TUBA1A tubulinopathy mutants disrupt neuron morphogenesis and override XMAP215/Stu2 regulation of microtubule dynamics

Heterozygous, missense mutations in α- or β-tubulin genes are associated with a wide range of human brain malformations, known as tubulinopathies. We seek to understand whether a mutation’s impact at the molecular and cellular levels scale with the severity of brain malformation. Here, we focus on two mutations at the valine 409 residue of TUBA1A, V409I, and V409A, identified in patients with pachygyria or lissencephaly, respectively. We find that ectopic expression of TUBA1A-V409I/A mutants disrupt neuronal migration in mice and promote excessive neurite branching and a decrease in the number of neurite retraction events in primary rat neuronal cultures. These neuronal phenotypes are accompanied by increased microtubule acetylation and polymerization rates. To determine the molecular mechanisms, we modeled the V409I/A mutants in budding yeast and found that they promote intrinsically faster microtubule polymerization rates in cells and in reconstitution experiments with purified tubulin. In addition, V409I/A mutants decrease the recruitment of XMAP215/Stu2 to plus ends in budding yeast and ablate tubulin binding to TOG (tumor overexpressed gene) domains. In each assay tested, the TUBA1A-V409I mutant exhibits an intermediate phenotype between wild type and the more severe TUBA1A-V409A, reflecting the severity observed in brain malformations. Together, our data support a model in which the V409I/A mutations disrupt microtubule regulation typically conferred by XMAP215 proteins during neuronal morphogenesis and migration, and this impact on tubulin activity at the molecular level scales with the impact at the cellular and tissue levels.


53
For example, the microtubule motor, kinesin-6, is important for establishing a leading process 54 that is required for the transition from a multipolar to bipolar state and subsequent neuronal 55 migration (Falnikar et al., 2013). Together this suggests the critical importance of regulating 56 microtubules in the right place and at the right time during the morphological transitions neurons 57 must undergo for migration and, ultimately, proper brain development.

182
These results suggest that the TUBA1A-V409I/A mutants act dominantly to disrupt neuron 183 migration, with TUBA1A-V409A being more severe than the TUBA1A-V409I mutant.

185
To further our understanding of how this migration defect occurs, we next used high-resolution  The microtubule cytoskeleton plays a crucial role in the series of morphological changes that 202 cortical neurons must undergo throughout the course of radial migration (Nadarajah et al., 2001; 203 Noctor et al., 2004;Tabata and Nakajima, 2003). When immature neurons are born, they 204 extend numerous neurites to probe their environment for directional cues. Once these cues 205 have been identified, neurons must retract most of their neurites and become bipolar, such that 206 one neurite becomes the axon and a neurite on the opposite side of the cell becomes the 207 leading process that guides radial migration. Therefore, it is particularly crucial that microtubules 208 are able to deftly respond to various cues throughout development that promote these different

214
To distinguish between these hypotheses, we examined the in vitro development of cultured 215 primary neurons ectopically expressing the 6X-His-tagged TUBA1A (WT or mutant) plasmids 216 described above. We first compared the number of primary, secondary, and tertiary neurites at  to WT cells (0.6 ± 0.2, 0.9 ± 0.3 branches/cell for WT and V409I, respectively; Figure 2B).

227
Strikingly, TUBA1A-V409A expressing cells have a significantly greater number of secondary 228 branches, and some also exhibit tertiary branches (Secondary: 0.6 ± 0.2, 2.2 ± 0.7 branches/cell, 229 Tertiary: 0.01 ± 0.03, 0.3 ± 0.2 branches/cell for WT and V409A, respectively; Figure 2B). These 230 data fail to support the hypothesis that TUBA1A-V409I/A mutants initiate ectopic primary neurite 231 growth because we find a similar number of primary neurites in WT and mutant expressing cells.

232
Our data instead lend support to the alternative hypotheses that there is either faster neurite 233 growth or insufficient neurite retraction, as evidenced by the increase in neurite branching.

