Fig 1.
Brain Tumor is a positive regulator of midline crossing.
(A-F) Stage 15–16 embryos of the indicated genotypes carrying eg-GAL4 and UAS-tauMycGFP transgenes, stained with anti-GFP antibody. Anti-GFP reveals cell bodies and axons of the eagle neurons (EG and EW). Anterior is up in all images. Scale bar represents 10μm (A). EG neurons project through the anterior commissure of each segment, while EW neurons project through the posterior commissure. Arrowheads indicate segments with non-crossing EW axons. (A) In wild-type embryos EW axons cross in the posterior commissure in 100% of segments. (B) In fra mutants EW axons fail to cross in 36% of segments (arrowheads). (C) EW axons fail to cross in 28% of segments when UAS-FraΔC is selectively expressed in eagle neurons. (D) In a FraΔC background the heterozygosity for brat enhances the EW crossing defects to 53%. (E) Complete loss of brat enhances the EW crossing defect to 69% of segments in FraΔC background. (F) EW crossing defects in brat/brat; FraΔC embryos are rescued (69% versus 41%) when UAS-Brat is expressed in eagle neurons. (G) Quantification of EW midline crossing defects in the genotypes shown in (B-F). Df (2L) Exel8040 is a chromosomal deficiency containing brat. Data are presented as mean ± SEM. 20 embryos were scored for each genotype. Significance was assessed by multiple comparisons using ANOVA (∗∗∗p < 0.001). (H) Schematic diagrams of the EW axon trajectories observed in each genotype; the EW axons can cross the midline (Cross), grow ipsilaterally (Ipsi) or stall (Stall). (I, K, M) When UAS-FraΔC is selectively expressed in eagle neurons, 72% of the EW axons cross the midline (I), 22% grow ipsilaterally (K) and 6% stall (M). (J,L,N) Heterozygosity for brat in a FraΔC background enhances the EW crossing defects, 47% of the EW cross the midline (J), 23% grow ipsilaterally (L) and 30% stall (N). (O) Quantification of the distribution of the EW axon trajectories in the genotypes shown in (I-N). The enhanced EW crossing defects in brat/+; FraΔC embryos are rescued when UAS-Brat is expressed in eagle neurons. Data are presented as mean ± SEM. 20 embryos were scored for each genotype. Significance was assessed using Chi-squared test (****p < 0.0001).
Fig 2.
Brat acts in parallel to the Netrin-Fra pathway.
(A-D) Stage 15–16 embryos of the indicated genotypes carrying eg-GAL4 and UAS-tauMycGFP transgenes, stained with anti-GFP (green) (A-D) or anti-HRP (magenta) (A’-D’) antibodies. Anti-GFP labels cell bodies and axons of the eagle neurons (EG and EW), Anti-HRP reveals all of the CNS axons. Scale bar represents 10μm (A). Arrowheads indicate segments with non-crossing EW axons (A-D) or thin commissures (A’-D’). (A) EW neurons cross in the posterior commissure in 100% of segments in wild-type embryos. (A’) In every segment thick anterior and posterior commissures are formed as axons cross the midline. (B) In fra mutants EW neurons fail to cross in 36% of segments. (B’) fra mutants show thinner commissures. (C) and (C’) brat homozygous mutants show no obvious signs of commissural guidance defects: EW neurons fail to cross in only 4% of segments. (D) In fra, brat double mutants EW axons fail to cross the midline in 56% of segments. (D’) fra, brat double mutants also show thinner commissures. (E) Quantification of EW midline crossing defects in the genotypes shown in (B-D). Data are presented as mean ± SEM. 20 embryos were scored for each genotype. Significance was assessed by multiple comparisons using ANOVA (∗∗∗∗p < 0.0001). (F) Schematic diagrams of the EW axon trajectories observed in each genotype; the EW axons can cross the midline (Cross), grow ipsilaterally (Ipsi) or stall (Stall). (G, I, K) In fra mutants, 55% of the EW axons cross the midline (G), 29% grow ipsilaterally (I) and 16% remain stalled (K). (H, J, L) In fra, brat double mutants, 38% of the EW axons cross the midline (H), 28% grow ipsilaterally (J) and 34% stall (L). (M) Quantification of the distribution of the EW axon trajectories in the genotypes shown in (G-L). Data are presented as mean ± SEM. 20 embryos were scored for each genotype. Significance was assessed using Chi-squared test (****p < 0.0001).
Fig 3.
Brat acts independently of the Nanos/Pumilio complex and of d4EHP.
(A-D) Stage 15–16 embryos of the indicated genotype carrying eg-GAL4 and UAS-tauMycGFP transgenes, stained with anti-GFP antibody. Anti-GFP labels cell bodies and axons of the eagle neurons (EG and EW). Scale bar represents 10μm (A). Arrowheads indicate segments with non-crossing EW axons. (A) Heterozygosity for brat enhances the EW crossing defects to 50% in a FraΔC background. (B-D) EW crossing defects in heterozygous brat mutants expressing FraΔC are rescued when (B) UAS-Brat (50% versus 34%), (C) UAS-BratGD (50% versus 36%) or (D) UAS-BratRD (50% versus 39%) are expressed in eagle neurons. (E) Schematic representation of Brat protein and its different domains. The G774D and R837D point mutations are indicated with arrows. (F) Quantification of EW midline crossing defects in the genotypes shown in (A-D). Data are presented as mean ± SEM. 20 embryos were scored for each genotype. Significance was assessed by multiple comparisons using ANOVA (∗∗∗p < 0.001).
