images of EGFP-SKIP-LZ-iLID^Lit coexpressed with CH-CH-tgRFP-SspB or tgRFP-SspB in S2 cells treated with DMSO as a control, or LatA (2 nM or 2 µM) for 1 h before imaging. Top and middle: EGFP and RFP channels, respectively. Bottom: Maximum projections of the EGFP channel (60 frames collected over 3 min). Bar, 5 µm. (B) EGFP comet velocity/cell (μm/min) for control cells (SKIP-LZ-iLID^Lit + SspB) and constitutively cross-linked cells (SKIP-LZ-iLID^Lit + CH-CH-SspB) treated with DMSO or LatA (2 nM or 2 µM). P-values were determined by two-way unpaired Student’s t test. (C) MT-void area as a percentage of the total cell area for constitutively cross-linked cells and control cells treated with DMSO, 2 nM LatA, or 2 µM LatA. Maximum projection images of 60 frames (collected over 3 min) were used to determine the area of the cell void of growing MT plus ends (representative image used for quantification shown in A). MT-void area decreased in constitutively cross-linked cells with the addition of LatA as compared with control DMSO treatment. P-values were determined using a two-tailed nonparametric Mann–Whitney U test. For plots in B and C, numbers in parentheses indicate (number of experiments, total number of cells quantified). Central lines represent the mean, and error bars indicate SD. *, P < 0.05; **, P < 0.005; ***, P < 0.0005; ****, P < 0.0001.

Figure 6.

CH-CH MT plus end recruitment is required to decrease MT comet velocities and generate a peripheral MT exclusion zone. (A) Representative still images of SKIP-LZ-iLID or SKIP-LZ-iLID^Lit coexpressed with CH-CH-tgRFP or tgRFP-SspB in S2 cells, showing EGFP and RFP channels. Right: Maximum projections of 60 frames (collected over 3 min) showing the area traversed by MT plus ends. Bar, 5 µm. (B) EGFP comet velocity/cell (μm/min) for cross-linked control cells (SKIP-LZ-iLID + SspB activated with blue light and SKIP-LZ-iLID^Lit + SspB) and cells transfected with a CH-CH-tgRFP construct that binds F-actin, but lacks the SspB domain that mediates cross-linking to MT plus ends (SKIP-LZ-iLID + CH-CH and SKIP-LZ-iLID^Lit + CH-CH). P-values were determined by two-way unpaired Student’s t test. (C) Area void of MT plus ends as a percentage of the total cell area for control cells (SKIP-LZ-iLID + SspB activated with blue light and SKIP-LZ-iLID^Lit + SspB) and cells in which the CH-CH construct engages the actin network, but is not cross-linked to MT plus ends (SKIP-LZ-iLID + CH-CH and SKIP-LZ-iLID^Lit + CH-CH). Maximum projection images of 60 frames (collected over 3 min) were used to determine the area of the cell void of MT plus ends (representative image used for quantification shown in A). P-values were determined using a two-tailed nonparametric Mann–Whitney U test. For plots in B and C, numbers in parentheses indicate (number of experiments, total number of cells quantified). Central lines represent the mean, and error bars indicate SD. *, P < 0.05; ***, P < 0.0005; ****, P < 0.0001.

Figure 7.

Cross-linking decreases MT comet velocities throughout the cell, limits MT entry into the peripheral zone, and increases the percent of comets that are swept along the peripheral zone boundary. (A) Schematic of MT comet trajectories categorized in S2 cells (left) and change in mean comet velocity in cross-linked compared with non–cross-linked cells (right, based on mean velocities presented in B). (B) EGFP comet velocities from control cells (cotransfected with SKIP-LZ-iLID^Lit + CH-CH, open circles) and constitutively cross-linked cells (cotransfected with SKIP-LZ-iLID^Lit + CH-CH-SspB, filled circles), grouped based on comet trajectory as delineated in A. Multiple individual EGFP-labeled comet velocities were determined and compiled across multiple cells. Numbers in parentheses indicate (number of independent experiments, total number of cells analyzed, total number of tracks annotated). Mean comet velocities are reported below the dotted line, and the relative percentages of comets in each category are shown at the top of the plot (bold percentages: >2-fold difference between cross-linked and non–cross-linked cells). Line represents the mean, and error bars indicate SD. P-values were determined by two-way unpaired Student’s t test. *, P < 0.05; **, P < 0.005; ***, P < 0.0005; ****, P < 0.0001.

Figure 8.

Schematic of MT exclusion upon optogenetically induced cross-linking. EB1 (yellow) tracks and recruits EGFP-SKIP-LZ-iLID to the growing MT plus end. CH-CH-tgRFP-SspB engages F-actin. Upon blue light activation, the iLID module engages CH-CH-tgRFP-SspB, cross-linking MTs and F-actin. In the F-actin–rich peripheral zone, the F-actin density and/or retrograde flow drives MT plus end exclusion from the region and “sweeps” MT comets along the peripheral zone boundary.