July 1, 2003;
Actomyosin transports microtubules and microtubules control actomyosin recruitment during Xenopus oocyte wound healing.
BACKGROUND: Interactions between microtubules and actin filaments (F-actin) are critical for cellular motility processes ranging from directed cell locomotion to cytokinesis. However, the cellular bases of these interactions remain poorly understood. We have analyzed the role of microtubules in generation of a contractile array comprised of F-actin and myosin-2 that forms around wounds made in Xenopus oocytes.
RESULTS: After wounding, microtubules are transported to the wound edge in association with F-actin that is itself recruited to wound borders via actomyosin-powered cortical flow. This transport generates sufficient force to buckle and break microtubules at the wound edge. Transport is complemented by local microtubule
assembly around wound borders. The region of microtubule
breakage and assembly coincides with a zone of actin assembly, and perturbation of the microtubule
cytoskeleton disrupts this zone as well as local recruitment of the Arp2/3 complex and myosin-2.
CONCLUSIONS: The results reveal transport of microtubules in association with F-actin that is pulled to wound borders via actomyosin-based contraction. Microtubules, in turn, focus zones of actin assembly and myosin-2 recruitment at the wound border. Thus, wounding triggers the formation of a spatially coordinated feedback loop in which transport and assembly of microtubules maintains actin and myosin-2 in close proximity to the closing contractile array. These results are surprisingly reminiscent of recent findings in locomoting cells, suggesting that similar feedback interactions may be generally employed in a variety of fundamental cell motility processes.
[+] show captions
Figure 1. Microtubule Organization around Wound Borders
All times are in min:sec.
(A) Confocal image of a laser-wounded oocyte. Microtubules (red), F-actin (green), and myosin-2 (blue) show complementary distribution
around the wound.
(B) Individual channels (MT, microtubule; FA, F-actin; and M2, myosin-2). F-actin and myosin-2 concentrate inside the area of highest
(C) Image from a movie of an OG-Tax-injected oocyte. Microtubules are organized into a radial array around the wound (W). See Movie 1
(Movies are available in the Supplemental Data available with this article on line).
(D) Image from a movie of a wound in an OG-Tax-injected oocyte. Microtubules are perpendicular to the wound border and translocate to
the wound. See Movie 2.
(D) Time course of microtubule movement toward the wound (indicated by an arrow). Asterisks mark the leading ends of individual microtubules,
and arrowheads mark the trailing ends.
(E) Kymograph of OG-Tax-labeled microtubule motility toward the wound edge. Horizontal arrow is 45 s; vertical arrow is 5 m. Microtubules
accelerate as they approach the wound border (w), then slow at its immediate edge.
(F) Images from a movie of an OG-Tax-injected oocyte. An arrowhead shows a sharply bent microtubule moving toward the wound.
(G) Images from a movie of an OG-Tax-injected oocyte. Microtubules buckle (arrowhead) at the wound border. The arrow indicates direction
Figure 2. Microtubule Transport is F-Actin Dependent
(A) Plot of instantaneous rates of microtubules before and after treatment with latrunculin B (20 M). Microtubule translocation to the wound
border was arrested after latrunculin treatment. For each point, the rates of movement of a minimum of 10 microtubules were quantified; bars
indicate standard error of the mean (SEM).
(B) Images from a movie of an OG-Tax-injected oocyte. Arrowheads show a microtubule translocating to the wound (W), until latrunclin
application (asterisk). See Movie 3.
(C) Double-labeled image from a movie of an oocyte injected with OG-Tax (green) and Alexa 568-phalloidin (red). See Movie 5.
(C) Microtubules move to the wound (arrow) in association with F-actin and buckle upon reaching the wound edge (arrowheads). See Movie 6.
(D) Dual-label images showing end-end association between F-actin (FA, red) and microtubules (MT, green); the overlap of the two is indicated
in two frames by arrowheads. F-actin is localized to the tip of microtubules moving to the wound. The horizontal arrow is 30 s; the vertical
arrow is 15 m. See Movie 7.
(E) Dual-label image from a movie showing microtubules (green) aligned with F-actin (red; alignment indicated by arrowheads).
Figure 3. Cotransport of F-Actin and Microtubules (A) Dual-label images showing examples of bent microtubules.
indicate direction of movement. Microtubules (green) bend
away from the direction ofmovement where they are not associated
with F-actin (red). -
(B) Images from a movie showing F-actin (red) and microtubules
(green). F-actin appears to link the ends of several microtubules
and drag them through the cortex. See Movie 8.
