Transcriptional repression of Plxnc1 by Lmx1a and Lmx1b directs topographic dopaminergic circuit formation

Abstract

Mesodiencephalic dopamine neurons play central roles in the regulation of a wide range of brain functions, including voluntary movement and behavioral processes. These functions are served by distinct subtypes of mesodiencephalic dopamine neurons located in the substantia nigra pars compacta and the ventral tegmental area, which form the nigrostriatal, mesolimbic, and mesocortical pathways. Until now, mechanisms involved in dopaminergic circuit formation remained largely unknown. Here, we show that Lmx1a, Lmx1b, and Otx2 transcription factors control subtype-specific mesodiencephalic dopamine neurons and their appropriate axon innervation. Our results revealed that the expression of Plxnc1, an axon guidance receptor, is repressed by Lmx1a/b and enhanced by Otx2. We also found that Sema7a/Plxnc1 interactions are responsible for the segregation of nigrostriatal and mesolimbic dopaminergic pathways. These findings identify Lmx1a/b, Otx2, and Plxnc1 as determinants of dopaminergic circuit formation and should assist in engineering mesodiencephalic dopamine neurons capable of regenerating appropriate connections for cell therapy.Midbrain dopaminergic neurons (mDAs) in the VTA and SNpc project to different regions and form distinct circuits. Here the authors show that transcription factors Lmx1a, Lmx1b, and Otx2 control the axon guidance of mDAs and the segregation of mesolimbic and nigrostriatal dopaminergic pathways.

Fig. 1

Ventral midbrain expression of transcription factors Lmx1a and Lmx1b at early stage of development. Representative confocal stitched images of immunohistochemical staining of the transcription factors Lmx1a and Lmx1b in TH-positive neurons in midbrain coronal sections at E15.5 (a) and P1 (b). Although Lmx1a and Lmx1b are present in all mDA neurons, the staining intensity for Lmx1a and Lmx1b varies in mDA neurons. Scale bars: a, 200 µm and 10 µm for high-magnification inserts; b, 200 µm, and 20 µm for high-magnification inserts. SNpc, substantia nigra pars compacta; VTA, ventral tegmental area

Fig. 2

Characterization of the phenotype of Lmx1a/b double conditional mutant mice at P1. a Schematic representation of coronal section of mouse brain at the midbrain level at post-natal day 1 (P1). Dashed line indicates the delimitation of the pictures shown in b. Distribution (b) and number (c) of mDA neurons in the midbrain were not different between controls and Lmx1a/b cKO at P1 (n = 4, two-tailed unpaired t test, p = 0.8134). d, e Schematic representations of axonal innervation in the striatum in control and Lmx1a/b cKO mice brains at P1. f, g Representative confocal images of control and Lmx1a/b cKO mice brain sections showing a loss of dopaminergic innervation in dorso-posterior striatum for Lmx1a/b cKO mice (TH in green and DAPI in blue). h, i Optical density measurements of TH axons in the striatum. Graphs in (i) show the ratio of TH intensity in dorsal vs. ventral striatum (n = 7, two-tailed unpaired t test, p ^{(Ant D)} = 0.0483, p ^{(Cent D)} = 0.3398, p ^{(Cent V)} = 0.9596, p ^{(Post D)} = 0.0003, p ^{(Post V)} = 0.5914, p ^{(Ant D/V)} = 0.0526, p ^{(Cent D/V)} = 0.1863, p ^{(Post D/V)} ≤ 0.0001, Mann–Whitney U, p ^{(Ant V)} = 0.1649). j Light-sheet scans of the TH immunostaining in the striatum of iDISCO-cleared brains from control and Lmx1a/b cKO mice. The panels show single optical planes in coronal and reconstructed sagittal and horizontal planes showing the lack of TH axons in the dorsal striatum of the Lmx1a/b cKO mice brains. Dotted lines delineate the border of the striatum. k Schematic representation of the location of electrophysiological recordings in the striatum. l, m Analysis of the frequency of miniature excitatory post-synaptic currents (mEPSC) and miniature inhibitory post-synaptic currents (mIPSC) in dorsal and ventral striatum (dorsal striatum: n = 9 cells for controls and n = 11 cells for Lmx1a/b cKO, Mann–Whitney U, p ^(mEPSC) = 0.0078; p ^(mIPSC) = 0.0275; ventral striatum: n = 9; two-tailed unpaired t test; p ^(mEPSC) = 0.0408; p ^(mIPSC) = 0.0444). Scale bars: 250 µm for (b), (f), (g) and 500 µm for (j). D, dorsal; L, lateral; N.S., not significant; Str, striatum; V, ventral

