First published online 17 June 2008
doi: 10.1242/jcs.024018
Journal of Cell Science 121, 2293-2300 (2008)
Published by The Company of Biologists 2008
A novel small-molecule inhibitor reveals a possible role of kinesin-5 in anastral spindle-pole assembly
Aaron C. Groen1,2,*,
Daniel Needleman1,2,
Clifford Brangwynne2,
Christain Gradinaru2,
Brandon Fowler3,
Ralph Mazitschek3 and
Timothy J. Mitchison1,2
1 Harvard Medical School, Department of Systems Biology, 200 Longwood Avenue, WAB536, Boston, MA 02115, USA
2 Marine Biological Laboratory, 7 MBL Street, Woods Hole, MA 02543, USA
3 Broad Institute of MIT and Harvard, 7 Cambridge Center, Cambridge, MA 02142, USA
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 1. FCPT induced tight binding of Eg5 to microtubules. (A) The structures of kinesin-5 inhibitors: 2-[1-(4-fluorophenyl)cyclopropyl]-4-(pyridin-4-yl)thiazole (FCPT), GSK246053, monastrol and S-Trityl-L-cysteine (STLC). (B) FCPT, similar to AMP-PNP, induced the binding of the recombinant kinesin-5 motor domain to microtubules (10 µM) in the presence of 1 mM ATP. The best-fit line of the dose-response curve identified the EC50 of AMP-PNP as 116 µM (±17 µM) and FCPT as 65 µM (±10 µM). Note that the concentration of FCPT required to bind 100% of the kinesin-5 motor domain to microtubules (1-2 mM) was above the solubility concentration for FCPT (approximately 500-600 µM). Error bars, s.e.m. (C) Coomassie-stained gels of lysates isolated from supernatants (S) and pellets (P) of either a DMSO (control)-, FCPT- or AMP-PNP-treated kinesin-5 motor domain/microtubule mix (in the presence of 1 mM ATP), showing kinesin-5 and tubulin. (D) FCPT enhanced the binding of kinesin-5 to microtubules. The best-fit line identified the apparent Kd of AMP-PNP (grey line) as 6.3 µM (±1.96 µM) and FCPT (black line) as 1.7 µM (±0.437 µM) – both in the presence of 1 mM ATP, 8 µM kinesin-5 and either 10 µM FCPT or 10 µM AMP-PNP. The fraction of kinesin-5 co-sedimenting with microtubules is scaled to reflect maximum binding as 1.0. Error bars, s.e.m. (E) Immunoblots show that FCPT (compared with DMSO as a control) enhanced the binding of kinesin-5 onto pelleted taxol-polymerized microtubules in clarified Xenopus egg extracts (approximately threefold); binding of either XCTK2 or MCAK to microtubules was unaffected. Tubulin is a loading control. (F) FCPT inhibited kinesin-5 motility on Xenopus egg-extract spindles. The kymographs (measured from the red broken line) of image sequences taken over 3.5 minutes (with one image every 3 seconds) show that X-rhodamine-labeled kinesin-5 is not motile when FCPT (right; 200 µM) was present (shown as straight lines). In the control (left; same time frame as the FCPT condition), kinesin-5 was dynamic, with speckles appearing and disappearing over time. The spindle is outlined with a white line. Scale bars: 5 µm.
|
|
View larger version (44K):
[in this window]
[in a new window]
|
Fig. 2. FCPT decreased spindle-pole microtubule density. (A) FCPT-treated (200 µM) Xenopus egg-extract spindles did not collapse, whereas STLC (non-competitive kinesin-5 inhibitor)-treated spindles collapsed within 20 minutes. Fixed images of different spindles at time points up to 20 minutes showing tubulin (red), NuMA (green) and DNA (blue) are shown. The spindle-pole marker NuMA remained localized to the spindle pole in the presence of FCPT. (B) FCPT-treated Xenopus egg-extract spindles maintained a constant length of about 35 µm, whereas STLC-treated spindles collapsed to approximately 0 µm over 20 minutes. Error bars, s.e.m.; n=6 for each time point. (C) The ratio of the fluorescence of the spindle-poles:mid-spindle decreased from approximately 0.9 to approximately 0.6 over 40 minutes. Error bars, s.e.m.; n=6 for each time point. (D) Spindles assembled in the presence of FCPT are elongated and asymmetric, compared with controls, which are bipolar. Fixed images of spindles showing tubulin (red), NuMA (green) and DNA (blue) are shown. NuMA was diffusely localized on FCPT-treated spindles. Scale bars: 10 µm.
|
|
View larger version (40K):
[in this window]
[in a new window]
|
Fig. 4. FCPT inhibited poleward microtubule movement and spindle-pole polymerization. (A) FCPT inhibition of poleward microtubule movement was titratable. Localizations of speckle level Alexa-Fluor-647-conjugated tubulin on Xenopus egg-extract spindles with 0, 40 and 200 µM FCPT. The kymographs (measured from colored broken lines) of image sequences obtained over the course of approximately 3.5 minutes (with one image every 3 seconds) show that poleward movement is only inhibited upon addition of 200 µM FCPT. Intermediate concentrations of FCPT had intermediate microtubule poleward rates (see B). Untreated control spindles had a microtubule poleward movement of approximately 1.9 ± 0.2 µm/minute (± s.e.m.). The spindle is outlined with a white line. Scale bar: 5 µm. (B) The reduction of spindle-pole density (as measured by the ratio of spindle-pole:mid-spindle fluorescence) only occurred at FCPT concentrations sufficient to completely inhibit poleward microtubule movement (200 µM). Error bars, s.e.m.; n=3 for each point. (C) Alexa-Fluor-488-EB1 localized on control and FCPT-treated spindles. FCPT removed spindle-pole localization of EB1 compared with controls (imaged approximately 4-5 minutes after FCPT addition; The spindle is outlined with a white line). See supplementary material Movies 1, 2. Scale bar: 10 µm.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2008
-tubulin from the spindle pole of bipolar spindles. (A-D) Fixed images of spindles treated with (and without) FCPT for approximately 20 minutes: tubulin (red), TPX2 (A; green), 
