Eileen Fong, Jin-Hong Kim,   Alexander Lin    

Microtubule Polymerization
Mitosis: In Vivo and In Vitro Microscopy  -  Final Project






   

.: Introduction

Microtubules (MT) play a significant role in almost all stages of mitosis.  Their functions include movement of organelles, transport of vesicles, protein attachment, as well as cytoskeleton support.   MT is composed of an a,β-tubulin heterodimer which are similar and approximately 55 kDa MW each.  When the tubulin are polymerized, they are joined from end to end to form a profilament.  13 profilaments assemble to form a helical arrangement that forms the MT as shown on the right.  It is important to note that tubulin can be polymerized at both ends in vitro at different rates where the faster rate is the plus-end and the slower rate is the minus-end.  These dynamic rates have been of particular interest to biophysicists as they appear to be fundamental to explain cellular organization.
Image courtesy of Cytoskeleton, Inc.  

.: Aims

The purpose of our study was to determine the dynamic instability of microtubule growth using fluorescent microtubules and fluorescent microscopy.  Using methods developed by Mitchison and Kirschner, growth will be arrested at multiple time points and MT lengths will be measured to determine the dynamics of MT polymerization.  Furthermore, the kinetics of MT polymerization will be determined by quantifying the relationship between MT length and tubulin concentration.

[top]

.: Methods

Tubulin Preparation:

  1. Using Fluorecent Microtubules Biochem Kit (Cat #BK007R), concentrations and measurements were pre-determined due to the fragility of the tubulin protein which begins to degrade when left at room temperature.  Therefore, all mixtures were stored on ice when not in use.

    GPEM: General tubulin buffer (GTB) and GTP at 1:100 ration
    Non-labeled tubulin: 50 ul @ 250 ug + 40ul of GPEM + 10 ul cushion buffer = 5 mg/mL
    Labeled tubulin: 5 ul @ 20 ug + 4 ul of GPEM + 1 ul of cushion buffer = 4mg/mL
     

    To obtain at 30% ratio of non-labeled to labeled tubulin:
    3.3 ul of unlabeled tubulin = 16.5 ug
    1.7 ul of labeled tubulin =      6.8 ug  --> 4.66 ug/ul
    5 ul                                    29.3 ug

  2. Put 500 ul of purified H2O into eppendorf and place on ice.
  3. Put 100 ul of cold H2O into GTP stock to reconstitute it and aliquot into 10 ul individual eppendorfs and store at -70 degC.
  4. Take 99ul of GTB and using 1 ul of one of the GTP aliquots to make 100 uL of GPEM
  5. Add 4ul of GPEM and 1ul of cushion buffer to 5 ul of labeled tubulin.
  6. Add 40ul of GPEM and 10ul of cushion buffer to 50 ul of unlabeled tubulin.
  7. Place 15 eppendorfs into ice and label as "30% labeled tubulin"
  8. From (6), take 3.3 ul of diluted unlabeled tubulin and add to one eppendorf from (7)
  9. From (5), take 1.7 ul of diluted labeled tubulin and add to eppendorf from (8)
  10. Repeat for all 15 aliquots and place on ice.
  11. In order to store for future use, snap freeze 30% tubulin mixture by dipping into LN2 with tweezers for approximately 7 seconds and store in -70 degC freezer.
  12. Dilute taxol into 100 ul of DMSO and aliquot into 12 50uL 2.2mM taxol and store in freezer.
  13. Prepare antifade by adding 1ml of 50% glycerol and purified H2O to reconstitute antifade if necessary, then add 50uL of 10x antifade (AF) to 450ul of GTB and aliquot, store at -70 degC.

[top]

EMCL Concentration Determination

  1. Put 500 uL of GTB into an eppendorf and heated it in waterbath at 35 degC for 15 minutes.
  2. Added 5ul of 2mM Taxol to GTB.
  3. Placed one aliquot of 30% labeled tubulin (46.6 mM concentration) into waterbath at 35 degC for 10 minutes.
  4. At 10 min, withdraw eppendorf from water bath and add 100 ul of GTB+Taxol mixture to arrest polymerization (both due to Taxol and due to dilution).
  5. Take 1uL of (4) and add 10 uL of antifade.
  6. Take 7uL of (5) and place on a slide, place coverslip, and seal edges with melted wax.
  7. Add a drop of microscope oil and view at 100x on the Zeiss Axio microscope using filter 5 (TexRed) and with fluorescence on.  Be careful not to keep fluorescence on for long as it will cause depolymerization of the microtubules.
  8. Repeat steps 5-7 except with 1uL, 2uL, 3uL, 5uL of EMCL and view to determine optimal EMCL concentration.  2uL EMCL seemed to work best.

