FIGURE SUMMARY
Title

The PAR complex controls the spatiotemporal dynamics of F-actin and the MTOC in directionally migrating leukocytes

Authors
Crespo, C.L., Vernieri, C., Keller, P.J., Garrè, M., Bender, J.R., Wittbrodt, J., Pardi, R.
Source
Full text @ J. Cell Sci.

The PAR complex promotes wound-directed migration of myeloid cells in vivo. (A) TG(FmpoP::memYFP) embryos were injected at the one-cell stage with a 1∶1 mixture of DNA coding for H2AmCherry as a nuclear reporter and one of the PAR transgenes, driven by the myeloid-specific Fmpo promoter and flanked by I-SceI integration sites (Sc), in the presence of I-SceI meganuclease. Wounded larvae with mosaic expression of H2AmCherry in tailfin myeloid cells were imaged. The dashed line represents the wound. The inset shows a transgenic cell (Cherry+; PAR) and the endogenous control (Cherry; CTR). Scale bars: 50 µm (left panel); 10 µm (inset). (B) Domains of interaction between members of the mammalian PAR complex. Connecting lines indicate regions of the proteins that interact with one another. PB1, phagocyte oxidase/Bem1 domain; Zn, Zinc finger motif; Kinase, catalytic domain; CRIB, Cdc42/Rac interactive binding motif; PDZ, PSD-95/Dlg/Zona occludens-1 domain; CR1, conserved region 1; aPKCBR, aPKC-binding region. A predicted coiled-coil region is also shown. (C) Schematics of the constructs used to perturb the function of the PAR complex in myeloid cells. Numbers refer to amino acid positions. *K to W mutation at codon 281. NT, N-terminal domain. (D) 2D tracks of individual leukocytes migrating in the tailfin of unwounded fish (left panel) or towards the tailfin wound (right panels). No wound, n = 11; control/PKC-ζ-WT, n = 12; control/PKC-ζ-KW, n = 10; control/PAR-6-NT, n = 13; control/PAR-3-aPKCBR, n = 13). Tracks are from one representative experiment of at least three independent experiments. (E–G) Quantification of 2D (E) speed, (F) path straightness and (G) directional speed ratio of myeloid cells during the wound response. Data are expressed as the mean±s.e.m. of at least three separate experiments (PKC-ζ-WT, n = 27 cells in three larvae; PKC-ζ-KW, n = 27 cells in three larvae; PAR-6-NT, n = 45 cells in four larvae; PAR-3-aPKCBR, n = 64 cells in five larvae); *P<0.05; **P<0.01; ***P<0.001; ns, non-significant (two-tailed unpaired Student's t-test). See also supplementary material Fig. S1; Movies 1, 2.

F-actin shows an oscillatory anteroposterior polarity during wound-directed migration of leukocytes in vivo. (A,B) Quantification of 2D (A) speed and (B) path straightness of memYFP+ or H2B–CFP+ EB3–EGFP+ RFP–Lifeact+ leukocytes migrating to wounds. Data are expressed as the mean±s.e.m. of all analyzed cells (memYFP+, n = 75 cells in three larvae; H2B–CFP+ EB3–EGFP+ RFP–Lifeact+, n = 64 cells in three larvae); ns, non-significant; two-tailed unpaired Student's t-test. (C) Two consecutive cycles (C1 and C2) of front-to-back F-actin waves (lower panels) were visualized in transgenic larvae coexpressing RFP–Lifeact and EB3–EGFP in myeloid cells [TG(FmpoP:: EB3–EGFP/FmpoP::RFP–Lifeact)], following transient expression of H2B–CFP as a nuclear marker (upper panels). Shown are frames from representative movies of myeloid cells migrating in the wounded tailfin. The white arrows indicate the direction of migration. Cell outlines are indicated by the gray line. Scale bar: 10 µm. (D) Schematics of the 2D geometric compartmentalization used to quantify F-actin subcellular distribution. Black dots indicate the cell centroid shown at three consecutive time-points, and the dashed red arrow indicates the direction of migration vector. (E) Timecourse of F-actin front and back fluorescence signal distribution in the cell depicted in C. Cycle C1 and C2 are those visualized in C. (F) Fourier spectra showing the time frequencies that compose the oscillatory signal of F-actin in the back and front regions of the cell shown in C. The peak with the highest amplitude value (main peak) corresponds to the predominant frequency of oscillation. Frequencies with lower amplitude values are also displayed and give a minor contribution to the oscillatory signal. a.u., arbitrary units. (G) Histograms showing the relative contribution of ranges of frequencies to back and front F-actin oscillations. Note that a predominant range of oscillatory frequencies (main range) emerges from ranges of secondary frequencies. Data are expressed as the mean±s.e.m. of all analyzed cells (n = 8 cells in four larvae); statistical analysis was performed between the main range and each of the remaining ranges of frequencies; **P<0.01; ***P<0.001 (two-tailed paired Student's t-test). See also supplementary material Fig. S2; Movie 4.

