Article
Reference
Transient deregulation of canonical Wnt signaling in developing pyramidal neurons leads to dendritic defects and impaired behavior
VIALE, Beatrice, et al.
Abstract
During development, the precise implementation of molecular programs is a key determinant of proper dendritic development. Here, we demonstrate that canonical Wnt signaling is active in dendritic bundle-forming layer II pyramidal neurons of the rat retrosplenial cortex during dendritic branching and spine formation. Transient downregulation of canonical Wnt transcriptional activity during the early postnatal period irreversibly reduces dendritic arbor architecture, leading to long-lasting deficits in spatial exploration and/or navigation and spatial memory in the adult. During the late phase of dendritogenesis, canonical Wnt-dependent transcription regulates spine formation and maturation. We identify neurotrophin-3 as canonical Wnt target gene in regulating dendritogenesis. Our findings demonstrate how temporary imbalance in canonical Wnt signaling during specific time windows can result in irreversible dendritic defects, leading to abnormal behavior in the adult.
VIALE, Beatrice,
et al. Transient deregulation of canonical Wnt signaling in developing
pyramidal neurons leads to dendritic defects and impaired behavior.
Cell Reports, 2019, vol.
27, no. 5, p. 1487-1502.e6
PMID : 31042475
DOI : 10.1016/j.celrep.2019.04.026
Available at:
http://archive-ouverte.unige.ch/unige:126064
Disclaimer: layout of this document may differ from the published version.
1 / 1
Cell Reports, Volume27
Supplemental Information
Transient Deregulation of Canonical Wnt Signaling in Developing Pyramidal Neurons Leads to
Dendritic Defects and Impaired Behavior
Beatrice Viale, Lin Song, Volodymyr Petrenko, Anne-Laure Wenger Combremont, Alessandro Contestabile, Riccardo Bocchi, Patrick Salmon, Alan Carleton, Lijia An, Laszlo Vutskits, and Jozsef Zoltan Kiss
Branch order
****
Apical dendrites
1 13
Number of branches 00 5 10 15
2 3 4 10 12
***
9
****
8 7
**
6
5 11
C Basal dendrites
Branch order 0
2 4 8
Number of branches 0
6 7 1 2
***
3
**
4 5 6
Figure S1
β-cateninControl
GFP LY
A GFP/β-catenin
E18 E21
LY analysisP21
β-catenin
pTF hPGK GFP E2A rtTA GFP
UBI
****
Apical dendrites branch points
0 100 180
120 140 160
0 100 160
Apical dendrites length (µm)
120 140
**
β-catenin Control
Basal dendrites length (µm) **
0 100 160
120 140
***
Basal dendrites branch points
0 100 180
120 140 160
B
0 10
Basal dendrites branch points
8
4 ns
****
6
2 ns
**
0 800
Basal dendrites length (µm)
600 400 200
***
ns
**
0 50
Apical dendrites branch points
40 30 20 10
**
ns
**
0 2500 2000 1500 1000 500
**
Apical dendrites length (µm)
dnTCF4/∆DVL2 E21
LY analysisP21 E18
D
∆DVL2dnTCF4
E
∆DVL2
pTF hPGK GFP E2A rtTA dnTCF4
pTF hPGK GFP E2A rtTA
∆DVL2 dnTCF4 Control
G
F GFP/dnTCF4
E21
LY analysisP21 E18
S1
****
0 3000
1000 2000
Apical dendrites length (µm) **
0 1500
Basal dendrites length (µm)
1000
500
****
0 15
Basal dendrites branch points10 5
Control S1 dnTCF4 S1
****
0 30
Apical dendrites branch points20 10
GFP UBI
dnTCF4
pTF hPGK GFP E2A rtTA
dnTCF4Control
S1
Figure S1. Wnt GOF increases dendritic complexity in RSC. Relative to Figure 1.
(A) Timeline of the experiment and plasmid used. Coronal slices of P21 brains electro- porated at E18 with control plasmid (top panel) or β-catenin (expressed from E21) (lower panel) showing cells iontophoretically injected with Lucifer Yellow (LY) and their neurolucida reconstructions.
(B) Quantification of total length and number of branch points in apical and basal den- drites of β-catenin-expressing neurons as a percentage of control cells at P21. Histo- grams represent mean ± SEM of n=4 brains for control, n=4 brains for β-catenin from at least 2 independent experiments (Student’s t-test).
