Genetic dissection of the function of hindbrain axonal commissures

Abstract

In Bilateria, many axons cross the midline of the central nervous system, forming well-defined commissures. Whereas in mammals the functions of commissures in the forebrain and in the visual system are well established, functions at other axial levels are less clearly understood. Here, we have dissected the function of several hindbrain commissures using genetic methods. By taking advantage of multiple Cre transgenic lines, we have induced site-specific deletions of the Robo3 receptor. These lines developed with the disruption of specific commissures in the sensory, motor, and sensorimotor systems, resulting in severe and permanent functional deficits. We show that mice with severely reduced commissures in rhombomeres 5 and 3 have abnormal lateral eye movements and auditory brainstem responses, respectively, whereas mice with a primarily uncrossed climbing fiber/Purkinje cell projection are strongly ataxic. Surprisingly, although rerouted axons remain ipsilateral, they still project to their appropriate neuronal targets. Moreover, some Cre;Robo3 lines represent potential models that can be used to study human syndromes, including horizontal gaze palsy with progressive scoliosis (HGPPS). To our knowledge, this study is one of the first to link defects in commissural axon guidance with specific cellular and behavioral phenotypes.

Figure 1. Rhombomere-specific deletion of Robo3.

Whole-mount control (A, C, and E) or Krox20::cre;Robo3^lox/lox (B, D, and F) E12 embryos hybridized with a Robo3 riboprobe covering exons 12–14 (A and B) or immunostained with anti-Robo3 (C and D) or anti-neurofilament (E and F) antibodies. (B) In Krox20::cre;Robo3^lox/lox embryos, Robo3^exon12-14 transcripts are not expressed in rhombomeres 3 (r3) and 5 (r5). (C and D) Likewise, there is a severe reduction of Robo3 immunoreactive commissural axons in r3 and r5. (E and F) Anti-neurofilament immunostaining confirms the strong reduction of commissures in r3 and r5. The arrowheads in (F) indicate axons that abnormally follow the midline. Scale bars represent 100 µm.

Figure 2. Rhombomere-specific deletion of commissures.

Coronal sections at the level of r3, r5, and r6 of E11 control (A–C′) or Krox20::cre;Robo3^lox/lox (D–F′) embryos immunostained with antibodies against Robo3, neurofilament, and Hb9. The midline is indicated by an arrow, and the Hb9-positive abducens motor neurons by an asterisk. In r3 and r5, there is an almost complete absence of Robo3 and neurofilament-positive commissural axons in Krox20::cre;Robo3^lox/lox (D–E′) compared to controls (A–B′). The density of neurofilament-positive axons is also strongly reduced. By contrast, there is no obvious reduction of commissural axons or Robo3 expression in r6 (C, C′, F, and F′). Note that at E11, abducens shape and position are not altered in mutants compared to controls (see Figure 3 for later stages). VII, migrating facial neurons and facial nerve. Scale bars represent 60 µm.

Figure 3. Normal projection of abducens motor axons in Robo3-deficient mice.

(A to I) show coronal hindbrain sections at the level of the abducens nucleus. (A and B) Robo3 transcripts are detected in the abducens nuclei of E12 (A) and E14 (B) embryos. The arrow in (B) points to the facial nerve. (C) abducens neurons coexpress BEN and GFP in Robo3^+/− E14 embryo. The floor plate (asterisk) also expresses BEN. (D) Hb9/GFP double labeling in E15 Robo3^+/− embryo. Note that some GFP+ cells (arrowhead) do not express Hb9. (E) abducens neurons (VI) and axons (arrow) express GFP in E14 Hb9::GFP transgenic embryo. (F–I) The abducens nucleus has an abnormal shape and is closer to the floor plate in Robo3^−/− (F) and Krox20::cre;Robo3^lox/lox (G) E14 embryos (compare with [E]). The arrow in (F) points to abducens axons. (H and I) This abnormal shape and position are also observed in adult animals with ChAT/Hb9 double immunostaining. (J to L) Coronal sections of P0 mouse head at eye cup (ec) level. GFP-positive abducens axons (arrows) still contact the lateral rectus muscle in Robo3-null embryo (K) and Krox20::cre;Robo3^lox/lox mutant (L). Scale bars represent 100 µm, except in (D), where it indicates 50 µm.

Figure 4. Reduced internuclear commissure in Robo3 knockout mice.

(A to I) show coronal hindbrain sections at the level of the abducens nucleus, visualized by Hb9 immunostaining (in A, D, E, F, and H). (A, B, and C) illustrate the projection of abducens axons (arrowheads) across the midline (dashed line) in Robo3^+/− E13 embryos. Some GFP+ axons originate from the abducens nucleus (VI) and are immunoreactive for Robo3 (B and C). (D) The internuclear commissure (arrowhead) is also observed in P0 controls, following DiI injection at the level of the oculomotor nucleus III. (E) This commissure is almost completely absent in P0 Krox20::cre;Robo3^lox/lox mice (arrowhead). (VIIn): Genu of facial nerve. (F to I) At E15, many neurofilament+ axons cross the midline at the VI level (arrowheads) in control embryo (F and G), whereas they are rare in Krox20::cre;Robo3^lox/lox embryo (H and I). Scale bars represent 100 µm, except in (G and I), where they indicate 50 µm.

Figure 5. Impaired horizontal compensatory eye movements in Krox20::cre;Robo3^lox/lox mice.

