18
microtubule tips led to the development of fluorescence speckled microscopy with significant
444
implications for the in vivo elucidation of microtubule dynamics (e.g., Vallotton et al., 2003).
445
Among others, it provided a tool to discriminate treadmilling-based and motor-driven
446
microtubule translocation (Waterman-Storer and Salmon, 1997; Salmon and Waterman, 2011).
447
In Arabidopsis thaliana hypocotyl cells labeled with GFP-TUA6 microtubule marker, cortical
448
microtubules were discontinuously labeled leading to their speckled appearance. Visual
449
inspection as well as quantitative fluorescence intensity profiling showed the alternation of bright
450
fluorescent islands and dark areas, separated at variable distances. Since such labeled
451
microtubules exhibited well resolved organization patterns and end-wise dynamics, it can be
452
assumed that the discontinuous microtubule labeling by GFP-TUA6 is owing to the stochastic
453
incorporation of GFP-TUA6-containing tubulin dimers and non-labeled dimers at the
454
microtubule ends.
455
Structure of cortical microtubule bundles and other subdiffraction details of the cortical
456
microtubule array
457
Microtubule bundle formation in interphase cortical arrays is mediated redundantly by members
458
of the MAP65 family including MAP65-1, MAP65-2 (e.g., Lucas and Shaw, 2012) and possibly
459
MAP65-5 (van Damme et al., 2004). Interestingly, MAP65-1 and MAP65-2 (along with the
460
cytokinesis specific MAP65-3) were previously identified as targets for MAPK phosphorylation
461
(Sasabe et al., 2011) which serves as a negative regulator of MAP65 affinity for the microtubule
462
surface (Smertenko et al., 2006). Importantly, MAP65-1 levels are significantly upregulated in
463
the
mpk4 mutant while the levels of phosphorylated MAP65-1 are
considerably lower compared
464
to the wild type (Beck et al., 2010). Theoretically, the above observations may explain the
465
mechanism behind extensive cortical microtubule bundling in the
mpk4 mutant.
466
The arrangement of individual microtubules within microtubule bundles was visually and
467
quantitatively very clearly discriminated, suggesting that high performance SIM may be used for
468
mapping the associations of MAPs with microtubules and perhaps resolve intricate microtubular
469
super-structures such as the preprophase microtubule band.
470
www.plantphysiol.org
on July 21, 2018 - Published by
Downloaded from
Copyright © 2014 American Society of Plant Biologists. All rights reserved.
19
Moreover, in comparison to WF and CLSM, SIM provided subdiffraction details of cortical
471
microtubule organization not resolvable with the other techniques. With high performance SIM it
472
was possible to capture the onset of branched microtubule formation and release. The
γ-tubulin-
473
mediated branch initiation (Chan et al., 2009; Nakamura and Hashimoto, 2009; Nakamura et al.,
474
2010) and the katanin-mediated branch release (Nakamura et al., 2010) are key mechanisms
475
supplementing the cortical array with nascent microtubules, thus SIM may provide a sub-
476
diffraction tool that will allow the spatiotemporal resolution of this successive nucleation and
477
severing mechanism.
478
Subdiffraction microtubule dynamics
479
Time-lapsed imaging
by SIM was traded for speed, thus the resolution of individual
480
microtubules was slightly inferior (135 and 133 nm for GFP-MBD and GFP-TUA6,
481
respectively) but still significantly outperformed all other imaging approaches (WF, CLSM,
482
TIRF and SD). Unfortunately, literature on dynamic plant cell studies is largely devoid of
483
resolution data. Thus there is no comparison measure other than the data included hereby.
484
However, the FWHM values for WF, CLSM, TIRF and SD are comparable to the resolutions
485
reported elsewhere (e.g., Zucker et al., 1999; Wang et al., 2005; Salmon and Waterman, 2011),
486
which optimally range within 250-270 nm. If this is also true for studies published previously in
487
plants then the difference between SIM and any other microscopic techniques is ca. 120 nm,
488
corresponding to roughly 195 tubulin subunits ([120 nm
× 13 protofilaments]/8 nm tubulin dimer
489
size) in terms of microtubule length.
490
Microtubule dynamics at the faster plus end or the slower minus end evolve with the linear
491
addition or removal of tubulin dimers (Desai and Mitchison, 1997). Since the addition of tubulin
492
dimers during growth or the removal during shrinkage are roughly linear function of time the
493
respective rates are not expected to be influenced by the resolution of the microscopy platform
494
used. Indeed this was the case of SIM time-lapsed recordings of microtubule growth and
495
shrinkage rates, which were within previously published ranges for both molecular markers
496
(GFP-MBD; e.g. van Damme et al., 2004; Vos et al., 2004; and GFP-TUA6; e.g. Dhonukshe and
497
Gadella 2003; Shaw et al., 2003). Similarly the elongation and shrinkage rates measured for
498
www.plantphysiol.org
on July 21, 2018 - Published by
Downloaded from
Copyright © 2014 American Society of Plant Biologists. All rights reserved.