Absolute values:
AD, apical dendrites; BD, basal dendrites
(C) Quantification of the total branch number for each branch order of control and β-cat- enin-expressing neurons at P21. Data represent mean ± SEM of n=4 brains for control, n=4 brains for β-catenin from at least 2 independent experiments (Two-way ANOVA followed by Bonferroni post-test).
(D) Timeline of the experiment. Neurolucida reconstructions of dnTCF4-expressing neu- rons and ΔDVL2-expressing neurons at P21 in RSC.
(E) Plasmids used. Quantification of total length and number of branch points in apical and basal dendrites of control, dnTCF4-expressing neurons and ΔDVL2-expressing neurons at P21. Histograms represent mean ± SEM of n=4 brains for control, n=5 brains for dnTCF4, n=3 brains for ΔDVL2 from at least 2 independent experiments (One-way ANOVA followed by Bonferroni post-test).
(F) Timeline of the experiment. Neurolucida reconstructions of control and dnTCF4-ex- pressing neurons at P21 in S1.
(G) Plasmids used. Quantification of total length and number of branch points in apical and basal dendrites of control and dnTCF4-expressing neurons at P21 in S1. Histo- grams represent mean ± SEM of n=6 brains for control, n=6 brains for dnTCF4 from at least 2 independent experiments (Student’s t-test).
** P<0.01; *** P<0.001; **** P<0.0001
Scale bar = 200 μm (A top panel), 50 μm (A bottom panel and C), 40 μm (F), 20 μm (A right panel, D)
Control β-catenin
Mean SEM Mean SEM
AD length (µm) 1097 31.63 1503 79.15 AD branch points 17.42 0.9335 25.02 0.6377 BD length (µm) 316.7 43.9 411.5 19.06 BD branch points 3.947 0.6078 5.798 0.3096
Figure S2. Dendritic defect induced by Wnt LOF does not depend on altered cell fate nor perturbed migration. Relative to Figure 2.
(A) Timeline of the experiment. Canonical Wnt activity reporter (TOPdGFP) and TOM co-expression at P7 and P11, corresponding to 0 and 4 days after the end of dnTCF4 induction, respectively. Quantification of TOPdGFP/TOM intensity ratio (dnTFC4/con- trol) shows that canonical Wnt activity levels are restored after 4 days without dnTCF4 induction. Histograms represent mean ± SEM of n=4 brains for P7, n=4 brains for P11 from at 2 independent experiments (Student’s t-test).
(B) Timeline of the experiment. Coronal slices of P56 brains electroporated at E18 with tomato (TOM) or dnTCF4 (GFP) showing positive SatB2 immunofluorescent staining.
(C) Timeline of the experiment. Neurolucida reconstructions of control cells and neurons expressing dnTCF4 from P4 to P7 analyzed at P21.
(D) Quantification of total length and number of branch points in apical and basal den- drites of control and dnTCF4-expressing neurons at P21, with dnTCF4 expression restricted from P4 to P7. Histograms represent mean ± SEM of n=11 brains for control, n=6 brains for dnTCF4 from at least 2 independent experiments (Student’s t-test).
* P<0.05; ** P<0.01; *** P<0.001; **** P<0.0001 Scale bar = 20 μm (A, B); 50 μm (C)
B
ControldnTCF4
P56 TOM
GFP
SatB2 TOM/dnTCF4
E18 E21 P56
immuno P7
C GFP/dnTCF4
E18 P4
LY analysisP21 P7
dnTCF4 P4-P7 Control
Figure S2
D
****
0 1500
500 1000
Apical dendrites length (µm) ****
0 20
Apical dendrites branch points
15 10 5
0 500
Basal dendrites length (µm)
400 300 200 100
**
0 5
Basal dendrites branch points
4 3 2 1
***
Control dnTCF4
A
dnTCF4 Control
P7P11
TOM/dnTCF4+TOPdGFP
E18 E21 P11
analysis analysisP7
*
0 1.5
0.5 1
Wnt activity ratio (dnTCF4/control) P7 P11
Figure S3. The early phase of dendritogenesis is activity-independent. Relative to Figure 2.
