Neuron-specific inactivation of Wt1 alters locomotion in mice and changes interneuron composition in the spinal cord

(A) Schematic illustration of the Wt1^fl/fl allele. loxP sites flanking exons 2 and 3 of the Wt1 coding sequence allow Cre-mediated excision and conditional knockout of Wt1. Confirmation of a functional conditional Wt1 knockout in Nes-Cre;Wt1^fl/fl at E12.5 using qRT–PCR (quantification to the right). Data expressed as mean ± SEM. n = 4–5 embryos. Significance determined by using pairwise reallocation randomisation test. (B) Loss of Wt1 immunopositive signals in Nes-Cre;Wt1^fl/fl embryos at E12.5 corroborates the loss of Wt1 protein. Schematic illustration shows the position where the pictures were taken. Stippled line represents the border between the ventricular and mantle zones. Scale bar: 40 μm. (C) X-ray radiograph of a walking mouse in lateral perspective. (D) Graphs displaying stride parameters collected in the X-ray radiograph. Stride frequency is significantly lower in female Nes-Cre;Wt1^fl/fl mice in both forelimbs and hindlimbs. The stride length in female Nes-Cre;Wt1^fl/fl mice is increased compared with female Wt1^fl/fl mice, whereas smaller differences are found in male mice. Box plots indicate the median of each group, n = 10 animals (bold white or black line), the 25^th and the 75^th percentile (box), and the data range (whiskers). Mann–Whitney U test was performed. Significance level of U: ***P < 0.001; **P < 0.01; *P < 0.05. (E, F) Interlimb coordination expressed as the time lag between footfall events in percent stride duration. Left limbs are reference limbs. The scheme (E) illustrates which phase relationships are shown by which graph. Phase relationships (F) between forelimbs (1) and between hindlimbs (2) illustrate overall symmetry of the walk. Timing of forelimb touchdown relative to hindlimb touchdown for ipsilateral (3) and contralateral (4) limbs show only minor differences between Wt1^fl/fl and Nes-Cre;Wt1^fl/fl mice. The timing of hindlimb footfalls relative to forelimb footfalls (5, 6) differ between Wt1^fl/fl and Nes-Cre;Wt1^fl/fl mice, particularly at the contralateral limbs. Box plots indicate the median of each group, n = 10 animals (bold white or black line), the 25^th and the 75^th percentile (box), and the data range (whiskers). Mann–Whitney U test or t test. Significance level of U or t_s: ***P < 0.001, **P < 0.01, *P < 0.05.

(A) Stride frequency and stride length of Nes-Cre;Wt1^fl/fl, Nes-Cre;Wt1^fl/+, and Wt1^fl/fl mice. Different parameters were measured using digital footprints of mice running on a transparent treadmill at distinct speed. Control Wt1^fl/fl and Nes-Cre;Wt1^fl/+ mice do not show significant changes in stride frequency and length in contrast to Nes-Cre;Wt1^fl/fl that exhibit a decline in stride frequency (P ≤ 0.001 for fore- and hind limbs at 15 cm/s and P ≤ 0.05 for fore- and hind limbs at 35 cm/s) and an increase in stride length (P ≤ 0.001 for fore- and hind limbs at 15 cm/s and P ≤ 0.05 for fore- and hind limbs at 35 cm/s). n = 20 Wt1^fl/fl control; n = 12 Nes-Cre;Wt1^fl/fl. (B) Experimental setup of high-resolution X-ray fluoroscopy showing treadmill and C-arm fluoroscope. (C) Absolute and nondimensional speed that animals run on the treadmill. A significant decline in absolute speed between Nes-Cre;Wt1^fl/fl mice and control animals (Wt1^fl/fl) in males and females was detected. For each of the sexes, n = 10 Wt1^fl/fl control; n = 10 Nes-Cre;Wt1^fl/fl. Significance level: ***P < 0.001, **P < 0.01, *P < 0.05 (according to one-way ANOVA). (D) Duty factors of forelimbs and hind limbs. The duty factor describes the contact duration of one foot being on the ground relative to the whole gait cycle and thus illustrates the temporal coordination between the stride phases. Duty factors for fore- and hind limbs are increased in Nes-Cre;Wt1^fl/fl mice pointing to a prolongation of the stance phase. Box plots indicate the median (bold white or black line), the 25^th and the 75^th percentile (box), and the data range (whiskers). For each of the sexes, n = 10 Wt1^fl/fl control; n = 10 Nes-Cre;Wt1^fl/fl. Significance level: ***P < 0.001, **P < 0.01, *P < 0.05 (according to one-way ANOVA). (E) Footfall sequences of a female Wt1^fl/fl control and a female Nes-Cre;Wt1^fl/fl mouse are shown across three consecutive strides of the left hind limb. Lines represent the stance phases and gaps the swing phases of all four limbs (R, right; L, left; F, forelimb; H, hind limb). (F) Schematic representation of a footfall pattern illustrating footfall sequences across two consecutive strides of the left forelimb that was used as the reference limb. Lines represent the stance phases and gaps the swing phases of all four limbs. One stride cycle starts with the stance phase and ends with the swing phase. The first stride cycle of a left limb (labelled in red)—here the left forelimb—was used as the reference cycle. The interlimb coordination was determined as the phase relationship between a respective limb (here RH, RF, or LH) to the reference limb (LF). It is expressed as the time lag (in percent) of the start of a stride cycle of one limb compared with the start of the reference cycle. For symmetric left–right alternation of limbs (RF to LF), the phase relationship is shifted about half of the reference cycle. The time lag is 50%. For ipsilateral coordination (LH to LF) in wild-type mice, the time lag is about 55% of the reference cycle. For wild-type contralateral coordination (RH to LF) in wild-type mice, the time lag is about 5% of the reference cycle. RH, right hind limb; RF, right forelimb; LF, left forelimb; LH, left hind limb.