The MTOC is highly mobile in the perinuclear compartment of leukocytes migrating to wounds in vivo. (A) The microtubules and the nucleus are visualized in a migrating leukocyte generated as described for Fig. 2. Upper panel, fluorescence images; lower panel, green dots represent a digitalized reconstruction of the MTOC position. The white arrows indicate the direction of migration. Cell outlines are shown as gray lines. Scale bar: 10 µm. (B) Schematic of the 2D analysis of MTOC perinuclear positioning during migration. The dashed red arrow indicates the direction of migration defined as in Fig. 2. The yellow arrow originating from the nuclear centroid (white dot) to the MTOC is the MTOC–nucleus vector. The angle α between the MTOC–nucleus vector and the direction-of-migration vector is the MTOC-nucleus angle (orange arc). (C) Rose diagram mapping the MTOC orientation and the respective spatial frequency of events in migrating myeloid cells (230 counts, seven leukocytes in five larvae). Light or dark gray areas correspond to 90° ranges for front or back orientations, respectively. CTR, control. (D) Quantification of MTOC perinuclear mobility in migrating myeloid cells. MTOC mobility for each cell is represented by the standard deviation associated with the MTOC-nucleus angle during the response to wounding. The red line shows the mean (±s.e.m.) for all cells analyzed (seven leukocytes in five larvae). (E) MTOC perinuclear orientation is plotted against cellular speed. For each range of speed, data represents the mean±s.e.m. (226 counts, seven leukocytes in five larvae); R Spearman = 0.008, P (two-tailed) = 0.8932 (non-significant). See also supplementary material Fig. S3A–D; Table S1; Movie 5.

The PAR complex regulates the anteroposterior polarity of F-actin during wound-directed migration of leukocytes in vivo. (A) F-actin distribution is visualized in transgenic larvae with RFP–Lifeact and EB3–EGFP in myeloid cells, following transient coexpression of nuclear H2B–CFP and each of the indicated PAR transgenes. Frames from representative movies of migrating leukocytes in the wounded tailfin are shown. The white arrows indicate the direction of migration. Outlines of cells are shown as a gray line. Scale bars: 10 µm. (B) Timecourse of F-actin front and back fluorescence intensity distribution in the cells shown in A. (C) Fourier spectra of F-actin oscillations in the back region of the cells depicted in A. a.u., arbitrary units. (D) Comparison of histograms of F-actin oscillatory frequencies in the back region of the cell [ranges (s−1) are the same as shown in Fig. 2G]. Data are expressed as the mean±s.e.m. of all analyzed cells (PKC-ζ-WT, eight leukocytes in five larvae; PKC-ζ-KW, eight leukocytes in six larvae; PAR-6-NT, seven leukocytes in four larvae; PAR-3-aPKCBR, six leukocytes in four larvae); *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001; ns, non-significant (two-tailed paired Student's t-test). See also supplementary material Fig. S4; Movie 6.

The PAR complex regulates the MTOC perinuclear positioning during wound-directed migration of leukocytes in vivo. (A) The microtubules and the nucleus are visualized in migrating leukocytes generated as described for Fig. 4. The white arrows indicate the direction of migration. Cell outlines are shown as gray lines. Scale bars: 10 µm. (B) Rose diagrams mapping the orientation of the MTOC and the respective spatial frequency in migrating myeloid cells [PKC-ζ-WT, 265 counts (seven leukocytes in six larvae); PKC-ζ-KW, 347 counts (seven leukocytes in five larvae); PAR-6-NT, 235 counts (seven leukocytes in five larvae); PAR-3-aPKCBR, 307 counts (seven leukocytes in five larvae)]. Gray areas indicate front or back orientation as in Fig. 3. (C) Comparison of MTOC perinuclear mobility in migrating myeloid cells as described in Fig. 3D (PKC-ζ-WT, seven leukocytes in six larvae; PKC-ζ-KW, seven leukocytes in five larvae; PAR-6-NT, seven leukocytes in five larvae; PAR-3-aPKCBR, seven leukocytes in five larvae); **P<0.01; ***P<0.001; ****P<0.0001 (two-tailed unpaired Student's t-test). See also supplementary material Table S1; Movie 7.