(A) Timeline of the experiment and plasmid used. Neurolucida reconstructions of control cells and neurons expressing Kir2.1 from E21 to P7 analyzed at P7. Quantification of total length and number of branch points in apical and basal dendrites of control and Kir2.1-expressing neurons at P7. Histograms represent mean ± SEM of n=10 brains for
ns
0 15
Basal dendrites branch points10 5 ns
0 400
Basal dendrites length (µm)
300 200 100 ns
0 15
Apical dendrites branch points10 5 ns
0 600
200 400
Apical dendrites length (µm)
Control Kir2.1
A GFP/Kir2.1
E21
LY analysisP7 E18
GFP/hM3Dq+CNO P0
LY analysisP7 E18
B Control
hM3Dq+CNO
ns
0 15
Basal dendrites branch points10 5 ns
0 800
200 400
Apical dendrites length (µm)
600
ns
0 15
Apical dendrites branch points10 5
ns
0 400
Basal dendrites length (µm)
300 200 100
Control dnTCF4
dnTCF4+hM3Dq+CNO
ns
0 600
200 400
Apical dendrites length (µm) *** **
0 15
Apical dendrites branch points10 5
* *ns
0 400
Basal dendrites length (µm)
300 200 100
nsns
0 15
Basal dendrites branch points10 5 GFP/dnTCF4/dnTCF+hM3Dq+CNO
E18 P0
LY analysisP7
C
Figure S3
hM3Dq
pTF hPGK GFP E2A rtTA Kir2.1
pTF hPGK GFP E2A rtTA
Kir2.1
Control hM3Dq+CNO
dnTCF4+hM3Dq+CNO dnTCF4
control, n=5 brains for Kir2.1 from at least 2 independent experiments (Student’s t-test).
(B) Timeline of the experiment and plasmid used. Neurolucida reconstructions of hM3Dq-expressing neurons at P7, with hM3Dq activation (CNO injections) from P0 to P7. Quantification of total length and number of branch points in apical and basal den- drites of control and hM3Dq-expressing neurons at P7. Histograms represent mean ± SEM of n=10 brains for control, n=4 brains for hM3Dq+CNO from at least 2 independent experiments (Student’s t-test).
(C) Timeline of the experiment. Neurolucida reconstructions of neurons expressing dnTCF4 from E21 to P7 and cells that co-express dnTCF4 and hM3Dq (activated from P0 to P7) (dnTCF4+hM3Dq+CNO) at P7. Quantification of total length and number of branch points in apical and basal dendrites of control, dnTCF4-expressing neurons and cells that co-express dnTCF4 and hM3Dq (dnTCF4+hM3Dq+CNO) at P7. Histograms represent mean ± SEM of n=10 brains for control, n=4 brains for dnTCF4, n=3 brains for dnTCF4+hM3Dq+CNO from at least 2 independent experiments (One-way ANOVA followed by Bonferroni post-test).
Controls in histograms in A, B, C, and dnTCF4 in histograms in C are the same as in Figure 4G.
* P<0.05; ** P<0.01; *** P<0.001 Scale bar = 20 μm (A, B, C).
Figure S4. During the late phase, spine development is influenced by activity. Rel- ative to Figure 3.
(A) Timeline of the experiment. Confocal images of representative apical dendritic seg- ments of P30 control cells and neurons expressing hM3Dq (activated from P21 to P30).
Quantification of number of spines per μm of control cells and neurons expressing hM3Dq activated with CNO from P21 to P30. Histograms represent mean ± SEM of n=6 brains for control (1810 spines for apical dendrite; 1110 spines for basal dendrites), n=3 brains for hM3Dq+CNO (1645 spines for apical dendrite; 1298 spines for basal den- drites) from at least 2 independent experiments (Student’s t-test).
(B) Timeline of the experiment. Confocal images of representative apical dendritic seg- ments of P30 neurons expressing dnTCF4 from P21 to P30 and cells that co-express dnTCF4 and hM3Dq (activated from P21 to P30). Quantification of number of spines per μm of control, dnTCF4-expressing neurons and cells that co-express dnTCF4 and hM3Dq (dnTCF4+hM3Dq+CNO) at P30 (activation of dnTCF4 and hM3Dq from P21 to P30). Histograms represent mean ± SEM of n=6 brains for control (1810 spines for apical dendrite; 1110 spines for basal dendrites), n=6 brains for dnTCF4 (1356 spines for apical dendrite; 1061 spines for basal dendrites), n=3 brains for dnTCF4+hM3Dq+C- NO (608 spines for apical dendrite; 396 spines for basal dendrites) from at least 2 inde- pendent experiments (One-way ANOVA followed by Bonferroni post-test).