(A) Images of Nes-Cre;Wt1^fl/fl and Wt1^fl/fl pups air-stepping. (B) Quantification of the air-stepping experiment. Nes-Cre;Wt1^fl/fl mice show a decreased number of alternating and increased number of uncoordinated hind limb steps during air-stepping compared with controls (n = 9 Wt1^fl/fl control; n = 7 Nes-Cre;Wt1^fl/fl). Data expressed as mean ± SD. Significance level: #P < 0.06.

(A) Representative traces showing locomotor-like activity during fictive locomotion from left and right lumbar (L) 2 and right L5 ventral roots from control (Nes-Cre;Wt1^+/+) and Wt1 conditional knockout (Nes-Cre;Wt1^fl/fl) mice. Rhythmic activity was induced by application of NMDA, serotonin, and dopamine. Raw traces in black; rectified, low-pass filtered signal of lL2 trace in blue; activity burst shown in green. Spinal cord schematic depicts the attached suction electrodes to the right (r) and left (l) L2 and rL5 ventral roots. Scale bar: 5 s. (B) Phase analysis and associated coherence power spectra of left/right (L2/L2) and flexor/extensor (L2/L5) recordings. Regions of persistent coherence emerge for control mice at 0.30 Hz, whereas spinal cords from Nes-Cre;Wt1^fl/fl mice show an intermittent coherence region at 0.18 Hz. Color-graded scale indicates normalized coherence. Scale bar: 125 s. (C) Locomotor patterns, analyzed from 20 consecutive bursts, reveal impaired and variable left/right and flexor/extensor alternation in Nes-Cre;Wt1^fl/fl mice (black dots). Normal left/right and flexor/extensor alternation is maintained in control (white dots) mice. Each dot represents one cord; arrows represent the mean phase. The length of the vector is a measure of the statistical significance of the preferred phase; dashed grey line indicates regions of significance and high significance at 0.5 and 0.8, respectively (Rayleigh test and Watson's U² test). n = 5 pups (control); n = 7 pups (Nes-Cre;Wt1^fl/fl). (D) Nes-Cre;Wt1^fl/fl mice have a slower locomotor frequency than control mice. Data are shown as mean ± SD. n = 5 pups (control); n = 7 pups (Nes-Cre;Wt1^fl/fl). Significance was tested using two-tailed Mann–Whitney U test. (E, F) The slower locomotor frequency in Nes-Cre;Wt1^fl/fl mice is mirrored by an increased cycle period, and burst and interburst duration in both L2 (E) and L5 (F) roots. Data are shown as mean ± SD. n = 5 pups (control); n = 7 pups (Nes-Cre;Wt1^fl/fl). Significance was tested using two-tailed Mann–Whitney U test. Significance level: **P < 0.01, ***P < 0.001.