Fig. 3

Anterograde and retrograde axonal tracing experiments showing aberrant dopaminergic axonal connections in Lmx1a/b conditional mutants. a, b Schematic description of the AAV-FLEx-EGFP experiment showing injection of the viral vector in the SNpc at P10 and the labeling of the axonal projections in the striatum. c, d Representative confocal images of the GPF-positive cells in the midbrain (AAV-FLEx-EGFP in green and DAPI in blue), and the resulting GFP-positive axons at the striatal level in the control (e) and in the Lmx1a/b cKO mice (f) 17 days after injection. g Schematic representation of the injection site of the AAV-retro-GFP in the ventral striatum. h Representative confocal images of the injection site in the ventral striatum. i Schematic representation of a coronal section of mouse brain at the midbrain level indicating the delimitation of the pictures shown in k–n. j Quantification of the percentage of retrogradely labeled neurons in SNpc (GFP⁺ TH⁺ in SNpc on total GFP⁺ TH⁺; n = 3, two-tailed unpaired t test, p < 0.0001). k–n Representative confocal images of the retrogradely labeled cells in control (k) and in Lmx1a/b cKO mice (m) 17 days after injection in the ventral striatum at P10 (TH in red, AAV-Retro-GFP in green and DAPI in blue). l, n Higher magnification in the VTA and SNpc as indicated by the dashed boxes in upper images k, m. Scale bars: c, d, e, f, 250 µm; h, k, m, 500 µm; l, n, 100 µm

Fig. 4

Lmx1a/b regulate the expression of the axon guidance receptor Plxnc1. a Images of brain tissue section following quick TH staining, before LCM and in the tube cap after LCM. b In situ hybridization for Plxnc1 expression in ventral midbrain coronal sections of control and Lmx1a/b cKO mice at P1. c RT-qPCR analysis of Plxnc1 expression specifically in SNpc and VTA neurons isolated with LCM (SNpc: n = 3, two-tailed unpaired t test, p = 0.0353; VTA: n = 3 controls and 4 Lmx1a/b cKO, one-tailed Mann–Whitney U, p = 0.0286). d RT-qPCR quantification of Plxnc1 expression after Lmx1a and Lmx1b overexpression in ventral midbrain primary cell cultures at P1 (n = 3 independent cultures; for Plxnc1: one-way ANOVA with Tukey’s post test, *p < 0.05; for Lmx1a and Lmx1b: one-tailed Mann–Whitney U, p = 0.05). Scale bars: 200 µm

Fig. 5

The axonal guidance repellent cue Sema7a shows higher expression in the dorsal than in the ventral striatum. a, b In situ hybridization for Sema7a on a sagittal section from embryonic day 18.5 and from post-natal day 4 mouse brains (from Allen Brain Atlas). The dorsal striatum (delineated in red) expresses more Sema7a than the ventral striatum. c, d Western blot and quantification of Sema7a protein level in the dorsal and ventral striatum at P1 (n = 3, two-tailed unpaired t test, p  = 0.0418). Scale bar: 500 µm

Fig. 6

In vitro experiments showing the effect of Sema7a on VTA and SNpc neurons. a, c Sholl analysis on ventral midbrain primary VTA or SNpc neuron cultures exposed to PBS or Sema7a. a Examples of confocal images of VTA and SNpc neurons with their respective Sholl intersection circle mask. b, c Quantitative analysis of the VTA and SNpc neurite intersection profiles, number of branches, and neurite length (SNpc: 71 neurons analyzed from 5 independent cultures; VTA: 114 neurons analyzed from 7 independent cultures; b: two-way ANOVA, and the Sidak test was used for post hoc comparisons, *p < 0.05; c: two-tailed unpaired t test, p ^{(VTA Branch)} = 0.0137, p ^{(VTA length)} = 0.0335, p ^{(SNpc Branch)} = 0.4370, p ^{(SNpc length)} = 0.4395). d Examples of embryonic ventral midbrain explants from E14.5 Pitx3-GFP embryos grown in collagen matrix then exposed to PBS (control) or Sema7a. e Sholl analysis of the neurite intersection profiles for the VTA and SNpc explants exposed to PBS or Sema7a (a total of 34 explants from 8 independent cultures, 20 explants from VTA and 14 from SNpc; two-way ANOVA, and the Sidak test was used for post hoc comparisons, *p < 0.05). f Confocal images of growth cones from VTA explants exposed to Sema7a or PBS. g Growth cone size measurements expressed as a percentage of control (n = 7 independent cultures; 65 growth cones for Sema7a and control were used in each of the 7 experiments, Mann–Whitney U, p = 0.0175). h–k Confocal images of stripe assay and quantification of the number of axons terminating on stripes for VTA explants grown on the control alternating IgG stripes and the Sema7a stripes alternating with IgG stripes. Dotted lines delineate the stripes (i, n = 29 explants in 4 independent experiments, two-tailed unpaired t test, p = 0.9349; k, n = 27 explants in 4 independent experiments, two-tailed unpaired t test, p = 0.0091). Scale bars: a, 20 µm; d, f, h, 250 µm; j, 10 µm