Image of control in A and image of dnTCF4 in B are the same as in Figure 3C.
* P<0.05; *** P<0.001 Scale bar = 5 μm (A, B).
Control dnTCF4
dnTCF4+hM3Dq+CNO ns
Apical dendrites
0 1.0
0.6 0.8
Nb of spines/µm
0.4 0.2
* nsns
Basal dendrites
0 1.0
0.6 0.8
Nb of spines/µm
0.4 0.2 GFP/hM3Dq+CNO
P21
LY analysisP30 E18
A GFP/dnTCF4/dnTCF+hM3Dq+CNO
P21 E18
LY analysisP30
B
Control hM3Dq+CNO
***
0 1.5
1.0
Nb of spines/µm
0.5
Basal dendrites
***
0 1.5
1.0
Nb of spines/µm
0.5
Apical dendrites
Figure S4
hM3Dq
Apical Dendrite Control
Apical Dendrite dnTCF4
hM3Dq dnTCF4
Figure S5. In vitro validation of shNT3 and NT3 LOF between P7 and P15. Relative to Figure 4.
(A) Timeline of the experiment and plasmids used. Quantification of NT3 mRNA expres- sion of HEK293T/17 cell cultures transfected at DIV1 with either a plasmid encoding for rat NT3 (NT3) alone or with both NT3 plasmid and dox-inducible shNT3 plasmid (NT3+shNT3 –dox when the expression of the shRNA was not induced; NT3+shNT3 +dox when shRNA expression was induced). Histogram represents mean ± SEM of n=3 experiments (One-way ANOVA followed by Bonferroni post-test).
(B) Timeline of the experiment. Neurolucida reconstructions of control cells and neurons expressing shNT3 from P7 to P15 analyzed at P15.
(C) Quantification of total length and number of branch points in apical and basal den- drites of control and shNT3-expressing neurons at P15. Histograms represent mean ± SEM of n=3 brains for control, n=3 brains for shNT3 from at least 2 independent experi- ments (Student’s t-test).
*** P<0.001; **** P<0.0001 Scale bar = 20 μm (B).
NT3/NT3+shNT3dox
qPCRDIV4 DIV1
A
Figure S5
shNT3
pTF hPGK GFP E2A rtTA NT3
UBI hPGK GFP
ns****
Relative NT3 mRNA expression 0 1.0
0.5
***
85% 19%
NT3
NT3+shNT3 -dox NT3+shNT3 +dox
Control shNT3 GFP/shNT3
E21
LY analysisP15 E18
B C
shNT3 Control
P7
0 500 1000 1500
Apical dendrites length (µm) ns
ns
0 25
Apical dendrites branch points
10 5 20 15
ns
0 400
Basal dendrites length (µm)
300 200 100
ns
0 15
Basal dendrites branch points10 5
50 pA 500 ms
dnTCF4
50 pA
Control
500 ms
Figure S6. The electrophysiological properties of dnTCF4 neurons are normal.
Relative to Figure 6.
(A-J) Comparison of electrophysiological properties as well as spontaneous and minia- ture postsynaptic currents of RSC layer II GFP+ neurons from acute slices of control and dnTCF4 electroporated animals.