(A) Spinal cords of Wt1-GFP embryos (stage E16.5) show a Wt1+ interneuron immunopositive for GFP. Wt1 is localized in the nucleus and GFP throughout the cell. Scale bar: 2 μm. (B) Wt1+ neurons (green) receive excitatory, inhibitory, and monoaminergic synaptic contacts. Synaptic terminals are identified with synaptophysin (blue). Glutamatergic terminals were immunolabelled for VGLUT2, inhibitory synapses immunolabelled for VGAT, and monoaminergic terminals immunolabelled for VMAT2. Arrows point to individual synaptic terminals (magenta) present on Wt1+ neurons (green). Boxed areas show higher magnification panels of separated channels. Scale bar: 2 μm. (C) GFP-immunolabelled dI6 neurons in the spinal cord of E16.5 Wt1-GFP embryos. GFP antibody staining (green; left panel) and merging with Hoechst (blue; right panel) is shown. Boxed areas represent location of higher magnification panels shown on the right of each panel. Contralateral projections crossing the midline (dashed line) of the spinal cord are visible (arrow heads in magnified images). Scale bar: 20 μm for overview images and 50 μm for magnified images. (D) Photomicrographs of transverse, 60 μm, lumbar, spinal cord sections with applied fluorescein dextran amine (FDA, green) and rhodamine dextran amine (RDA, red) tracers. Higher magnification images (insets) of wild-type (Nes-Cre;Wt1^+/+) and homozygous (Nes-Cre;Wt1^fl/fl) segments, showing intersegmental retrograde FDA (white arrow), RDA (open arrow), and double-labelled (triangle arrow) neurons. Scale bar: 200 μm. (E) Schematic illustration of FDA (lumbar [L]1) and RDA (L4/5) application sites tracing descending (green), ascending (red), and bifurcating (yellow) neurons. The area of analysis (L2/L3) is indicated by black dashed line. (F–H) Quantification of descending FDA-labelled neurons (F), ascending RDA-labelled neurons (G), and bifurcating, double-labelled neurons per section (H). Descending, ascending, and bifurcating CINs are significantly fewer in homozygous spinal cords compared with control cords (according to Kruskal–Wallis test and followed by a Dunn’s post hoc test comparing all groups). Data expressed as mean ± SEM (Wt1^fl/fl control: 3,975 total cells, 215 sections, and nine spinal cords; Nes-Cre;Wt1^fl/fl: 3,421 total cells, 228 sections, and seven spinal cords). Significance level: *P < 0.05, **P < 0.01, ***P < 0.001.

(A) Schematic illustration of fluorescein dextran amine (FDA) application sites to trace commissural Wt1+ neurons. FDA was applied at lumbar segment L1. The spinal cord was sectioned and Wt1+ neurons were labelled using an antibody against Wt1. The area of analysis (L2/L3) is indicated by a dashed box. (B) Axonal projection tracing of commissural Wt1-positive interneurons. Transverse, lumbar spinal cord sections of P3 newborn mice were labelled using an antibody against Wt1 (red) after an FDA tracer; (green) had been inserted into the contralateral hemisphere and retrogradely transported. This retrograde tracing of commissural interneurons reveals that a fraction of Wt1-positive dI6 interneurons projects commissurally. Higher magnification images at the bottom row show Wt1-positive dI6 neurons (white arrow) that overlap with the FDA tracer. Scale bar: 100 μm.

(A) Average cell number of Wt1, Dmrt3, and Wt1/Dmrt3 neurons per 12 μm spinal cord section from different embryonic and postnatal stages of control (Wt1^fl/fl) and Wt1 conditional knockout (Nes-Cre;Wt1^fl/fl) mice. Number of Dmrt3 neurons significantly decrease in Nes-Cre;Wt1^fl/fl. No Wt1/Dmrt3 neurons are detected in Nes-Cre;Wt1^fl/fl animals. Data expressed as mean ± SD. n = 3 embryos per developmental stage and genotype. Significance determined by t test. (B) Average cell number of Lmx1b, Evx1, Foxp2, Chx10, Gata3, and ventral Islet1/2 neurons per 12 μm spinal cord section from control (Wt1^fl/fl) and homozygous (Nes-Cre;Wt1^fl/fl) mice at E12.5 and E16.5. Number of Evx1+ V0 neurons is significantly increased. Data expressed as mean ± SD. n = 3–4 embryos per developmental stage and genotype. Significance determined by t test. (C) Average cell number of Wt1, Dmrt3, and Wt1/Dmrt3 dI6 neurons and Evx1 V0 neurons per 12 μm spinal cord section from E16.5 control (Wt1^fl/fl) and Wt1 conditional knockout (Lbx1-Cre;Wt1^fl/fl) mice. Lbx1-Cre–based conditional Wt1 knockouts show a decrease in the amount of dI6 neurons and an increase in the cell number of Evx1 neurons as for Nes-Cre;Wt1^fl/fl animals. Data expressed as mean ± SD. n = 3–5 embryos per genotype. Significance determined by t test. (D) Schematic illustration of Wt1-GFP-DTA allele. Cassette consisting of loxP sites flanking GFP coding sequence upstream of DTA was inserted into the Wt1 locus. This cassette allows Cre-mediated ablation of Wt1+ neurons. Graph shows average cell number of Wt1, Dmrt3, and Wt1/Dmrt3 dI6 neurons and Evx1+ V0 neurons per 12 μm spinal cord section from E16.5 wild-type control and Lbx1-Cre;Wt1-GFP-DTA mice. Nearly all Wt1+ neurons are absent. The number of Dmrt3 neurons is significantly decreased. Population size of Evx1 neurons is not altered after ablation of Wt1+ neurons. Data expressed as mean ± SD. n = 3 embryos per genotype. Significance determined by t test. Significance level: *P < 0.05, **P < 0.01, ***P < 0.001.