DEV029330 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Lrig1 and Lrig3 are both present in multiple sites of netrin 1 expression. (A-C) E9 whole embryos. (D-F) Transverse sections through the E12 neural tube. (G-I) Sections through the E16 cochlea. Expression patterns were revealed by in situ hybridization of probes to Lrig1 (A,D,G) and Ntn1 (C,F,I), and by X-gal staining for Lrig3-βgeo activity (B,E,H). Extensive overlap between Lrig1 and Lrig3 in somites (A,B), in the ventricular zone and midline of the neural tube (D,E), and in the inner sulcus of the developing cochlea (asterisk, G,H), suggest that Lrig1 may compensate for the loss of Lrig3. For example, the unique expression of Lrig3 in the branchial arches (asterisk, A,B), structures that give rise to the jaw and other facial features, could explain the craniofacial abnormalities evident in Lrig3 mutants. As regions of Lrig1/3 overlap also coincide with Ntn1 expression in other tissues, Lrig-mediated control of Ntn1 expression might not be limited to inner ear morphogenesis. NT, neural tube; D, dorsal; L, lateral.

• Supplemental Figure S2 -

  Fig. S2. Hearing is normal in Lrig3 mutant mice. Auditory brainstem responses (ABR) to pure tones ranging from 5 kHz to 45 kHz were obtained as previously described (Liberman et al., 2002). Tucker-Davis Technologies System3 software was used to generate the stimulus and to acquire and analyze the auditory brainstem responses. Thresholds to pure tone stimuli were determined by the lowest Sound Pressure Level (SPL) measured in decibels (dB) at which each peak could be detected. Results were analyzed using Student's two-tailed test with a significance level of 0.05. Two ABR recordings were performed on 5-week-old heterozygous (n=5) and mutant (n=5) animals presented with pure tone stimuli at the frequencies indicated (x-axis). Thresholds (y-axis) were defined as the lowest sound pressure level (SPL) that elicited a reponse. There is no significant change in the threshold at any frequency. Data are presented as mean±standard deviation and were analyzed using the Student's t-test, with P<0.05. kHz, kilohertz; dB, decibels.

• Supplemental Figure S3 -

  Fig. S3. Lrig3 is not essential for otic vesicle patterning. (A-D) In situ hybridization for Otx1 (A), Otx2 (B), Dlx5 (C) and Hmx3 (D) on E11.5 Lrig3^+/− (A-D) and Lrig3^−/− (A′-D′) littermates. Sections are in the transverse plane, with dorsal upwards and lateral towards the right. The otic vesicle is outlined in A. The region that will develop as the lateral canal is indicated with a bracket in each panel. Note that, at this stage, the lateral pouch has not yet grown out and is identified by gene expression patterns, not by morphology. All markers are expressed normally in homozygotes. ED, endolymphatic duct. Scale bar: 100 µm.

• Supplemental Figure S4 -

  Fig. S4. Cell death rates are unchanged in the lateral pouch of Lrig3 homozygotes. (A,B) Transverse sections through the lateral pouch of E12 Lrig3^+/− (A) and Lrig3^−/− (B) embryos. Immunofluorescent detection of activated cleaved caspase 3 (green; Cell Signaling #9661) in DAPI-positive nuclei (blue) reveals apoptotic cells (arrowheads) in the base of the pouch in embryos of both genotypes. D, dorsal; L, lateral. Scale bar: 50 µm. (C) To assess the amount of cell death in the lateral pouch of Lrig3^+/− and Lrig3^−/− E12 embryos, caspase 3-positive nuclei were counted on 20 consecutive 3 sections (5 µms each) through the lateral pouch. Consistent with qualitative observation, there is no significant difference in the number of caspase 3-positive cell nuclei (n=3 for each genotype). Data are represented as the mean±s.d. and were analyzed by Student's t-test with P<0.05.

