Formation of the substantia nigra requires Reelin-mediated fast, laterally-directed migration of dopaminergic neurons

Midbrain dopaminergic (mDA) neurons migrate to form the laterally-located substantia nigra pars compacta (SN) and medially-located ventral tegmental area (VTA), but little is known about the underlying cellular and molecular processes. Reelin signaling regulates tangential migration of SN-mDA neurons, but whether Reelin acts directly on SN-mDA neurons and how it affects their cellular morphology and migratory behavior has not been explored. Here we visualize the dynamic cell morphologies of tangentially migrating SN-mDA neurons with 3D-time-lapse imaging and identify two distinct migration modes. Slow migration is the default mode in SN-mDA neurons, while fast, laterally-directed migration occurs infrequently and is strongly associated with bipolar cell morphology. By speci1cally inactivating Reelin signaling in mDA neurons we demonstrate its direct role in SN-mDA tangential migration. We show that Reelin signaling promotes laterally-biased movements in mDA neurons during their slow migration mode, stabilizes leading process morphology and increases the probability of fast, laterally-directed migration.


Introduction
Dopaminergic neurons in the ventral midbrain (mDA neurons) are the major source of dopamine in 27 the mammalian brain. Dysfunction in the dopaminergic system is associated with schizophrenia, 28 addiction and depression, and degeneration of mDA neurons in the substantia nigra pars com-     . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; Manuscript submitted to eLife  . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; infrequent, fast movements that are promoted by Reelin signaling 178 The role of Reelin signaling has been studied extensively in the cortex and hippocampus. 179 However, only few studies have examined Reelin function in regulating the speed of migrating 180 neurons. These studies have shown that the effect of Reelin varies depending on the brain region 181 and type of neuron analyzed Simó et al. ( 204 increasing the proportion of both non-migratory and 'slow' cells. 205 Next, we asked how frequently migrating SN-mDA neurons moved with soma speeds comparable 206 to their max-speeds and whether the fraction of total time-points spent in high migratory speeds 207 was different in control and Dab1 −∕− populations. To evaluate this, we used the criteria previously 208 defined for max-speeds, but applied them to individual soma speeds for each cell at each time 209 point. For example, we analyzed the fraction of time (percentage of total time-points) spent by each 210 'fast' cell with a soma speed of more than 60 m/hr (fast migratory phase), 30-60 m/hr (moderate 211 migratory phase), 10-30 m/hr (slow migratory phase) and less than 10 m/hr (resting phase). In 212 control slices, 'fast', 'moderate' and 'slow' cells spent a predominant fraction of time at rest (62.6 213 +/-20%; 68.5 +/-18.2%, 85.7 +/-11.1%, respectively) and were frequently in a slow migratory phase 214 (26.8 +/-17.4%, 25.1 +/-16.3%, 14.2 +/-11.1%, respectively). 'Fast' and 'moderate' cells achieved the 215 moderate migratory phase in only a few frames (5.5 +/-5.5% and 6.3 +/-3.9, respectively), and the . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; directed cell displacements 229 We next asked whether max-speeds and directionality of migration were linked. We computed 230 directionality as the ratio of total displacement (the 3D displacement between the initial and Reelin promotes preference for laterally-directed migration in mDA neurons 242 As tangential migration ultimately results in SN-mDA migration away from the midline, we 243 analyzed the trajectories of migratory SN-mDA neurons in the presence and absence of Reelin 244 signaling. We determined the "trajectory angle" for each cell as the angle between the midline 245 (y-axis in live-images) and the cell's displacement vector ( Figure 6A). Thus, a trajectory angle of 90 • 246 indicates a cell whose total movement is precisely aligned to the lateral axis (x-axis in live-images). 247 We defined a cell as migrating laterally if its trajectory angle was between 45 -135 • . We then showed differences in their preference for lateral migration, we analyzed their trajectories separately. 255 We found that trajectories of all three SN-mDA groups were anisotropic in controls, favoring 256 migration towards lateral directions, but this anisotropy was greater in 'fast' and 'moderate' cells . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; Manuscript submitted to eLife  . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; (C) Control mDA neuron transitioning between bipolar and multipolar morphology. At t = 0 min, the cell has a branched leading process (LP), the cell soma moves along the LP to reach the branch-point and takes up a multipolar morphology (t = 20 min). The cell remains multipolar until t =90 min, after which one process is retracted (t = 120min) and the cell resumes a bipolar morphology (t=150 min). Bipolar phase: one or two processes arise directly from the soma. Multipolar phase: more than two processes arise directly from the soma. (A,C) Colored circles: soma as defined by the tracking process. Scale bar: 25 m. (D) Soma speed for the neuron in (C) is higher during its bipolar phase. (E,F) Proportion of bipolar unbranched and branched morphology is significantly higher during fast and moderate phases of migration, while slow phase shows higher proportion of multipolar cells in both control (E) and Dab1 −∕− (F) mDA neurons (* p < 0.05, ** p < 0.005, *** p < 0.001, **** p < 0.0001; two-way ANOVA).    . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; of tangential neuronal migration in the brain. CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.

Reelin protein is localized in the lateral ventral midbrain
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; contrast, slow cells, which are weakly anisotropic in controls are significantly more isotropic in . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.  CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; domain. Since we observed that in both, Dab1 CKO and Dab1 −∕− mice, a few TH-positive cells of the 669 lateral most SN lateralis were consistently present (yellow arrowheads ( Figure 1D . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; Manuscript submitted to eLife . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018;  . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; Manuscript submitted to eLife . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint . http://dx.doi.org/10.1101/413708 doi: bioRxiv preprint first posted online Sep. 10, 2018; CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.

Control Control
Dab1-/- . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.

Movie 3. Morphology as detected by YFP mosaic labelling is similar to morphology detected by TH antibody.
Example SN-mDA neuron from fixed, cleared whole-mount embryonic brain of the same age as used in �me-lapse experiments (E14.5) shows similar morphology with YFP (green) and TH (magenta) immunostaining. . CC-BY 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.