C P21 sEPSCs
dnTCF4
500 ms 10 pA
Control
500 ms 10 pA
ns
Amplitude(pA)
0 5 10 ns
0 200 400 600
IEI(ms)
D P21 sIPSCs ns
0 20 40
Amplitude(pA)
ns
0 500 1000
IEI(ms)
50 pA 500 ms
Control
50 pA 500 ms
dnTCF4
A
Figure S6
ns
0 1000 1500 2000
IEI(ms)
500
ns
Amplitude(pA)
0 20 40 60 ns 80
0 200 400 600
IEI(ms)
ns
Amplitude(pA)
0 10
20 ns
0 500 1000 1500
IEI(ms)
ns
Amplitude(pA)
0 10 ns 20
0 500 1000
IEI(ms)
ns
Amplitude(pA)
0 20 40 60 80 ns
0 50 100 150
Rheobase (pA)
ns
0 20 40 60 80
Capacitance (pF)
ns
0 0.1 0.2 0.3 0.4
Input resistance (GΩ)
ns
-80 -70 -50
-60
RMP (mV)
Control dnTCF4
P21 B
E P21 mEPSCs F P21 mIPSCs
ns
0 500 1000 1500
IEI(ms)
ns
0 1000 1500 2000
500
IEI(ms)
ns
0 20 40 60 80
Amplitude(pA)
ns
0 10 20
Amplitude(pA)
G P45 sEPSCs H P45 sIPSCs I P45 mEPSCs J P45 mIPSCs P45
ns
0 0.1 0.2 0.3 0.4
Input resistance (GΩ)
ns
-80 -70 -50
-60
RMP (mV)
ns
0 10 20 30 40
Capacitance (pF)
ns
0 50 100 150
Rheobase (pA)
10 pA 500 ms
dnTCF4
10 pA 500 ms
Control
(A) Graphs comparing typical electrophysiological properties between dnTCF4-ex- pressing neurons and control cells at P21. Histograms represent mean + SEM of n=12 and 9 cells for control and dnTCF4, respectively, (Mann-Whitney test). RMP= resting membrane potential.
(B) Same as in (A) at P45. Histograms represent mean + SEM of n=7 cells/condition, (Mann-Whitney test).
(C) Example of spontaneous EPSCs recording together with quantification of interev- ent-interval (IEI) and amplitude of the events at P21. Histograms represent mean + SEM of n=6 cells/condition, (Mann-Whitney test).
(D) Same as C but for spontaneous inhibitory postsynaptic currents. Histograms repre- sent mean + SEM of n=7 cells/condition, (Mann-Whitney test).
(E) Example of miniature EPSCs recording together with quantification of interevent-in- terval (IEI) and amplitude of the events at P21. Histograms represent mean + SEM of n=8 cells/condition, (Mann-Whitney test).
(F) Same as E but for miniature IPSCs. Histograms represent mean + SEM of n=6 and 8 cells for control and dnTCF4, respectively (Mann-Whitney test).
(G) Quantification of interevent-interval and amplitude of spontaneous EPSCs at P45.
Histograms represent mean + SEM of n=7 cells/condition, (Mann-Whitney test).
(H) Same as G for spontaneous IPSCs at P45. Histograms represent mean + SEM of n=5 and 6 cells for control and dnTCF4, respectively, (Mann-Whitney test).
(I) Quantification of interevent-interval and amplitude of miniature EPSCs at P45. Histo- grams represent mean + SEM of n=7 cells/condition, (Mann-Whitney test).
(J) Same as I for miniature IPSCs at P45. Histograms represent mean + SEM of n=7 and 8 cells for control and dnTCF4, respectively, (Mann-Whitney test).
Figure S7. dnTCF4 electroporated animals show no defect in motor activity. Rela- tive to Figure 7.
(A) Extent of the electroporated area for behavioral experiments.
(B) Barnes Maze Test showing no difference in number of errors/latency and in percent of time spent in the target quadrant during the habituation phase. Histograms represent mean ± SEM of n=24 animals for control, n=18 animals for dnTCF4 (Student’s t-test).
(C) Cylinder Test showing no motor defect. Histogram represents mean ± SEM of n=31 animals for control, n=24 animals for dnTCF4 (Student’s t-test).
(D) Footprint Test showing no motor defect. Histograms represent mean ± SEM of n=23 animals for control, n=24 animals for dnTCF4 (Student’s t-test).
ns
nb rearing/min
0 20 15 10 5
caudal
rostral
dnTCF4 Control
Footprint Test Cylinder Test
C
stride length
step width stance length
D A
ns
Nb errors/latency
0 0.5
0.3 0.4
0.2 0.1
ns
% time in target quadrant 0 200 150 100 50
Habituation
B
Barnes Maze Test
Forelimb
ns
Length/weight (cm/g)
0 0.02
0.01
Step Width
ns
Length/weight (cm/g)
0 0.10
0.04
Stride Length
0.02 0.08 0.06
ns
Length/weight (cm/g)
0 0.05
0.02
Stance Length
0.01 0.04 0.03
Figure S7
Control dnTCF4