II-βV spectrin bridges the plasma membrane and cortical lattice in the lateral wall of the auditory outer hair cells

JCS028134 Supplementary Material

Files in this Data Supplement:

• Supplemental Table S1 - Adobe PDF

• Supplemental Figure S1 -

  Fig. S1. (A) Diagrams illustrating the positions and organization of vestibular and auditory sensory epithelia, and the lateral wall of outer hair cells (OHCs). (Left panels) The inner ear is made up of two sense organs: the vestibule, which senses equilibrium, and the cochlea, the auditory organ. (Middle panel) The auditory sensory epithelium, the organ of Corti, is made up of a highly organized mosaic of sensory inner (IHC) and outer (OHC) hair cells, flanked by various types of supporting cells. The IHCs are bulbar in shape and are arranged into a single row. They are the genuine sensory cells. Their synapses transmit information on sound temporal structure and intensity to the central nervous system. OHCs are cylindrical and arranged into three rows along the outer edge of the organ of Corti. They act as amplifiers of the sound stimuli. The right panel illustrates the organization of the OHC trilaminate lateral wall (100-nm thick). It is made of (1) the plasma membrane, containing the electromotility protein prestin, (2) the actin- and spectrin-based cortical lattice, and (3) the subsurface cisternae. Pillars of unknown composition connect the cortical lattice to the overlying plasma membrane lipid bilayer.

• Supplemental Figure S2 -

  Fig. S2. Distribution of the spectrin subunits in the auditory organ. Sections of auditory sensory epithelia from mice at embryonic day 18 (E18, A) and postnatal day 8 (P8, B). Besides αII, only two of the five β-spectrin subunits, βII and βV, are detected in the auditory hair cells (labelled using an anti-myosin-VIIa antibody). The αII- and βII-spectrin subunits are present both in sensory hair cells and supporting cells, whereas βV spectrin is present only in the hair cells. No βI-spectrin labelling is observed in the hair cells; a non-specific labelling is observed over the acellular basilar membrane (arrow in B, right panel). The βIII-spectrin subunit displays a cytoplasmic labelling in neuronal cell bodies. The βIV-spectrin subunit is mainly detected at nodes of Ranvier along the auditory nerve fibers. Bars: 5 µm (A); 10 µm (B).

• Supplemental Figure S3 -

  Fig. S3. Immunofluorescence and western blot characterization of anti-βV-spectrin antibodies. (A) The upper panel is a graphic representation of the βV-spectrin polypeptide, indicating the positions of the two subfragments (Pep and Rec) that were used to produce anti-βV-spectrin antibodies (see Materials and Methods). (B) In transfected HeLa cells, both the anti-hβVpep and anti-hβVrec antibodies recognize the overexpressed myc-tagged βVR29 fragment. However, only the anti-hβVrec antibody labels mouse native βV spectrin, and was thus used for all further experiments. Bars: 10 µm. (C) Human embryonic kidney HEK293 cells were transfected using Lipofectamin with Plus Reagent (Invitrogen), according to the manufacturer’s instructions. In HEK293 cell lysates containing myc-tagged βVCH, myc-tagged βVPH or myc-tagged βVR29, the anti-hβVrec antibody recognizes only the βV-spectrin C-terminal fragments, βVR29 and βVPH. The anti-myc antibody, which labels all three fragments, was used as a control. (D) In protein extracts from the mouse brain, inner ear or eye, the anti-βV-spectrin antibody recognizes two clear major bands at ∼100 kDa and ∼280-300 kDa. Sequence analysis reveals no evidence for the existence of βV-spectrin splice variants. These two major bands, not detected by the preimmune serum, are probably proteolytic degradation products of the βV-spectrin endogeneous protein (∼417 kDa). (E) The proteolytic degradation may have occurred in vivo or during the initial steps of cell extraction, since the same pattern was obtained with freshly prepared brain and Caco2 cells lysates, or those kept at room temperature for an additional 3 or 24 hours before SDS-PAGE analysis. (F) Because the upper βV-spectrin band migrates at the zone of the conventional spectrins, we tested possible cross-reactivity of the anti-hβVrec antibody with conventional β-spectrin subunits. A previously described monoclonal antibody that recognizes βI spectrin (NCLSPEC2; Novocastra) and that labels an immunoreactive band at ∼250-280 kDa was used. Caco2 cells lysates were incubated with the anti-βV-spectrin (βV), anti-βI-spectrin (βI), or a combination of the two antibodies (βV + βI). The bands detected by each of the subunit antibodies are clearly different.

• Supplemental Figure S4 -

  Fig. S4. βV-spectrin-subunit distribution in inner-ear endothelial cells. (A) The left and middle panels show schematic representations illustrating the positions of the inner-ear regions used for RT-PCR experiments; e.g. the five vestibular sensory epithelia (VE), the organ of Corti (OC) and the stria vascularis (SV). (B) RT-PCR analysis of different inner-ear regions. βV-spectrin transcripts are detected in the organ of Corti, vestibular sensory epithelia and stria vascularis. +/− RT indicates presence (+ RT) or absence (− RT) of reverse transcriptase in the sample. (C) Section of the stria vascularis from P8 mice, showing βV-spectrin distribution in the vessels within and nearby the stria vascularis epithelium. (D,E) Section (D) and whole-mount (E) preparations of mouse auditory epithelia. As early as E14, a strong βV-spectrin staining is detected in endothelial cells of the developing inner ear. The βV-spectrin labelling in endothelial cells persists at postnatal stages, as shown by whole-mount experiments of organs of Corti from P8 mice, in the region beneath the cochlear spiral limbus (E). cd, cochlear duct. DAPI labels all cell nuclei (blue). Bars: 100 µm.

• Supplemental Figure S5 -

  Fig. S5. Distribution of βV spectrin in the auditory inner and outer hair cells. (A,B) Section (A) and whole-mount (B) preparations of organs of Corti from P8 mice. βV spectrin (green) localizes essentially to the OHC lateral wall. No βV spectrin is detected in supporting cells, or in the apical and basal synaptic zone of hair cells. βIII tubulin (red in A) labels all hair cell nerve terminals. (B) Confocal sections taken at different levels of the organ of Corti, illustrating the differences in βV-spectrin fluorescence staining intensity between IHCs and OHCs. This could also be illustrated by the pseudocolored confocal image of the organ of Corti (lower middle panel). Fluorescence intensity is coded on a linear pseudocolor scale. Bars: 10 µm.

• Supplemental Figure S6 -

  Fig. S6. βV spectrin labels the entire lateral wall of OHCs, irrespective of their position in cochlea. From this stage onwards, the typical ‘ring’ pattern of βV-spectrin staining spreads along the entire lateral wall. In IHCs, the ‘ring’ pattern is, however, observed only in the apical neck region. (B) In OHCs, the αII- and βII-spectrin subunits co-distribute in the cuticular plate, whereas the αII- and β- spectrin subunits are detected along the lateral wall. Bars: 5 µm.

• Supplemental Figure S7 -

  Fig. S7. Ultrastructural distribution of βV spectrin in OHCs. Gold particles are not observed in the hair bundle (HB), the cuticular plate (CP) or at the apical junctional complex (AJC). A quantitative analysis of βV-spectrin immunogold labelling was carried out on 17 different OHCs, subdivided into three regions (1, 2 and 3) as indicated in the schematic diagram (lower left panel). M, mitochondria; N, nucleus. Bars: 100 nm.

• Supplemental Figure S8 -

  Fig. S8. Schematic diagrams showing the domains of molecular interaction between the βV- and αII-spectrin subunits, band 4.1, F-actin, and the indirect link with the membrane motor protein prestin.