Fig 4.
The B-box domains of Brat are required for its midline crossing function.
(A-F) Stage 15–16 embryos of the indicated genotype carrying eg-GAL4 and UAS-tauMycGFP transgenes, stained with anti-GFP antibody. Anti-GFP labels cell bodies and axons of the eagle neurons (EG and EW). Scale bar represents 10μm (A). Arrowheads indicate segments with non-crossing EW axons. (A-C) EW crossing defects in the heterozygous brat mutant expressing FraΔC are rescued when (A) UAS-Brat (50% versus 24%), (B) UAS-BratΔNHL (50% versus 32%) or (C) UAS-BratΔCC (50% versus 29%) are expressed in eagle neurons. (D-F) In the heterozygous brat mutant expressing FraΔC, expression of (D) UAS-BratΔBB, (E) UAS-BratΔBB1 or (F) UAS-BratΔBB2 fail to rescue the EW midline crossing defects (respectively for (D) (E) and (F): 50% versus 54%, 50% versus 40% and 50% versus 37%). (G) Schematic representation of Brat full-length protein and Brat deletion domain mutants used to identify the domain required for midline crossing. (H) Quantification of EW midline crossing defects in the genotypes shown in (A-F). Data are presented as mean ± SEM. 20 embryos were scored for each genotype. Significance was assessed by multiple comparisons using ANOVA (∗∗∗∗p < 0.0001).
Fig 5.
Midline crossing is sensitive to reduced Apc2 and Arm function and does not require Arm transcriptional activity.
(A-D) Stage 15–16 embryos of the indicated genotype carrying eg-GAL4 and UAS-tauMycGFP transgenes, stained with anti-GFP (grey or green) (A-C) or anti-HRP (magenta) (D) antibodies. Anti-GFP labels cell bodies and axons of the eagle neurons (EG and EW), Anti-HRP reveals all of the CNS axons. Scale bar represents 10μm (A). Arrowheads indicate segments with non-crossing EW axons (A-C) or thin commissures (D). (A) In a FraΔC background the heterozygosity for Apc2 enhances the EW crossing defects to 59%. (B) In the embryos double heterozygous for Apc2 and brat expressing UAS-FraΔC selectively in eagle neurons, EW axons fail to cross in the posterior commissure in 72% of segments. (C) In Apc2 and brat double mutant embryos, EW axons fail to cross in the posterior commissure in 20% of segments and show thinner commissures in some segments (D). (E) Quantification of EW midline crossing defects in the genotypes shown in (A-D). Df (2L) Exel6168 is a chromosomal deficiency containing Apc2. Data are presented as mean ± SEM. 20 embryos were scored for each genotype. Significance was assessed by multiple comparisons using ANOVA (∗∗∗∗p< 0.0001). (F) Quantification of EW midline crossing defects in the indicated genotypes. Data are presented as mean ± SEM. 20 embryos were scored for each genotype. Significance was assessed by multiple comparisons using ANOVA (∗∗∗p < 0.001).
Fig 6.
Apc2 expression co-localizes with EB1 in growing axon and cell bodies of Eagle neurons and is reduced in brat mutant embryos.
(A-F’) Stage 13 and 16 embryos carrying eg-GAL4, UAS-Apc2GFP and UAS-EB1RFP transgenes, stained with anti-GFP (green) and anti-RFP (red) antibodies. Anti-GFP and anti-RFP label cell bodies and axons of the eagle neurons (EG and EW). Scale bar represents 10μm (A) or 2μm (C’ and F’). (A-C’) At stage 13, Apc2 and EB1 expression co-localize in the growing axon and the cell body of the Eagle neurons. (D-F’) At stage 16, Apc2 and EB1 expression co-localize in the elongated axon of the Eagle neurons. (G-H’) Stage 15–16 embryos of the indicated genotype carrying eg-GAL4 and UAS-Apc2GFP transgenes, stained with anti-GFP antibodies. Anti-GFP labels cell bodies and axons of the eagle neurons (EG and EW). Scale bar represents 10μm (G) or 5 μm (G’). (G) and (G’) In control embryos the average of the GFP signal intensity reflecting the Apc2 transgene expression, corresponds to 89% in cell bodies and 79% in axons. (H) and (H’) brat homozygous mutant embryos, show a decrease of the GFP signal intensity to 50% in cell bodies and 25% in axons, reflecting a reduction of the Apc2 transgene expression. (I) Quantification of the GFP staining signal intensity shown in (G-H’). Data are presented as mean ± SEM. 10 embryos were scored for each genotype. Significance was assessed using the Student’s t-test (∗∗∗∗p < 0.0001).
Fig 7.
Model for how brain tumor interacts with Apc2 to promote axon growth across the midline.
We propose that Brat maintains Apc2 at the plus-ends of microtubules at the periphery of the growth cone resulting in axon extension across the midline.