Rho-kinase-dependent actomyosin contraction is required for MTOC dynamic positioning in leukocytes migrating to wounds in vivo. (A) F-actin (upper and middle panels), the microtubules and the nucleus (upper and lower panels) are visualized in control (CTR) or Y-27632-treated transgenic larvae generated as described for Fig. 2. Frames from representative movies of migrating myeloid cells in wounded tailfins are shown. The white arrows indicate the direction of migration. The arrowheads point to F-actin accumulation at the trailing edge of the cell. Cell outlines are shown as a gray line. Scale bars: 10 µm. (B) Comparison of histograms of F-actin oscillatory frequencies in the back region of the cell [ranges (s−1) as shown in Fig. 2G]. Data are expressed as the mean±s.e.m. of all analyzed cells (control, eight leukocytes in four larvae; Y-27632, seven leukocytes in three larvae); *P<0.05; **P<0.01; ***P<0.001; ns, non-significant (two-tailed paired Student's t-test). (C) Comparison of MTOC mobility as described for Fig. 3D (control, seven leukocytes in five larvae; Y-27632, six leukocytes in four larvae); **P<0.01 (two-tailed unpaired Student's t-test). (D) Rose diagrams mapping the orientation of the MTOC and its spatial frequency [control, 230 counts (seven leukocytes in five larvae); Y-27632, 524 counts (six leukocytes in four larvae)]. Gray areas indicate front or back orientation as in Fig. 3. (E,F) The strength of the main frequency of F-actin oscillations in the back region (E) is plotted against directional speed ratio. Shown is the normalized amplitude of the main frequency peak in the Fourier spectra (44 leukocytes, R Spearman = 0.5149). MTOC perinuclear mobility (F) is plotted against directional speed ratio (41 leukocytes, R Spearman = 0.5416). ***P (two-tailed)<0.001. See also supplementary material Fig. S3E,F; Table S1; Movies 8, 9.

The catalytic activity of PKC-ζ is essential for its polarized localization in leukocytes migrating in vivo. For A,C and D, Frames from representative movies of migrating myeloid cells in wounded tailfins are shown. The white arrows indicate the direction of migration. Cell outlines are drawn as a gray line. Scale bars: 10 µm. (A) Upper panel, mCherry and GFP–PKC-ζ were visualized in TG(FmpoP::mCherry) larvae transiently expressing GFP–PKC-ζ in myeloid cells. Lower panel, ratiometric GFP–PKC-ζ∶mCherry images were generated using mCherry as a volumetric control. (B) Schematics of the 2D analysis of protein asymmetry during cell migration. Black dots indicate the cell centroid shown at three consecutive time-points, and the dashed black arrow indicates the direction of migration. The asterisk represents the fluorescence center of ratiometric signal, and the dashed red arrow from the cell centroid to the fluorescence center is the asymmetry vector. The angle α between the asymmetry vector and the direction of migration (blue arc) is defined as the asymmetry-migration angle, which approaches 180° for a protein that is located at the back of the cell. (C) Upper panel, mCherry and GFP–PKC-ζ were visualized in TG(FmpoP::mCherry) transgenic larvae transiently expressing GFP–PKC-ζ together with PKC-ζ-WT or PKC-ζ-KW in myeloid cells. Lower panel, ratiometric GFP–PKC-ζ∶mCherry images were created. (D) Upper panel, mCherry and GFP–PKC-ζ were visualized in control or Y-27632-treated transgenic larvae established as described for A. Lower panel, ratiometric images GFP–PKC-ζ∶mCherry were generated. (E,F) Histograms show the polarized distribution of ratiometric GFP–PKC-ζ∶mCherry images in migrating cells assessed using the asymmetry-migration angles. CTR, control. Data are expressed as the mean±s.e.m. of all analyzed cells (control, 15 leukocytes in four larvae; PKC-ζ-WT, eight leukocytes in four larvae; PKC-ζ-KW, ten leukocytes in three larvae; Y-27632, 11 leukocytes in four larvae); *P<0.05; ***P<0.001; ****P<0.0001 (two-tailed unpaired Student's t-test). See also supplementary material Movies 10, 11.

