Figure 1. Conditional ablation of Fig4 in neurons or OLs leads to CNS hypomyelination.

(A-D) Coronal sections of juvenile (P21-30) mouse forebrain stained with FluoroMyelin Green. (A) Fig4 control mice (harboring at least one Fig4 WT allele), (B) Fig4 germline null mice (Fig4^-/-), (C) Fig4^-/flox,SynCre mice and (D) Fig4^-flox,Olig2Cre mice. Thinning of the corpus callosum and internal capsule (white arrowheads) is observed in Fig4^-/-, Fig4^-/flox,SynCre, and Fig4^-flox,Olig2Cre mice. (A’-D’) Higher magnification images of the corpus callosum. Scale bar (A-D), 1 mm and (A’-D’), 400 µm. (E) Representative Western blots of P21 brain membranes prepared from Fig4^+/+(WT), Fig4^-/-, Fig4^-/flox,SynCre and Fig4^-/flox,Olig2Cre mice probed with antibodies specific for the myelin proteins MAG, CNPase, PLP, and MBP. To control for protein loading, membranes were probed for the neuronal marker class III β-tubulin (βIII Tub). (F-I) Quantification of Western blot signals for MAG, MBP, CNPase, and PLP in Fig4^+/+ (black bars), Fig4^-/- (purple bars), Fig4^-/flox,SynCre (light blue bars), and Fig4^-flox,Olig2Cre (red bars) brain membranes. Quantification of myelin protein signals is normalized to βIII Tub. Relative protein intensities compared to WT brain are shown as mean value ± SEM. For each of the four genotypes, three independent membrane preparations were carried out. One-way ANOVA with multiple comparisons, Dunnett posthoc test; **p<0.01, ***p<0.001 and ****p<0.0001. An independent strategy for OL-specific Fig4 deletion results in a similar phenotype as shown in Figure 1—figure supplement 1. Histochemical staining of brain, spinal cord and dorsal root ganglion tissue sections of Fig4 conditional knock-out mice, as well as Kaplan-Meier plots for Fig4^-/flox,SynCre and Fig4^-flox,Olig2Cre mice are shown in Figure 1—figure supplement 2.

DOI: http://dx.doi.org/10.7554/eLife.13023.003

Figure 1—figure supplement 1. CNS hypomyelination in Fig4^-/flox,PdgfrαCreER mice.

Representative Western blots of (A) P21 forebrain and (B) P21 hindbrain (cerebellum/brainstem) lysates prepared from Fig4 control littermate mice (Fig4^+/flox,PdgfrαCre-ER) and Fig4^-/flox,PdgfrαCreER mutant mice, probed with antibodies specific for the myelin proteins MAG, CNPase and MBP. To control for protein loading, blots were probed for the neuronal marker class III β-tubulin (βIII Tub). (A’ and B’) Quantification of Western blot signals for MAG, CNPase and MBP in (A’) forebrain and (B’) cerebellum/brainstem lysates. Relative protein intensities compared to control tissue are shown as mean value ± SEM. Six pairs of control littermate and Fig4^-/flox,PdgfrαCreER mice were analyzed and quantified. Unpaired Student’s t-test, *p=0.0323 (A', MAG), ***p=0.0006 (A', CNPase), **p=0.0096 (A’, MBP), **p=0.027 (B', MAG), **p=0.038 (B’, CNPase), ***p=0.0004 (B’, MBP). (C and C’) Sagittal sections of P21 forebrain of control littermate (Fig4^+/flox,PdgfrαCreER) and Fig4^-/flox,PdgfrαCreER mutant mice probed for Mbp mRNA expression. (D and D’) Sagittal sections of P21 cerebellum of control littermate (Fig4^+/flox,PdgfrαCreER) and Fig4^-/flox,PdgfrαCreER mutant mice probed for Mbp mRNA expression. (E and E’) Longitudinal optic nerve sections of P21 littermate control and Fig4^-/flox,PdgfraCreER mice probed for Plp1 mRNA expression. Scale bar (C-D’), 500 μm and (E and E’), 200 μm.

DOI: http://dx.doi.org/10.7554/eLife.13023.004

Figure 1—figure supplement 2. Loss of Fig4 in the OL-lineage or neurons differentially affects spongiform degeneration and lifespan.

(A-D’’’) Hematoxylin/eosin stained tissue sections of P30 mouse neocortex, cerebellum, dorsal root ganglion (DRG) and spinal cord ventral horn. Tissue sections of mice with the following genotypes are shown: (A, B, C, D) control mice (Fig4^flox/-), (A’, B’, C’, D’) Fig4 germline null mice (Fig4^-/-), (A”, B”, C”, D”)Fig4^-/flox,SynCre conditional mutants and (A’’’, B’’’, C’’’, D’’’) Fig4^-/flox,Olig2Cre conditional mutants. Most notable are the large vacuolar (sponge-like) structures in different regions of the Fig4^-/- nervous system, including (A’) deep layers of the neocortex, (B’) deep cerebellar nuclei, (C’) DRGs and (D’) ventral horn of the spinal cord. (A”-D”) A milder but similar phenotype is observed in Fig4^-/flox,SynCre mice. (A’’’) In the Fig4^-/flox,Olig2Cre neocortex small vacuoles are observed in all layers of the neocortex. (B’’’ and C’’’) In Fig4^-/flox,Olig2Cre mice deep cerebellar nuclei and DRGs look largely normal. (D’’’) The large vacuoles in the spinal cord ventral horn of Fig4^-/flox,Olig2Cre mice likely represent motoneurons, as the Olig2 promoter is known to drive cre expression in motoneurons and the OL-linage. (E) The Hb9-cre driver line was used for conditional deletion of Fig4 specifically in motoneurons. Toluidine blue stained section of Fig4^-/flox,Hb9Cre ventral horn shows multiple large vacuolar structures within the gray and white matter of the spinal cord. Examples of vacuolar structures are labeled with asterisks. Apparently normal motoneuron profiles are indicated by arrows. (F) Electron micrograph of Fig4^-/flox,Hb9Cre ventral horn with large vacuolar structures labeled by asterisks. Vacuolar structures are mostly devoid of electron-dense material and found in axons surrounded by thin myelin sheaths (arrows). Vacuoles cause peripheral displacement of axoplasm and mitochondria. The arrowhead points to a dystrophic axon with accumulation of numerous smaller vesicles. (G) Viability of Fig4 conditional mutants. Kaplan-Meier plot shows an average life-span of 6 months for Fig4^-/flox,SynCre mice (n = 15), while Fig4^-/flox,Olig2Cre mice (n = 5) are viable for 12–14 months (the oldest mice currently in our colony).

DOI: http://dx.doi.org/10.7554/eLife.13023.005