Kölsch, A., Windoffer, R., Würflinger, T., Aach, T. and Leube, R. E.
The keratin filament cycle of assembly and disassembly.
Journal of Cell Science. 123:2266-2272.
Continuous
and regulated remodelling of the cytoskeleton is crucial for many basic
cell functions. In contrast to actin filaments and microtubules it is
not understood how this is accomplished for the third major
cytoskeletal filament system consisting of intermediate filament
polypeptides. Using time-lapse fluorescence microscopy of living
interphase cells, in combination with photobleaching, photoactivation
and quantitative fluorescence measurements, we observe that epithelial
keratin intermediate filaments constantly release non-filamentous
subunits, which are reutilized in the cell periphery for filament
assembly. This cycle is independent of protein biosynthesis. The
different stages of the cycle occur in defined cellular subdomains:
assembly takes place in the cell periphery, newly formed filaments are
constantly transported toward the perinuclear region while disassembly
occurs giving rise to diffusible subunits for another round of
peripheral assembly. Remaining juxtanuclear filaments stabilize and
encage the nucleus. Our data suggest that the keratin filament cycle of
assembly and disassembly is a major mechanism of intermediate filament
network plasticity allowing rapid adaptation to specific requirements,
notably in migrating cells.
click animated previews to open full size
 |  |  | Movie 1
| Movie 2
| Movie 3
|  |  |  | Movie 4
| Movie 5
| Movie 6
|
Movie 7
|
Movie 8
|
|
Movie 1. KF network reorganization after
scratch-wounding (see corresponding Fig. 1A).
Time-lapse recording (three-minute intervals; display rate 30 frames/s) of HK18-YFP
fluorescence in a confluent monolayer of EK18-1 cells after scratch-wounding.
Movie 2. KFP appearance in
the leading edge of migrating cells (see corresponding Fig. 1B-D).
Tableau of time-lapse recordings of HK18-YFP fluorescence (left) and
corresponding phase contrast (right) of a migrating EK18-1 cell displaying multiple
emerging KFPs in the proceeding lamellipodium. The bottom panel shows high
power views of part of the lamellipodium. In this instance, cellular movement
was compensated for with an image intensity-based method. The images were
acquired every 30 s and are displayed at 30 frames/s.
Movie 3. Persistence of KF
network formation in the presence of the protein biosynthesis inhibitor
cycloheximide (see corresponding Fig. 2A-B, E)
Time-lapse fluorescence microscopy of a PK18-5 cell treated with the
protein biosynthesis inhibitor cycloheximide (17 µM). Images were acquired
every 20 s and are displayed at 30 frames/s. The inhibitor was added after
recording of 50 pictures. Note the continuous formation of new precursors.
Movie 4. Persistence
of KF network formation in the presence
of the protein biosynthesis inhibitor puromycin (see corresponding Fig. 2C-E’).
Time-lapse fluorescence recording of a PK18-5 cell treated with the
protein biosynthesis inhibitor puromycin (1 µg/ml). Images were taken every 20
s and are displayed at 30 frames/s. The inhibitor was added after recording of
50 pictures. Note the continuous formation of new KFPs in the cell periphery.
Movie 5. Detection
of continuous inward-directed KF network motility (see corresponding Fig. 3).
Time-lapse fluorescence recording of HK18-YFP in a section of a PK18-5
cell (projected images of 22 focal planes). Recording intervals were 60 s
(display rate 25 frames/s). The frames are aligned to the first frame to
compensate for cell movement. Note the continuous inward movement of KFs within
the network.
Movie 6. Detection
of inward-directed KF network motility and loss of KFs by ROI-tracking (see
corresponding Fig. 4A-F).
Time-lapse fluorescence recording of HK18-YFP in a segment of a PK18-5
cell (periphery at left, nucleus at right). Projection views of 11 focal planes
are shown for each time point (recording intervals 60 s; 30 frames/s). ROIs
demarcated in the lower panel were defined manually and the respective margins
are shown by coloured lines. Note the continuous inward-directed flow and loss
of fluorescent filaments during inward translocation of each ROI.
Movie 7 Detection
of inward-directed KF network motility and loss of KFs by ROI-tracking (see
corresponding Fig. 4M-M').
Time-lapse fluorescence recording of a HK18-YFP-producing PK18-5 cell
after bleaching of three centripetal segments. Shown are projections of 25
focal planes, recording intervals were 2 min (display rate 50 frames/s). Note
the inward movement of unbleached KF bundles and continuous loss of
fluorescence without fragmentation, which are best seen at higher magnification
at right.
Movie 8. Detection of a
continuous KF network turnover cycle by FRAP (see corresponding Fig. 5B)
Time-lapse fluorescence recordings were prepared from two PK18-5 cells producing
HK18-YFP after bleaching of half of one of the cells. The movie presents a series
of projected images (10 focal planes) after registration at a display rate 25
frames/s. Note the peripheral recovery of fluorescence and the continuous
centripetal motility of the keratin system. Image stacks were recorded every 5
min.
| 