(C) Dual-label and split-channel images showing buckling and
breakage of a microtubule that is associated with F-actin at both
leading and trailing edges. F-actin at the trailing edge accelerates
between the 00:45 and 01:00 time points, whereas the F-actin at
the leading edge does not (white lines). Microtubule buckling is
observed in precisely this time window. Later (at 01:30), the microtu-
bule breaks (arrow). See Movie 9.
(D) Quantification of specific microtubule-F-actin overlap (see Supplemental Data for details). The ratio of yellow pixels in unrotated:
rotated images is significantly higher in the region where microtu-
bule-F-actin cotransport occurs (10–30 m) than at the wound edge
(0–10 m) or in areas more than 30 m away from wound. Results
are mean SEM from four independent experiments; an asterisk
(E) Double-labeled kymograph. F-actin (red) and microtubules
(green) show same patterns of acceleration and deceleration. W
indicates the wound; the horizontal arrow is 90 s; the vertical arrow
is 10 m.
Figure 4. Local Actin Assembly Coincides with Microtubule Breakage and Assembly
(A) Image from a movie of an oocyte injected with OG-taxol and Alexa 568 G-actin. Microtubules (green) move to the zone of actin polymerization
(red; indicated by a double-headed arrow). Microtubule disappearance into the zone is correlated with bending and buckling. See Movie 10.
(B) Dual-label image from a movie showing microtubules translocating toward the wound (an arrow indicates the direction of the wound).
Arrowheads show individual microtubules buckling.
(C) Plot of microtubule buckling and relative actin intensity versus distance from the wound border. Microtubule buckling is most common
near the actin assembly zone.
(D) Image showing e-GFP-Xtub-labeled microtubules (arrowheads) near the wound.
(E) Images from a movie of an oocyte injected with e-GFP-Xtub. Microtubules polymerize perpendicular to (left panels) and away from the
wound (right panels). Arrows indicate the direction of flow; arrowheads show the start and stop of polymerization; frames are shown at 4 s
intervals. See Movies 11 and 12.
(F) Graph showing that microtubule polymerization rates are inversely correlated with distance from the wound.
(G) Plot of microtubule polymerization events and relative actin intensity versus distance from the wound border. Microtubule assembly events
are most common near the zone of actin assembly.
(H) Image from a movie of an oocyte wounded after injection with e-GFP-X tub and treated with latrunculin B (20 M). Microtubules accumulate
around the wound border.
Figure 5. Microtubule Perturbation Defocuses
the Actin Assembly Zone.
(A) Images from movies of oocytes injected
with OG-actin and wounded after no treatment
(control), nocodazole treatment (Nocodazole;
5 m; 1 hr), or taxol treatment (Taxol;
20 m; 1 hr). In controls, the zone of F-actin
polymerization remains tightly focused around
wounds; in nocodazole-treated oocytes it becomes
uneven and broadens; in taxol-treated
samples it spreads and then splits into two
rings (arrows). See Movies 13, 14, and 15.
(B) Confocal fluorescence images showing
distribution of F-actin (red) and microtubules
(green) in a taxol-treated oocyte subjected to
wounding. The microtubule and actin arrays
have split into concentric rings.
Figure 6. Microtubule Perturbation Disrupts
Normal Patterns of Arp2/3 Complex and Myosin-
2 Localization and Uncouples the Zone
of Actin Polymerization from the Contractile
(A) Confocal images showing distribution of
p16 around wounds in control and taxoltreated
oocytes. The region of p16 accumulation
is broader in the taxol sample. Quantification
shows that the width of the region
enriched in p16 increases in the presence of
taxol. Results are mean SEM from three
independent experiments; an asterisk indicates
(B) Confocal images showing examples of
taxol-induced spreading (Taxol-Spread) or
splitting (Taxol-Split) of the region of myosin-
2 enrichment around wounds and, for
comparison, controls. Quantification shows
that the width of the region enriched in actin
and myosin-2 around wounds increased in
the presence of taxol. Results are mean
SEM from three independent experiments; an
asterisk indicates p 0.01.
(C) Double-labeled images from 4D movies
of oocytes injected with Alexa 568-G-actin
and OG-phalloidin, treated with taxol (taxol)
or nothing (control), and wounded. Lower
panels show images taken from the regions
indicated by the arrows in the upper image.
In control, the zone of assembly (red) encompasses
the contractile ring that contains stable
F-actin (green). However, in the taxol sample,
dynamic actin moves to the outer ring,
whereas stable F-actin remains concentrated
at the wound border. See Movies 16 and 17.
Figure 7. Schematic Diagram of Microtubule-Actomyosin Interac-tions during Oocyte Wound Healing