PKC-ζ regulates RhoA activity in leukocytes migrating to wounds in vivo. (A) Frames from representative movies of migrating myeloid cells in wounded tailfins are shown. The white arrows indicate the direction of migration. Cell outlines are shown as white lines. Scale bars: 10 µm. See also supplementary material Movie 12. RhoA activity was visualized in TG(FmpoP::mCherry) larvae transiently expressing cytosolic RhoA-FRET biosensor together with PKC-ζ-WT or PKC-ζ-KW in myeloid cells. Ratiometric images of YFP∶CFP emission for each cell are shown. (B) Average activation level of RhoA (mean emission ratio of YFP∶CFP for the entire cell during migration) in wound-activated leukocytes. Data are expressed as the mean±s.e.m. of all analyzed cells (PKC-ζ-WT, 25 leukocytes in 15 larvae; PKC-ζ-KW, 15 leukocytes in 5 larvae); **P<0.01 (two-tailed unpaired Student's t-test). (C) A model illustrating the mechanism by which the PAR complex controls wound-directed leukocyte migration in vivo. The PAR complex coordinately controls Rho-dependent F-actin dynamics and MTOC perinuclear mobility to support the persistent migration of leukocytes to wounds.

Transient expression of transgenes in myeloid cells in vivo, related to Fig. 1. (A) Sudan black (SB) staining of zebrafish MPO::GFP (left panel) or medaka FmpoP::memYFP larvae (right panel), respectively at 3 or 11dpf. White arrows indicate GFP+ SB+ (left panel) or memYFP+ SB+ (right panel) leukocytes. Scale bars = 20 μm. (B) Outline of the experimental strategy used to co-express two transgenes in medaka myeloid cells. Wild-type Cab embryos were injected at the one-cell stage with a 1:1 mixture of DNA coding for H2AmCherry as a nuclear reporter and membrane-tethered YFP, driven by the myeloid cell-specific Fmpo promoter and flanked by I-SceI integration sites (Sc), in the presence of I-SceI meganuclease. (C) Top, illustration of the location in the fish larval tailfin imaged for the analysis of coexpression of two transgenes in live myeloid cells. Bottom, representative images of larvae containing single (YFP+; Cherry+) and double (YFP+ Cherry+14 ) positive cells within the region highlighted. Scale bar = 50 μm. (D) Quantification of the degree of coexpression of two transgenes (YFP+ Cherry+ - gray bar) in live myeloid cells. Data are expressed as means ± s.e.m. of three independent experiments (173 leukocytes from 3 separate larvae). (E) Top, schematics of the reporter construct used to express mCherry and memYFP linked by a self-cleavable viral P2A peptide in medaka myeloid cells. Bottom, independent expression of mCherry and memYFP protein products in medaka leukocytes from the mCherry-P2A-memYFP transgene. Scale bar = 20 μm. (F) Schematics of the constructs used to express mCherry and PKC-ζ variants linked by a self-cleavable viral P2A peptide in medaka myeloid cells. (G and H) Quantitation of 2D (G) path straightness and (H) directional speed ratio (Vy/Vx) of mCherry- P2A+ PKC-ζ+ myeloid cells during the wound response. Data are expressed as means ± s.e.m. of at least two separate experiments (PKC-ζ-WT: n = 30 cells in 2 larvae, PKC-ζ-KW: n = 53 cells in 4 larvae; * p < 0.05, two-tailed unpaired Student’s t-test).

F-actin anteroposterior polarity in live myeloid cells correlates with cell speed, related to Fig. 2. (A) Top panel, one cycle of front-rear F-actin waves is visualized in wounded TG(FmpoP::memYFP) larvae co-expressing transiently RFP-Lifeact and PKC-ζ-WT in myeloid cells. Shown are frames taken at the indicated times from representative movies of migrating leukocytes in the wounded tailfin. The white arrows indicate direction of migration. Scale bar = 10 μm. Bottom panel, time profile of cell’s speed. Black dots correspond to the indicated time points. (B) Fluorescence signal distribution of F-actin in “back” and “front” regions of the cell at each time point ti is plotted against cellular speed (determined using the positions at ti-1 and ti+1 as described in the Materials and Methods). For each range of speed, data represents means ± s.e.m. (521 counts, 16 leukocytes from 3 larvae; Back: Spearman r = 0.61, Front: Spearman r = - 0.57, **** p (two-tailed) < 0.0001). (C) Fluorescence signal distribution in (B) is normalized to memYFP intensity levels and plotted against cellular speed. For each range of speed, data represents means ± s.e.m. (Back: Spearman r = 0.69, Front: Spearman r = -0.67, **** p (two41 tailed) < 0.0001).

Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image. Full text @ J. Cell Sci.