235
To assess potential differences in neurite extension or retraction, we imaged living neurons 236 expressing TUBA1A-WT, -V409I or -V409A over time to measure neurite dynamics ( Figure 2C; 237 Figure 2 -video 1). We find that neurite growth rates and retraction rates are similar between 238 TUBA1A-WT, -V409I, and -V409A expressing cells (Table 1). The duration of growth and 239 retraction events is also not significantly different; however, neurites in V409I expressing cells,

280
(reviewed in Westermann and Weber, 2003). Acetylated tubulin is a marker of stable 281 microtubule polymer and is highly localized to the axon, but largely absent in the dendrites.

302
However, we observed that some V409I/A cells exhibit increased microtubule acetylation in the 303 dendrite regions that are closer to the soma, which was not observed in WT-expressing cells 304 ( Figure 3A). These data suggest that cells expressing V409I/A mutants have increased levels of 305 microtubule acetylation as compared to WT, and V409A cells in particular have abnormally high 306 levels of acetylation at the distal axon tip.

328
Using this system, we created strains in which all the α-tubulin expressed in the cell was either 329 WT, V410I, or V410A. We find that compared to WT, V410I cells have no significant change in

350
Additionally, we find that astral microtubules in tub1/tub3-V410A mutant cells are retained for a 351 longer time at 4°C than those in WT control cells, and tub1/tub3-V410I microtubules have an 352 intermediate phenotype ( Figure 4H). Summarizing these dynamics data, we find that α-tubulin-

353
V410I/A microtubules, and particularly the tub1/tub3-V410A variant, exhibit faster microtubule 354 polymerization rates and decrease how often the microtubule catastrophes. However, when 355 these mutant microtubules do catastrophe, they depolymerize at a faster rate than WT.

357
To understand how the different microtubule parameters described above work together to 358 influence microtubule activity, we calculated the dynamicity of WT and tub1/tub3-V410I/A 359 microtubules. Dynamicity is defined as the total change in microtubule length divided by the 360 change in time (Jordan et al., 1993). We find that tub1/tub3-V410A microtubules have the

439
To test this, we used interference reflection microscopy (IRM) to measure the intrinsic 440 polymerization activity of our purified yeast Tub1-WT, -V410I, and -V410A heterodimers in an in 441 vitro dynamics assay. We used GMPCPP-stabilized, rhodamine-labeled porcine microtubule 442 'seeds' attached to the surface of the coverslip via anti-rhodamine antibodies to nucleate the 443 assembly of microtubules from the purified tubulin in the reaction ( Figure 6A). We used three 444 concentrations of purified tubulin, between 0.5 and 0.9µM, to measure microtubule dynamics in 445 vitro. Within this concentration range, we find that soluble tubulin assembles from the stabilized 446 seeds and forms dynamic microtubules. If tubulin concentration is too high, the tubulin will 447 spontaneously nucleate away from the stabilized seeds (to form microtubules, oligomers, 448 aggregates, etc.) and will not be visible on the microscope. If the tubulin concentration is too low, 449 the tubulin will not sustain assembly from the seeds. In our experimental set up, we are unable

454
In contrast to our results in budding yeast cells, we find that purified tub1-V410I microtubules do 455 not have significantly increased polymerization rates as compared to WT at any of the three 456 concentrations tested ( Figure 6C). However, with purified tub1-V410A tubulin, we were unable 457 to observe microtubule dynamics at these three concentrations, nor did we observe any at 458 higher concentrations up to 1.5µM or lower concentrations down to 0.1µM (data not shown).

459
Based on our previous data, we predict that tub1-V410A tubulin exhibits increased assembly 460 activity and would be more dynamic at lower concentrations. If tub1-V410A tubulin exhibits 461 higher assembly activity than WT tubulin, even the low end of our usable concentration range 462 may be too high for this mutant, and we would be unable to visualize microtubule dynamics 463 because the mutant may readily assemble away from seeds (similar to top panel of Figure 6A).

464
To circumvent these issues while still testing the intrinsic capabilities of tub1-V410A, we mixed 465 WT and tub1-V410A tubulin in a one-to-one ratio, for a total tubulin concentration of 0.5, 0.7, or 466 0.9µM. At this WT:V410A one-to-one ratio, we find an increase in microtubule polymerization 467 rates at each concentration tested ( Figure 6C). Since the amount of either WT or V410A tubulin 468 that is present in each of these one-to-one mixtures (i.e., 0.25, 0.35, or 0.45µM) is not sufficient 469 to support microtubule assembly on its own, we conclude that the increased microtubule 470 polymerization is a synergistic effect of the blend of WT and V410A mutant tubulin.