[top]

Microtubule Polymerization and Visualization

  1. The following concentrations of 30% labeled MT were prepared:

    46.6 uM = 0.5 uL of MT, No GTB added (x5)
    20.0 uM = 2 uL of MT, 3 uL of GTB (x5)
    10.0 uM = 1 uL of MT, 4 uL of GTB (x5)
    5.0 uM   = 0.5 uL of MT, 4.61 uL of GTB (x5)
    2.5 uM   = 0.5 uL of MT, 8.82 ul of GTB (x3), aliquot to 5 uL each
    1 uM      = 0.5 uL of MT, 22.8 uL of GTB (x1), aliquot to 5 uL each
     

  2. Place each concentration in water bath for the following times: 5, 10, 15, 20, and 30 minutes.  In the 46.6uM solution, times were 1, 2, 5, 10, 20, and 30 minutes.
  3. When time has been reached, add 100 ul of GTB+Taxol mixture to arrest polymerization.
  4. Take 1uL of (3) and add 10 uL of antifade and 2uL of EMCL.
  5. Take 7uL of (4) and place on a slide, place coverslip, and seal edges with melted wax.
  6. Add a drop of microscope oil and view at 100x on the Zeiss Axio microscope using filter 5 (TexRed) and with fluorescence on.  Be careful not to keep fluorescence on for long as it will cause depolymerization of the microtubules.
  7. Acquire  fluorescent images with AxioCam and using AxioVision v4.4 to capture images.
  8. Multiple locations throughout the slide were selected for a total of five images per concentration per time point.
  9. 150ms exposure time was used for all images.
  10. Save images as 16-bit images and store on snowdome.caltech.edu/AEK.

[top]

Data Analysis

  1. MT images all time points from 20uM and 46.6uM concentrations were analyzed.
  2. Manual measurement of MT thickness was made to ensure that all MT are the same width.
  3. Generated code in Matlab to measure MT area for each concentration and at each time point.  Assuming that (2) is true for all MT, then area is proportional to length.


    Original Image before Matlab processing                                                                 Segmented image after Matlab image processing

     

  4. Bins were set at lengths of 50 and analysis was restricted to 100<area<3000 pixels to eliminate background noise and out-of-focus fluorescent MT.
  5. Lengths were then plotted versus time.
  6. Concentration was plotted vs. lengths.

[top]

.: Results

Microtubule Images and Length Measurements at 46.6 uM

   
   Image of 47uM concentration taken at 1 minute                           Histogram of MT length at 1 minute

  
   Image of 47uM concentration taken at 2 minutes                           Histogram of MT length at 2 minutes

 
   Image of 47uM concentration taken at 5 minutes                           Histogram of MT length at 5 minutes

 
   Image of 47uM concentration taken at 10 minutes                           Histogram of MT length at10 minutes

 
   Image of 47uM concentration taken at 20 minutes                           Histogram of MT length at 20 minutes

 
   Image of 47uM concentration taken at 30 minutes                           Histogram of MT length at 30 minutes


[top]


Mean MT length vs. Polymerization Time for 47uM

MT length appeared to reach steady state after 5 minutes at this concentration based on the histograms and visual inspection of the images.  Mean length however does not reflect steady-state as demonstrates in the plot versus time.  This is likely due to uneven sampling of the population of microtubules in the images. One source of error was that at high concentration there were a significant number of MT in the field of view and therefore significant time was spent focusing the images.  The longer the fluorescence is on, the more MT are degraded.  Furthermore, at high concentration there is a shearing effect which may also be reflected in the lack of steady-state.

[top]


MT Lengths over time at 20uM


MT Lengths at 20 uM


MT length over time at 20uM


At 20uM steady state is achieved after 10 minutes.  This demonstrates the dynamic instability of MT growth.

[top]

Growth Rate at Different Concentrations:


These figures clearly demonstrate that as tubulin concentration increases, the growth rate increases linearly.  The line fit was a least squares fit.  This reflects a simple first-order rate equation from which the rates of microtuble assembly can be calculated.  Unfortunately, we did not measure depolymerization rates which would need to be taken into consideration for accurate kinetic calculations.

Images were acquired (but now shown) of concentrations at 10uM and lower because there was little to no microtubules present in the slide.  We believe that this error arises from the low volumes of labeled tubulin that we were pipetting into the polymerization solution.  It is likely that the tubulin stuck to the sides of the pipette tip and were not successfully transferred to the ependorf tubes.  At higher concentrations, the small loss of tubulin would not have a siginficant effect but at low concentrations, the loss can significant, causing the solution to fall under the critical concentration necessary for polymerization.

.: Conclusion

We have successfully replicated the Mitchison and Kirschner study of the dynamic instability of microtubule growth.  The data reflects steady-state dynamics and the kinetics that were obtained with electron microscopy in the earlier work.

[top]

 




Web site contents © Copyright AEK 2006, All rights reserved.
Website templates