• Supplemental Figure S5 -

  Fig. S5. Inner ear morphogenesis does not require Unc5hb or integrin α6. (A) In situ hybridization with a probe to Unc5hb on transverse sections through the otic vesicle at E12. Like Lrig3, Unc5hb is expressed in the non-fusing epithelium of the lateral pouch (arrowhead). Unc5hb is also expressed in the dorsal canal pouch epithelium (bracket). (B,B′) Lateral (B) and dorsal (B′) views of a paintfilled inner ear from an E13.5 Unc5hb homozygote. Although most Unc5hb homozygotes die by E10.5, lethality is delayed until E13.5 on a CD1 background. However, Unc5hb is not required for inner ear morphogenesis. (C) Immunostaining for integrin α6 (1:500, Abcam, ab19765) on transverse sections through the otic vesicle at E12. Integrin α6 protein is present throughout the otic epithelium, with prominent localization in the basement membrane. As for all components of the basement membrane in the inner ear, integrin α6 distribution is disrupted above the fusion plate (arrowhead). (D,D′) Lateral (D) and dorsal (D′) views of a paintfilled inner ear from an E14.5 integrin α6 homozygote (Mitchell et al., 2001). All three canals form normally in mutant animals. Complementary to our results, recent work by Matilainen et al. (Matilainen et al., 2007) demonstrates that Unc5ha and Unc5hd are not expressed in the inner ear, and that DCC, neogenin 1, Unc5hc and integrin α3 mutants do not show any inner ear malformations.

• Supplemental Figure S6 -

  Fig. S6. Analysis of Lrig3 and Ntn1 in 3D cultures does not provide evidence for a direct interaction. (A) We generated cell lines expressing gapEGFP, myc-tagged Ntn1, FLAG-tagged Lrig3 (FL-Lrig3) or both Ntn1-myc and FL-Lrig3. Western blot analysis reveals that similar levels of Ntn1-myc and FL-Lrig3 protein are detectable in cell lines expressing either protein alone or in combination. Non-clonal lines of MCF-10A cells were produced via retroviral infection as previously described (Debnath, 2003). Constructs for gapEGFP (Inoue et al., 2001) and chicken Ntn1-myc have been previously described (Serafini et al., 1994). The FL-Lrig3 fusion contains an in-frame insertion of the FLAG epitope after the signal sequence of mouse Lrig3; the epitope with flanking protein sequence is HGAPGMDYKDDDDKGQLLDD. Ntn1-myc expressing cells were infected with a FL-Lrig3 virus to generate a co-expressing cell line. Western blot analysis was performed on monolayer cultures using mouse anti-FLAG-epitope (M2, Sigma-Aldrich), rat anti-mouse netrin 1 monoclonal (MAB1109, R&D systems) or mouse anti-actin monoclonal antibody (mAbcam8226). (B) The distribution of Ntn1 and Lrig3 proteins was assessed in acini, three-dimensional structures assembled from MCF-10A cells. Images through the center of the acini reveal even distribution of gapEGFP on the cell surface. By contrast, the majority of Ntn1-myc is deposited in the basement membrane, with weak cell surface staining. FL-Lrig3 is present in the secretory pathway and on the cell surface. Co-expression of Ntn1-myc and FL-Lrig3 does not alter their respective localization. To generate the acini, MCF-10A cell lines were cultured on Matrigel (BD Biosciences) in eight-well chamber slides for 15 days as previously described (Debnath, 2003). Cultures were fixed with 4% paraformaldehyde in PBS, and permeabilized and immunostained in PBS+1% BSA+0.1%TritonX-100. Protein was localized using rabbit anti-GFP polyclonal (Molecular Probes), mouse anti-myc epitope monoclonal (sc-40, Santa Cruz Biotechnology) or mouse anti-FLAG-epitope monoclonal antibody and AlexaFluor secondary antibodies (Molecular Probes). Nuclei are labeled with DAPI (blue). Scale bar: 50 µm. (C) Detection of Ntn1-myc secreted from acinar cultures in the presence and absence of FL-Lrig3. Co-expression of Ntn1-myc and FLLrig3 does not affect the amount of Ntn1 secreted. Acinar culture medium was collected from days 4-8 and 8-12 and concentrated approximately fourfold using a 10MWCO filter (UFC901024, Millipore). An equivalent amount of starting material per cell lines was separated on a 7.5% acrylamide gel and blotted with a rat anti-mouse Ntn1 monoclonal (R&D systems).