472
The fitted lines of the data collected at each tubulin concentration provide us with important 473 information about the intrinsic properties of the WT and mutant microtubules formed in vitro. The 474 slopes of the fitted lines represent the concentration-dependent polymerization rate, and we find 475 that WT/V410A has a ~2-fold increased rate as compared to WT and V410I (11.6 compared to 476 5.8 and 6.5µm/hr/µM). The x-and y-intercepts represent the critical concentration and apparent 477 off rate constants, respectively (x-intercepts = 0.01, 0.02, and 0.02µM; y-intercepts = -0.07, -478 0.13, and -0.18µm/hr for WT, V410I, and WT/V410A, respectively). In contrast to the increase 479 observed in microtubule polymerization rates, the microtubule depolymerization rates were not 480 different between WT and either the V410I or V410A mutants ( Figure 6D). These data indicate 481 that tub1-V410A has significantly increased intrinsic microtubule polymerization as compared to

482
WT. Together, our data explain the intrinsic mechanism by which the V409/V410A mutant has 483 the most severe phenotypes across scales, while V409/V410I has more mild effects.   cultures, will provide insight into whether neuronal migration is completely impaired or simply 526 delayed. To the best of our knowledge, there are no tubulinopathy mutations to date that disrupt 527 neuron migration via impaired neuron morphology transitions. Therefore, future work will be 528 required to determine whether this is a common mechanism for particular types of brain 529 malformations.

531
Our findings support the idea that microtubule stability plays a determining role in neuron  and G). We also find that the duration of polymerization events is shorter when V409I and 549 V409A is expressed, compared to WT ( Figure 3H). Whether these shorter polymerization 550 events are followed by transitions to depolymerization (i.e., catastrophe) or to a stable, non-551 polymerizing state is unclear. Our data suggest that the V409 mutants do not promote increased 552 nucleation as compared to WT, but rather V409I, and more strongly V409A, increase the 553 content of stable microtubule polymer in neurons. This may be due in part to increased 554 polymerization rates; however, we cannot rule out changes in transition frequencies or 555 depolymerization rate, neither of which are accessible in our dynamics data.

557
Our results from modeling V409 mutants in budding yeast further reveal the mechanistic origins 558 of the highly stable microtubules in neurons. Similar to our results in neurons we find that, 559 compared to WT, tub1/tub3-V410I/A microtubules have faster polymerization rates ( Figure 4D).

560
In yeast we are able to analyze additional microtubule dynamics parameters, and we find that 561 tub1/tub3-V410I/A microtubules also have increased depolymerization rates, decreased

587
We find that V410I/A tubulins have weaker affinity for Stu2 TOG1/TOG2 domains in our in vitro 588 binding assays, and that V410I microtubules, and more significantly -V410A microtubules, have 589 less Stu2 at their plus ends in yeast cells (Figure 5C

634
It is interesting that in our in vitro system, we were unable to identify a concentration at which 635 homogeneous V410A polymerizes from the GMPCPP-stabilized porcine tubulin seeds. However,

636
blending V410A with WT tubulin in a one-to-one ratio supports the assembly of microtubules 637 with significantly increased polymerization rates as compared to WT at the same concentrations 638 of total tubulin. This suggests that V410A may be unable to polymerize from GMPCPP seeds on 639 its own, but it drives fast polymerization when mixed with WT tubulin. Based on our data 640 acquired from neuron cultures, and particularly from budding yeast in which all the α-tubulin in 641 the cell is V410A, we know that this mutant assembles into microtubules that polymerize faster           Purification of GST-TOG1/2 followed previously described methods and is described briefly

885
The cells were chased with 50ml lysis buffer supplemented with 50µM GTP and 1mM PMSF.

1398
Bars are mean ± 95% confidence interval. Statistical analysis was done using an unpaired t-test.

1399
All statistics with p≤0.05 are indicated on graph. F) Representative images of neurons 1400 expressing the above plasmids, exposed to 4°C for indicated time, and stained with TUBB2A/B.

1406
The proportion of neurites that have retraction bulbs per cell measured every two hours over a          F. C.

G.
I.