(A) During optokinetic stimulation horizontal gains are reduced most prominently at the lower frequencies in in Krox20::cre;Robo3^lox/lox mice (p = 0.043 ANOVA; n = 6 versus n = 4 for controls; see Table S1). (B) At higher frequencies, the VOR is severely impaired (p = 0.001 versus control mice curve; ANOVA for repeated measurements; Table S1), confirming the importance of the commissural connections in large-amplitude eye movements. (C) When horizontal visual and vestibular inputs are combined in the VVOR (visual vestibulo-ocular reflex), it results in lower gains over the entire range of frequencies tested (p = 0.004 versus control mice curve; ANOVA for repeated measurements). (D) OKR deficits are strongly correlated to the amplitude of stimulation. (E–G) In marked contrast, in the vertical plane, no significant differences were observed in OKR (E), VOR (F), or VVOR (G), supporting the concept that primarily horizontal eye movements require the presence of commissural connections (see Table S1 for all statistics). Error bars indicate standard error of the mean. Results were obtained from four control and six Krox20::cre;Robo3^lox/lox mice.

Figure 6. Uncrossed aVCN-MNTB projections and abnormal ABRs in Krox20::cre;Robo3^lox/lox mice.

(A to C) Robo3 is expressed by neurons of the cochlear nucleus (CN). Coronal sections of a E14 embryo (A) or side view of whole-mount E14 embryos (B and C) hybridized with a Robo3 probe (A and B) or labeled with anti-Robo3 antibodies (C). The arrowheads in (B) indicate migrating pontine neurons. The arrow in (C) points to cochlear axons. (D–G) Coronal sections of P12 mice injected with DiI in the cochlear nucleus (Hoechst counterstaining). In control (D and E), DiI-labeled axons end in the ipsilateral superior olive (SO) and the contralateral MNTB, whereas in Krox20::cre;Robo3^lox/lox mutant (F and G), all axons project ipsilaterally. The arrow in (D and F) indicates the midline. (E and G) are high magnification pictures of the MNTB showing DiI-labeled calyces of Held (arrows). (H) ABRs collected from the ipsilateral or contralateral mastoid electrode in response to 60 dB SPL clicks. Ipsilaterally, only three waves were observed in mutant instead of four in controls. Mutant wave III exhibited a mean latency of about 4.3 ms, much longer than that of control wave III (3.6 ms; n = 13 controls versus n = 12 mutants, average latency difference 0.77 ms for wave III; p<0.0001; unpaired Student t-test), yet too short for matching the latency of wave IV in controls. Contralaterally, in mutants, the mean latency of IVc was 0.42 ms longer than in controls (n = 12 controls versus n = 8 mutants; p = 0.006; unpaired Student t-test). The asterisk marks a recording artifact (see Materials and Methods). Scale bars represent 300 µm except in (A), where it indicates 150 µm, and in (E and G), where it indicates 50 µm.

Figure 7. Uncrossed inferior olivary axons cause ataxic gait.

(A) Ptf1a::cre;Robo3^lox/lox mice are severely ataxic, which is demonstrated by a very short latency to fall in the Rotarod test (n = 7 mutants versus n = 9 for controls). (B) Image of a 4-mo-old Ptf1a::cre;Robo3^lox/lox mouse displaying an ataxic gait. (C) The step time of Ptf1a::cre;Robo3^lox/lox mice (light blue; n = 6) on the Erasmus Ladder is longer than in controls (violet and red; n = 4; p<0.001; multiple comparisons) or in Lurcher mice (orange; p<0.001; multiple comparisons). (D) Ptf1a::cre;Robo3^lox/lox mice (light blue) also show significantly less successful trials per session than Lurcher mice (orange) (p<0.001 for all sessions; Mann-Whitney U-test, one-tailed). A trial was defined as successful if the mice were able to walk on the ladder without disruption (twisting, turning, walking backwards, etc.). (E to J) The morphology of the cerebellum in controls (E to G) and Ptf1a::cre;Robo3^lox/lox mice (H to J) is comparable, and in both cases, their climbing fibers (the terminal arbors of inferior olivary axons) innervate the cerebellar cortex. (F and I) The size of the cerebellum and its foliation are similar as shown on sagittal sections of P32 control and Ptf1a::cre;Robo3^lox/lox mice labeled with anti-calbindin antibodies and Hoechst. (G and J) VGLUT2-positive climbing fibers properly innervate CaBP+ Purkinje cell dendrites. (K and L) Coronal sections of P1 mice with unilateral injection of DiI in the cerebellum. The arrow marks the midline. In control (K), DiI-labeled inferior olivary (IO) neurons are exclusively found on the contralateral side, whereas in Ptf1a::cre;Robo3^lox/lox mice (L), most IO neurons are traced on the ipsilateral side, and their axons do not cross the midline. Scale bars represent 500 µm except in (G and J), where they indicate 25 µm, and in (K and L), where they indicate 150 µm.

Figure 8. Analysis of Ptf1a::cre;Robo3^lox/lox;Tau^mGFP mice.

(A and B) Coronal sections of the hindbrain at the level of inferior olive of E16 Ptf1a::cre;Tau^mGFP embryo labeled with anti-βgal and Brn3.2 antibodies. Some Brn3.2-positive neurons in the inferior olive do not express the βgal (arrowheads and inset). (C–L) Coronal sections of the spinal cord at the level of the forelimbs in E13 Ptf1a::cre;Robo3^lox/+;Tau^mGFP (C–E, K, and L) or Ptf1a::cre;Robo3^lox/lox;Tau^mGFP (F–H, I, and J) embryos labeled with anti-GFP and anti-Robo3 antibodies. Most GFP-positive axons are in the dorsal spinal cord, and only a small subset of GFP-positive axons (short arrows) cross the floor plate in Ptf1a::cre;Robo3^lox/+;Tau^mGFP (C and K) but do not express Robo3 (E and L). This subset of GFP-positive commissural axons is still observed in Ptf1a::cre;Robo3^lox/lox;Tau^mGFP embryos (F and J). Scale bars represent 100 µm except in (B and I–L), where they indicate 50 µm.