Figure 5

A balance between Netrin and Slit/Robo signaling sets the in vivo position of medial longitudinal fasciculus axons. Combined mutants between Netrin1, Robo1, and Robo2. (A-F) Whole-mount open book preparations stained with βIII-tubulin antibody. (Key genotypes with significant phenotypes are shown; several classes of heterozygous combinations did not cause medial longitudinal fasciculus (MLF) shifts, and are not shown). (A-C) Compared to the control MLF (wild type in (A), n = 4), the MLF shifted away from the midline in Netrin1^−/− mutants (B, n = 3). In Robo1/2 mutants, the MLF shifted toward the midline and, in addition, many axons entered and grew longitudinally within the midline (C, n = 4). (D) When Robo1/2 function was reduced in a Netrin1 mutant background, Netrin1^−/−; Robo1^+/−;2^+/−, the MLF maintained a similar dorsal shift (n = 1). (E) When Netrin1 function was reduced in a Robo1/2 mutant background, Netrin1^+/−; Robo1^−/−;2^−/−, the MLF tracts shifted to a near-normal position, and the number of midline axon bundles was reduced (n = 7). (F) A single triple mutant embryo was isolated, showing an MLF pattern similar to wild-type, with neither dorsal or ventral shifts (n = 1). (G) Schematic summary of longitudinal pioneer patterns in key genotypes. (H) Graph of average normalized distance from the midline to the most ventral MLF fibers. All means differed significantly by analysis of variance (P < 10⁻⁷), and by pair-wise t-tests. Scale bar in (F) also applies to all: 100 μm.