Fig. 7

Sema7a/Plxnc1 interaction regulates mDA axon guidance. a, b Schematic representations of coronal views of the axonal projections in the striatum showing that Plxnc1 mDA axons (in red) are more numerous in the dorso-posterior striatal region of Sema7a KO mice (b) compared to control animals (a). c, d Confocal images of TH and Plxnc1 immunostained striatal sections from control and Sema7a KO mice. e–g Quantification of TH relative optical density at different striatum levels (n = 3, two-tailed unpaired t test, p ^{(Ant D)} = 0.2855, p ^{(Ant V)} = 0.5355, p ^{(Cent D)} = 0.7461, p ^{(Cent V)} = 0.3933, p ^{(Post D)} = 0.0491, p ^{(Post V)} = 0.9514, p ^{(Ant D/V)} = 0.0023, p ^{(Cent D/V)} = 0.4418, p ^{(Post D/V)} = 0.0023). h Schematic representation of coronal section of mouse brain at the midbrain level. Dashed line indicates the boundaries of the pictures shown in i. i Representative confocal images of the VTA and SNpc for Control and Sema7a KO mice. j Stereological counts of mDA neurons in control and Sema7a KO mice at P1 (n = 3, two-tailed unpaired t test, p = 0.4843). k Schematic representation of the injection site of the AAV-retro-GFP in the dorsal striatum. l Representative confocal images of the injection site in the dorsal striatum. m–p Representative confocal images of the retrogradely labeled cells in control (m, n) and in Sema7a KO mice (o, p) 17 days after injection in the dorsal striatum at P30 (TH in red, AAV-Retro-GFP in green and DAPI in blue). n, p Higher magnifications in the VTA as indicated by the dashed boxes in m, o. q Quantification of the percentage of retrogradely labeled neurons in VTA (GFP⁺ TH⁺ in VTA on total GFP⁺ TH⁺; n = 4 for controls and n = 7 for Sema7a KO mice; two-tailed unpaired t test, p = 0.0245). r, s Schematic and confocal images of TH labeling in Plxnc1 overexpression mice showing a loss of DA innervation in dorso-posterior striatum. t RT-qPCR quantification of Plxnc1 from the ventral midbrain section of control and Plxnc1 overexpression mice at P1 (n = 3, one-tailed unpaired t test, p = 0.0322). u Stereological counts of mDA neurons in control and Plxnc1 overexpression mice at P1 (n = 3, two-tailed unpaired t test, p = 0.5431). v, w Schematic and confocal images of TH and td-Tomato in the VTA and SNpc for one hemisphere. The reporter tdTomato indicates that all mDA neurons express the transgene (Plxnc1-ires-tdTomato). x Optical density measurements of TH-positive axons in the striatum at three antero–posterior levels (n = 4, two-tailed unpaired t test, n = 4, Mann–Whitney U, p ^{(Ant D/V)} = 0.6857, two-tailed unpaired t test, p ^{(Cent D/V)} = 0.1945, p ^{(Post D/V)} = 0.0019). Scale bars: 200 µm except n, p, 100 µm

Fig. 8

Otx2 and Lmx1a/b control Plxnc1 expression in the VTA. a, b Confocal images of Otx2 immunostaining in TH-positive cells in coronal midbrain sections of control and Lmx1a/b cKO mutant mice showing ectopic Otx2 expression in SNpc mDA neurons. c Quantification of the number of mDA neurons in SNpc with ectopic expression of Otx2 (n = 3, two-tailed unpaired t test, p = 0.0015). d Rt-qPCR quantification of Otx2 in VTA and SNpc isolated from control and Lmx1a/b cKO with LCM (VTA: n = 3 for control and n = 5 for Lmx1a/b cKO, two-tailed unpaired t test, p = 0.9128; in SNpc n = 3, one-tailed Mann–Whitney U, p = 0.05). e Overexpression of Otx2 in primary cell cultures leads to an increase of Plxnc1 expression as quantified by RT-qPCR (n = 3 independent cultures; for Plxnc1: two-tailed unpaired t test, p = 0.03; for Otx2: Mann–Whitney U, p = 0.0286). f RT-qPCR quantification of Otx2 in mDA primary cell cultures following Lmx1a and Lmx1b overexpression (n = 8 for control, n = 7 for Lmx1a and n = 5 for Lmx1b; one-way ANOVA with Tukey’s multiple comparison test, not significant). g Representative images of TH and Otx2 immunolabeling from E15.5 Lmx1a/b cKO embryo at E15.5 showing SNpc neurons co-expressing Otx2 and TH (arrows) before and after LCM. Right panels are high magnification of the left panels. h RT-qPCR quantification of Plxnc1 expression from LCM isolated cells co-expressing Otx2 and TH in VTA and SNpc from control and Lmx1a/b cKO embryos. For control SNpc, TH-positive cells were isolated (n = 3, 5–8 cells per embryo, one-way ANOVA with Tukey’s multiple comparison test, not significant between groups expressing Plxnc1). Image on the right is a representative image of the gel showing the RT-qPCR product. Plxnc1 was not detected in mDA neurons of SNpc from control. Scale bars: a, 200 µm; b, 10 µm; g (left), 100 µm, (right), 10 µm